MXS-2732 Add sqlite3 version 3110100
This commit is contained in:
Johan Wikman
@ -0,0 +1,178 @@
|
||||
|
||||
FTS4 CONTENT OPTION
|
||||
|
||||
Normally, in order to create a full-text index on a dataset, the FTS4
|
||||
module stores a copy of all indexed documents in a specially created
|
||||
database table.
|
||||
|
||||
As of SQLite version 3.7.9, FTS4 supports a new option - "content" -
|
||||
designed to extend FTS4 to support the creation of full-text indexes where:
|
||||
|
||||
* The indexed documents are not stored within the SQLite database
|
||||
at all (a "contentless" FTS4 table), or
|
||||
|
||||
* The indexed documents are stored in a database table created and
|
||||
managed by the user (an "external content" FTS4 table).
|
||||
|
||||
Because the indexed documents themselves are usually much larger than
|
||||
the full-text index, the content option can sometimes be used to achieve
|
||||
significant space savings.
|
||||
|
||||
CONTENTLESS FTS4 TABLES
|
||||
|
||||
In order to create an FTS4 table that does not store a copy of the indexed
|
||||
documents at all, the content option should be set to an empty string.
|
||||
For example, the following SQL creates such an FTS4 table with three
|
||||
columns - "a", "b", and "c":
|
||||
|
||||
CREATE VIRTUAL TABLE t1 USING fts4(content="", a, b, c);
|
||||
|
||||
Data can be inserted into such an FTS4 table using an INSERT statements.
|
||||
However, unlike ordinary FTS4 tables, the user must supply an explicit
|
||||
integer docid value. For example:
|
||||
|
||||
-- This statement is Ok:
|
||||
INSERT INTO t1(docid, a, b, c) VALUES(1, 'a b c', 'd e f', 'g h i');
|
||||
|
||||
-- This statement causes an error, as no docid value has been provided:
|
||||
INSERT INTO t1(a, b, c) VALUES('j k l', 'm n o', 'p q r');
|
||||
|
||||
It is not possible to UPDATE or DELETE a row stored in a contentless FTS4
|
||||
table. Attempting to do so is an error.
|
||||
|
||||
Contentless FTS4 tables also support SELECT statements. However, it is
|
||||
an error to attempt to retrieve the value of any table column other than
|
||||
the docid column. The auxiliary function matchinfo() may be used, but
|
||||
snippet() and offsets() may not. For example:
|
||||
|
||||
-- The following statements are Ok:
|
||||
SELECT docid FROM t1 WHERE t1 MATCH 'xxx';
|
||||
SELECT docid FROM t1 WHERE a MATCH 'xxx';
|
||||
SELECT matchinfo(t1) FROM t1 WHERE t1 MATCH 'xxx';
|
||||
|
||||
-- The following statements all cause errors, as the value of columns
|
||||
-- other than docid are required to evaluate them.
|
||||
SELECT * FROM t1;
|
||||
SELECT a, b FROM t1 WHERE t1 MATCH 'xxx';
|
||||
SELECT docid FROM t1 WHERE a LIKE 'xxx%';
|
||||
SELECT snippet(t1) FROM t1 WHERE t1 MATCH 'xxx';
|
||||
|
||||
Errors related to attempting to retrieve column values other than docid
|
||||
are runtime errors that occur within sqlite3_step(). In some cases, for
|
||||
example if the MATCH expression in a SELECT query matches zero rows, there
|
||||
may be no error at all even if a statement does refer to column values
|
||||
other than docid.
|
||||
|
||||
EXTERNAL CONTENT FTS4 TABLES
|
||||
|
||||
An "external content" FTS4 table is similar to a contentless table, except
|
||||
that if evaluation of a query requires the value of a column other than
|
||||
docid, FTS4 attempts to retrieve that value from a table (or view, or
|
||||
virtual table) nominated by the user (hereafter referred to as the "content
|
||||
table"). The FTS4 module never writes to the content table, and writing
|
||||
to the content table does not affect the full-text index. It is the
|
||||
responsibility of the user to ensure that the content table and the
|
||||
full-text index are consistent.
|
||||
|
||||
An external content FTS4 table is created by setting the content option
|
||||
to the name of a table (or view, or virtual table) that may be queried by
|
||||
FTS4 to retrieve column values when required. If the nominated table does
|
||||
not exist, then an external content table behaves in the same way as
|
||||
a contentless table. For example:
|
||||
|
||||
CREATE TABLE t2(id INTEGER PRIMARY KEY, a, b, c);
|
||||
CREATE VIRTUAL TABLE t3 USING fts4(content="t2", a, c);
|
||||
|
||||
Assuming the nominated table does exist, then its columns must be the same
|
||||
as or a superset of those defined for the FTS table.
|
||||
|
||||
When a users query on the FTS table requires a column value other than
|
||||
docid, FTS attempts to read this value from the corresponding column of
|
||||
the row in the content table with a rowid value equal to the current FTS
|
||||
docid. Or, if such a row cannot be found in the content table, a NULL
|
||||
value is used instead. For example:
|
||||
|
||||
CREATE TABLE t2(id INTEGER PRIMARY KEY, a, b, c, d);
|
||||
CREATE VIRTUAL TABLE t3 USING fts4(content="t2", b, c);
|
||||
|
||||
INSERT INTO t2 VALUES(2, 'a b', 'c d', 'e f');
|
||||
INSERT INTO t2 VALUES(3, 'g h', 'i j', 'k l');
|
||||
INSERT INTO t3(docid, b, c) SELECT id, b, c FROM t2;
|
||||
|
||||
-- The following query returns a single row with two columns containing
|
||||
-- the text values "i j" and "k l".
|
||||
--
|
||||
-- The query uses the full-text index to discover that the MATCH
|
||||
-- term matches the row with docid=3. It then retrieves the values
|
||||
-- of columns b and c from the row with rowid=3 in the content table
|
||||
-- to return.
|
||||
--
|
||||
SELECT * FROM t3 WHERE t3 MATCH 'k';
|
||||
|
||||
-- Following the UPDATE, the query still returns a single row, this
|
||||
-- time containing the text values "xxx" and "yyy". This is because the
|
||||
-- full-text index still indicates that the row with docid=3 matches
|
||||
-- the FTS4 query 'k', even though the documents stored in the content
|
||||
-- table have been modified.
|
||||
--
|
||||
UPDATE t2 SET b = 'xxx', c = 'yyy' WHERE rowid = 3;
|
||||
SELECT * FROM t3 WHERE t3 MATCH 'k';
|
||||
|
||||
-- Following the DELETE below, the query returns one row containing two
|
||||
-- NULL values. NULL values are returned because FTS is unable to find
|
||||
-- a row with rowid=3 within the content table.
|
||||
--
|
||||
DELETE FROM t2;
|
||||
SELECT * FROM t3 WHERE t3 MATCH 'k';
|
||||
|
||||
When a row is deleted from an external content FTS4 table, FTS4 needs to
|
||||
retrieve the column values of the row being deleted from the content table.
|
||||
This is so that FTS4 can update the full-text index entries for each token
|
||||
that occurs within the deleted row to indicate that that row has been
|
||||
deleted. If the content table row cannot be found, or if it contains values
|
||||
inconsistent with the contents of the FTS index, the results can be difficult
|
||||
to predict. The FTS index may be left containing entries corresponding to the
|
||||
deleted row, which can lead to seemingly nonsensical results being returned
|
||||
by subsequent SELECT queries. The same applies when a row is updated, as
|
||||
internally an UPDATE is the same as a DELETE followed by an INSERT.
|
||||
|
||||
Instead of writing separately to the full-text index and the content table,
|
||||
some users may wish to use database triggers to keep the full-text index
|
||||
up to date with respect to the set of documents stored in the content table.
|
||||
For example, using the tables from earlier examples:
|
||||
|
||||
CREATE TRIGGER t2_bu BEFORE UPDATE ON t2 BEGIN
|
||||
DELETE FROM t3 WHERE docid=old.rowid;
|
||||
END;
|
||||
CREATE TRIGGER t2_bd BEFORE DELETE ON t2 BEGIN
|
||||
DELETE FROM t3 WHERE docid=old.rowid;
|
||||
END;
|
||||
|
||||
CREATE TRIGGER t2_bu AFTER UPDATE ON t2 BEGIN
|
||||
INSERT INTO t3(docid, b, c) VALUES(new.rowid, new.b, new.c);
|
||||
END;
|
||||
CREATE TRIGGER t2_bd AFTER INSERT ON t2 BEGIN
|
||||
INSERT INTO t3(docid, b, c) VALUES(new.rowid, new.b, new.c);
|
||||
END;
|
||||
|
||||
The DELETE trigger must be fired before the actual delete takes place
|
||||
on the content table. This is so that FTS4 can still retrieve the original
|
||||
values in order to update the full-text index. And the INSERT trigger must
|
||||
be fired after the new row is inserted, so as to handle the case where the
|
||||
rowid is assigned automatically within the system. The UPDATE trigger must
|
||||
be split into two parts, one fired before and one after the update of the
|
||||
content table, for the same reasons.
|
||||
|
||||
FTS4 features a special command similar to the 'optimize' command that
|
||||
deletes the entire full-text index and rebuilds it based on the current
|
||||
set of documents in the content table. Assuming again that "t3" is the
|
||||
name of the external content FTS4 table, the command is:
|
||||
|
||||
INSERT INTO t3(t3) VALUES('rebuild');
|
||||
|
||||
This command may also be used with ordinary FTS4 tables, although it may
|
||||
only be useful if the full-text index has somehow become corrupt. It is an
|
||||
error to attempt to rebuild the full-text index maintained by a contentless
|
||||
FTS4 table.
|
||||
|
||||
|
||||
@ -0,0 +1,209 @@
|
||||
|
||||
1. OVERVIEW
|
||||
|
||||
This README file describes the syntax of the arguments that may be passed to
|
||||
the FTS3 MATCH operator used for full-text queries. For example, if table
|
||||
"t1" is an Fts3 virtual table, the following SQL query:
|
||||
|
||||
SELECT * FROM t1 WHERE <col> MATCH <full-text query>
|
||||
|
||||
may be used to retrieve all rows that match a specified for full-text query.
|
||||
The text "<col>" should be replaced by either the name of the fts3 table
|
||||
(in this case "t1"), or by the name of one of the columns of the fts3
|
||||
table. <full-text-query> should be replaced by an SQL expression that
|
||||
computes to a string containing an Fts3 query.
|
||||
|
||||
If the left-hand-side of the MATCH operator is set to the name of the
|
||||
fts3 table, then by default the query may be matched against any column
|
||||
of the table. If it is set to a column name, then by default the query
|
||||
may only match the specified column. In both cases this may be overriden
|
||||
as part of the query text (see sections 2 and 3 below).
|
||||
|
||||
As of SQLite version 3.6.8, Fts3 supports two slightly different query
|
||||
formats; the standard syntax, which is used by default, and the enhanced
|
||||
query syntax which can be selected by compiling with the pre-processor
|
||||
symbol SQLITE_ENABLE_FTS3_PARENTHESIS defined.
|
||||
|
||||
-DSQLITE_ENABLE_FTS3_PARENTHESIS
|
||||
|
||||
2. STANDARD QUERY SYNTAX
|
||||
|
||||
When using the standard Fts3 query syntax, a query usually consists of a
|
||||
list of terms (words) separated by white-space characters. To match a
|
||||
query, a row (or column) of an Fts3 table must contain each of the specified
|
||||
terms. For example, the following query:
|
||||
|
||||
<col> MATCH 'hello world'
|
||||
|
||||
matches rows (or columns, if <col> is the name of a column name) that
|
||||
contain at least one instance of the token "hello", and at least one
|
||||
instance of the token "world". Tokens may be grouped into phrases using
|
||||
quotation marks. In this case, a matching row or column must contain each
|
||||
of the tokens in the phrase in the order specified, with no intervening
|
||||
tokens. For example, the query:
|
||||
|
||||
<col> MATCH '"hello world" joe"
|
||||
|
||||
matches the first of the following two documents, but not the second or
|
||||
third:
|
||||
|
||||
"'Hello world', said Joe."
|
||||
"One should always greet the world with a cheery hello, thought Joe."
|
||||
"How many hello world programs could their be?"
|
||||
|
||||
As well as grouping tokens together by phrase, the binary NEAR operator
|
||||
may be used to search for rows that contain two or more specified tokens
|
||||
or phrases within a specified proximity of each other. The NEAR operator
|
||||
must always be specified in upper case. The word "near" in lower or mixed
|
||||
case is treated as an ordinary token. For example, the following query:
|
||||
|
||||
<col> MATCH 'engineering NEAR consultancy'
|
||||
|
||||
matches rows that contain both the "engineering" and "consultancy" tokens
|
||||
in the same column with not more than 10 other words between them. It does
|
||||
not matter which of the two terms occurs first in the document, only that
|
||||
they be seperated by only 10 tokens or less. The user may also specify
|
||||
a different required proximity by adding "/N" immediately after the NEAR
|
||||
operator, where N is an integer. For example:
|
||||
|
||||
<col> MATCH 'engineering NEAR/5 consultancy'
|
||||
|
||||
searches for a row containing an instance of each specified token seperated
|
||||
by not more than 5 other tokens. More than one NEAR operator can be used
|
||||
in as sequence. For example this query:
|
||||
|
||||
<col> MATCH 'reliable NEAR/2 engineering NEAR/5 consultancy'
|
||||
|
||||
searches for a row that contains an instance of the token "reliable"
|
||||
seperated by not more than two tokens from an instance of "engineering",
|
||||
which is in turn separated by not more than 5 other tokens from an
|
||||
instance of the term "consultancy". Phrases enclosed in quotes may
|
||||
also be used as arguments to the NEAR operator.
|
||||
|
||||
Similar to the NEAR operator, one or more tokens or phrases may be
|
||||
separated by OR operators. In this case, only one of the specified tokens
|
||||
or phrases must appear in the document. For example, the query:
|
||||
|
||||
<col> MATCH 'hello OR world'
|
||||
|
||||
matches rows that contain either the term "hello", or the term "world",
|
||||
or both. Note that unlike in many programming languages, the OR operator
|
||||
has a higher precedence than the AND operators implied between white-space
|
||||
separated tokens. The following query matches documents that contain the
|
||||
term 'sqlite' and at least one of the terms 'fantastic' or 'impressive',
|
||||
not those that contain both 'sqlite' and 'fantastic' or 'impressive':
|
||||
|
||||
<col> MATCH 'sqlite fantastic OR impressive'
|
||||
|
||||
Any token that is part of an Fts3 query expression, whether or not it is
|
||||
part of a phrase enclosed in quotes, may have a '*' character appended to
|
||||
it. In this case, the token matches all terms that begin with the characters
|
||||
of the token, not just those that exactly match it. For example, the
|
||||
following query:
|
||||
|
||||
<col> MATCH 'sql*'
|
||||
|
||||
matches all rows that contain the term "SQLite", as well as those that
|
||||
contain "SQL".
|
||||
|
||||
A token that is not part of a quoted phrase may be preceded by a '-'
|
||||
character, which indicates that matching rows must not contain the
|
||||
specified term. For example, the following:
|
||||
|
||||
<col> MATCH '"database engine" -sqlite'
|
||||
|
||||
matches rows that contain the phrase "database engine" but do not contain
|
||||
the term "sqlite". If the '-' character occurs inside a quoted phrase,
|
||||
it is ignored. It is possible to use both the '-' prefix and the '*' postfix
|
||||
on a single term. At this time, all Fts3 queries must contain at least
|
||||
one term or phrase that is not preceded by the '-' prefix.
|
||||
|
||||
Regardless of whether or not a table name or column name is used on the
|
||||
left hand side of the MATCH operator, a specific column of the fts3 table
|
||||
may be associated with each token in a query by preceding a token with
|
||||
a column name followed by a ':' character. For example, regardless of what
|
||||
is specified for <col>, the following query requires that column "col1"
|
||||
of the table contains the term "hello", and that column "col2" of the
|
||||
table contains the term "world". If the table does not contain columns
|
||||
named "col1" and "col2", then an error is returned and the query is
|
||||
not run.
|
||||
|
||||
<col> MATCH 'col1:hello col2:world'
|
||||
|
||||
It is not possible to associate a specific table column with a quoted
|
||||
phrase or a term preceded by a '-' operator. A '*' character may be
|
||||
appended to a term associated with a specific column for prefix matching.
|
||||
|
||||
3. ENHANCED QUERY SYNTAX
|
||||
|
||||
The enhanced query syntax is quite similar to the standard query syntax,
|
||||
with the following four differences:
|
||||
|
||||
1) Parenthesis are supported. When using the enhanced query syntax,
|
||||
parenthesis may be used to overcome the built-in precedence of the
|
||||
supplied binary operators. For example, the following query:
|
||||
|
||||
<col> MATCH '(hello world) OR (simple example)'
|
||||
|
||||
matches documents that contain both "hello" and "world", and documents
|
||||
that contain both "simple" and "example". It is not possible to forumlate
|
||||
such a query using the standard syntax.
|
||||
|
||||
2) Instead of separating tokens and phrases by whitespace, an AND operator
|
||||
may be explicitly specified. This does not change query processing at
|
||||
all, but may be used to improve readability. For example, the following
|
||||
query is handled identically to the one above:
|
||||
|
||||
<col> MATCH '(hello AND world) OR (simple AND example)'
|
||||
|
||||
As with the OR and NEAR operators, the AND operator must be specified
|
||||
in upper case. The word "and" specified in lower or mixed case is
|
||||
handled as a regular token.
|
||||
|
||||
3) The '-' token prefix is not supported. Instead, a new binary operator,
|
||||
NOT, is included. The NOT operator requires that the query specified
|
||||
as its left-hand operator matches, but that the query specified as the
|
||||
right-hand operator does not. For example, to query for all rows that
|
||||
contain the term "example" but not the term "simple", the following
|
||||
query could be used:
|
||||
|
||||
<col> MATCH 'example NOT simple'
|
||||
|
||||
As for all other operators, the NOT operator must be specified in
|
||||
upper case. Otherwise it will be treated as a regular token.
|
||||
|
||||
4) Unlike in the standard syntax, where the OR operator has a higher
|
||||
precedence than the implicit AND operator, when using the enhanced
|
||||
syntax implicit and explict AND operators have a higher precedence
|
||||
than OR operators. Using the enhanced syntax, the following two
|
||||
queries are equivalent:
|
||||
|
||||
<col> MATCH 'sqlite fantastic OR impressive'
|
||||
<col> MATCH '(sqlite AND fantastic) OR impressive'
|
||||
|
||||
however, when using the standard syntax, the query:
|
||||
|
||||
<col> MATCH 'sqlite fantastic OR impressive'
|
||||
|
||||
is equivalent to the enhanced syntax query:
|
||||
|
||||
<col> MATCH 'sqlite AND (fantastic OR impressive)'
|
||||
|
||||
The precedence of all enhanced syntax operators, in order from highest
|
||||
to lowest, is:
|
||||
|
||||
NEAR (highest precedence, tightest grouping)
|
||||
NOT
|
||||
AND
|
||||
OR (lowest precedence, loosest grouping)
|
||||
|
||||
Using the advanced syntax, it is possible to specify expressions enclosed
|
||||
in parenthesis as operands to the NOT, AND and OR operators. However both
|
||||
the left and right hand side operands of NEAR operators must be either
|
||||
tokens or phrases. Attempting the following query will return an error:
|
||||
|
||||
<col> MATCH 'sqlite NEAR (fantastic OR impressive)'
|
||||
|
||||
Queries of this form must be re-written as:
|
||||
|
||||
<col> MATCH 'sqlite NEAR fantastic OR sqlite NEAR impressive'
|
||||
@ -0,0 +1,133 @@
|
||||
|
||||
1. FTS3 Tokenizers
|
||||
|
||||
When creating a new full-text table, FTS3 allows the user to select
|
||||
the text tokenizer implementation to be used when indexing text
|
||||
by specifying a "tokenize" clause as part of the CREATE VIRTUAL TABLE
|
||||
statement:
|
||||
|
||||
CREATE VIRTUAL TABLE <table-name> USING fts3(
|
||||
<columns ...> [, tokenize <tokenizer-name> [<tokenizer-args>]]
|
||||
);
|
||||
|
||||
The built-in tokenizers (valid values to pass as <tokenizer name>) are
|
||||
"simple", "porter" and "unicode".
|
||||
|
||||
<tokenizer-args> should consist of zero or more white-space separated
|
||||
arguments to pass to the selected tokenizer implementation. The
|
||||
interpretation of the arguments, if any, depends on the individual
|
||||
tokenizer.
|
||||
|
||||
2. Custom Tokenizers
|
||||
|
||||
FTS3 allows users to provide custom tokenizer implementations. The
|
||||
interface used to create a new tokenizer is defined and described in
|
||||
the fts3_tokenizer.h source file.
|
||||
|
||||
Registering a new FTS3 tokenizer is similar to registering a new
|
||||
virtual table module with SQLite. The user passes a pointer to a
|
||||
structure containing pointers to various callback functions that
|
||||
make up the implementation of the new tokenizer type. For tokenizers,
|
||||
the structure (defined in fts3_tokenizer.h) is called
|
||||
"sqlite3_tokenizer_module".
|
||||
|
||||
FTS3 does not expose a C-function that users call to register new
|
||||
tokenizer types with a database handle. Instead, the pointer must
|
||||
be encoded as an SQL blob value and passed to FTS3 through the SQL
|
||||
engine by evaluating a special scalar function, "fts3_tokenizer()".
|
||||
The fts3_tokenizer() function may be called with one or two arguments,
|
||||
as follows:
|
||||
|
||||
SELECT fts3_tokenizer(<tokenizer-name>);
|
||||
SELECT fts3_tokenizer(<tokenizer-name>, <sqlite3_tokenizer_module ptr>);
|
||||
|
||||
Where <tokenizer-name> is a string identifying the tokenizer and
|
||||
<sqlite3_tokenizer_module ptr> is a pointer to an sqlite3_tokenizer_module
|
||||
structure encoded as an SQL blob. If the second argument is present,
|
||||
it is registered as tokenizer <tokenizer-name> and a copy of it
|
||||
returned. If only one argument is passed, a pointer to the tokenizer
|
||||
implementation currently registered as <tokenizer-name> is returned,
|
||||
encoded as a blob. Or, if no such tokenizer exists, an SQL exception
|
||||
(error) is raised.
|
||||
|
||||
SECURITY: If the fts3 extension is used in an environment where potentially
|
||||
malicious users may execute arbitrary SQL (i.e. gears), they should be
|
||||
prevented from invoking the fts3_tokenizer() function, possibly using the
|
||||
authorisation callback.
|
||||
|
||||
See "Sample code" below for an example of calling the fts3_tokenizer()
|
||||
function from C code.
|
||||
|
||||
3. ICU Library Tokenizers
|
||||
|
||||
If this extension is compiled with the SQLITE_ENABLE_ICU pre-processor
|
||||
symbol defined, then there exists a built-in tokenizer named "icu"
|
||||
implemented using the ICU library. The first argument passed to the
|
||||
xCreate() method (see fts3_tokenizer.h) of this tokenizer may be
|
||||
an ICU locale identifier. For example "tr_TR" for Turkish as used
|
||||
in Turkey, or "en_AU" for English as used in Australia. For example:
|
||||
|
||||
"CREATE VIRTUAL TABLE thai_text USING fts3(text, tokenizer icu th_TH)"
|
||||
|
||||
The ICU tokenizer implementation is very simple. It splits the input
|
||||
text according to the ICU rules for finding word boundaries and discards
|
||||
any tokens that consist entirely of white-space. This may be suitable
|
||||
for some applications in some locales, but not all. If more complex
|
||||
processing is required, for example to implement stemming or
|
||||
discard punctuation, this can be done by creating a tokenizer
|
||||
implementation that uses the ICU tokenizer as part of its implementation.
|
||||
|
||||
When using the ICU tokenizer this way, it is safe to overwrite the
|
||||
contents of the strings returned by the xNext() method (see
|
||||
fts3_tokenizer.h).
|
||||
|
||||
4. Sample code.
|
||||
|
||||
The following two code samples illustrate the way C code should invoke
|
||||
the fts3_tokenizer() scalar function:
|
||||
|
||||
int registerTokenizer(
|
||||
sqlite3 *db,
|
||||
char *zName,
|
||||
const sqlite3_tokenizer_module *p
|
||||
){
|
||||
int rc;
|
||||
sqlite3_stmt *pStmt;
|
||||
const char zSql[] = "SELECT fts3_tokenizer(?, ?)";
|
||||
|
||||
rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, 0);
|
||||
if( rc!=SQLITE_OK ){
|
||||
return rc;
|
||||
}
|
||||
|
||||
sqlite3_bind_text(pStmt, 1, zName, -1, SQLITE_STATIC);
|
||||
sqlite3_bind_blob(pStmt, 2, &p, sizeof(p), SQLITE_STATIC);
|
||||
sqlite3_step(pStmt);
|
||||
|
||||
return sqlite3_finalize(pStmt);
|
||||
}
|
||||
|
||||
int queryTokenizer(
|
||||
sqlite3 *db,
|
||||
char *zName,
|
||||
const sqlite3_tokenizer_module **pp
|
||||
){
|
||||
int rc;
|
||||
sqlite3_stmt *pStmt;
|
||||
const char zSql[] = "SELECT fts3_tokenizer(?)";
|
||||
|
||||
*pp = 0;
|
||||
rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, 0);
|
||||
if( rc!=SQLITE_OK ){
|
||||
return rc;
|
||||
}
|
||||
|
||||
sqlite3_bind_text(pStmt, 1, zName, -1, SQLITE_STATIC);
|
||||
if( SQLITE_ROW==sqlite3_step(pStmt) ){
|
||||
if( sqlite3_column_type(pStmt, 0)==SQLITE_BLOB ){
|
||||
memcpy(pp, sqlite3_column_blob(pStmt, 0), sizeof(*pp));
|
||||
}
|
||||
}
|
||||
|
||||
return sqlite3_finalize(pStmt);
|
||||
}
|
||||
@ -0,0 +1,4 @@
|
||||
This folder contains source code to the second full-text search
|
||||
extension for SQLite. While the API is the same, this version uses a
|
||||
substantially different storage schema from fts1, so tables will need
|
||||
to be rebuilt.
|
||||
5924
query_classifier/qc_sqlite/sqlite-src-3110100/ext/fts3/fts3.c
Normal file
5924
query_classifier/qc_sqlite/sqlite-src-3110100/ext/fts3/fts3.c
Normal file
File diff suppressed because it is too large
Load Diff
@ -0,0 +1,26 @@
|
||||
/*
|
||||
** 2006 Oct 10
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
** This header file is used by programs that want to link against the
|
||||
** FTS3 library. All it does is declare the sqlite3Fts3Init() interface.
|
||||
*/
|
||||
#include "sqlite3.h"
|
||||
|
||||
#ifdef __cplusplus
|
||||
extern "C" {
|
||||
#endif /* __cplusplus */
|
||||
|
||||
int sqlite3Fts3Init(sqlite3 *db);
|
||||
|
||||
#ifdef __cplusplus
|
||||
} /* extern "C" */
|
||||
#endif /* __cplusplus */
|
||||
617
query_classifier/qc_sqlite/sqlite-src-3110100/ext/fts3/fts3Int.h
Normal file
617
query_classifier/qc_sqlite/sqlite-src-3110100/ext/fts3/fts3Int.h
Normal file
@ -0,0 +1,617 @@
|
||||
/*
|
||||
** 2009 Nov 12
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
*/
|
||||
#ifndef _FTSINT_H
|
||||
#define _FTSINT_H
|
||||
|
||||
#if !defined(NDEBUG) && !defined(SQLITE_DEBUG)
|
||||
# define NDEBUG 1
|
||||
#endif
|
||||
|
||||
/* FTS3/FTS4 require virtual tables */
|
||||
#ifdef SQLITE_OMIT_VIRTUALTABLE
|
||||
# undef SQLITE_ENABLE_FTS3
|
||||
# undef SQLITE_ENABLE_FTS4
|
||||
#endif
|
||||
|
||||
/*
|
||||
** FTS4 is really an extension for FTS3. It is enabled using the
|
||||
** SQLITE_ENABLE_FTS3 macro. But to avoid confusion we also all
|
||||
** the SQLITE_ENABLE_FTS4 macro to serve as an alisse for SQLITE_ENABLE_FTS3.
|
||||
*/
|
||||
#if defined(SQLITE_ENABLE_FTS4) && !defined(SQLITE_ENABLE_FTS3)
|
||||
# define SQLITE_ENABLE_FTS3
|
||||
#endif
|
||||
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
|
||||
/* If not building as part of the core, include sqlite3ext.h. */
|
||||
#ifndef SQLITE_CORE
|
||||
# include "sqlite3ext.h"
|
||||
SQLITE_EXTENSION_INIT3
|
||||
#endif
|
||||
|
||||
#include "sqlite3.h"
|
||||
#include "fts3_tokenizer.h"
|
||||
#include "fts3_hash.h"
|
||||
|
||||
/*
|
||||
** This constant determines the maximum depth of an FTS expression tree
|
||||
** that the library will create and use. FTS uses recursion to perform
|
||||
** various operations on the query tree, so the disadvantage of a large
|
||||
** limit is that it may allow very large queries to use large amounts
|
||||
** of stack space (perhaps causing a stack overflow).
|
||||
*/
|
||||
#ifndef SQLITE_FTS3_MAX_EXPR_DEPTH
|
||||
# define SQLITE_FTS3_MAX_EXPR_DEPTH 12
|
||||
#endif
|
||||
|
||||
|
||||
/*
|
||||
** This constant controls how often segments are merged. Once there are
|
||||
** FTS3_MERGE_COUNT segments of level N, they are merged into a single
|
||||
** segment of level N+1.
|
||||
*/
|
||||
#define FTS3_MERGE_COUNT 16
|
||||
|
||||
/*
|
||||
** This is the maximum amount of data (in bytes) to store in the
|
||||
** Fts3Table.pendingTerms hash table. Normally, the hash table is
|
||||
** populated as documents are inserted/updated/deleted in a transaction
|
||||
** and used to create a new segment when the transaction is committed.
|
||||
** However if this limit is reached midway through a transaction, a new
|
||||
** segment is created and the hash table cleared immediately.
|
||||
*/
|
||||
#define FTS3_MAX_PENDING_DATA (1*1024*1024)
|
||||
|
||||
/*
|
||||
** Macro to return the number of elements in an array. SQLite has a
|
||||
** similar macro called ArraySize(). Use a different name to avoid
|
||||
** a collision when building an amalgamation with built-in FTS3.
|
||||
*/
|
||||
#define SizeofArray(X) ((int)(sizeof(X)/sizeof(X[0])))
|
||||
|
||||
|
||||
#ifndef MIN
|
||||
# define MIN(x,y) ((x)<(y)?(x):(y))
|
||||
#endif
|
||||
#ifndef MAX
|
||||
# define MAX(x,y) ((x)>(y)?(x):(y))
|
||||
#endif
|
||||
|
||||
/*
|
||||
** Maximum length of a varint encoded integer. The varint format is different
|
||||
** from that used by SQLite, so the maximum length is 10, not 9.
|
||||
*/
|
||||
#define FTS3_VARINT_MAX 10
|
||||
|
||||
/*
|
||||
** FTS4 virtual tables may maintain multiple indexes - one index of all terms
|
||||
** in the document set and zero or more prefix indexes. All indexes are stored
|
||||
** as one or more b+-trees in the %_segments and %_segdir tables.
|
||||
**
|
||||
** It is possible to determine which index a b+-tree belongs to based on the
|
||||
** value stored in the "%_segdir.level" column. Given this value L, the index
|
||||
** that the b+-tree belongs to is (L<<10). In other words, all b+-trees with
|
||||
** level values between 0 and 1023 (inclusive) belong to index 0, all levels
|
||||
** between 1024 and 2047 to index 1, and so on.
|
||||
**
|
||||
** It is considered impossible for an index to use more than 1024 levels. In
|
||||
** theory though this may happen, but only after at least
|
||||
** (FTS3_MERGE_COUNT^1024) separate flushes of the pending-terms tables.
|
||||
*/
|
||||
#define FTS3_SEGDIR_MAXLEVEL 1024
|
||||
#define FTS3_SEGDIR_MAXLEVEL_STR "1024"
|
||||
|
||||
/*
|
||||
** The testcase() macro is only used by the amalgamation. If undefined,
|
||||
** make it a no-op.
|
||||
*/
|
||||
#ifndef testcase
|
||||
# define testcase(X)
|
||||
#endif
|
||||
|
||||
/*
|
||||
** Terminator values for position-lists and column-lists.
|
||||
*/
|
||||
#define POS_COLUMN (1) /* Column-list terminator */
|
||||
#define POS_END (0) /* Position-list terminator */
|
||||
|
||||
/*
|
||||
** This section provides definitions to allow the
|
||||
** FTS3 extension to be compiled outside of the
|
||||
** amalgamation.
|
||||
*/
|
||||
#ifndef SQLITE_AMALGAMATION
|
||||
/*
|
||||
** Macros indicating that conditional expressions are always true or
|
||||
** false.
|
||||
*/
|
||||
#ifdef SQLITE_COVERAGE_TEST
|
||||
# define ALWAYS(x) (1)
|
||||
# define NEVER(X) (0)
|
||||
#elif defined(SQLITE_DEBUG)
|
||||
# define ALWAYS(x) sqlite3Fts3Always((x)!=0)
|
||||
# define NEVER(x) sqlite3Fts3Never((x)!=0)
|
||||
int sqlite3Fts3Always(int b);
|
||||
int sqlite3Fts3Never(int b);
|
||||
#else
|
||||
# define ALWAYS(x) (x)
|
||||
# define NEVER(x) (x)
|
||||
#endif
|
||||
|
||||
/*
|
||||
** Internal types used by SQLite.
|
||||
*/
|
||||
typedef unsigned char u8; /* 1-byte (or larger) unsigned integer */
|
||||
typedef short int i16; /* 2-byte (or larger) signed integer */
|
||||
typedef unsigned int u32; /* 4-byte unsigned integer */
|
||||
typedef sqlite3_uint64 u64; /* 8-byte unsigned integer */
|
||||
typedef sqlite3_int64 i64; /* 8-byte signed integer */
|
||||
|
||||
/*
|
||||
** Macro used to suppress compiler warnings for unused parameters.
|
||||
*/
|
||||
#define UNUSED_PARAMETER(x) (void)(x)
|
||||
|
||||
/*
|
||||
** Activate assert() only if SQLITE_TEST is enabled.
|
||||
*/
|
||||
#if !defined(NDEBUG) && !defined(SQLITE_DEBUG)
|
||||
# define NDEBUG 1
|
||||
#endif
|
||||
|
||||
/*
|
||||
** The TESTONLY macro is used to enclose variable declarations or
|
||||
** other bits of code that are needed to support the arguments
|
||||
** within testcase() and assert() macros.
|
||||
*/
|
||||
#if defined(SQLITE_DEBUG) || defined(SQLITE_COVERAGE_TEST)
|
||||
# define TESTONLY(X) X
|
||||
#else
|
||||
# define TESTONLY(X)
|
||||
#endif
|
||||
|
||||
#endif /* SQLITE_AMALGAMATION */
|
||||
|
||||
#ifdef SQLITE_DEBUG
|
||||
int sqlite3Fts3Corrupt(void);
|
||||
# define FTS_CORRUPT_VTAB sqlite3Fts3Corrupt()
|
||||
#else
|
||||
# define FTS_CORRUPT_VTAB SQLITE_CORRUPT_VTAB
|
||||
#endif
|
||||
|
||||
typedef struct Fts3Table Fts3Table;
|
||||
typedef struct Fts3Cursor Fts3Cursor;
|
||||
typedef struct Fts3Expr Fts3Expr;
|
||||
typedef struct Fts3Phrase Fts3Phrase;
|
||||
typedef struct Fts3PhraseToken Fts3PhraseToken;
|
||||
|
||||
typedef struct Fts3Doclist Fts3Doclist;
|
||||
typedef struct Fts3SegFilter Fts3SegFilter;
|
||||
typedef struct Fts3DeferredToken Fts3DeferredToken;
|
||||
typedef struct Fts3SegReader Fts3SegReader;
|
||||
typedef struct Fts3MultiSegReader Fts3MultiSegReader;
|
||||
|
||||
typedef struct MatchinfoBuffer MatchinfoBuffer;
|
||||
|
||||
/*
|
||||
** A connection to a fulltext index is an instance of the following
|
||||
** structure. The xCreate and xConnect methods create an instance
|
||||
** of this structure and xDestroy and xDisconnect free that instance.
|
||||
** All other methods receive a pointer to the structure as one of their
|
||||
** arguments.
|
||||
*/
|
||||
struct Fts3Table {
|
||||
sqlite3_vtab base; /* Base class used by SQLite core */
|
||||
sqlite3 *db; /* The database connection */
|
||||
const char *zDb; /* logical database name */
|
||||
const char *zName; /* virtual table name */
|
||||
int nColumn; /* number of named columns in virtual table */
|
||||
char **azColumn; /* column names. malloced */
|
||||
u8 *abNotindexed; /* True for 'notindexed' columns */
|
||||
sqlite3_tokenizer *pTokenizer; /* tokenizer for inserts and queries */
|
||||
char *zContentTbl; /* content=xxx option, or NULL */
|
||||
char *zLanguageid; /* languageid=xxx option, or NULL */
|
||||
int nAutoincrmerge; /* Value configured by 'automerge' */
|
||||
u32 nLeafAdd; /* Number of leaf blocks added this trans */
|
||||
|
||||
/* Precompiled statements used by the implementation. Each of these
|
||||
** statements is run and reset within a single virtual table API call.
|
||||
*/
|
||||
sqlite3_stmt *aStmt[40];
|
||||
|
||||
char *zReadExprlist;
|
||||
char *zWriteExprlist;
|
||||
|
||||
int nNodeSize; /* Soft limit for node size */
|
||||
u8 bFts4; /* True for FTS4, false for FTS3 */
|
||||
u8 bHasStat; /* True if %_stat table exists (2==unknown) */
|
||||
u8 bHasDocsize; /* True if %_docsize table exists */
|
||||
u8 bDescIdx; /* True if doclists are in reverse order */
|
||||
u8 bIgnoreSavepoint; /* True to ignore xSavepoint invocations */
|
||||
int nPgsz; /* Page size for host database */
|
||||
char *zSegmentsTbl; /* Name of %_segments table */
|
||||
sqlite3_blob *pSegments; /* Blob handle open on %_segments table */
|
||||
|
||||
/*
|
||||
** The following array of hash tables is used to buffer pending index
|
||||
** updates during transactions. All pending updates buffered at any one
|
||||
** time must share a common language-id (see the FTS4 langid= feature).
|
||||
** The current language id is stored in variable iPrevLangid.
|
||||
**
|
||||
** A single FTS4 table may have multiple full-text indexes. For each index
|
||||
** there is an entry in the aIndex[] array. Index 0 is an index of all the
|
||||
** terms that appear in the document set. Each subsequent index in aIndex[]
|
||||
** is an index of prefixes of a specific length.
|
||||
**
|
||||
** Variable nPendingData contains an estimate the memory consumed by the
|
||||
** pending data structures, including hash table overhead, but not including
|
||||
** malloc overhead. When nPendingData exceeds nMaxPendingData, all hash
|
||||
** tables are flushed to disk. Variable iPrevDocid is the docid of the most
|
||||
** recently inserted record.
|
||||
*/
|
||||
int nIndex; /* Size of aIndex[] */
|
||||
struct Fts3Index {
|
||||
int nPrefix; /* Prefix length (0 for main terms index) */
|
||||
Fts3Hash hPending; /* Pending terms table for this index */
|
||||
} *aIndex;
|
||||
int nMaxPendingData; /* Max pending data before flush to disk */
|
||||
int nPendingData; /* Current bytes of pending data */
|
||||
sqlite_int64 iPrevDocid; /* Docid of most recently inserted document */
|
||||
int iPrevLangid; /* Langid of recently inserted document */
|
||||
int bPrevDelete; /* True if last operation was a delete */
|
||||
|
||||
#if defined(SQLITE_DEBUG) || defined(SQLITE_COVERAGE_TEST)
|
||||
/* State variables used for validating that the transaction control
|
||||
** methods of the virtual table are called at appropriate times. These
|
||||
** values do not contribute to FTS functionality; they are used for
|
||||
** verifying the operation of the SQLite core.
|
||||
*/
|
||||
int inTransaction; /* True after xBegin but before xCommit/xRollback */
|
||||
int mxSavepoint; /* Largest valid xSavepoint integer */
|
||||
#endif
|
||||
|
||||
#ifdef SQLITE_TEST
|
||||
/* True to disable the incremental doclist optimization. This is controled
|
||||
** by special insert command 'test-no-incr-doclist'. */
|
||||
int bNoIncrDoclist;
|
||||
#endif
|
||||
};
|
||||
|
||||
/*
|
||||
** When the core wants to read from the virtual table, it creates a
|
||||
** virtual table cursor (an instance of the following structure) using
|
||||
** the xOpen method. Cursors are destroyed using the xClose method.
|
||||
*/
|
||||
struct Fts3Cursor {
|
||||
sqlite3_vtab_cursor base; /* Base class used by SQLite core */
|
||||
i16 eSearch; /* Search strategy (see below) */
|
||||
u8 isEof; /* True if at End Of Results */
|
||||
u8 isRequireSeek; /* True if must seek pStmt to %_content row */
|
||||
sqlite3_stmt *pStmt; /* Prepared statement in use by the cursor */
|
||||
Fts3Expr *pExpr; /* Parsed MATCH query string */
|
||||
int iLangid; /* Language being queried for */
|
||||
int nPhrase; /* Number of matchable phrases in query */
|
||||
Fts3DeferredToken *pDeferred; /* Deferred search tokens, if any */
|
||||
sqlite3_int64 iPrevId; /* Previous id read from aDoclist */
|
||||
char *pNextId; /* Pointer into the body of aDoclist */
|
||||
char *aDoclist; /* List of docids for full-text queries */
|
||||
int nDoclist; /* Size of buffer at aDoclist */
|
||||
u8 bDesc; /* True to sort in descending order */
|
||||
int eEvalmode; /* An FTS3_EVAL_XX constant */
|
||||
int nRowAvg; /* Average size of database rows, in pages */
|
||||
sqlite3_int64 nDoc; /* Documents in table */
|
||||
i64 iMinDocid; /* Minimum docid to return */
|
||||
i64 iMaxDocid; /* Maximum docid to return */
|
||||
int isMatchinfoNeeded; /* True when aMatchinfo[] needs filling in */
|
||||
MatchinfoBuffer *pMIBuffer; /* Buffer for matchinfo data */
|
||||
};
|
||||
|
||||
#define FTS3_EVAL_FILTER 0
|
||||
#define FTS3_EVAL_NEXT 1
|
||||
#define FTS3_EVAL_MATCHINFO 2
|
||||
|
||||
/*
|
||||
** The Fts3Cursor.eSearch member is always set to one of the following.
|
||||
** Actualy, Fts3Cursor.eSearch can be greater than or equal to
|
||||
** FTS3_FULLTEXT_SEARCH. If so, then Fts3Cursor.eSearch - 2 is the index
|
||||
** of the column to be searched. For example, in
|
||||
**
|
||||
** CREATE VIRTUAL TABLE ex1 USING fts3(a,b,c,d);
|
||||
** SELECT docid FROM ex1 WHERE b MATCH 'one two three';
|
||||
**
|
||||
** Because the LHS of the MATCH operator is 2nd column "b",
|
||||
** Fts3Cursor.eSearch will be set to FTS3_FULLTEXT_SEARCH+1. (+0 for a,
|
||||
** +1 for b, +2 for c, +3 for d.) If the LHS of MATCH were "ex1"
|
||||
** indicating that all columns should be searched,
|
||||
** then eSearch would be set to FTS3_FULLTEXT_SEARCH+4.
|
||||
*/
|
||||
#define FTS3_FULLSCAN_SEARCH 0 /* Linear scan of %_content table */
|
||||
#define FTS3_DOCID_SEARCH 1 /* Lookup by rowid on %_content table */
|
||||
#define FTS3_FULLTEXT_SEARCH 2 /* Full-text index search */
|
||||
|
||||
/*
|
||||
** The lower 16-bits of the sqlite3_index_info.idxNum value set by
|
||||
** the xBestIndex() method contains the Fts3Cursor.eSearch value described
|
||||
** above. The upper 16-bits contain a combination of the following
|
||||
** bits, used to describe extra constraints on full-text searches.
|
||||
*/
|
||||
#define FTS3_HAVE_LANGID 0x00010000 /* languageid=? */
|
||||
#define FTS3_HAVE_DOCID_GE 0x00020000 /* docid>=? */
|
||||
#define FTS3_HAVE_DOCID_LE 0x00040000 /* docid<=? */
|
||||
|
||||
struct Fts3Doclist {
|
||||
char *aAll; /* Array containing doclist (or NULL) */
|
||||
int nAll; /* Size of a[] in bytes */
|
||||
char *pNextDocid; /* Pointer to next docid */
|
||||
|
||||
sqlite3_int64 iDocid; /* Current docid (if pList!=0) */
|
||||
int bFreeList; /* True if pList should be sqlite3_free()d */
|
||||
char *pList; /* Pointer to position list following iDocid */
|
||||
int nList; /* Length of position list */
|
||||
};
|
||||
|
||||
/*
|
||||
** A "phrase" is a sequence of one or more tokens that must match in
|
||||
** sequence. A single token is the base case and the most common case.
|
||||
** For a sequence of tokens contained in double-quotes (i.e. "one two three")
|
||||
** nToken will be the number of tokens in the string.
|
||||
*/
|
||||
struct Fts3PhraseToken {
|
||||
char *z; /* Text of the token */
|
||||
int n; /* Number of bytes in buffer z */
|
||||
int isPrefix; /* True if token ends with a "*" character */
|
||||
int bFirst; /* True if token must appear at position 0 */
|
||||
|
||||
/* Variables above this point are populated when the expression is
|
||||
** parsed (by code in fts3_expr.c). Below this point the variables are
|
||||
** used when evaluating the expression. */
|
||||
Fts3DeferredToken *pDeferred; /* Deferred token object for this token */
|
||||
Fts3MultiSegReader *pSegcsr; /* Segment-reader for this token */
|
||||
};
|
||||
|
||||
struct Fts3Phrase {
|
||||
/* Cache of doclist for this phrase. */
|
||||
Fts3Doclist doclist;
|
||||
int bIncr; /* True if doclist is loaded incrementally */
|
||||
int iDoclistToken;
|
||||
|
||||
/* Used by sqlite3Fts3EvalPhrasePoslist() if this is a descendent of an
|
||||
** OR condition. */
|
||||
char *pOrPoslist;
|
||||
i64 iOrDocid;
|
||||
|
||||
/* Variables below this point are populated by fts3_expr.c when parsing
|
||||
** a MATCH expression. Everything above is part of the evaluation phase.
|
||||
*/
|
||||
int nToken; /* Number of tokens in the phrase */
|
||||
int iColumn; /* Index of column this phrase must match */
|
||||
Fts3PhraseToken aToken[1]; /* One entry for each token in the phrase */
|
||||
};
|
||||
|
||||
/*
|
||||
** A tree of these objects forms the RHS of a MATCH operator.
|
||||
**
|
||||
** If Fts3Expr.eType is FTSQUERY_PHRASE and isLoaded is true, then aDoclist
|
||||
** points to a malloced buffer, size nDoclist bytes, containing the results
|
||||
** of this phrase query in FTS3 doclist format. As usual, the initial
|
||||
** "Length" field found in doclists stored on disk is omitted from this
|
||||
** buffer.
|
||||
**
|
||||
** Variable aMI is used only for FTSQUERY_NEAR nodes to store the global
|
||||
** matchinfo data. If it is not NULL, it points to an array of size nCol*3,
|
||||
** where nCol is the number of columns in the queried FTS table. The array
|
||||
** is populated as follows:
|
||||
**
|
||||
** aMI[iCol*3 + 0] = Undefined
|
||||
** aMI[iCol*3 + 1] = Number of occurrences
|
||||
** aMI[iCol*3 + 2] = Number of rows containing at least one instance
|
||||
**
|
||||
** The aMI array is allocated using sqlite3_malloc(). It should be freed
|
||||
** when the expression node is.
|
||||
*/
|
||||
struct Fts3Expr {
|
||||
int eType; /* One of the FTSQUERY_XXX values defined below */
|
||||
int nNear; /* Valid if eType==FTSQUERY_NEAR */
|
||||
Fts3Expr *pParent; /* pParent->pLeft==this or pParent->pRight==this */
|
||||
Fts3Expr *pLeft; /* Left operand */
|
||||
Fts3Expr *pRight; /* Right operand */
|
||||
Fts3Phrase *pPhrase; /* Valid if eType==FTSQUERY_PHRASE */
|
||||
|
||||
/* The following are used by the fts3_eval.c module. */
|
||||
sqlite3_int64 iDocid; /* Current docid */
|
||||
u8 bEof; /* True this expression is at EOF already */
|
||||
u8 bStart; /* True if iDocid is valid */
|
||||
u8 bDeferred; /* True if this expression is entirely deferred */
|
||||
|
||||
/* The following are used by the fts3_snippet.c module. */
|
||||
int iPhrase; /* Index of this phrase in matchinfo() results */
|
||||
u32 *aMI; /* See above */
|
||||
};
|
||||
|
||||
/*
|
||||
** Candidate values for Fts3Query.eType. Note that the order of the first
|
||||
** four values is in order of precedence when parsing expressions. For
|
||||
** example, the following:
|
||||
**
|
||||
** "a OR b AND c NOT d NEAR e"
|
||||
**
|
||||
** is equivalent to:
|
||||
**
|
||||
** "a OR (b AND (c NOT (d NEAR e)))"
|
||||
*/
|
||||
#define FTSQUERY_NEAR 1
|
||||
#define FTSQUERY_NOT 2
|
||||
#define FTSQUERY_AND 3
|
||||
#define FTSQUERY_OR 4
|
||||
#define FTSQUERY_PHRASE 5
|
||||
|
||||
|
||||
/* fts3_write.c */
|
||||
int sqlite3Fts3UpdateMethod(sqlite3_vtab*,int,sqlite3_value**,sqlite3_int64*);
|
||||
int sqlite3Fts3PendingTermsFlush(Fts3Table *);
|
||||
void sqlite3Fts3PendingTermsClear(Fts3Table *);
|
||||
int sqlite3Fts3Optimize(Fts3Table *);
|
||||
int sqlite3Fts3SegReaderNew(int, int, sqlite3_int64,
|
||||
sqlite3_int64, sqlite3_int64, const char *, int, Fts3SegReader**);
|
||||
int sqlite3Fts3SegReaderPending(
|
||||
Fts3Table*,int,const char*,int,int,Fts3SegReader**);
|
||||
void sqlite3Fts3SegReaderFree(Fts3SegReader *);
|
||||
int sqlite3Fts3AllSegdirs(Fts3Table*, int, int, int, sqlite3_stmt **);
|
||||
int sqlite3Fts3ReadBlock(Fts3Table*, sqlite3_int64, char **, int*, int*);
|
||||
|
||||
int sqlite3Fts3SelectDoctotal(Fts3Table *, sqlite3_stmt **);
|
||||
int sqlite3Fts3SelectDocsize(Fts3Table *, sqlite3_int64, sqlite3_stmt **);
|
||||
|
||||
#ifndef SQLITE_DISABLE_FTS4_DEFERRED
|
||||
void sqlite3Fts3FreeDeferredTokens(Fts3Cursor *);
|
||||
int sqlite3Fts3DeferToken(Fts3Cursor *, Fts3PhraseToken *, int);
|
||||
int sqlite3Fts3CacheDeferredDoclists(Fts3Cursor *);
|
||||
void sqlite3Fts3FreeDeferredDoclists(Fts3Cursor *);
|
||||
int sqlite3Fts3DeferredTokenList(Fts3DeferredToken *, char **, int *);
|
||||
#else
|
||||
# define sqlite3Fts3FreeDeferredTokens(x)
|
||||
# define sqlite3Fts3DeferToken(x,y,z) SQLITE_OK
|
||||
# define sqlite3Fts3CacheDeferredDoclists(x) SQLITE_OK
|
||||
# define sqlite3Fts3FreeDeferredDoclists(x)
|
||||
# define sqlite3Fts3DeferredTokenList(x,y,z) SQLITE_OK
|
||||
#endif
|
||||
|
||||
void sqlite3Fts3SegmentsClose(Fts3Table *);
|
||||
int sqlite3Fts3MaxLevel(Fts3Table *, int *);
|
||||
|
||||
/* Special values interpreted by sqlite3SegReaderCursor() */
|
||||
#define FTS3_SEGCURSOR_PENDING -1
|
||||
#define FTS3_SEGCURSOR_ALL -2
|
||||
|
||||
int sqlite3Fts3SegReaderStart(Fts3Table*, Fts3MultiSegReader*, Fts3SegFilter*);
|
||||
int sqlite3Fts3SegReaderStep(Fts3Table *, Fts3MultiSegReader *);
|
||||
void sqlite3Fts3SegReaderFinish(Fts3MultiSegReader *);
|
||||
|
||||
int sqlite3Fts3SegReaderCursor(Fts3Table *,
|
||||
int, int, int, const char *, int, int, int, Fts3MultiSegReader *);
|
||||
|
||||
/* Flags allowed as part of the 4th argument to SegmentReaderIterate() */
|
||||
#define FTS3_SEGMENT_REQUIRE_POS 0x00000001
|
||||
#define FTS3_SEGMENT_IGNORE_EMPTY 0x00000002
|
||||
#define FTS3_SEGMENT_COLUMN_FILTER 0x00000004
|
||||
#define FTS3_SEGMENT_PREFIX 0x00000008
|
||||
#define FTS3_SEGMENT_SCAN 0x00000010
|
||||
#define FTS3_SEGMENT_FIRST 0x00000020
|
||||
|
||||
/* Type passed as 4th argument to SegmentReaderIterate() */
|
||||
struct Fts3SegFilter {
|
||||
const char *zTerm;
|
||||
int nTerm;
|
||||
int iCol;
|
||||
int flags;
|
||||
};
|
||||
|
||||
struct Fts3MultiSegReader {
|
||||
/* Used internally by sqlite3Fts3SegReaderXXX() calls */
|
||||
Fts3SegReader **apSegment; /* Array of Fts3SegReader objects */
|
||||
int nSegment; /* Size of apSegment array */
|
||||
int nAdvance; /* How many seg-readers to advance */
|
||||
Fts3SegFilter *pFilter; /* Pointer to filter object */
|
||||
char *aBuffer; /* Buffer to merge doclists in */
|
||||
int nBuffer; /* Allocated size of aBuffer[] in bytes */
|
||||
|
||||
int iColFilter; /* If >=0, filter for this column */
|
||||
int bRestart;
|
||||
|
||||
/* Used by fts3.c only. */
|
||||
int nCost; /* Cost of running iterator */
|
||||
int bLookup; /* True if a lookup of a single entry. */
|
||||
|
||||
/* Output values. Valid only after Fts3SegReaderStep() returns SQLITE_ROW. */
|
||||
char *zTerm; /* Pointer to term buffer */
|
||||
int nTerm; /* Size of zTerm in bytes */
|
||||
char *aDoclist; /* Pointer to doclist buffer */
|
||||
int nDoclist; /* Size of aDoclist[] in bytes */
|
||||
};
|
||||
|
||||
int sqlite3Fts3Incrmerge(Fts3Table*,int,int);
|
||||
|
||||
#define fts3GetVarint32(p, piVal) ( \
|
||||
(*(u8*)(p)&0x80) ? sqlite3Fts3GetVarint32(p, piVal) : (*piVal=*(u8*)(p), 1) \
|
||||
)
|
||||
|
||||
/* fts3.c */
|
||||
void sqlite3Fts3ErrMsg(char**,const char*,...);
|
||||
int sqlite3Fts3PutVarint(char *, sqlite3_int64);
|
||||
int sqlite3Fts3GetVarint(const char *, sqlite_int64 *);
|
||||
int sqlite3Fts3GetVarint32(const char *, int *);
|
||||
int sqlite3Fts3VarintLen(sqlite3_uint64);
|
||||
void sqlite3Fts3Dequote(char *);
|
||||
void sqlite3Fts3DoclistPrev(int,char*,int,char**,sqlite3_int64*,int*,u8*);
|
||||
int sqlite3Fts3EvalPhraseStats(Fts3Cursor *, Fts3Expr *, u32 *);
|
||||
int sqlite3Fts3FirstFilter(sqlite3_int64, char *, int, char *);
|
||||
void sqlite3Fts3CreateStatTable(int*, Fts3Table*);
|
||||
int sqlite3Fts3EvalTestDeferred(Fts3Cursor *pCsr, int *pRc);
|
||||
|
||||
/* fts3_tokenizer.c */
|
||||
const char *sqlite3Fts3NextToken(const char *, int *);
|
||||
int sqlite3Fts3InitHashTable(sqlite3 *, Fts3Hash *, const char *);
|
||||
int sqlite3Fts3InitTokenizer(Fts3Hash *pHash, const char *,
|
||||
sqlite3_tokenizer **, char **
|
||||
);
|
||||
int sqlite3Fts3IsIdChar(char);
|
||||
|
||||
/* fts3_snippet.c */
|
||||
void sqlite3Fts3Offsets(sqlite3_context*, Fts3Cursor*);
|
||||
void sqlite3Fts3Snippet(sqlite3_context *, Fts3Cursor *, const char *,
|
||||
const char *, const char *, int, int
|
||||
);
|
||||
void sqlite3Fts3Matchinfo(sqlite3_context *, Fts3Cursor *, const char *);
|
||||
void sqlite3Fts3MIBufferFree(MatchinfoBuffer *p);
|
||||
|
||||
/* fts3_expr.c */
|
||||
int sqlite3Fts3ExprParse(sqlite3_tokenizer *, int,
|
||||
char **, int, int, int, const char *, int, Fts3Expr **, char **
|
||||
);
|
||||
void sqlite3Fts3ExprFree(Fts3Expr *);
|
||||
#ifdef SQLITE_TEST
|
||||
int sqlite3Fts3ExprInitTestInterface(sqlite3 *db);
|
||||
int sqlite3Fts3InitTerm(sqlite3 *db);
|
||||
#endif
|
||||
|
||||
int sqlite3Fts3OpenTokenizer(sqlite3_tokenizer *, int, const char *, int,
|
||||
sqlite3_tokenizer_cursor **
|
||||
);
|
||||
|
||||
/* fts3_aux.c */
|
||||
int sqlite3Fts3InitAux(sqlite3 *db);
|
||||
|
||||
void sqlite3Fts3EvalPhraseCleanup(Fts3Phrase *);
|
||||
|
||||
int sqlite3Fts3MsrIncrStart(
|
||||
Fts3Table*, Fts3MultiSegReader*, int, const char*, int);
|
||||
int sqlite3Fts3MsrIncrNext(
|
||||
Fts3Table *, Fts3MultiSegReader *, sqlite3_int64 *, char **, int *);
|
||||
int sqlite3Fts3EvalPhrasePoslist(Fts3Cursor *, Fts3Expr *, int iCol, char **);
|
||||
int sqlite3Fts3MsrOvfl(Fts3Cursor *, Fts3MultiSegReader *, int *);
|
||||
int sqlite3Fts3MsrIncrRestart(Fts3MultiSegReader *pCsr);
|
||||
|
||||
/* fts3_tokenize_vtab.c */
|
||||
int sqlite3Fts3InitTok(sqlite3*, Fts3Hash *);
|
||||
|
||||
/* fts3_unicode2.c (functions generated by parsing unicode text files) */
|
||||
#ifndef SQLITE_DISABLE_FTS3_UNICODE
|
||||
int sqlite3FtsUnicodeFold(int, int);
|
||||
int sqlite3FtsUnicodeIsalnum(int);
|
||||
int sqlite3FtsUnicodeIsdiacritic(int);
|
||||
#endif
|
||||
|
||||
#endif /* !SQLITE_CORE || SQLITE_ENABLE_FTS3 */
|
||||
#endif /* _FTSINT_H */
|
||||
@ -0,0 +1,550 @@
|
||||
/*
|
||||
** 2011 Jan 27
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
*/
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
|
||||
#include <string.h>
|
||||
#include <assert.h>
|
||||
|
||||
typedef struct Fts3auxTable Fts3auxTable;
|
||||
typedef struct Fts3auxCursor Fts3auxCursor;
|
||||
|
||||
struct Fts3auxTable {
|
||||
sqlite3_vtab base; /* Base class used by SQLite core */
|
||||
Fts3Table *pFts3Tab;
|
||||
};
|
||||
|
||||
struct Fts3auxCursor {
|
||||
sqlite3_vtab_cursor base; /* Base class used by SQLite core */
|
||||
Fts3MultiSegReader csr; /* Must be right after "base" */
|
||||
Fts3SegFilter filter;
|
||||
char *zStop;
|
||||
int nStop; /* Byte-length of string zStop */
|
||||
int iLangid; /* Language id to query */
|
||||
int isEof; /* True if cursor is at EOF */
|
||||
sqlite3_int64 iRowid; /* Current rowid */
|
||||
|
||||
int iCol; /* Current value of 'col' column */
|
||||
int nStat; /* Size of aStat[] array */
|
||||
struct Fts3auxColstats {
|
||||
sqlite3_int64 nDoc; /* 'documents' values for current csr row */
|
||||
sqlite3_int64 nOcc; /* 'occurrences' values for current csr row */
|
||||
} *aStat;
|
||||
};
|
||||
|
||||
/*
|
||||
** Schema of the terms table.
|
||||
*/
|
||||
#define FTS3_AUX_SCHEMA \
|
||||
"CREATE TABLE x(term, col, documents, occurrences, languageid HIDDEN)"
|
||||
|
||||
/*
|
||||
** This function does all the work for both the xConnect and xCreate methods.
|
||||
** These tables have no persistent representation of their own, so xConnect
|
||||
** and xCreate are identical operations.
|
||||
*/
|
||||
static int fts3auxConnectMethod(
|
||||
sqlite3 *db, /* Database connection */
|
||||
void *pUnused, /* Unused */
|
||||
int argc, /* Number of elements in argv array */
|
||||
const char * const *argv, /* xCreate/xConnect argument array */
|
||||
sqlite3_vtab **ppVtab, /* OUT: New sqlite3_vtab object */
|
||||
char **pzErr /* OUT: sqlite3_malloc'd error message */
|
||||
){
|
||||
char const *zDb; /* Name of database (e.g. "main") */
|
||||
char const *zFts3; /* Name of fts3 table */
|
||||
int nDb; /* Result of strlen(zDb) */
|
||||
int nFts3; /* Result of strlen(zFts3) */
|
||||
int nByte; /* Bytes of space to allocate here */
|
||||
int rc; /* value returned by declare_vtab() */
|
||||
Fts3auxTable *p; /* Virtual table object to return */
|
||||
|
||||
UNUSED_PARAMETER(pUnused);
|
||||
|
||||
/* The user should invoke this in one of two forms:
|
||||
**
|
||||
** CREATE VIRTUAL TABLE xxx USING fts4aux(fts4-table);
|
||||
** CREATE VIRTUAL TABLE xxx USING fts4aux(fts4-table-db, fts4-table);
|
||||
*/
|
||||
if( argc!=4 && argc!=5 ) goto bad_args;
|
||||
|
||||
zDb = argv[1];
|
||||
nDb = (int)strlen(zDb);
|
||||
if( argc==5 ){
|
||||
if( nDb==4 && 0==sqlite3_strnicmp("temp", zDb, 4) ){
|
||||
zDb = argv[3];
|
||||
nDb = (int)strlen(zDb);
|
||||
zFts3 = argv[4];
|
||||
}else{
|
||||
goto bad_args;
|
||||
}
|
||||
}else{
|
||||
zFts3 = argv[3];
|
||||
}
|
||||
nFts3 = (int)strlen(zFts3);
|
||||
|
||||
rc = sqlite3_declare_vtab(db, FTS3_AUX_SCHEMA);
|
||||
if( rc!=SQLITE_OK ) return rc;
|
||||
|
||||
nByte = sizeof(Fts3auxTable) + sizeof(Fts3Table) + nDb + nFts3 + 2;
|
||||
p = (Fts3auxTable *)sqlite3_malloc(nByte);
|
||||
if( !p ) return SQLITE_NOMEM;
|
||||
memset(p, 0, nByte);
|
||||
|
||||
p->pFts3Tab = (Fts3Table *)&p[1];
|
||||
p->pFts3Tab->zDb = (char *)&p->pFts3Tab[1];
|
||||
p->pFts3Tab->zName = &p->pFts3Tab->zDb[nDb+1];
|
||||
p->pFts3Tab->db = db;
|
||||
p->pFts3Tab->nIndex = 1;
|
||||
|
||||
memcpy((char *)p->pFts3Tab->zDb, zDb, nDb);
|
||||
memcpy((char *)p->pFts3Tab->zName, zFts3, nFts3);
|
||||
sqlite3Fts3Dequote((char *)p->pFts3Tab->zName);
|
||||
|
||||
*ppVtab = (sqlite3_vtab *)p;
|
||||
return SQLITE_OK;
|
||||
|
||||
bad_args:
|
||||
sqlite3Fts3ErrMsg(pzErr, "invalid arguments to fts4aux constructor");
|
||||
return SQLITE_ERROR;
|
||||
}
|
||||
|
||||
/*
|
||||
** This function does the work for both the xDisconnect and xDestroy methods.
|
||||
** These tables have no persistent representation of their own, so xDisconnect
|
||||
** and xDestroy are identical operations.
|
||||
*/
|
||||
static int fts3auxDisconnectMethod(sqlite3_vtab *pVtab){
|
||||
Fts3auxTable *p = (Fts3auxTable *)pVtab;
|
||||
Fts3Table *pFts3 = p->pFts3Tab;
|
||||
int i;
|
||||
|
||||
/* Free any prepared statements held */
|
||||
for(i=0; i<SizeofArray(pFts3->aStmt); i++){
|
||||
sqlite3_finalize(pFts3->aStmt[i]);
|
||||
}
|
||||
sqlite3_free(pFts3->zSegmentsTbl);
|
||||
sqlite3_free(p);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
#define FTS4AUX_EQ_CONSTRAINT 1
|
||||
#define FTS4AUX_GE_CONSTRAINT 2
|
||||
#define FTS4AUX_LE_CONSTRAINT 4
|
||||
|
||||
/*
|
||||
** xBestIndex - Analyze a WHERE and ORDER BY clause.
|
||||
*/
|
||||
static int fts3auxBestIndexMethod(
|
||||
sqlite3_vtab *pVTab,
|
||||
sqlite3_index_info *pInfo
|
||||
){
|
||||
int i;
|
||||
int iEq = -1;
|
||||
int iGe = -1;
|
||||
int iLe = -1;
|
||||
int iLangid = -1;
|
||||
int iNext = 1; /* Next free argvIndex value */
|
||||
|
||||
UNUSED_PARAMETER(pVTab);
|
||||
|
||||
/* This vtab delivers always results in "ORDER BY term ASC" order. */
|
||||
if( pInfo->nOrderBy==1
|
||||
&& pInfo->aOrderBy[0].iColumn==0
|
||||
&& pInfo->aOrderBy[0].desc==0
|
||||
){
|
||||
pInfo->orderByConsumed = 1;
|
||||
}
|
||||
|
||||
/* Search for equality and range constraints on the "term" column.
|
||||
** And equality constraints on the hidden "languageid" column. */
|
||||
for(i=0; i<pInfo->nConstraint; i++){
|
||||
if( pInfo->aConstraint[i].usable ){
|
||||
int op = pInfo->aConstraint[i].op;
|
||||
int iCol = pInfo->aConstraint[i].iColumn;
|
||||
|
||||
if( iCol==0 ){
|
||||
if( op==SQLITE_INDEX_CONSTRAINT_EQ ) iEq = i;
|
||||
if( op==SQLITE_INDEX_CONSTRAINT_LT ) iLe = i;
|
||||
if( op==SQLITE_INDEX_CONSTRAINT_LE ) iLe = i;
|
||||
if( op==SQLITE_INDEX_CONSTRAINT_GT ) iGe = i;
|
||||
if( op==SQLITE_INDEX_CONSTRAINT_GE ) iGe = i;
|
||||
}
|
||||
if( iCol==4 ){
|
||||
if( op==SQLITE_INDEX_CONSTRAINT_EQ ) iLangid = i;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if( iEq>=0 ){
|
||||
pInfo->idxNum = FTS4AUX_EQ_CONSTRAINT;
|
||||
pInfo->aConstraintUsage[iEq].argvIndex = iNext++;
|
||||
pInfo->estimatedCost = 5;
|
||||
}else{
|
||||
pInfo->idxNum = 0;
|
||||
pInfo->estimatedCost = 20000;
|
||||
if( iGe>=0 ){
|
||||
pInfo->idxNum += FTS4AUX_GE_CONSTRAINT;
|
||||
pInfo->aConstraintUsage[iGe].argvIndex = iNext++;
|
||||
pInfo->estimatedCost /= 2;
|
||||
}
|
||||
if( iLe>=0 ){
|
||||
pInfo->idxNum += FTS4AUX_LE_CONSTRAINT;
|
||||
pInfo->aConstraintUsage[iLe].argvIndex = iNext++;
|
||||
pInfo->estimatedCost /= 2;
|
||||
}
|
||||
}
|
||||
if( iLangid>=0 ){
|
||||
pInfo->aConstraintUsage[iLangid].argvIndex = iNext++;
|
||||
pInfo->estimatedCost--;
|
||||
}
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xOpen - Open a cursor.
|
||||
*/
|
||||
static int fts3auxOpenMethod(sqlite3_vtab *pVTab, sqlite3_vtab_cursor **ppCsr){
|
||||
Fts3auxCursor *pCsr; /* Pointer to cursor object to return */
|
||||
|
||||
UNUSED_PARAMETER(pVTab);
|
||||
|
||||
pCsr = (Fts3auxCursor *)sqlite3_malloc(sizeof(Fts3auxCursor));
|
||||
if( !pCsr ) return SQLITE_NOMEM;
|
||||
memset(pCsr, 0, sizeof(Fts3auxCursor));
|
||||
|
||||
*ppCsr = (sqlite3_vtab_cursor *)pCsr;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xClose - Close a cursor.
|
||||
*/
|
||||
static int fts3auxCloseMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3Table *pFts3 = ((Fts3auxTable *)pCursor->pVtab)->pFts3Tab;
|
||||
Fts3auxCursor *pCsr = (Fts3auxCursor *)pCursor;
|
||||
|
||||
sqlite3Fts3SegmentsClose(pFts3);
|
||||
sqlite3Fts3SegReaderFinish(&pCsr->csr);
|
||||
sqlite3_free((void *)pCsr->filter.zTerm);
|
||||
sqlite3_free(pCsr->zStop);
|
||||
sqlite3_free(pCsr->aStat);
|
||||
sqlite3_free(pCsr);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
static int fts3auxGrowStatArray(Fts3auxCursor *pCsr, int nSize){
|
||||
if( nSize>pCsr->nStat ){
|
||||
struct Fts3auxColstats *aNew;
|
||||
aNew = (struct Fts3auxColstats *)sqlite3_realloc(pCsr->aStat,
|
||||
sizeof(struct Fts3auxColstats) * nSize
|
||||
);
|
||||
if( aNew==0 ) return SQLITE_NOMEM;
|
||||
memset(&aNew[pCsr->nStat], 0,
|
||||
sizeof(struct Fts3auxColstats) * (nSize - pCsr->nStat)
|
||||
);
|
||||
pCsr->aStat = aNew;
|
||||
pCsr->nStat = nSize;
|
||||
}
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xNext - Advance the cursor to the next row, if any.
|
||||
*/
|
||||
static int fts3auxNextMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3auxCursor *pCsr = (Fts3auxCursor *)pCursor;
|
||||
Fts3Table *pFts3 = ((Fts3auxTable *)pCursor->pVtab)->pFts3Tab;
|
||||
int rc;
|
||||
|
||||
/* Increment our pretend rowid value. */
|
||||
pCsr->iRowid++;
|
||||
|
||||
for(pCsr->iCol++; pCsr->iCol<pCsr->nStat; pCsr->iCol++){
|
||||
if( pCsr->aStat[pCsr->iCol].nDoc>0 ) return SQLITE_OK;
|
||||
}
|
||||
|
||||
rc = sqlite3Fts3SegReaderStep(pFts3, &pCsr->csr);
|
||||
if( rc==SQLITE_ROW ){
|
||||
int i = 0;
|
||||
int nDoclist = pCsr->csr.nDoclist;
|
||||
char *aDoclist = pCsr->csr.aDoclist;
|
||||
int iCol;
|
||||
|
||||
int eState = 0;
|
||||
|
||||
if( pCsr->zStop ){
|
||||
int n = (pCsr->nStop<pCsr->csr.nTerm) ? pCsr->nStop : pCsr->csr.nTerm;
|
||||
int mc = memcmp(pCsr->zStop, pCsr->csr.zTerm, n);
|
||||
if( mc<0 || (mc==0 && pCsr->csr.nTerm>pCsr->nStop) ){
|
||||
pCsr->isEof = 1;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
}
|
||||
|
||||
if( fts3auxGrowStatArray(pCsr, 2) ) return SQLITE_NOMEM;
|
||||
memset(pCsr->aStat, 0, sizeof(struct Fts3auxColstats) * pCsr->nStat);
|
||||
iCol = 0;
|
||||
|
||||
while( i<nDoclist ){
|
||||
sqlite3_int64 v = 0;
|
||||
|
||||
i += sqlite3Fts3GetVarint(&aDoclist[i], &v);
|
||||
switch( eState ){
|
||||
/* State 0. In this state the integer just read was a docid. */
|
||||
case 0:
|
||||
pCsr->aStat[0].nDoc++;
|
||||
eState = 1;
|
||||
iCol = 0;
|
||||
break;
|
||||
|
||||
/* State 1. In this state we are expecting either a 1, indicating
|
||||
** that the following integer will be a column number, or the
|
||||
** start of a position list for column 0.
|
||||
**
|
||||
** The only difference between state 1 and state 2 is that if the
|
||||
** integer encountered in state 1 is not 0 or 1, then we need to
|
||||
** increment the column 0 "nDoc" count for this term.
|
||||
*/
|
||||
case 1:
|
||||
assert( iCol==0 );
|
||||
if( v>1 ){
|
||||
pCsr->aStat[1].nDoc++;
|
||||
}
|
||||
eState = 2;
|
||||
/* fall through */
|
||||
|
||||
case 2:
|
||||
if( v==0 ){ /* 0x00. Next integer will be a docid. */
|
||||
eState = 0;
|
||||
}else if( v==1 ){ /* 0x01. Next integer will be a column number. */
|
||||
eState = 3;
|
||||
}else{ /* 2 or greater. A position. */
|
||||
pCsr->aStat[iCol+1].nOcc++;
|
||||
pCsr->aStat[0].nOcc++;
|
||||
}
|
||||
break;
|
||||
|
||||
/* State 3. The integer just read is a column number. */
|
||||
default: assert( eState==3 );
|
||||
iCol = (int)v;
|
||||
if( fts3auxGrowStatArray(pCsr, iCol+2) ) return SQLITE_NOMEM;
|
||||
pCsr->aStat[iCol+1].nDoc++;
|
||||
eState = 2;
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
pCsr->iCol = 0;
|
||||
rc = SQLITE_OK;
|
||||
}else{
|
||||
pCsr->isEof = 1;
|
||||
}
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** xFilter - Initialize a cursor to point at the start of its data.
|
||||
*/
|
||||
static int fts3auxFilterMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* The cursor used for this query */
|
||||
int idxNum, /* Strategy index */
|
||||
const char *idxStr, /* Unused */
|
||||
int nVal, /* Number of elements in apVal */
|
||||
sqlite3_value **apVal /* Arguments for the indexing scheme */
|
||||
){
|
||||
Fts3auxCursor *pCsr = (Fts3auxCursor *)pCursor;
|
||||
Fts3Table *pFts3 = ((Fts3auxTable *)pCursor->pVtab)->pFts3Tab;
|
||||
int rc;
|
||||
int isScan = 0;
|
||||
int iLangVal = 0; /* Language id to query */
|
||||
|
||||
int iEq = -1; /* Index of term=? value in apVal */
|
||||
int iGe = -1; /* Index of term>=? value in apVal */
|
||||
int iLe = -1; /* Index of term<=? value in apVal */
|
||||
int iLangid = -1; /* Index of languageid=? value in apVal */
|
||||
int iNext = 0;
|
||||
|
||||
UNUSED_PARAMETER(nVal);
|
||||
UNUSED_PARAMETER(idxStr);
|
||||
|
||||
assert( idxStr==0 );
|
||||
assert( idxNum==FTS4AUX_EQ_CONSTRAINT || idxNum==0
|
||||
|| idxNum==FTS4AUX_LE_CONSTRAINT || idxNum==FTS4AUX_GE_CONSTRAINT
|
||||
|| idxNum==(FTS4AUX_LE_CONSTRAINT|FTS4AUX_GE_CONSTRAINT)
|
||||
);
|
||||
|
||||
if( idxNum==FTS4AUX_EQ_CONSTRAINT ){
|
||||
iEq = iNext++;
|
||||
}else{
|
||||
isScan = 1;
|
||||
if( idxNum & FTS4AUX_GE_CONSTRAINT ){
|
||||
iGe = iNext++;
|
||||
}
|
||||
if( idxNum & FTS4AUX_LE_CONSTRAINT ){
|
||||
iLe = iNext++;
|
||||
}
|
||||
}
|
||||
if( iNext<nVal ){
|
||||
iLangid = iNext++;
|
||||
}
|
||||
|
||||
/* In case this cursor is being reused, close and zero it. */
|
||||
testcase(pCsr->filter.zTerm);
|
||||
sqlite3Fts3SegReaderFinish(&pCsr->csr);
|
||||
sqlite3_free((void *)pCsr->filter.zTerm);
|
||||
sqlite3_free(pCsr->aStat);
|
||||
memset(&pCsr->csr, 0, ((u8*)&pCsr[1]) - (u8*)&pCsr->csr);
|
||||
|
||||
pCsr->filter.flags = FTS3_SEGMENT_REQUIRE_POS|FTS3_SEGMENT_IGNORE_EMPTY;
|
||||
if( isScan ) pCsr->filter.flags |= FTS3_SEGMENT_SCAN;
|
||||
|
||||
if( iEq>=0 || iGe>=0 ){
|
||||
const unsigned char *zStr = sqlite3_value_text(apVal[0]);
|
||||
assert( (iEq==0 && iGe==-1) || (iEq==-1 && iGe==0) );
|
||||
if( zStr ){
|
||||
pCsr->filter.zTerm = sqlite3_mprintf("%s", zStr);
|
||||
pCsr->filter.nTerm = sqlite3_value_bytes(apVal[0]);
|
||||
if( pCsr->filter.zTerm==0 ) return SQLITE_NOMEM;
|
||||
}
|
||||
}
|
||||
|
||||
if( iLe>=0 ){
|
||||
pCsr->zStop = sqlite3_mprintf("%s", sqlite3_value_text(apVal[iLe]));
|
||||
pCsr->nStop = sqlite3_value_bytes(apVal[iLe]);
|
||||
if( pCsr->zStop==0 ) return SQLITE_NOMEM;
|
||||
}
|
||||
|
||||
if( iLangid>=0 ){
|
||||
iLangVal = sqlite3_value_int(apVal[iLangid]);
|
||||
|
||||
/* If the user specified a negative value for the languageid, use zero
|
||||
** instead. This works, as the "languageid=?" constraint will also
|
||||
** be tested by the VDBE layer. The test will always be false (since
|
||||
** this module will not return a row with a negative languageid), and
|
||||
** so the overall query will return zero rows. */
|
||||
if( iLangVal<0 ) iLangVal = 0;
|
||||
}
|
||||
pCsr->iLangid = iLangVal;
|
||||
|
||||
rc = sqlite3Fts3SegReaderCursor(pFts3, iLangVal, 0, FTS3_SEGCURSOR_ALL,
|
||||
pCsr->filter.zTerm, pCsr->filter.nTerm, 0, isScan, &pCsr->csr
|
||||
);
|
||||
if( rc==SQLITE_OK ){
|
||||
rc = sqlite3Fts3SegReaderStart(pFts3, &pCsr->csr, &pCsr->filter);
|
||||
}
|
||||
|
||||
if( rc==SQLITE_OK ) rc = fts3auxNextMethod(pCursor);
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** xEof - Return true if the cursor is at EOF, or false otherwise.
|
||||
*/
|
||||
static int fts3auxEofMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3auxCursor *pCsr = (Fts3auxCursor *)pCursor;
|
||||
return pCsr->isEof;
|
||||
}
|
||||
|
||||
/*
|
||||
** xColumn - Return a column value.
|
||||
*/
|
||||
static int fts3auxColumnMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* Cursor to retrieve value from */
|
||||
sqlite3_context *pCtx, /* Context for sqlite3_result_xxx() calls */
|
||||
int iCol /* Index of column to read value from */
|
||||
){
|
||||
Fts3auxCursor *p = (Fts3auxCursor *)pCursor;
|
||||
|
||||
assert( p->isEof==0 );
|
||||
switch( iCol ){
|
||||
case 0: /* term */
|
||||
sqlite3_result_text(pCtx, p->csr.zTerm, p->csr.nTerm, SQLITE_TRANSIENT);
|
||||
break;
|
||||
|
||||
case 1: /* col */
|
||||
if( p->iCol ){
|
||||
sqlite3_result_int(pCtx, p->iCol-1);
|
||||
}else{
|
||||
sqlite3_result_text(pCtx, "*", -1, SQLITE_STATIC);
|
||||
}
|
||||
break;
|
||||
|
||||
case 2: /* documents */
|
||||
sqlite3_result_int64(pCtx, p->aStat[p->iCol].nDoc);
|
||||
break;
|
||||
|
||||
case 3: /* occurrences */
|
||||
sqlite3_result_int64(pCtx, p->aStat[p->iCol].nOcc);
|
||||
break;
|
||||
|
||||
default: /* languageid */
|
||||
assert( iCol==4 );
|
||||
sqlite3_result_int(pCtx, p->iLangid);
|
||||
break;
|
||||
}
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xRowid - Return the current rowid for the cursor.
|
||||
*/
|
||||
static int fts3auxRowidMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* Cursor to retrieve value from */
|
||||
sqlite_int64 *pRowid /* OUT: Rowid value */
|
||||
){
|
||||
Fts3auxCursor *pCsr = (Fts3auxCursor *)pCursor;
|
||||
*pRowid = pCsr->iRowid;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Register the fts3aux module with database connection db. Return SQLITE_OK
|
||||
** if successful or an error code if sqlite3_create_module() fails.
|
||||
*/
|
||||
int sqlite3Fts3InitAux(sqlite3 *db){
|
||||
static const sqlite3_module fts3aux_module = {
|
||||
0, /* iVersion */
|
||||
fts3auxConnectMethod, /* xCreate */
|
||||
fts3auxConnectMethod, /* xConnect */
|
||||
fts3auxBestIndexMethod, /* xBestIndex */
|
||||
fts3auxDisconnectMethod, /* xDisconnect */
|
||||
fts3auxDisconnectMethod, /* xDestroy */
|
||||
fts3auxOpenMethod, /* xOpen */
|
||||
fts3auxCloseMethod, /* xClose */
|
||||
fts3auxFilterMethod, /* xFilter */
|
||||
fts3auxNextMethod, /* xNext */
|
||||
fts3auxEofMethod, /* xEof */
|
||||
fts3auxColumnMethod, /* xColumn */
|
||||
fts3auxRowidMethod, /* xRowid */
|
||||
0, /* xUpdate */
|
||||
0, /* xBegin */
|
||||
0, /* xSync */
|
||||
0, /* xCommit */
|
||||
0, /* xRollback */
|
||||
0, /* xFindFunction */
|
||||
0, /* xRename */
|
||||
0, /* xSavepoint */
|
||||
0, /* xRelease */
|
||||
0 /* xRollbackTo */
|
||||
};
|
||||
int rc; /* Return code */
|
||||
|
||||
rc = sqlite3_create_module(db, "fts4aux", &fts3aux_module, 0);
|
||||
return rc;
|
||||
}
|
||||
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
1309
query_classifier/qc_sqlite/sqlite-src-3110100/ext/fts3/fts3_expr.c
Normal file
1309
query_classifier/qc_sqlite/sqlite-src-3110100/ext/fts3/fts3_expr.c
Normal file
File diff suppressed because it is too large
Load Diff
@ -0,0 +1,383 @@
|
||||
/*
|
||||
** 2001 September 22
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
*************************************************************************
|
||||
** This is the implementation of generic hash-tables used in SQLite.
|
||||
** We've modified it slightly to serve as a standalone hash table
|
||||
** implementation for the full-text indexing module.
|
||||
*/
|
||||
|
||||
/*
|
||||
** The code in this file is only compiled if:
|
||||
**
|
||||
** * The FTS3 module is being built as an extension
|
||||
** (in which case SQLITE_CORE is not defined), or
|
||||
**
|
||||
** * The FTS3 module is being built into the core of
|
||||
** SQLite (in which case SQLITE_ENABLE_FTS3 is defined).
|
||||
*/
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
|
||||
#include <assert.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
#include "fts3_hash.h"
|
||||
|
||||
/*
|
||||
** Malloc and Free functions
|
||||
*/
|
||||
static void *fts3HashMalloc(int n){
|
||||
void *p = sqlite3_malloc(n);
|
||||
if( p ){
|
||||
memset(p, 0, n);
|
||||
}
|
||||
return p;
|
||||
}
|
||||
static void fts3HashFree(void *p){
|
||||
sqlite3_free(p);
|
||||
}
|
||||
|
||||
/* Turn bulk memory into a hash table object by initializing the
|
||||
** fields of the Hash structure.
|
||||
**
|
||||
** "pNew" is a pointer to the hash table that is to be initialized.
|
||||
** keyClass is one of the constants
|
||||
** FTS3_HASH_BINARY or FTS3_HASH_STRING. The value of keyClass
|
||||
** determines what kind of key the hash table will use. "copyKey" is
|
||||
** true if the hash table should make its own private copy of keys and
|
||||
** false if it should just use the supplied pointer.
|
||||
*/
|
||||
void sqlite3Fts3HashInit(Fts3Hash *pNew, char keyClass, char copyKey){
|
||||
assert( pNew!=0 );
|
||||
assert( keyClass>=FTS3_HASH_STRING && keyClass<=FTS3_HASH_BINARY );
|
||||
pNew->keyClass = keyClass;
|
||||
pNew->copyKey = copyKey;
|
||||
pNew->first = 0;
|
||||
pNew->count = 0;
|
||||
pNew->htsize = 0;
|
||||
pNew->ht = 0;
|
||||
}
|
||||
|
||||
/* Remove all entries from a hash table. Reclaim all memory.
|
||||
** Call this routine to delete a hash table or to reset a hash table
|
||||
** to the empty state.
|
||||
*/
|
||||
void sqlite3Fts3HashClear(Fts3Hash *pH){
|
||||
Fts3HashElem *elem; /* For looping over all elements of the table */
|
||||
|
||||
assert( pH!=0 );
|
||||
elem = pH->first;
|
||||
pH->first = 0;
|
||||
fts3HashFree(pH->ht);
|
||||
pH->ht = 0;
|
||||
pH->htsize = 0;
|
||||
while( elem ){
|
||||
Fts3HashElem *next_elem = elem->next;
|
||||
if( pH->copyKey && elem->pKey ){
|
||||
fts3HashFree(elem->pKey);
|
||||
}
|
||||
fts3HashFree(elem);
|
||||
elem = next_elem;
|
||||
}
|
||||
pH->count = 0;
|
||||
}
|
||||
|
||||
/*
|
||||
** Hash and comparison functions when the mode is FTS3_HASH_STRING
|
||||
*/
|
||||
static int fts3StrHash(const void *pKey, int nKey){
|
||||
const char *z = (const char *)pKey;
|
||||
unsigned h = 0;
|
||||
if( nKey<=0 ) nKey = (int) strlen(z);
|
||||
while( nKey > 0 ){
|
||||
h = (h<<3) ^ h ^ *z++;
|
||||
nKey--;
|
||||
}
|
||||
return (int)(h & 0x7fffffff);
|
||||
}
|
||||
static int fts3StrCompare(const void *pKey1, int n1, const void *pKey2, int n2){
|
||||
if( n1!=n2 ) return 1;
|
||||
return strncmp((const char*)pKey1,(const char*)pKey2,n1);
|
||||
}
|
||||
|
||||
/*
|
||||
** Hash and comparison functions when the mode is FTS3_HASH_BINARY
|
||||
*/
|
||||
static int fts3BinHash(const void *pKey, int nKey){
|
||||
int h = 0;
|
||||
const char *z = (const char *)pKey;
|
||||
while( nKey-- > 0 ){
|
||||
h = (h<<3) ^ h ^ *(z++);
|
||||
}
|
||||
return h & 0x7fffffff;
|
||||
}
|
||||
static int fts3BinCompare(const void *pKey1, int n1, const void *pKey2, int n2){
|
||||
if( n1!=n2 ) return 1;
|
||||
return memcmp(pKey1,pKey2,n1);
|
||||
}
|
||||
|
||||
/*
|
||||
** Return a pointer to the appropriate hash function given the key class.
|
||||
**
|
||||
** The C syntax in this function definition may be unfamilar to some
|
||||
** programmers, so we provide the following additional explanation:
|
||||
**
|
||||
** The name of the function is "ftsHashFunction". The function takes a
|
||||
** single parameter "keyClass". The return value of ftsHashFunction()
|
||||
** is a pointer to another function. Specifically, the return value
|
||||
** of ftsHashFunction() is a pointer to a function that takes two parameters
|
||||
** with types "const void*" and "int" and returns an "int".
|
||||
*/
|
||||
static int (*ftsHashFunction(int keyClass))(const void*,int){
|
||||
if( keyClass==FTS3_HASH_STRING ){
|
||||
return &fts3StrHash;
|
||||
}else{
|
||||
assert( keyClass==FTS3_HASH_BINARY );
|
||||
return &fts3BinHash;
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
** Return a pointer to the appropriate hash function given the key class.
|
||||
**
|
||||
** For help in interpreted the obscure C code in the function definition,
|
||||
** see the header comment on the previous function.
|
||||
*/
|
||||
static int (*ftsCompareFunction(int keyClass))(const void*,int,const void*,int){
|
||||
if( keyClass==FTS3_HASH_STRING ){
|
||||
return &fts3StrCompare;
|
||||
}else{
|
||||
assert( keyClass==FTS3_HASH_BINARY );
|
||||
return &fts3BinCompare;
|
||||
}
|
||||
}
|
||||
|
||||
/* Link an element into the hash table
|
||||
*/
|
||||
static void fts3HashInsertElement(
|
||||
Fts3Hash *pH, /* The complete hash table */
|
||||
struct _fts3ht *pEntry, /* The entry into which pNew is inserted */
|
||||
Fts3HashElem *pNew /* The element to be inserted */
|
||||
){
|
||||
Fts3HashElem *pHead; /* First element already in pEntry */
|
||||
pHead = pEntry->chain;
|
||||
if( pHead ){
|
||||
pNew->next = pHead;
|
||||
pNew->prev = pHead->prev;
|
||||
if( pHead->prev ){ pHead->prev->next = pNew; }
|
||||
else { pH->first = pNew; }
|
||||
pHead->prev = pNew;
|
||||
}else{
|
||||
pNew->next = pH->first;
|
||||
if( pH->first ){ pH->first->prev = pNew; }
|
||||
pNew->prev = 0;
|
||||
pH->first = pNew;
|
||||
}
|
||||
pEntry->count++;
|
||||
pEntry->chain = pNew;
|
||||
}
|
||||
|
||||
|
||||
/* Resize the hash table so that it cantains "new_size" buckets.
|
||||
** "new_size" must be a power of 2. The hash table might fail
|
||||
** to resize if sqliteMalloc() fails.
|
||||
**
|
||||
** Return non-zero if a memory allocation error occurs.
|
||||
*/
|
||||
static int fts3Rehash(Fts3Hash *pH, int new_size){
|
||||
struct _fts3ht *new_ht; /* The new hash table */
|
||||
Fts3HashElem *elem, *next_elem; /* For looping over existing elements */
|
||||
int (*xHash)(const void*,int); /* The hash function */
|
||||
|
||||
assert( (new_size & (new_size-1))==0 );
|
||||
new_ht = (struct _fts3ht *)fts3HashMalloc( new_size*sizeof(struct _fts3ht) );
|
||||
if( new_ht==0 ) return 1;
|
||||
fts3HashFree(pH->ht);
|
||||
pH->ht = new_ht;
|
||||
pH->htsize = new_size;
|
||||
xHash = ftsHashFunction(pH->keyClass);
|
||||
for(elem=pH->first, pH->first=0; elem; elem = next_elem){
|
||||
int h = (*xHash)(elem->pKey, elem->nKey) & (new_size-1);
|
||||
next_elem = elem->next;
|
||||
fts3HashInsertElement(pH, &new_ht[h], elem);
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
/* This function (for internal use only) locates an element in an
|
||||
** hash table that matches the given key. The hash for this key has
|
||||
** already been computed and is passed as the 4th parameter.
|
||||
*/
|
||||
static Fts3HashElem *fts3FindElementByHash(
|
||||
const Fts3Hash *pH, /* The pH to be searched */
|
||||
const void *pKey, /* The key we are searching for */
|
||||
int nKey,
|
||||
int h /* The hash for this key. */
|
||||
){
|
||||
Fts3HashElem *elem; /* Used to loop thru the element list */
|
||||
int count; /* Number of elements left to test */
|
||||
int (*xCompare)(const void*,int,const void*,int); /* comparison function */
|
||||
|
||||
if( pH->ht ){
|
||||
struct _fts3ht *pEntry = &pH->ht[h];
|
||||
elem = pEntry->chain;
|
||||
count = pEntry->count;
|
||||
xCompare = ftsCompareFunction(pH->keyClass);
|
||||
while( count-- && elem ){
|
||||
if( (*xCompare)(elem->pKey,elem->nKey,pKey,nKey)==0 ){
|
||||
return elem;
|
||||
}
|
||||
elem = elem->next;
|
||||
}
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
/* Remove a single entry from the hash table given a pointer to that
|
||||
** element and a hash on the element's key.
|
||||
*/
|
||||
static void fts3RemoveElementByHash(
|
||||
Fts3Hash *pH, /* The pH containing "elem" */
|
||||
Fts3HashElem* elem, /* The element to be removed from the pH */
|
||||
int h /* Hash value for the element */
|
||||
){
|
||||
struct _fts3ht *pEntry;
|
||||
if( elem->prev ){
|
||||
elem->prev->next = elem->next;
|
||||
}else{
|
||||
pH->first = elem->next;
|
||||
}
|
||||
if( elem->next ){
|
||||
elem->next->prev = elem->prev;
|
||||
}
|
||||
pEntry = &pH->ht[h];
|
||||
if( pEntry->chain==elem ){
|
||||
pEntry->chain = elem->next;
|
||||
}
|
||||
pEntry->count--;
|
||||
if( pEntry->count<=0 ){
|
||||
pEntry->chain = 0;
|
||||
}
|
||||
if( pH->copyKey && elem->pKey ){
|
||||
fts3HashFree(elem->pKey);
|
||||
}
|
||||
fts3HashFree( elem );
|
||||
pH->count--;
|
||||
if( pH->count<=0 ){
|
||||
assert( pH->first==0 );
|
||||
assert( pH->count==0 );
|
||||
fts3HashClear(pH);
|
||||
}
|
||||
}
|
||||
|
||||
Fts3HashElem *sqlite3Fts3HashFindElem(
|
||||
const Fts3Hash *pH,
|
||||
const void *pKey,
|
||||
int nKey
|
||||
){
|
||||
int h; /* A hash on key */
|
||||
int (*xHash)(const void*,int); /* The hash function */
|
||||
|
||||
if( pH==0 || pH->ht==0 ) return 0;
|
||||
xHash = ftsHashFunction(pH->keyClass);
|
||||
assert( xHash!=0 );
|
||||
h = (*xHash)(pKey,nKey);
|
||||
assert( (pH->htsize & (pH->htsize-1))==0 );
|
||||
return fts3FindElementByHash(pH,pKey,nKey, h & (pH->htsize-1));
|
||||
}
|
||||
|
||||
/*
|
||||
** Attempt to locate an element of the hash table pH with a key
|
||||
** that matches pKey,nKey. Return the data for this element if it is
|
||||
** found, or NULL if there is no match.
|
||||
*/
|
||||
void *sqlite3Fts3HashFind(const Fts3Hash *pH, const void *pKey, int nKey){
|
||||
Fts3HashElem *pElem; /* The element that matches key (if any) */
|
||||
|
||||
pElem = sqlite3Fts3HashFindElem(pH, pKey, nKey);
|
||||
return pElem ? pElem->data : 0;
|
||||
}
|
||||
|
||||
/* Insert an element into the hash table pH. The key is pKey,nKey
|
||||
** and the data is "data".
|
||||
**
|
||||
** If no element exists with a matching key, then a new
|
||||
** element is created. A copy of the key is made if the copyKey
|
||||
** flag is set. NULL is returned.
|
||||
**
|
||||
** If another element already exists with the same key, then the
|
||||
** new data replaces the old data and the old data is returned.
|
||||
** The key is not copied in this instance. If a malloc fails, then
|
||||
** the new data is returned and the hash table is unchanged.
|
||||
**
|
||||
** If the "data" parameter to this function is NULL, then the
|
||||
** element corresponding to "key" is removed from the hash table.
|
||||
*/
|
||||
void *sqlite3Fts3HashInsert(
|
||||
Fts3Hash *pH, /* The hash table to insert into */
|
||||
const void *pKey, /* The key */
|
||||
int nKey, /* Number of bytes in the key */
|
||||
void *data /* The data */
|
||||
){
|
||||
int hraw; /* Raw hash value of the key */
|
||||
int h; /* the hash of the key modulo hash table size */
|
||||
Fts3HashElem *elem; /* Used to loop thru the element list */
|
||||
Fts3HashElem *new_elem; /* New element added to the pH */
|
||||
int (*xHash)(const void*,int); /* The hash function */
|
||||
|
||||
assert( pH!=0 );
|
||||
xHash = ftsHashFunction(pH->keyClass);
|
||||
assert( xHash!=0 );
|
||||
hraw = (*xHash)(pKey, nKey);
|
||||
assert( (pH->htsize & (pH->htsize-1))==0 );
|
||||
h = hraw & (pH->htsize-1);
|
||||
elem = fts3FindElementByHash(pH,pKey,nKey,h);
|
||||
if( elem ){
|
||||
void *old_data = elem->data;
|
||||
if( data==0 ){
|
||||
fts3RemoveElementByHash(pH,elem,h);
|
||||
}else{
|
||||
elem->data = data;
|
||||
}
|
||||
return old_data;
|
||||
}
|
||||
if( data==0 ) return 0;
|
||||
if( (pH->htsize==0 && fts3Rehash(pH,8))
|
||||
|| (pH->count>=pH->htsize && fts3Rehash(pH, pH->htsize*2))
|
||||
){
|
||||
pH->count = 0;
|
||||
return data;
|
||||
}
|
||||
assert( pH->htsize>0 );
|
||||
new_elem = (Fts3HashElem*)fts3HashMalloc( sizeof(Fts3HashElem) );
|
||||
if( new_elem==0 ) return data;
|
||||
if( pH->copyKey && pKey!=0 ){
|
||||
new_elem->pKey = fts3HashMalloc( nKey );
|
||||
if( new_elem->pKey==0 ){
|
||||
fts3HashFree(new_elem);
|
||||
return data;
|
||||
}
|
||||
memcpy((void*)new_elem->pKey, pKey, nKey);
|
||||
}else{
|
||||
new_elem->pKey = (void*)pKey;
|
||||
}
|
||||
new_elem->nKey = nKey;
|
||||
pH->count++;
|
||||
assert( pH->htsize>0 );
|
||||
assert( (pH->htsize & (pH->htsize-1))==0 );
|
||||
h = hraw & (pH->htsize-1);
|
||||
fts3HashInsertElement(pH, &pH->ht[h], new_elem);
|
||||
new_elem->data = data;
|
||||
return 0;
|
||||
}
|
||||
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
@ -0,0 +1,112 @@
|
||||
/*
|
||||
** 2001 September 22
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
*************************************************************************
|
||||
** This is the header file for the generic hash-table implementation
|
||||
** used in SQLite. We've modified it slightly to serve as a standalone
|
||||
** hash table implementation for the full-text indexing module.
|
||||
**
|
||||
*/
|
||||
#ifndef _FTS3_HASH_H_
|
||||
#define _FTS3_HASH_H_
|
||||
|
||||
/* Forward declarations of structures. */
|
||||
typedef struct Fts3Hash Fts3Hash;
|
||||
typedef struct Fts3HashElem Fts3HashElem;
|
||||
|
||||
/* A complete hash table is an instance of the following structure.
|
||||
** The internals of this structure are intended to be opaque -- client
|
||||
** code should not attempt to access or modify the fields of this structure
|
||||
** directly. Change this structure only by using the routines below.
|
||||
** However, many of the "procedures" and "functions" for modifying and
|
||||
** accessing this structure are really macros, so we can't really make
|
||||
** this structure opaque.
|
||||
*/
|
||||
struct Fts3Hash {
|
||||
char keyClass; /* HASH_INT, _POINTER, _STRING, _BINARY */
|
||||
char copyKey; /* True if copy of key made on insert */
|
||||
int count; /* Number of entries in this table */
|
||||
Fts3HashElem *first; /* The first element of the array */
|
||||
int htsize; /* Number of buckets in the hash table */
|
||||
struct _fts3ht { /* the hash table */
|
||||
int count; /* Number of entries with this hash */
|
||||
Fts3HashElem *chain; /* Pointer to first entry with this hash */
|
||||
} *ht;
|
||||
};
|
||||
|
||||
/* Each element in the hash table is an instance of the following
|
||||
** structure. All elements are stored on a single doubly-linked list.
|
||||
**
|
||||
** Again, this structure is intended to be opaque, but it can't really
|
||||
** be opaque because it is used by macros.
|
||||
*/
|
||||
struct Fts3HashElem {
|
||||
Fts3HashElem *next, *prev; /* Next and previous elements in the table */
|
||||
void *data; /* Data associated with this element */
|
||||
void *pKey; int nKey; /* Key associated with this element */
|
||||
};
|
||||
|
||||
/*
|
||||
** There are 2 different modes of operation for a hash table:
|
||||
**
|
||||
** FTS3_HASH_STRING pKey points to a string that is nKey bytes long
|
||||
** (including the null-terminator, if any). Case
|
||||
** is respected in comparisons.
|
||||
**
|
||||
** FTS3_HASH_BINARY pKey points to binary data nKey bytes long.
|
||||
** memcmp() is used to compare keys.
|
||||
**
|
||||
** A copy of the key is made if the copyKey parameter to fts3HashInit is 1.
|
||||
*/
|
||||
#define FTS3_HASH_STRING 1
|
||||
#define FTS3_HASH_BINARY 2
|
||||
|
||||
/*
|
||||
** Access routines. To delete, insert a NULL pointer.
|
||||
*/
|
||||
void sqlite3Fts3HashInit(Fts3Hash *pNew, char keyClass, char copyKey);
|
||||
void *sqlite3Fts3HashInsert(Fts3Hash*, const void *pKey, int nKey, void *pData);
|
||||
void *sqlite3Fts3HashFind(const Fts3Hash*, const void *pKey, int nKey);
|
||||
void sqlite3Fts3HashClear(Fts3Hash*);
|
||||
Fts3HashElem *sqlite3Fts3HashFindElem(const Fts3Hash *, const void *, int);
|
||||
|
||||
/*
|
||||
** Shorthand for the functions above
|
||||
*/
|
||||
#define fts3HashInit sqlite3Fts3HashInit
|
||||
#define fts3HashInsert sqlite3Fts3HashInsert
|
||||
#define fts3HashFind sqlite3Fts3HashFind
|
||||
#define fts3HashClear sqlite3Fts3HashClear
|
||||
#define fts3HashFindElem sqlite3Fts3HashFindElem
|
||||
|
||||
/*
|
||||
** Macros for looping over all elements of a hash table. The idiom is
|
||||
** like this:
|
||||
**
|
||||
** Fts3Hash h;
|
||||
** Fts3HashElem *p;
|
||||
** ...
|
||||
** for(p=fts3HashFirst(&h); p; p=fts3HashNext(p)){
|
||||
** SomeStructure *pData = fts3HashData(p);
|
||||
** // do something with pData
|
||||
** }
|
||||
*/
|
||||
#define fts3HashFirst(H) ((H)->first)
|
||||
#define fts3HashNext(E) ((E)->next)
|
||||
#define fts3HashData(E) ((E)->data)
|
||||
#define fts3HashKey(E) ((E)->pKey)
|
||||
#define fts3HashKeysize(E) ((E)->nKey)
|
||||
|
||||
/*
|
||||
** Number of entries in a hash table
|
||||
*/
|
||||
#define fts3HashCount(H) ((H)->count)
|
||||
|
||||
#endif /* _FTS3_HASH_H_ */
|
||||
@ -0,0 +1,262 @@
|
||||
/*
|
||||
** 2007 June 22
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
*************************************************************************
|
||||
** This file implements a tokenizer for fts3 based on the ICU library.
|
||||
*/
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
#ifdef SQLITE_ENABLE_ICU
|
||||
|
||||
#include <assert.h>
|
||||
#include <string.h>
|
||||
#include "fts3_tokenizer.h"
|
||||
|
||||
#include <unicode/ubrk.h>
|
||||
#include <unicode/ucol.h>
|
||||
#include <unicode/ustring.h>
|
||||
#include <unicode/utf16.h>
|
||||
|
||||
typedef struct IcuTokenizer IcuTokenizer;
|
||||
typedef struct IcuCursor IcuCursor;
|
||||
|
||||
struct IcuTokenizer {
|
||||
sqlite3_tokenizer base;
|
||||
char *zLocale;
|
||||
};
|
||||
|
||||
struct IcuCursor {
|
||||
sqlite3_tokenizer_cursor base;
|
||||
|
||||
UBreakIterator *pIter; /* ICU break-iterator object */
|
||||
int nChar; /* Number of UChar elements in pInput */
|
||||
UChar *aChar; /* Copy of input using utf-16 encoding */
|
||||
int *aOffset; /* Offsets of each character in utf-8 input */
|
||||
|
||||
int nBuffer;
|
||||
char *zBuffer;
|
||||
|
||||
int iToken;
|
||||
};
|
||||
|
||||
/*
|
||||
** Create a new tokenizer instance.
|
||||
*/
|
||||
static int icuCreate(
|
||||
int argc, /* Number of entries in argv[] */
|
||||
const char * const *argv, /* Tokenizer creation arguments */
|
||||
sqlite3_tokenizer **ppTokenizer /* OUT: Created tokenizer */
|
||||
){
|
||||
IcuTokenizer *p;
|
||||
int n = 0;
|
||||
|
||||
if( argc>0 ){
|
||||
n = strlen(argv[0])+1;
|
||||
}
|
||||
p = (IcuTokenizer *)sqlite3_malloc(sizeof(IcuTokenizer)+n);
|
||||
if( !p ){
|
||||
return SQLITE_NOMEM;
|
||||
}
|
||||
memset(p, 0, sizeof(IcuTokenizer));
|
||||
|
||||
if( n ){
|
||||
p->zLocale = (char *)&p[1];
|
||||
memcpy(p->zLocale, argv[0], n);
|
||||
}
|
||||
|
||||
*ppTokenizer = (sqlite3_tokenizer *)p;
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Destroy a tokenizer
|
||||
*/
|
||||
static int icuDestroy(sqlite3_tokenizer *pTokenizer){
|
||||
IcuTokenizer *p = (IcuTokenizer *)pTokenizer;
|
||||
sqlite3_free(p);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Prepare to begin tokenizing a particular string. The input
|
||||
** string to be tokenized is pInput[0..nBytes-1]. A cursor
|
||||
** used to incrementally tokenize this string is returned in
|
||||
** *ppCursor.
|
||||
*/
|
||||
static int icuOpen(
|
||||
sqlite3_tokenizer *pTokenizer, /* The tokenizer */
|
||||
const char *zInput, /* Input string */
|
||||
int nInput, /* Length of zInput in bytes */
|
||||
sqlite3_tokenizer_cursor **ppCursor /* OUT: Tokenization cursor */
|
||||
){
|
||||
IcuTokenizer *p = (IcuTokenizer *)pTokenizer;
|
||||
IcuCursor *pCsr;
|
||||
|
||||
const int32_t opt = U_FOLD_CASE_DEFAULT;
|
||||
UErrorCode status = U_ZERO_ERROR;
|
||||
int nChar;
|
||||
|
||||
UChar32 c;
|
||||
int iInput = 0;
|
||||
int iOut = 0;
|
||||
|
||||
*ppCursor = 0;
|
||||
|
||||
if( zInput==0 ){
|
||||
nInput = 0;
|
||||
zInput = "";
|
||||
}else if( nInput<0 ){
|
||||
nInput = strlen(zInput);
|
||||
}
|
||||
nChar = nInput+1;
|
||||
pCsr = (IcuCursor *)sqlite3_malloc(
|
||||
sizeof(IcuCursor) + /* IcuCursor */
|
||||
((nChar+3)&~3) * sizeof(UChar) + /* IcuCursor.aChar[] */
|
||||
(nChar+1) * sizeof(int) /* IcuCursor.aOffset[] */
|
||||
);
|
||||
if( !pCsr ){
|
||||
return SQLITE_NOMEM;
|
||||
}
|
||||
memset(pCsr, 0, sizeof(IcuCursor));
|
||||
pCsr->aChar = (UChar *)&pCsr[1];
|
||||
pCsr->aOffset = (int *)&pCsr->aChar[(nChar+3)&~3];
|
||||
|
||||
pCsr->aOffset[iOut] = iInput;
|
||||
U8_NEXT(zInput, iInput, nInput, c);
|
||||
while( c>0 ){
|
||||
int isError = 0;
|
||||
c = u_foldCase(c, opt);
|
||||
U16_APPEND(pCsr->aChar, iOut, nChar, c, isError);
|
||||
if( isError ){
|
||||
sqlite3_free(pCsr);
|
||||
return SQLITE_ERROR;
|
||||
}
|
||||
pCsr->aOffset[iOut] = iInput;
|
||||
|
||||
if( iInput<nInput ){
|
||||
U8_NEXT(zInput, iInput, nInput, c);
|
||||
}else{
|
||||
c = 0;
|
||||
}
|
||||
}
|
||||
|
||||
pCsr->pIter = ubrk_open(UBRK_WORD, p->zLocale, pCsr->aChar, iOut, &status);
|
||||
if( !U_SUCCESS(status) ){
|
||||
sqlite3_free(pCsr);
|
||||
return SQLITE_ERROR;
|
||||
}
|
||||
pCsr->nChar = iOut;
|
||||
|
||||
ubrk_first(pCsr->pIter);
|
||||
*ppCursor = (sqlite3_tokenizer_cursor *)pCsr;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Close a tokenization cursor previously opened by a call to icuOpen().
|
||||
*/
|
||||
static int icuClose(sqlite3_tokenizer_cursor *pCursor){
|
||||
IcuCursor *pCsr = (IcuCursor *)pCursor;
|
||||
ubrk_close(pCsr->pIter);
|
||||
sqlite3_free(pCsr->zBuffer);
|
||||
sqlite3_free(pCsr);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Extract the next token from a tokenization cursor.
|
||||
*/
|
||||
static int icuNext(
|
||||
sqlite3_tokenizer_cursor *pCursor, /* Cursor returned by simpleOpen */
|
||||
const char **ppToken, /* OUT: *ppToken is the token text */
|
||||
int *pnBytes, /* OUT: Number of bytes in token */
|
||||
int *piStartOffset, /* OUT: Starting offset of token */
|
||||
int *piEndOffset, /* OUT: Ending offset of token */
|
||||
int *piPosition /* OUT: Position integer of token */
|
||||
){
|
||||
IcuCursor *pCsr = (IcuCursor *)pCursor;
|
||||
|
||||
int iStart = 0;
|
||||
int iEnd = 0;
|
||||
int nByte = 0;
|
||||
|
||||
while( iStart==iEnd ){
|
||||
UChar32 c;
|
||||
|
||||
iStart = ubrk_current(pCsr->pIter);
|
||||
iEnd = ubrk_next(pCsr->pIter);
|
||||
if( iEnd==UBRK_DONE ){
|
||||
return SQLITE_DONE;
|
||||
}
|
||||
|
||||
while( iStart<iEnd ){
|
||||
int iWhite = iStart;
|
||||
U16_NEXT(pCsr->aChar, iWhite, pCsr->nChar, c);
|
||||
if( u_isspace(c) ){
|
||||
iStart = iWhite;
|
||||
}else{
|
||||
break;
|
||||
}
|
||||
}
|
||||
assert(iStart<=iEnd);
|
||||
}
|
||||
|
||||
do {
|
||||
UErrorCode status = U_ZERO_ERROR;
|
||||
if( nByte ){
|
||||
char *zNew = sqlite3_realloc(pCsr->zBuffer, nByte);
|
||||
if( !zNew ){
|
||||
return SQLITE_NOMEM;
|
||||
}
|
||||
pCsr->zBuffer = zNew;
|
||||
pCsr->nBuffer = nByte;
|
||||
}
|
||||
|
||||
u_strToUTF8(
|
||||
pCsr->zBuffer, pCsr->nBuffer, &nByte, /* Output vars */
|
||||
&pCsr->aChar[iStart], iEnd-iStart, /* Input vars */
|
||||
&status /* Output success/failure */
|
||||
);
|
||||
} while( nByte>pCsr->nBuffer );
|
||||
|
||||
*ppToken = pCsr->zBuffer;
|
||||
*pnBytes = nByte;
|
||||
*piStartOffset = pCsr->aOffset[iStart];
|
||||
*piEndOffset = pCsr->aOffset[iEnd];
|
||||
*piPosition = pCsr->iToken++;
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** The set of routines that implement the simple tokenizer
|
||||
*/
|
||||
static const sqlite3_tokenizer_module icuTokenizerModule = {
|
||||
0, /* iVersion */
|
||||
icuCreate, /* xCreate */
|
||||
icuDestroy, /* xCreate */
|
||||
icuOpen, /* xOpen */
|
||||
icuClose, /* xClose */
|
||||
icuNext, /* xNext */
|
||||
0, /* xLanguageid */
|
||||
};
|
||||
|
||||
/*
|
||||
** Set *ppModule to point at the implementation of the ICU tokenizer.
|
||||
*/
|
||||
void sqlite3Fts3IcuTokenizerModule(
|
||||
sqlite3_tokenizer_module const**ppModule
|
||||
){
|
||||
*ppModule = &icuTokenizerModule;
|
||||
}
|
||||
|
||||
#endif /* defined(SQLITE_ENABLE_ICU) */
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
@ -0,0 +1,662 @@
|
||||
/*
|
||||
** 2006 September 30
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
*************************************************************************
|
||||
** Implementation of the full-text-search tokenizer that implements
|
||||
** a Porter stemmer.
|
||||
*/
|
||||
|
||||
/*
|
||||
** The code in this file is only compiled if:
|
||||
**
|
||||
** * The FTS3 module is being built as an extension
|
||||
** (in which case SQLITE_CORE is not defined), or
|
||||
**
|
||||
** * The FTS3 module is being built into the core of
|
||||
** SQLite (in which case SQLITE_ENABLE_FTS3 is defined).
|
||||
*/
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
|
||||
#include <assert.h>
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
|
||||
#include "fts3_tokenizer.h"
|
||||
|
||||
/*
|
||||
** Class derived from sqlite3_tokenizer
|
||||
*/
|
||||
typedef struct porter_tokenizer {
|
||||
sqlite3_tokenizer base; /* Base class */
|
||||
} porter_tokenizer;
|
||||
|
||||
/*
|
||||
** Class derived from sqlite3_tokenizer_cursor
|
||||
*/
|
||||
typedef struct porter_tokenizer_cursor {
|
||||
sqlite3_tokenizer_cursor base;
|
||||
const char *zInput; /* input we are tokenizing */
|
||||
int nInput; /* size of the input */
|
||||
int iOffset; /* current position in zInput */
|
||||
int iToken; /* index of next token to be returned */
|
||||
char *zToken; /* storage for current token */
|
||||
int nAllocated; /* space allocated to zToken buffer */
|
||||
} porter_tokenizer_cursor;
|
||||
|
||||
|
||||
/*
|
||||
** Create a new tokenizer instance.
|
||||
*/
|
||||
static int porterCreate(
|
||||
int argc, const char * const *argv,
|
||||
sqlite3_tokenizer **ppTokenizer
|
||||
){
|
||||
porter_tokenizer *t;
|
||||
|
||||
UNUSED_PARAMETER(argc);
|
||||
UNUSED_PARAMETER(argv);
|
||||
|
||||
t = (porter_tokenizer *) sqlite3_malloc(sizeof(*t));
|
||||
if( t==NULL ) return SQLITE_NOMEM;
|
||||
memset(t, 0, sizeof(*t));
|
||||
*ppTokenizer = &t->base;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Destroy a tokenizer
|
||||
*/
|
||||
static int porterDestroy(sqlite3_tokenizer *pTokenizer){
|
||||
sqlite3_free(pTokenizer);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Prepare to begin tokenizing a particular string. The input
|
||||
** string to be tokenized is zInput[0..nInput-1]. A cursor
|
||||
** used to incrementally tokenize this string is returned in
|
||||
** *ppCursor.
|
||||
*/
|
||||
static int porterOpen(
|
||||
sqlite3_tokenizer *pTokenizer, /* The tokenizer */
|
||||
const char *zInput, int nInput, /* String to be tokenized */
|
||||
sqlite3_tokenizer_cursor **ppCursor /* OUT: Tokenization cursor */
|
||||
){
|
||||
porter_tokenizer_cursor *c;
|
||||
|
||||
UNUSED_PARAMETER(pTokenizer);
|
||||
|
||||
c = (porter_tokenizer_cursor *) sqlite3_malloc(sizeof(*c));
|
||||
if( c==NULL ) return SQLITE_NOMEM;
|
||||
|
||||
c->zInput = zInput;
|
||||
if( zInput==0 ){
|
||||
c->nInput = 0;
|
||||
}else if( nInput<0 ){
|
||||
c->nInput = (int)strlen(zInput);
|
||||
}else{
|
||||
c->nInput = nInput;
|
||||
}
|
||||
c->iOffset = 0; /* start tokenizing at the beginning */
|
||||
c->iToken = 0;
|
||||
c->zToken = NULL; /* no space allocated, yet. */
|
||||
c->nAllocated = 0;
|
||||
|
||||
*ppCursor = &c->base;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Close a tokenization cursor previously opened by a call to
|
||||
** porterOpen() above.
|
||||
*/
|
||||
static int porterClose(sqlite3_tokenizer_cursor *pCursor){
|
||||
porter_tokenizer_cursor *c = (porter_tokenizer_cursor *) pCursor;
|
||||
sqlite3_free(c->zToken);
|
||||
sqlite3_free(c);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
/*
|
||||
** Vowel or consonant
|
||||
*/
|
||||
static const char cType[] = {
|
||||
0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0,
|
||||
1, 1, 1, 2, 1
|
||||
};
|
||||
|
||||
/*
|
||||
** isConsonant() and isVowel() determine if their first character in
|
||||
** the string they point to is a consonant or a vowel, according
|
||||
** to Porter ruls.
|
||||
**
|
||||
** A consonate is any letter other than 'a', 'e', 'i', 'o', or 'u'.
|
||||
** 'Y' is a consonant unless it follows another consonant,
|
||||
** in which case it is a vowel.
|
||||
**
|
||||
** In these routine, the letters are in reverse order. So the 'y' rule
|
||||
** is that 'y' is a consonant unless it is followed by another
|
||||
** consonent.
|
||||
*/
|
||||
static int isVowel(const char*);
|
||||
static int isConsonant(const char *z){
|
||||
int j;
|
||||
char x = *z;
|
||||
if( x==0 ) return 0;
|
||||
assert( x>='a' && x<='z' );
|
||||
j = cType[x-'a'];
|
||||
if( j<2 ) return j;
|
||||
return z[1]==0 || isVowel(z + 1);
|
||||
}
|
||||
static int isVowel(const char *z){
|
||||
int j;
|
||||
char x = *z;
|
||||
if( x==0 ) return 0;
|
||||
assert( x>='a' && x<='z' );
|
||||
j = cType[x-'a'];
|
||||
if( j<2 ) return 1-j;
|
||||
return isConsonant(z + 1);
|
||||
}
|
||||
|
||||
/*
|
||||
** Let any sequence of one or more vowels be represented by V and let
|
||||
** C be sequence of one or more consonants. Then every word can be
|
||||
** represented as:
|
||||
**
|
||||
** [C] (VC){m} [V]
|
||||
**
|
||||
** In prose: A word is an optional consonant followed by zero or
|
||||
** vowel-consonant pairs followed by an optional vowel. "m" is the
|
||||
** number of vowel consonant pairs. This routine computes the value
|
||||
** of m for the first i bytes of a word.
|
||||
**
|
||||
** Return true if the m-value for z is 1 or more. In other words,
|
||||
** return true if z contains at least one vowel that is followed
|
||||
** by a consonant.
|
||||
**
|
||||
** In this routine z[] is in reverse order. So we are really looking
|
||||
** for an instance of a consonant followed by a vowel.
|
||||
*/
|
||||
static int m_gt_0(const char *z){
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
return *z!=0;
|
||||
}
|
||||
|
||||
/* Like mgt0 above except we are looking for a value of m which is
|
||||
** exactly 1
|
||||
*/
|
||||
static int m_eq_1(const char *z){
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 1;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
return *z==0;
|
||||
}
|
||||
|
||||
/* Like mgt0 above except we are looking for a value of m>1 instead
|
||||
** or m>0
|
||||
*/
|
||||
static int m_gt_1(const char *z){
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isVowel(z) ){ z++; }
|
||||
if( *z==0 ) return 0;
|
||||
while( isConsonant(z) ){ z++; }
|
||||
return *z!=0;
|
||||
}
|
||||
|
||||
/*
|
||||
** Return TRUE if there is a vowel anywhere within z[0..n-1]
|
||||
*/
|
||||
static int hasVowel(const char *z){
|
||||
while( isConsonant(z) ){ z++; }
|
||||
return *z!=0;
|
||||
}
|
||||
|
||||
/*
|
||||
** Return TRUE if the word ends in a double consonant.
|
||||
**
|
||||
** The text is reversed here. So we are really looking at
|
||||
** the first two characters of z[].
|
||||
*/
|
||||
static int doubleConsonant(const char *z){
|
||||
return isConsonant(z) && z[0]==z[1];
|
||||
}
|
||||
|
||||
/*
|
||||
** Return TRUE if the word ends with three letters which
|
||||
** are consonant-vowel-consonent and where the final consonant
|
||||
** is not 'w', 'x', or 'y'.
|
||||
**
|
||||
** The word is reversed here. So we are really checking the
|
||||
** first three letters and the first one cannot be in [wxy].
|
||||
*/
|
||||
static int star_oh(const char *z){
|
||||
return
|
||||
isConsonant(z) &&
|
||||
z[0]!='w' && z[0]!='x' && z[0]!='y' &&
|
||||
isVowel(z+1) &&
|
||||
isConsonant(z+2);
|
||||
}
|
||||
|
||||
/*
|
||||
** If the word ends with zFrom and xCond() is true for the stem
|
||||
** of the word that preceeds the zFrom ending, then change the
|
||||
** ending to zTo.
|
||||
**
|
||||
** The input word *pz and zFrom are both in reverse order. zTo
|
||||
** is in normal order.
|
||||
**
|
||||
** Return TRUE if zFrom matches. Return FALSE if zFrom does not
|
||||
** match. Not that TRUE is returned even if xCond() fails and
|
||||
** no substitution occurs.
|
||||
*/
|
||||
static int stem(
|
||||
char **pz, /* The word being stemmed (Reversed) */
|
||||
const char *zFrom, /* If the ending matches this... (Reversed) */
|
||||
const char *zTo, /* ... change the ending to this (not reversed) */
|
||||
int (*xCond)(const char*) /* Condition that must be true */
|
||||
){
|
||||
char *z = *pz;
|
||||
while( *zFrom && *zFrom==*z ){ z++; zFrom++; }
|
||||
if( *zFrom!=0 ) return 0;
|
||||
if( xCond && !xCond(z) ) return 1;
|
||||
while( *zTo ){
|
||||
*(--z) = *(zTo++);
|
||||
}
|
||||
*pz = z;
|
||||
return 1;
|
||||
}
|
||||
|
||||
/*
|
||||
** This is the fallback stemmer used when the porter stemmer is
|
||||
** inappropriate. The input word is copied into the output with
|
||||
** US-ASCII case folding. If the input word is too long (more
|
||||
** than 20 bytes if it contains no digits or more than 6 bytes if
|
||||
** it contains digits) then word is truncated to 20 or 6 bytes
|
||||
** by taking 10 or 3 bytes from the beginning and end.
|
||||
*/
|
||||
static void copy_stemmer(const char *zIn, int nIn, char *zOut, int *pnOut){
|
||||
int i, mx, j;
|
||||
int hasDigit = 0;
|
||||
for(i=0; i<nIn; i++){
|
||||
char c = zIn[i];
|
||||
if( c>='A' && c<='Z' ){
|
||||
zOut[i] = c - 'A' + 'a';
|
||||
}else{
|
||||
if( c>='0' && c<='9' ) hasDigit = 1;
|
||||
zOut[i] = c;
|
||||
}
|
||||
}
|
||||
mx = hasDigit ? 3 : 10;
|
||||
if( nIn>mx*2 ){
|
||||
for(j=mx, i=nIn-mx; i<nIn; i++, j++){
|
||||
zOut[j] = zOut[i];
|
||||
}
|
||||
i = j;
|
||||
}
|
||||
zOut[i] = 0;
|
||||
*pnOut = i;
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Stem the input word zIn[0..nIn-1]. Store the output in zOut.
|
||||
** zOut is at least big enough to hold nIn bytes. Write the actual
|
||||
** size of the output word (exclusive of the '\0' terminator) into *pnOut.
|
||||
**
|
||||
** Any upper-case characters in the US-ASCII character set ([A-Z])
|
||||
** are converted to lower case. Upper-case UTF characters are
|
||||
** unchanged.
|
||||
**
|
||||
** Words that are longer than about 20 bytes are stemmed by retaining
|
||||
** a few bytes from the beginning and the end of the word. If the
|
||||
** word contains digits, 3 bytes are taken from the beginning and
|
||||
** 3 bytes from the end. For long words without digits, 10 bytes
|
||||
** are taken from each end. US-ASCII case folding still applies.
|
||||
**
|
||||
** If the input word contains not digits but does characters not
|
||||
** in [a-zA-Z] then no stemming is attempted and this routine just
|
||||
** copies the input into the input into the output with US-ASCII
|
||||
** case folding.
|
||||
**
|
||||
** Stemming never increases the length of the word. So there is
|
||||
** no chance of overflowing the zOut buffer.
|
||||
*/
|
||||
static void porter_stemmer(const char *zIn, int nIn, char *zOut, int *pnOut){
|
||||
int i, j;
|
||||
char zReverse[28];
|
||||
char *z, *z2;
|
||||
if( nIn<3 || nIn>=(int)sizeof(zReverse)-7 ){
|
||||
/* The word is too big or too small for the porter stemmer.
|
||||
** Fallback to the copy stemmer */
|
||||
copy_stemmer(zIn, nIn, zOut, pnOut);
|
||||
return;
|
||||
}
|
||||
for(i=0, j=sizeof(zReverse)-6; i<nIn; i++, j--){
|
||||
char c = zIn[i];
|
||||
if( c>='A' && c<='Z' ){
|
||||
zReverse[j] = c + 'a' - 'A';
|
||||
}else if( c>='a' && c<='z' ){
|
||||
zReverse[j] = c;
|
||||
}else{
|
||||
/* The use of a character not in [a-zA-Z] means that we fallback
|
||||
** to the copy stemmer */
|
||||
copy_stemmer(zIn, nIn, zOut, pnOut);
|
||||
return;
|
||||
}
|
||||
}
|
||||
memset(&zReverse[sizeof(zReverse)-5], 0, 5);
|
||||
z = &zReverse[j+1];
|
||||
|
||||
|
||||
/* Step 1a */
|
||||
if( z[0]=='s' ){
|
||||
if(
|
||||
!stem(&z, "sess", "ss", 0) &&
|
||||
!stem(&z, "sei", "i", 0) &&
|
||||
!stem(&z, "ss", "ss", 0)
|
||||
){
|
||||
z++;
|
||||
}
|
||||
}
|
||||
|
||||
/* Step 1b */
|
||||
z2 = z;
|
||||
if( stem(&z, "dee", "ee", m_gt_0) ){
|
||||
/* Do nothing. The work was all in the test */
|
||||
}else if(
|
||||
(stem(&z, "gni", "", hasVowel) || stem(&z, "de", "", hasVowel))
|
||||
&& z!=z2
|
||||
){
|
||||
if( stem(&z, "ta", "ate", 0) ||
|
||||
stem(&z, "lb", "ble", 0) ||
|
||||
stem(&z, "zi", "ize", 0) ){
|
||||
/* Do nothing. The work was all in the test */
|
||||
}else if( doubleConsonant(z) && (*z!='l' && *z!='s' && *z!='z') ){
|
||||
z++;
|
||||
}else if( m_eq_1(z) && star_oh(z) ){
|
||||
*(--z) = 'e';
|
||||
}
|
||||
}
|
||||
|
||||
/* Step 1c */
|
||||
if( z[0]=='y' && hasVowel(z+1) ){
|
||||
z[0] = 'i';
|
||||
}
|
||||
|
||||
/* Step 2 */
|
||||
switch( z[1] ){
|
||||
case 'a':
|
||||
if( !stem(&z, "lanoita", "ate", m_gt_0) ){
|
||||
stem(&z, "lanoit", "tion", m_gt_0);
|
||||
}
|
||||
break;
|
||||
case 'c':
|
||||
if( !stem(&z, "icne", "ence", m_gt_0) ){
|
||||
stem(&z, "icna", "ance", m_gt_0);
|
||||
}
|
||||
break;
|
||||
case 'e':
|
||||
stem(&z, "rezi", "ize", m_gt_0);
|
||||
break;
|
||||
case 'g':
|
||||
stem(&z, "igol", "log", m_gt_0);
|
||||
break;
|
||||
case 'l':
|
||||
if( !stem(&z, "ilb", "ble", m_gt_0)
|
||||
&& !stem(&z, "illa", "al", m_gt_0)
|
||||
&& !stem(&z, "iltne", "ent", m_gt_0)
|
||||
&& !stem(&z, "ile", "e", m_gt_0)
|
||||
){
|
||||
stem(&z, "ilsuo", "ous", m_gt_0);
|
||||
}
|
||||
break;
|
||||
case 'o':
|
||||
if( !stem(&z, "noitazi", "ize", m_gt_0)
|
||||
&& !stem(&z, "noita", "ate", m_gt_0)
|
||||
){
|
||||
stem(&z, "rota", "ate", m_gt_0);
|
||||
}
|
||||
break;
|
||||
case 's':
|
||||
if( !stem(&z, "msila", "al", m_gt_0)
|
||||
&& !stem(&z, "ssenevi", "ive", m_gt_0)
|
||||
&& !stem(&z, "ssenluf", "ful", m_gt_0)
|
||||
){
|
||||
stem(&z, "ssensuo", "ous", m_gt_0);
|
||||
}
|
||||
break;
|
||||
case 't':
|
||||
if( !stem(&z, "itila", "al", m_gt_0)
|
||||
&& !stem(&z, "itivi", "ive", m_gt_0)
|
||||
){
|
||||
stem(&z, "itilib", "ble", m_gt_0);
|
||||
}
|
||||
break;
|
||||
}
|
||||
|
||||
/* Step 3 */
|
||||
switch( z[0] ){
|
||||
case 'e':
|
||||
if( !stem(&z, "etaci", "ic", m_gt_0)
|
||||
&& !stem(&z, "evita", "", m_gt_0)
|
||||
){
|
||||
stem(&z, "ezila", "al", m_gt_0);
|
||||
}
|
||||
break;
|
||||
case 'i':
|
||||
stem(&z, "itici", "ic", m_gt_0);
|
||||
break;
|
||||
case 'l':
|
||||
if( !stem(&z, "laci", "ic", m_gt_0) ){
|
||||
stem(&z, "luf", "", m_gt_0);
|
||||
}
|
||||
break;
|
||||
case 's':
|
||||
stem(&z, "ssen", "", m_gt_0);
|
||||
break;
|
||||
}
|
||||
|
||||
/* Step 4 */
|
||||
switch( z[1] ){
|
||||
case 'a':
|
||||
if( z[0]=='l' && m_gt_1(z+2) ){
|
||||
z += 2;
|
||||
}
|
||||
break;
|
||||
case 'c':
|
||||
if( z[0]=='e' && z[2]=='n' && (z[3]=='a' || z[3]=='e') && m_gt_1(z+4) ){
|
||||
z += 4;
|
||||
}
|
||||
break;
|
||||
case 'e':
|
||||
if( z[0]=='r' && m_gt_1(z+2) ){
|
||||
z += 2;
|
||||
}
|
||||
break;
|
||||
case 'i':
|
||||
if( z[0]=='c' && m_gt_1(z+2) ){
|
||||
z += 2;
|
||||
}
|
||||
break;
|
||||
case 'l':
|
||||
if( z[0]=='e' && z[2]=='b' && (z[3]=='a' || z[3]=='i') && m_gt_1(z+4) ){
|
||||
z += 4;
|
||||
}
|
||||
break;
|
||||
case 'n':
|
||||
if( z[0]=='t' ){
|
||||
if( z[2]=='a' ){
|
||||
if( m_gt_1(z+3) ){
|
||||
z += 3;
|
||||
}
|
||||
}else if( z[2]=='e' ){
|
||||
if( !stem(&z, "tneme", "", m_gt_1)
|
||||
&& !stem(&z, "tnem", "", m_gt_1)
|
||||
){
|
||||
stem(&z, "tne", "", m_gt_1);
|
||||
}
|
||||
}
|
||||
}
|
||||
break;
|
||||
case 'o':
|
||||
if( z[0]=='u' ){
|
||||
if( m_gt_1(z+2) ){
|
||||
z += 2;
|
||||
}
|
||||
}else if( z[3]=='s' || z[3]=='t' ){
|
||||
stem(&z, "noi", "", m_gt_1);
|
||||
}
|
||||
break;
|
||||
case 's':
|
||||
if( z[0]=='m' && z[2]=='i' && m_gt_1(z+3) ){
|
||||
z += 3;
|
||||
}
|
||||
break;
|
||||
case 't':
|
||||
if( !stem(&z, "eta", "", m_gt_1) ){
|
||||
stem(&z, "iti", "", m_gt_1);
|
||||
}
|
||||
break;
|
||||
case 'u':
|
||||
if( z[0]=='s' && z[2]=='o' && m_gt_1(z+3) ){
|
||||
z += 3;
|
||||
}
|
||||
break;
|
||||
case 'v':
|
||||
case 'z':
|
||||
if( z[0]=='e' && z[2]=='i' && m_gt_1(z+3) ){
|
||||
z += 3;
|
||||
}
|
||||
break;
|
||||
}
|
||||
|
||||
/* Step 5a */
|
||||
if( z[0]=='e' ){
|
||||
if( m_gt_1(z+1) ){
|
||||
z++;
|
||||
}else if( m_eq_1(z+1) && !star_oh(z+1) ){
|
||||
z++;
|
||||
}
|
||||
}
|
||||
|
||||
/* Step 5b */
|
||||
if( m_gt_1(z) && z[0]=='l' && z[1]=='l' ){
|
||||
z++;
|
||||
}
|
||||
|
||||
/* z[] is now the stemmed word in reverse order. Flip it back
|
||||
** around into forward order and return.
|
||||
*/
|
||||
*pnOut = i = (int)strlen(z);
|
||||
zOut[i] = 0;
|
||||
while( *z ){
|
||||
zOut[--i] = *(z++);
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
** Characters that can be part of a token. We assume any character
|
||||
** whose value is greater than 0x80 (any UTF character) can be
|
||||
** part of a token. In other words, delimiters all must have
|
||||
** values of 0x7f or lower.
|
||||
*/
|
||||
static const char porterIdChar[] = {
|
||||
/* x0 x1 x2 x3 x4 x5 x6 x7 x8 x9 xA xB xC xD xE xF */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, /* 3x */
|
||||
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 4x */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, /* 5x */
|
||||
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 6x */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, /* 7x */
|
||||
};
|
||||
#define isDelim(C) (((ch=C)&0x80)==0 && (ch<0x30 || !porterIdChar[ch-0x30]))
|
||||
|
||||
/*
|
||||
** Extract the next token from a tokenization cursor. The cursor must
|
||||
** have been opened by a prior call to porterOpen().
|
||||
*/
|
||||
static int porterNext(
|
||||
sqlite3_tokenizer_cursor *pCursor, /* Cursor returned by porterOpen */
|
||||
const char **pzToken, /* OUT: *pzToken is the token text */
|
||||
int *pnBytes, /* OUT: Number of bytes in token */
|
||||
int *piStartOffset, /* OUT: Starting offset of token */
|
||||
int *piEndOffset, /* OUT: Ending offset of token */
|
||||
int *piPosition /* OUT: Position integer of token */
|
||||
){
|
||||
porter_tokenizer_cursor *c = (porter_tokenizer_cursor *) pCursor;
|
||||
const char *z = c->zInput;
|
||||
|
||||
while( c->iOffset<c->nInput ){
|
||||
int iStartOffset, ch;
|
||||
|
||||
/* Scan past delimiter characters */
|
||||
while( c->iOffset<c->nInput && isDelim(z[c->iOffset]) ){
|
||||
c->iOffset++;
|
||||
}
|
||||
|
||||
/* Count non-delimiter characters. */
|
||||
iStartOffset = c->iOffset;
|
||||
while( c->iOffset<c->nInput && !isDelim(z[c->iOffset]) ){
|
||||
c->iOffset++;
|
||||
}
|
||||
|
||||
if( c->iOffset>iStartOffset ){
|
||||
int n = c->iOffset-iStartOffset;
|
||||
if( n>c->nAllocated ){
|
||||
char *pNew;
|
||||
c->nAllocated = n+20;
|
||||
pNew = sqlite3_realloc(c->zToken, c->nAllocated);
|
||||
if( !pNew ) return SQLITE_NOMEM;
|
||||
c->zToken = pNew;
|
||||
}
|
||||
porter_stemmer(&z[iStartOffset], n, c->zToken, pnBytes);
|
||||
*pzToken = c->zToken;
|
||||
*piStartOffset = iStartOffset;
|
||||
*piEndOffset = c->iOffset;
|
||||
*piPosition = c->iToken++;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
}
|
||||
return SQLITE_DONE;
|
||||
}
|
||||
|
||||
/*
|
||||
** The set of routines that implement the porter-stemmer tokenizer
|
||||
*/
|
||||
static const sqlite3_tokenizer_module porterTokenizerModule = {
|
||||
0,
|
||||
porterCreate,
|
||||
porterDestroy,
|
||||
porterOpen,
|
||||
porterClose,
|
||||
porterNext,
|
||||
0
|
||||
};
|
||||
|
||||
/*
|
||||
** Allocate a new porter tokenizer. Return a pointer to the new
|
||||
** tokenizer in *ppModule
|
||||
*/
|
||||
void sqlite3Fts3PorterTokenizerModule(
|
||||
sqlite3_tokenizer_module const**ppModule
|
||||
){
|
||||
*ppModule = &porterTokenizerModule;
|
||||
}
|
||||
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
File diff suppressed because it is too large
Load Diff
@ -0,0 +1,373 @@
|
||||
/*
|
||||
** 2011 Jan 27
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
** This file is not part of the production FTS code. It is only used for
|
||||
** testing. It contains a virtual table implementation that provides direct
|
||||
** access to the full-text index of an FTS table.
|
||||
*/
|
||||
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
#ifdef SQLITE_TEST
|
||||
|
||||
#include <string.h>
|
||||
#include <assert.h>
|
||||
#include <stdlib.h>
|
||||
|
||||
typedef struct Fts3termTable Fts3termTable;
|
||||
typedef struct Fts3termCursor Fts3termCursor;
|
||||
|
||||
struct Fts3termTable {
|
||||
sqlite3_vtab base; /* Base class used by SQLite core */
|
||||
int iIndex; /* Index for Fts3Table.aIndex[] */
|
||||
Fts3Table *pFts3Tab;
|
||||
};
|
||||
|
||||
struct Fts3termCursor {
|
||||
sqlite3_vtab_cursor base; /* Base class used by SQLite core */
|
||||
Fts3MultiSegReader csr; /* Must be right after "base" */
|
||||
Fts3SegFilter filter;
|
||||
|
||||
int isEof; /* True if cursor is at EOF */
|
||||
char *pNext;
|
||||
|
||||
sqlite3_int64 iRowid; /* Current 'rowid' value */
|
||||
sqlite3_int64 iDocid; /* Current 'docid' value */
|
||||
int iCol; /* Current 'col' value */
|
||||
int iPos; /* Current 'pos' value */
|
||||
};
|
||||
|
||||
/*
|
||||
** Schema of the terms table.
|
||||
*/
|
||||
#define FTS3_TERMS_SCHEMA "CREATE TABLE x(term, docid, col, pos)"
|
||||
|
||||
/*
|
||||
** This function does all the work for both the xConnect and xCreate methods.
|
||||
** These tables have no persistent representation of their own, so xConnect
|
||||
** and xCreate are identical operations.
|
||||
*/
|
||||
static int fts3termConnectMethod(
|
||||
sqlite3 *db, /* Database connection */
|
||||
void *pCtx, /* Non-zero for an fts4prefix table */
|
||||
int argc, /* Number of elements in argv array */
|
||||
const char * const *argv, /* xCreate/xConnect argument array */
|
||||
sqlite3_vtab **ppVtab, /* OUT: New sqlite3_vtab object */
|
||||
char **pzErr /* OUT: sqlite3_malloc'd error message */
|
||||
){
|
||||
char const *zDb; /* Name of database (e.g. "main") */
|
||||
char const *zFts3; /* Name of fts3 table */
|
||||
int nDb; /* Result of strlen(zDb) */
|
||||
int nFts3; /* Result of strlen(zFts3) */
|
||||
int nByte; /* Bytes of space to allocate here */
|
||||
int rc; /* value returned by declare_vtab() */
|
||||
Fts3termTable *p; /* Virtual table object to return */
|
||||
int iIndex = 0;
|
||||
|
||||
UNUSED_PARAMETER(pCtx);
|
||||
if( argc==5 ){
|
||||
iIndex = atoi(argv[4]);
|
||||
argc--;
|
||||
}
|
||||
|
||||
/* The user should specify a single argument - the name of an fts3 table. */
|
||||
if( argc!=4 ){
|
||||
sqlite3Fts3ErrMsg(pzErr,
|
||||
"wrong number of arguments to fts4term constructor"
|
||||
);
|
||||
return SQLITE_ERROR;
|
||||
}
|
||||
|
||||
zDb = argv[1];
|
||||
nDb = (int)strlen(zDb);
|
||||
zFts3 = argv[3];
|
||||
nFts3 = (int)strlen(zFts3);
|
||||
|
||||
rc = sqlite3_declare_vtab(db, FTS3_TERMS_SCHEMA);
|
||||
if( rc!=SQLITE_OK ) return rc;
|
||||
|
||||
nByte = sizeof(Fts3termTable) + sizeof(Fts3Table) + nDb + nFts3 + 2;
|
||||
p = (Fts3termTable *)sqlite3_malloc(nByte);
|
||||
if( !p ) return SQLITE_NOMEM;
|
||||
memset(p, 0, nByte);
|
||||
|
||||
p->pFts3Tab = (Fts3Table *)&p[1];
|
||||
p->pFts3Tab->zDb = (char *)&p->pFts3Tab[1];
|
||||
p->pFts3Tab->zName = &p->pFts3Tab->zDb[nDb+1];
|
||||
p->pFts3Tab->db = db;
|
||||
p->pFts3Tab->nIndex = iIndex+1;
|
||||
p->iIndex = iIndex;
|
||||
|
||||
memcpy((char *)p->pFts3Tab->zDb, zDb, nDb);
|
||||
memcpy((char *)p->pFts3Tab->zName, zFts3, nFts3);
|
||||
sqlite3Fts3Dequote((char *)p->pFts3Tab->zName);
|
||||
|
||||
*ppVtab = (sqlite3_vtab *)p;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** This function does the work for both the xDisconnect and xDestroy methods.
|
||||
** These tables have no persistent representation of their own, so xDisconnect
|
||||
** and xDestroy are identical operations.
|
||||
*/
|
||||
static int fts3termDisconnectMethod(sqlite3_vtab *pVtab){
|
||||
Fts3termTable *p = (Fts3termTable *)pVtab;
|
||||
Fts3Table *pFts3 = p->pFts3Tab;
|
||||
int i;
|
||||
|
||||
/* Free any prepared statements held */
|
||||
for(i=0; i<SizeofArray(pFts3->aStmt); i++){
|
||||
sqlite3_finalize(pFts3->aStmt[i]);
|
||||
}
|
||||
sqlite3_free(pFts3->zSegmentsTbl);
|
||||
sqlite3_free(p);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
#define FTS4AUX_EQ_CONSTRAINT 1
|
||||
#define FTS4AUX_GE_CONSTRAINT 2
|
||||
#define FTS4AUX_LE_CONSTRAINT 4
|
||||
|
||||
/*
|
||||
** xBestIndex - Analyze a WHERE and ORDER BY clause.
|
||||
*/
|
||||
static int fts3termBestIndexMethod(
|
||||
sqlite3_vtab *pVTab,
|
||||
sqlite3_index_info *pInfo
|
||||
){
|
||||
UNUSED_PARAMETER(pVTab);
|
||||
|
||||
/* This vtab naturally does "ORDER BY term, docid, col, pos". */
|
||||
if( pInfo->nOrderBy ){
|
||||
int i;
|
||||
for(i=0; i<pInfo->nOrderBy; i++){
|
||||
if( pInfo->aOrderBy[i].iColumn!=i || pInfo->aOrderBy[i].desc ) break;
|
||||
}
|
||||
if( i==pInfo->nOrderBy ){
|
||||
pInfo->orderByConsumed = 1;
|
||||
}
|
||||
}
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xOpen - Open a cursor.
|
||||
*/
|
||||
static int fts3termOpenMethod(sqlite3_vtab *pVTab, sqlite3_vtab_cursor **ppCsr){
|
||||
Fts3termCursor *pCsr; /* Pointer to cursor object to return */
|
||||
|
||||
UNUSED_PARAMETER(pVTab);
|
||||
|
||||
pCsr = (Fts3termCursor *)sqlite3_malloc(sizeof(Fts3termCursor));
|
||||
if( !pCsr ) return SQLITE_NOMEM;
|
||||
memset(pCsr, 0, sizeof(Fts3termCursor));
|
||||
|
||||
*ppCsr = (sqlite3_vtab_cursor *)pCsr;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xClose - Close a cursor.
|
||||
*/
|
||||
static int fts3termCloseMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3Table *pFts3 = ((Fts3termTable *)pCursor->pVtab)->pFts3Tab;
|
||||
Fts3termCursor *pCsr = (Fts3termCursor *)pCursor;
|
||||
|
||||
sqlite3Fts3SegmentsClose(pFts3);
|
||||
sqlite3Fts3SegReaderFinish(&pCsr->csr);
|
||||
sqlite3_free(pCsr);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xNext - Advance the cursor to the next row, if any.
|
||||
*/
|
||||
static int fts3termNextMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3termCursor *pCsr = (Fts3termCursor *)pCursor;
|
||||
Fts3Table *pFts3 = ((Fts3termTable *)pCursor->pVtab)->pFts3Tab;
|
||||
int rc;
|
||||
sqlite3_int64 v;
|
||||
|
||||
/* Increment our pretend rowid value. */
|
||||
pCsr->iRowid++;
|
||||
|
||||
/* Advance to the next term in the full-text index. */
|
||||
if( pCsr->csr.aDoclist==0
|
||||
|| pCsr->pNext>=&pCsr->csr.aDoclist[pCsr->csr.nDoclist-1]
|
||||
){
|
||||
rc = sqlite3Fts3SegReaderStep(pFts3, &pCsr->csr);
|
||||
if( rc!=SQLITE_ROW ){
|
||||
pCsr->isEof = 1;
|
||||
return rc;
|
||||
}
|
||||
|
||||
pCsr->iCol = 0;
|
||||
pCsr->iPos = 0;
|
||||
pCsr->iDocid = 0;
|
||||
pCsr->pNext = pCsr->csr.aDoclist;
|
||||
|
||||
/* Read docid */
|
||||
pCsr->pNext += sqlite3Fts3GetVarint(pCsr->pNext, &pCsr->iDocid);
|
||||
}
|
||||
|
||||
pCsr->pNext += sqlite3Fts3GetVarint(pCsr->pNext, &v);
|
||||
if( v==0 ){
|
||||
pCsr->pNext += sqlite3Fts3GetVarint(pCsr->pNext, &v);
|
||||
pCsr->iDocid += v;
|
||||
pCsr->pNext += sqlite3Fts3GetVarint(pCsr->pNext, &v);
|
||||
pCsr->iCol = 0;
|
||||
pCsr->iPos = 0;
|
||||
}
|
||||
|
||||
if( v==1 ){
|
||||
pCsr->pNext += sqlite3Fts3GetVarint(pCsr->pNext, &v);
|
||||
pCsr->iCol += (int)v;
|
||||
pCsr->iPos = 0;
|
||||
pCsr->pNext += sqlite3Fts3GetVarint(pCsr->pNext, &v);
|
||||
}
|
||||
|
||||
pCsr->iPos += (int)(v - 2);
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xFilter - Initialize a cursor to point at the start of its data.
|
||||
*/
|
||||
static int fts3termFilterMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* The cursor used for this query */
|
||||
int idxNum, /* Strategy index */
|
||||
const char *idxStr, /* Unused */
|
||||
int nVal, /* Number of elements in apVal */
|
||||
sqlite3_value **apVal /* Arguments for the indexing scheme */
|
||||
){
|
||||
Fts3termCursor *pCsr = (Fts3termCursor *)pCursor;
|
||||
Fts3termTable *p = (Fts3termTable *)pCursor->pVtab;
|
||||
Fts3Table *pFts3 = p->pFts3Tab;
|
||||
int rc;
|
||||
|
||||
UNUSED_PARAMETER(nVal);
|
||||
UNUSED_PARAMETER(idxNum);
|
||||
UNUSED_PARAMETER(idxStr);
|
||||
UNUSED_PARAMETER(apVal);
|
||||
|
||||
assert( idxStr==0 && idxNum==0 );
|
||||
|
||||
/* In case this cursor is being reused, close and zero it. */
|
||||
testcase(pCsr->filter.zTerm);
|
||||
sqlite3Fts3SegReaderFinish(&pCsr->csr);
|
||||
memset(&pCsr->csr, 0, ((u8*)&pCsr[1]) - (u8*)&pCsr->csr);
|
||||
|
||||
pCsr->filter.flags = FTS3_SEGMENT_REQUIRE_POS|FTS3_SEGMENT_IGNORE_EMPTY;
|
||||
pCsr->filter.flags |= FTS3_SEGMENT_SCAN;
|
||||
|
||||
rc = sqlite3Fts3SegReaderCursor(pFts3, 0, p->iIndex, FTS3_SEGCURSOR_ALL,
|
||||
pCsr->filter.zTerm, pCsr->filter.nTerm, 0, 1, &pCsr->csr
|
||||
);
|
||||
if( rc==SQLITE_OK ){
|
||||
rc = sqlite3Fts3SegReaderStart(pFts3, &pCsr->csr, &pCsr->filter);
|
||||
}
|
||||
if( rc==SQLITE_OK ){
|
||||
rc = fts3termNextMethod(pCursor);
|
||||
}
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** xEof - Return true if the cursor is at EOF, or false otherwise.
|
||||
*/
|
||||
static int fts3termEofMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3termCursor *pCsr = (Fts3termCursor *)pCursor;
|
||||
return pCsr->isEof;
|
||||
}
|
||||
|
||||
/*
|
||||
** xColumn - Return a column value.
|
||||
*/
|
||||
static int fts3termColumnMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* Cursor to retrieve value from */
|
||||
sqlite3_context *pCtx, /* Context for sqlite3_result_xxx() calls */
|
||||
int iCol /* Index of column to read value from */
|
||||
){
|
||||
Fts3termCursor *p = (Fts3termCursor *)pCursor;
|
||||
|
||||
assert( iCol>=0 && iCol<=3 );
|
||||
switch( iCol ){
|
||||
case 0:
|
||||
sqlite3_result_text(pCtx, p->csr.zTerm, p->csr.nTerm, SQLITE_TRANSIENT);
|
||||
break;
|
||||
case 1:
|
||||
sqlite3_result_int64(pCtx, p->iDocid);
|
||||
break;
|
||||
case 2:
|
||||
sqlite3_result_int64(pCtx, p->iCol);
|
||||
break;
|
||||
default:
|
||||
sqlite3_result_int64(pCtx, p->iPos);
|
||||
break;
|
||||
}
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xRowid - Return the current rowid for the cursor.
|
||||
*/
|
||||
static int fts3termRowidMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* Cursor to retrieve value from */
|
||||
sqlite_int64 *pRowid /* OUT: Rowid value */
|
||||
){
|
||||
Fts3termCursor *pCsr = (Fts3termCursor *)pCursor;
|
||||
*pRowid = pCsr->iRowid;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Register the fts3term module with database connection db. Return SQLITE_OK
|
||||
** if successful or an error code if sqlite3_create_module() fails.
|
||||
*/
|
||||
int sqlite3Fts3InitTerm(sqlite3 *db){
|
||||
static const sqlite3_module fts3term_module = {
|
||||
0, /* iVersion */
|
||||
fts3termConnectMethod, /* xCreate */
|
||||
fts3termConnectMethod, /* xConnect */
|
||||
fts3termBestIndexMethod, /* xBestIndex */
|
||||
fts3termDisconnectMethod, /* xDisconnect */
|
||||
fts3termDisconnectMethod, /* xDestroy */
|
||||
fts3termOpenMethod, /* xOpen */
|
||||
fts3termCloseMethod, /* xClose */
|
||||
fts3termFilterMethod, /* xFilter */
|
||||
fts3termNextMethod, /* xNext */
|
||||
fts3termEofMethod, /* xEof */
|
||||
fts3termColumnMethod, /* xColumn */
|
||||
fts3termRowidMethod, /* xRowid */
|
||||
0, /* xUpdate */
|
||||
0, /* xBegin */
|
||||
0, /* xSync */
|
||||
0, /* xCommit */
|
||||
0, /* xRollback */
|
||||
0, /* xFindFunction */
|
||||
0, /* xRename */
|
||||
0, /* xSavepoint */
|
||||
0, /* xRelease */
|
||||
0 /* xRollbackTo */
|
||||
};
|
||||
int rc; /* Return code */
|
||||
|
||||
rc = sqlite3_create_module(db, "fts4term", &fts3term_module, 0);
|
||||
return rc;
|
||||
}
|
||||
|
||||
#endif
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
@ -0,0 +1,584 @@
|
||||
/*
|
||||
** 2011 Jun 13
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
** This file is not part of the production FTS code. It is only used for
|
||||
** testing. It contains a Tcl command that can be used to test if a document
|
||||
** matches an FTS NEAR expression.
|
||||
**
|
||||
** As of March 2012, it also contains a version 1 tokenizer used for testing
|
||||
** that the sqlite3_tokenizer_module.xLanguage() method is invoked correctly.
|
||||
*/
|
||||
|
||||
#include <tcl.h>
|
||||
#include <string.h>
|
||||
#include <assert.h>
|
||||
|
||||
#if defined(SQLITE_TEST)
|
||||
#if defined(SQLITE_ENABLE_FTS3) || defined(SQLITE_ENABLE_FTS4)
|
||||
|
||||
/* Required so that the "ifdef SQLITE_ENABLE_FTS3" below works */
|
||||
#include "fts3Int.h"
|
||||
|
||||
#define NM_MAX_TOKEN 12
|
||||
|
||||
typedef struct NearPhrase NearPhrase;
|
||||
typedef struct NearDocument NearDocument;
|
||||
typedef struct NearToken NearToken;
|
||||
|
||||
struct NearDocument {
|
||||
int nToken; /* Length of token in bytes */
|
||||
NearToken *aToken; /* Token array */
|
||||
};
|
||||
|
||||
struct NearToken {
|
||||
int n; /* Length of token in bytes */
|
||||
const char *z; /* Pointer to token string */
|
||||
};
|
||||
|
||||
struct NearPhrase {
|
||||
int nNear; /* Preceding NEAR value */
|
||||
int nToken; /* Number of tokens in this phrase */
|
||||
NearToken aToken[NM_MAX_TOKEN]; /* Array of tokens in this phrase */
|
||||
};
|
||||
|
||||
static int nm_phrase_match(
|
||||
NearPhrase *p,
|
||||
NearToken *aToken
|
||||
){
|
||||
int ii;
|
||||
|
||||
for(ii=0; ii<p->nToken; ii++){
|
||||
NearToken *pToken = &p->aToken[ii];
|
||||
if( pToken->n>0 && pToken->z[pToken->n-1]=='*' ){
|
||||
if( aToken[ii].n<(pToken->n-1) ) return 0;
|
||||
if( memcmp(aToken[ii].z, pToken->z, pToken->n-1) ) return 0;
|
||||
}else{
|
||||
if( aToken[ii].n!=pToken->n ) return 0;
|
||||
if( memcmp(aToken[ii].z, pToken->z, pToken->n) ) return 0;
|
||||
}
|
||||
}
|
||||
|
||||
return 1;
|
||||
}
|
||||
|
||||
static int nm_near_chain(
|
||||
int iDir, /* Direction to iterate through aPhrase[] */
|
||||
NearDocument *pDoc, /* Document to match against */
|
||||
int iPos, /* Position at which iPhrase was found */
|
||||
int nPhrase, /* Size of phrase array */
|
||||
NearPhrase *aPhrase, /* Phrase array */
|
||||
int iPhrase /* Index of phrase found */
|
||||
){
|
||||
int iStart;
|
||||
int iStop;
|
||||
int ii;
|
||||
int nNear;
|
||||
int iPhrase2;
|
||||
NearPhrase *p;
|
||||
NearPhrase *pPrev;
|
||||
|
||||
assert( iDir==1 || iDir==-1 );
|
||||
|
||||
if( iDir==1 ){
|
||||
if( (iPhrase+1)==nPhrase ) return 1;
|
||||
nNear = aPhrase[iPhrase+1].nNear;
|
||||
}else{
|
||||
if( iPhrase==0 ) return 1;
|
||||
nNear = aPhrase[iPhrase].nNear;
|
||||
}
|
||||
pPrev = &aPhrase[iPhrase];
|
||||
iPhrase2 = iPhrase+iDir;
|
||||
p = &aPhrase[iPhrase2];
|
||||
|
||||
iStart = iPos - nNear - p->nToken;
|
||||
iStop = iPos + nNear + pPrev->nToken;
|
||||
|
||||
if( iStart<0 ) iStart = 0;
|
||||
if( iStop > pDoc->nToken - p->nToken ) iStop = pDoc->nToken - p->nToken;
|
||||
|
||||
for(ii=iStart; ii<=iStop; ii++){
|
||||
if( nm_phrase_match(p, &pDoc->aToken[ii]) ){
|
||||
if( nm_near_chain(iDir, pDoc, ii, nPhrase, aPhrase, iPhrase2) ) return 1;
|
||||
}
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
static int nm_match_count(
|
||||
NearDocument *pDoc, /* Document to match against */
|
||||
int nPhrase, /* Size of phrase array */
|
||||
NearPhrase *aPhrase, /* Phrase array */
|
||||
int iPhrase /* Index of phrase to count matches for */
|
||||
){
|
||||
int nOcc = 0;
|
||||
int ii;
|
||||
NearPhrase *p = &aPhrase[iPhrase];
|
||||
|
||||
for(ii=0; ii<(pDoc->nToken + 1 - p->nToken); ii++){
|
||||
if( nm_phrase_match(p, &pDoc->aToken[ii]) ){
|
||||
/* Test forward NEAR chain (i>iPhrase) */
|
||||
if( 0==nm_near_chain(1, pDoc, ii, nPhrase, aPhrase, iPhrase) ) continue;
|
||||
|
||||
/* Test reverse NEAR chain (i<iPhrase) */
|
||||
if( 0==nm_near_chain(-1, pDoc, ii, nPhrase, aPhrase, iPhrase) ) continue;
|
||||
|
||||
/* This is a real match. Increment the counter. */
|
||||
nOcc++;
|
||||
}
|
||||
}
|
||||
|
||||
return nOcc;
|
||||
}
|
||||
|
||||
/*
|
||||
** Tclcmd: fts3_near_match DOCUMENT EXPR ?OPTIONS?
|
||||
*/
|
||||
static int fts3_near_match_cmd(
|
||||
ClientData clientData,
|
||||
Tcl_Interp *interp,
|
||||
int objc,
|
||||
Tcl_Obj *CONST objv[]
|
||||
){
|
||||
int nTotal = 0;
|
||||
int rc;
|
||||
int ii;
|
||||
int nPhrase;
|
||||
NearPhrase *aPhrase = 0;
|
||||
NearDocument doc = {0, 0};
|
||||
Tcl_Obj **apDocToken;
|
||||
Tcl_Obj *pRet;
|
||||
Tcl_Obj *pPhrasecount = 0;
|
||||
|
||||
Tcl_Obj **apExprToken;
|
||||
int nExprToken;
|
||||
|
||||
UNUSED_PARAMETER(clientData);
|
||||
|
||||
/* Must have 3 or more arguments. */
|
||||
if( objc<3 || (objc%2)==0 ){
|
||||
Tcl_WrongNumArgs(interp, 1, objv, "DOCUMENT EXPR ?OPTION VALUE?...");
|
||||
rc = TCL_ERROR;
|
||||
goto near_match_out;
|
||||
}
|
||||
|
||||
for(ii=3; ii<objc; ii+=2){
|
||||
enum NM_enum { NM_PHRASECOUNTS };
|
||||
struct TestnmSubcmd {
|
||||
char *zName;
|
||||
enum NM_enum eOpt;
|
||||
} aOpt[] = {
|
||||
{ "-phrasecountvar", NM_PHRASECOUNTS },
|
||||
{ 0, 0 }
|
||||
};
|
||||
int iOpt;
|
||||
if( Tcl_GetIndexFromObjStruct(
|
||||
interp, objv[ii], aOpt, sizeof(aOpt[0]), "option", 0, &iOpt)
|
||||
){
|
||||
return TCL_ERROR;
|
||||
}
|
||||
|
||||
switch( aOpt[iOpt].eOpt ){
|
||||
case NM_PHRASECOUNTS:
|
||||
pPhrasecount = objv[ii+1];
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
rc = Tcl_ListObjGetElements(interp, objv[1], &doc.nToken, &apDocToken);
|
||||
if( rc!=TCL_OK ) goto near_match_out;
|
||||
doc.aToken = (NearToken *)ckalloc(doc.nToken*sizeof(NearToken));
|
||||
for(ii=0; ii<doc.nToken; ii++){
|
||||
doc.aToken[ii].z = Tcl_GetStringFromObj(apDocToken[ii], &doc.aToken[ii].n);
|
||||
}
|
||||
|
||||
rc = Tcl_ListObjGetElements(interp, objv[2], &nExprToken, &apExprToken);
|
||||
if( rc!=TCL_OK ) goto near_match_out;
|
||||
|
||||
nPhrase = (nExprToken + 1) / 2;
|
||||
aPhrase = (NearPhrase *)ckalloc(nPhrase * sizeof(NearPhrase));
|
||||
memset(aPhrase, 0, nPhrase * sizeof(NearPhrase));
|
||||
for(ii=0; ii<nPhrase; ii++){
|
||||
Tcl_Obj *pPhrase = apExprToken[ii*2];
|
||||
Tcl_Obj **apToken;
|
||||
int nToken;
|
||||
int jj;
|
||||
|
||||
rc = Tcl_ListObjGetElements(interp, pPhrase, &nToken, &apToken);
|
||||
if( rc!=TCL_OK ) goto near_match_out;
|
||||
if( nToken>NM_MAX_TOKEN ){
|
||||
Tcl_AppendResult(interp, "Too many tokens in phrase", 0);
|
||||
rc = TCL_ERROR;
|
||||
goto near_match_out;
|
||||
}
|
||||
for(jj=0; jj<nToken; jj++){
|
||||
NearToken *pT = &aPhrase[ii].aToken[jj];
|
||||
pT->z = Tcl_GetStringFromObj(apToken[jj], &pT->n);
|
||||
}
|
||||
aPhrase[ii].nToken = nToken;
|
||||
}
|
||||
for(ii=1; ii<nPhrase; ii++){
|
||||
Tcl_Obj *pNear = apExprToken[2*ii-1];
|
||||
int nNear;
|
||||
rc = Tcl_GetIntFromObj(interp, pNear, &nNear);
|
||||
if( rc!=TCL_OK ) goto near_match_out;
|
||||
aPhrase[ii].nNear = nNear;
|
||||
}
|
||||
|
||||
pRet = Tcl_NewObj();
|
||||
Tcl_IncrRefCount(pRet);
|
||||
for(ii=0; ii<nPhrase; ii++){
|
||||
int nOcc = nm_match_count(&doc, nPhrase, aPhrase, ii);
|
||||
Tcl_ListObjAppendElement(interp, pRet, Tcl_NewIntObj(nOcc));
|
||||
nTotal += nOcc;
|
||||
}
|
||||
if( pPhrasecount ){
|
||||
Tcl_ObjSetVar2(interp, pPhrasecount, 0, pRet, 0);
|
||||
}
|
||||
Tcl_DecrRefCount(pRet);
|
||||
Tcl_SetObjResult(interp, Tcl_NewBooleanObj(nTotal>0));
|
||||
|
||||
near_match_out:
|
||||
ckfree((char *)aPhrase);
|
||||
ckfree((char *)doc.aToken);
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** Tclcmd: fts3_configure_incr_load ?CHUNKSIZE THRESHOLD?
|
||||
**
|
||||
** Normally, FTS uses hard-coded values to determine the minimum doclist
|
||||
** size eligible for incremental loading, and the size of the chunks loaded
|
||||
** when a doclist is incrementally loaded. This command allows the built-in
|
||||
** values to be overridden for testing purposes.
|
||||
**
|
||||
** If present, the first argument is the chunksize in bytes to load doclists
|
||||
** in. The second argument is the minimum doclist size in bytes to use
|
||||
** incremental loading with.
|
||||
**
|
||||
** Whether or not the arguments are present, this command returns a list of
|
||||
** two integers - the initial chunksize and threshold when the command is
|
||||
** invoked. This can be used to restore the default behavior after running
|
||||
** tests. For example:
|
||||
**
|
||||
** # Override incr-load settings for testing:
|
||||
** set cfg [fts3_configure_incr_load $new_chunksize $new_threshold]
|
||||
**
|
||||
** .... run tests ....
|
||||
**
|
||||
** # Restore initial incr-load settings:
|
||||
** eval fts3_configure_incr_load $cfg
|
||||
*/
|
||||
static int fts3_configure_incr_load_cmd(
|
||||
ClientData clientData,
|
||||
Tcl_Interp *interp,
|
||||
int objc,
|
||||
Tcl_Obj *CONST objv[]
|
||||
){
|
||||
#ifdef SQLITE_ENABLE_FTS3
|
||||
extern int test_fts3_node_chunksize;
|
||||
extern int test_fts3_node_chunk_threshold;
|
||||
Tcl_Obj *pRet;
|
||||
|
||||
if( objc!=1 && objc!=3 ){
|
||||
Tcl_WrongNumArgs(interp, 1, objv, "?CHUNKSIZE THRESHOLD?");
|
||||
return TCL_ERROR;
|
||||
}
|
||||
|
||||
pRet = Tcl_NewObj();
|
||||
Tcl_IncrRefCount(pRet);
|
||||
Tcl_ListObjAppendElement(
|
||||
interp, pRet, Tcl_NewIntObj(test_fts3_node_chunksize));
|
||||
Tcl_ListObjAppendElement(
|
||||
interp, pRet, Tcl_NewIntObj(test_fts3_node_chunk_threshold));
|
||||
|
||||
if( objc==3 ){
|
||||
int iArg1;
|
||||
int iArg2;
|
||||
if( Tcl_GetIntFromObj(interp, objv[1], &iArg1)
|
||||
|| Tcl_GetIntFromObj(interp, objv[2], &iArg2)
|
||||
){
|
||||
Tcl_DecrRefCount(pRet);
|
||||
return TCL_ERROR;
|
||||
}
|
||||
test_fts3_node_chunksize = iArg1;
|
||||
test_fts3_node_chunk_threshold = iArg2;
|
||||
}
|
||||
|
||||
Tcl_SetObjResult(interp, pRet);
|
||||
Tcl_DecrRefCount(pRet);
|
||||
#endif
|
||||
UNUSED_PARAMETER(clientData);
|
||||
return TCL_OK;
|
||||
}
|
||||
|
||||
#ifdef SQLITE_ENABLE_FTS3
|
||||
/**************************************************************************
|
||||
** Beginning of test tokenizer code.
|
||||
**
|
||||
** For language 0, this tokenizer is similar to the default 'simple'
|
||||
** tokenizer. For other languages L, the following:
|
||||
**
|
||||
** * Odd numbered languages are case-sensitive. Even numbered
|
||||
** languages are not.
|
||||
**
|
||||
** * Language ids 100 or greater are considered an error.
|
||||
**
|
||||
** The implementation assumes that the input contains only ASCII characters
|
||||
** (i.e. those that may be encoded in UTF-8 using a single byte).
|
||||
*/
|
||||
typedef struct test_tokenizer {
|
||||
sqlite3_tokenizer base;
|
||||
} test_tokenizer;
|
||||
|
||||
typedef struct test_tokenizer_cursor {
|
||||
sqlite3_tokenizer_cursor base;
|
||||
const char *aInput; /* Input being tokenized */
|
||||
int nInput; /* Size of the input in bytes */
|
||||
int iInput; /* Current offset in aInput */
|
||||
int iToken; /* Index of next token to be returned */
|
||||
char *aBuffer; /* Buffer containing current token */
|
||||
int nBuffer; /* Number of bytes allocated at pToken */
|
||||
int iLangid; /* Configured language id */
|
||||
} test_tokenizer_cursor;
|
||||
|
||||
static int testTokenizerCreate(
|
||||
int argc, const char * const *argv,
|
||||
sqlite3_tokenizer **ppTokenizer
|
||||
){
|
||||
test_tokenizer *pNew;
|
||||
UNUSED_PARAMETER(argc);
|
||||
UNUSED_PARAMETER(argv);
|
||||
|
||||
pNew = sqlite3_malloc(sizeof(test_tokenizer));
|
||||
if( !pNew ) return SQLITE_NOMEM;
|
||||
memset(pNew, 0, sizeof(test_tokenizer));
|
||||
|
||||
*ppTokenizer = (sqlite3_tokenizer *)pNew;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
static int testTokenizerDestroy(sqlite3_tokenizer *pTokenizer){
|
||||
test_tokenizer *p = (test_tokenizer *)pTokenizer;
|
||||
sqlite3_free(p);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
static int testTokenizerOpen(
|
||||
sqlite3_tokenizer *pTokenizer, /* The tokenizer */
|
||||
const char *pInput, int nBytes, /* String to be tokenized */
|
||||
sqlite3_tokenizer_cursor **ppCursor /* OUT: Tokenization cursor */
|
||||
){
|
||||
int rc = SQLITE_OK; /* Return code */
|
||||
test_tokenizer_cursor *pCsr; /* New cursor object */
|
||||
|
||||
UNUSED_PARAMETER(pTokenizer);
|
||||
|
||||
pCsr = (test_tokenizer_cursor *)sqlite3_malloc(sizeof(test_tokenizer_cursor));
|
||||
if( pCsr==0 ){
|
||||
rc = SQLITE_NOMEM;
|
||||
}else{
|
||||
memset(pCsr, 0, sizeof(test_tokenizer_cursor));
|
||||
pCsr->aInput = pInput;
|
||||
if( nBytes<0 ){
|
||||
pCsr->nInput = (int)strlen(pInput);
|
||||
}else{
|
||||
pCsr->nInput = nBytes;
|
||||
}
|
||||
}
|
||||
|
||||
*ppCursor = (sqlite3_tokenizer_cursor *)pCsr;
|
||||
return rc;
|
||||
}
|
||||
|
||||
static int testTokenizerClose(sqlite3_tokenizer_cursor *pCursor){
|
||||
test_tokenizer_cursor *pCsr = (test_tokenizer_cursor *)pCursor;
|
||||
sqlite3_free(pCsr->aBuffer);
|
||||
sqlite3_free(pCsr);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
static int testIsTokenChar(char c){
|
||||
return (c>='a' && c<='z') || (c>='A' && c<='Z');
|
||||
}
|
||||
static int testTolower(char c){
|
||||
char ret = c;
|
||||
if( ret>='A' && ret<='Z') ret = ret - ('A'-'a');
|
||||
return ret;
|
||||
}
|
||||
|
||||
static int testTokenizerNext(
|
||||
sqlite3_tokenizer_cursor *pCursor, /* Cursor returned by testTokenizerOpen */
|
||||
const char **ppToken, /* OUT: *ppToken is the token text */
|
||||
int *pnBytes, /* OUT: Number of bytes in token */
|
||||
int *piStartOffset, /* OUT: Starting offset of token */
|
||||
int *piEndOffset, /* OUT: Ending offset of token */
|
||||
int *piPosition /* OUT: Position integer of token */
|
||||
){
|
||||
test_tokenizer_cursor *pCsr = (test_tokenizer_cursor *)pCursor;
|
||||
int rc = SQLITE_OK;
|
||||
const char *p;
|
||||
const char *pEnd;
|
||||
|
||||
p = &pCsr->aInput[pCsr->iInput];
|
||||
pEnd = &pCsr->aInput[pCsr->nInput];
|
||||
|
||||
/* Skip past any white-space */
|
||||
assert( p<=pEnd );
|
||||
while( p<pEnd && testIsTokenChar(*p)==0 ) p++;
|
||||
|
||||
if( p==pEnd ){
|
||||
rc = SQLITE_DONE;
|
||||
}else{
|
||||
/* Advance to the end of the token */
|
||||
const char *pToken = p;
|
||||
int nToken;
|
||||
while( p<pEnd && testIsTokenChar(*p) ) p++;
|
||||
nToken = (int)(p-pToken);
|
||||
|
||||
/* Copy the token into the buffer */
|
||||
if( nToken>pCsr->nBuffer ){
|
||||
sqlite3_free(pCsr->aBuffer);
|
||||
pCsr->aBuffer = sqlite3_malloc(nToken);
|
||||
}
|
||||
if( pCsr->aBuffer==0 ){
|
||||
rc = SQLITE_NOMEM;
|
||||
}else{
|
||||
int i;
|
||||
|
||||
if( pCsr->iLangid & 0x00000001 ){
|
||||
for(i=0; i<nToken; i++) pCsr->aBuffer[i] = pToken[i];
|
||||
}else{
|
||||
for(i=0; i<nToken; i++) pCsr->aBuffer[i] = testTolower(pToken[i]);
|
||||
}
|
||||
pCsr->iToken++;
|
||||
pCsr->iInput = (int)(p - pCsr->aInput);
|
||||
|
||||
*ppToken = pCsr->aBuffer;
|
||||
*pnBytes = nToken;
|
||||
*piStartOffset = (int)(pToken - pCsr->aInput);
|
||||
*piEndOffset = (int)(p - pCsr->aInput);
|
||||
*piPosition = pCsr->iToken;
|
||||
}
|
||||
}
|
||||
|
||||
return rc;
|
||||
}
|
||||
|
||||
static int testTokenizerLanguage(
|
||||
sqlite3_tokenizer_cursor *pCursor,
|
||||
int iLangid
|
||||
){
|
||||
int rc = SQLITE_OK;
|
||||
test_tokenizer_cursor *pCsr = (test_tokenizer_cursor *)pCursor;
|
||||
pCsr->iLangid = iLangid;
|
||||
if( pCsr->iLangid>=100 ){
|
||||
rc = SQLITE_ERROR;
|
||||
}
|
||||
return rc;
|
||||
}
|
||||
#endif
|
||||
|
||||
static int fts3_test_tokenizer_cmd(
|
||||
ClientData clientData,
|
||||
Tcl_Interp *interp,
|
||||
int objc,
|
||||
Tcl_Obj *CONST objv[]
|
||||
){
|
||||
#ifdef SQLITE_ENABLE_FTS3
|
||||
static const sqlite3_tokenizer_module testTokenizerModule = {
|
||||
1,
|
||||
testTokenizerCreate,
|
||||
testTokenizerDestroy,
|
||||
testTokenizerOpen,
|
||||
testTokenizerClose,
|
||||
testTokenizerNext,
|
||||
testTokenizerLanguage
|
||||
};
|
||||
const sqlite3_tokenizer_module *pPtr = &testTokenizerModule;
|
||||
if( objc!=1 ){
|
||||
Tcl_WrongNumArgs(interp, 1, objv, "");
|
||||
return TCL_ERROR;
|
||||
}
|
||||
Tcl_SetObjResult(interp, Tcl_NewByteArrayObj(
|
||||
(const unsigned char *)&pPtr, sizeof(sqlite3_tokenizer_module *)
|
||||
));
|
||||
#endif
|
||||
UNUSED_PARAMETER(clientData);
|
||||
return TCL_OK;
|
||||
}
|
||||
|
||||
static int fts3_test_varint_cmd(
|
||||
ClientData clientData,
|
||||
Tcl_Interp *interp,
|
||||
int objc,
|
||||
Tcl_Obj *CONST objv[]
|
||||
){
|
||||
#ifdef SQLITE_ENABLE_FTS3
|
||||
char aBuf[24];
|
||||
int rc;
|
||||
Tcl_WideInt w, w2;
|
||||
int nByte, nByte2;
|
||||
|
||||
if( objc!=2 ){
|
||||
Tcl_WrongNumArgs(interp, 1, objv, "INTEGER");
|
||||
return TCL_ERROR;
|
||||
}
|
||||
|
||||
rc = Tcl_GetWideIntFromObj(interp, objv[1], &w);
|
||||
if( rc!=TCL_OK ) return rc;
|
||||
|
||||
nByte = sqlite3Fts3PutVarint(aBuf, w);
|
||||
nByte2 = sqlite3Fts3GetVarint(aBuf, &w2);
|
||||
if( w!=w2 || nByte!=nByte2 ){
|
||||
char *zErr = sqlite3_mprintf("error testing %lld", w);
|
||||
Tcl_ResetResult(interp);
|
||||
Tcl_AppendResult(interp, zErr, 0);
|
||||
return TCL_ERROR;
|
||||
}
|
||||
|
||||
if( w<=2147483647 && w>=0 ){
|
||||
int i;
|
||||
nByte2 = fts3GetVarint32(aBuf, &i);
|
||||
if( (int)w!=i || nByte!=nByte2 ){
|
||||
char *zErr = sqlite3_mprintf("error testing %lld (32-bit)", w);
|
||||
Tcl_ResetResult(interp);
|
||||
Tcl_AppendResult(interp, zErr, 0);
|
||||
return TCL_ERROR;
|
||||
}
|
||||
}
|
||||
|
||||
#endif
|
||||
UNUSED_PARAMETER(clientData);
|
||||
return TCL_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** End of tokenizer code.
|
||||
**************************************************************************/
|
||||
|
||||
int Sqlitetestfts3_Init(Tcl_Interp *interp){
|
||||
Tcl_CreateObjCommand(interp, "fts3_near_match", fts3_near_match_cmd, 0, 0);
|
||||
Tcl_CreateObjCommand(interp,
|
||||
"fts3_configure_incr_load", fts3_configure_incr_load_cmd, 0, 0
|
||||
);
|
||||
Tcl_CreateObjCommand(
|
||||
interp, "fts3_test_tokenizer", fts3_test_tokenizer_cmd, 0, 0
|
||||
);
|
||||
|
||||
Tcl_CreateObjCommand(
|
||||
interp, "fts3_test_varint", fts3_test_varint_cmd, 0, 0
|
||||
);
|
||||
return TCL_OK;
|
||||
}
|
||||
#endif /* SQLITE_ENABLE_FTS3 || SQLITE_ENABLE_FTS4 */
|
||||
#endif /* ifdef SQLITE_TEST */
|
||||
@ -0,0 +1,454 @@
|
||||
/*
|
||||
** 2013 Apr 22
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
** This file contains code for the "fts3tokenize" virtual table module.
|
||||
** An fts3tokenize virtual table is created as follows:
|
||||
**
|
||||
** CREATE VIRTUAL TABLE <tbl> USING fts3tokenize(
|
||||
** <tokenizer-name>, <arg-1>, ...
|
||||
** );
|
||||
**
|
||||
** The table created has the following schema:
|
||||
**
|
||||
** CREATE TABLE <tbl>(input, token, start, end, position)
|
||||
**
|
||||
** When queried, the query must include a WHERE clause of type:
|
||||
**
|
||||
** input = <string>
|
||||
**
|
||||
** The virtual table module tokenizes this <string>, using the FTS3
|
||||
** tokenizer specified by the arguments to the CREATE VIRTUAL TABLE
|
||||
** statement and returns one row for each token in the result. With
|
||||
** fields set as follows:
|
||||
**
|
||||
** input: Always set to a copy of <string>
|
||||
** token: A token from the input.
|
||||
** start: Byte offset of the token within the input <string>.
|
||||
** end: Byte offset of the byte immediately following the end of the
|
||||
** token within the input string.
|
||||
** pos: Token offset of token within input.
|
||||
**
|
||||
*/
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
|
||||
#include <string.h>
|
||||
#include <assert.h>
|
||||
|
||||
typedef struct Fts3tokTable Fts3tokTable;
|
||||
typedef struct Fts3tokCursor Fts3tokCursor;
|
||||
|
||||
/*
|
||||
** Virtual table structure.
|
||||
*/
|
||||
struct Fts3tokTable {
|
||||
sqlite3_vtab base; /* Base class used by SQLite core */
|
||||
const sqlite3_tokenizer_module *pMod;
|
||||
sqlite3_tokenizer *pTok;
|
||||
};
|
||||
|
||||
/*
|
||||
** Virtual table cursor structure.
|
||||
*/
|
||||
struct Fts3tokCursor {
|
||||
sqlite3_vtab_cursor base; /* Base class used by SQLite core */
|
||||
char *zInput; /* Input string */
|
||||
sqlite3_tokenizer_cursor *pCsr; /* Cursor to iterate through zInput */
|
||||
int iRowid; /* Current 'rowid' value */
|
||||
const char *zToken; /* Current 'token' value */
|
||||
int nToken; /* Size of zToken in bytes */
|
||||
int iStart; /* Current 'start' value */
|
||||
int iEnd; /* Current 'end' value */
|
||||
int iPos; /* Current 'pos' value */
|
||||
};
|
||||
|
||||
/*
|
||||
** Query FTS for the tokenizer implementation named zName.
|
||||
*/
|
||||
static int fts3tokQueryTokenizer(
|
||||
Fts3Hash *pHash,
|
||||
const char *zName,
|
||||
const sqlite3_tokenizer_module **pp,
|
||||
char **pzErr
|
||||
){
|
||||
sqlite3_tokenizer_module *p;
|
||||
int nName = (int)strlen(zName);
|
||||
|
||||
p = (sqlite3_tokenizer_module *)sqlite3Fts3HashFind(pHash, zName, nName+1);
|
||||
if( !p ){
|
||||
sqlite3Fts3ErrMsg(pzErr, "unknown tokenizer: %s", zName);
|
||||
return SQLITE_ERROR;
|
||||
}
|
||||
|
||||
*pp = p;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** The second argument, argv[], is an array of pointers to nul-terminated
|
||||
** strings. This function makes a copy of the array and strings into a
|
||||
** single block of memory. It then dequotes any of the strings that appear
|
||||
** to be quoted.
|
||||
**
|
||||
** If successful, output parameter *pazDequote is set to point at the
|
||||
** array of dequoted strings and SQLITE_OK is returned. The caller is
|
||||
** responsible for eventually calling sqlite3_free() to free the array
|
||||
** in this case. Or, if an error occurs, an SQLite error code is returned.
|
||||
** The final value of *pazDequote is undefined in this case.
|
||||
*/
|
||||
static int fts3tokDequoteArray(
|
||||
int argc, /* Number of elements in argv[] */
|
||||
const char * const *argv, /* Input array */
|
||||
char ***pazDequote /* Output array */
|
||||
){
|
||||
int rc = SQLITE_OK; /* Return code */
|
||||
if( argc==0 ){
|
||||
*pazDequote = 0;
|
||||
}else{
|
||||
int i;
|
||||
int nByte = 0;
|
||||
char **azDequote;
|
||||
|
||||
for(i=0; i<argc; i++){
|
||||
nByte += (int)(strlen(argv[i]) + 1);
|
||||
}
|
||||
|
||||
*pazDequote = azDequote = sqlite3_malloc(sizeof(char *)*argc + nByte);
|
||||
if( azDequote==0 ){
|
||||
rc = SQLITE_NOMEM;
|
||||
}else{
|
||||
char *pSpace = (char *)&azDequote[argc];
|
||||
for(i=0; i<argc; i++){
|
||||
int n = (int)strlen(argv[i]);
|
||||
azDequote[i] = pSpace;
|
||||
memcpy(pSpace, argv[i], n+1);
|
||||
sqlite3Fts3Dequote(pSpace);
|
||||
pSpace += (n+1);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** Schema of the tokenizer table.
|
||||
*/
|
||||
#define FTS3_TOK_SCHEMA "CREATE TABLE x(input, token, start, end, position)"
|
||||
|
||||
/*
|
||||
** This function does all the work for both the xConnect and xCreate methods.
|
||||
** These tables have no persistent representation of their own, so xConnect
|
||||
** and xCreate are identical operations.
|
||||
**
|
||||
** argv[0]: module name
|
||||
** argv[1]: database name
|
||||
** argv[2]: table name
|
||||
** argv[3]: first argument (tokenizer name)
|
||||
*/
|
||||
static int fts3tokConnectMethod(
|
||||
sqlite3 *db, /* Database connection */
|
||||
void *pHash, /* Hash table of tokenizers */
|
||||
int argc, /* Number of elements in argv array */
|
||||
const char * const *argv, /* xCreate/xConnect argument array */
|
||||
sqlite3_vtab **ppVtab, /* OUT: New sqlite3_vtab object */
|
||||
char **pzErr /* OUT: sqlite3_malloc'd error message */
|
||||
){
|
||||
Fts3tokTable *pTab = 0;
|
||||
const sqlite3_tokenizer_module *pMod = 0;
|
||||
sqlite3_tokenizer *pTok = 0;
|
||||
int rc;
|
||||
char **azDequote = 0;
|
||||
int nDequote;
|
||||
|
||||
rc = sqlite3_declare_vtab(db, FTS3_TOK_SCHEMA);
|
||||
if( rc!=SQLITE_OK ) return rc;
|
||||
|
||||
nDequote = argc-3;
|
||||
rc = fts3tokDequoteArray(nDequote, &argv[3], &azDequote);
|
||||
|
||||
if( rc==SQLITE_OK ){
|
||||
const char *zModule;
|
||||
if( nDequote<1 ){
|
||||
zModule = "simple";
|
||||
}else{
|
||||
zModule = azDequote[0];
|
||||
}
|
||||
rc = fts3tokQueryTokenizer((Fts3Hash*)pHash, zModule, &pMod, pzErr);
|
||||
}
|
||||
|
||||
assert( (rc==SQLITE_OK)==(pMod!=0) );
|
||||
if( rc==SQLITE_OK ){
|
||||
const char * const *azArg = (const char * const *)&azDequote[1];
|
||||
rc = pMod->xCreate((nDequote>1 ? nDequote-1 : 0), azArg, &pTok);
|
||||
}
|
||||
|
||||
if( rc==SQLITE_OK ){
|
||||
pTab = (Fts3tokTable *)sqlite3_malloc(sizeof(Fts3tokTable));
|
||||
if( pTab==0 ){
|
||||
rc = SQLITE_NOMEM;
|
||||
}
|
||||
}
|
||||
|
||||
if( rc==SQLITE_OK ){
|
||||
memset(pTab, 0, sizeof(Fts3tokTable));
|
||||
pTab->pMod = pMod;
|
||||
pTab->pTok = pTok;
|
||||
*ppVtab = &pTab->base;
|
||||
}else{
|
||||
if( pTok ){
|
||||
pMod->xDestroy(pTok);
|
||||
}
|
||||
}
|
||||
|
||||
sqlite3_free(azDequote);
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** This function does the work for both the xDisconnect and xDestroy methods.
|
||||
** These tables have no persistent representation of their own, so xDisconnect
|
||||
** and xDestroy are identical operations.
|
||||
*/
|
||||
static int fts3tokDisconnectMethod(sqlite3_vtab *pVtab){
|
||||
Fts3tokTable *pTab = (Fts3tokTable *)pVtab;
|
||||
|
||||
pTab->pMod->xDestroy(pTab->pTok);
|
||||
sqlite3_free(pTab);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xBestIndex - Analyze a WHERE and ORDER BY clause.
|
||||
*/
|
||||
static int fts3tokBestIndexMethod(
|
||||
sqlite3_vtab *pVTab,
|
||||
sqlite3_index_info *pInfo
|
||||
){
|
||||
int i;
|
||||
UNUSED_PARAMETER(pVTab);
|
||||
|
||||
for(i=0; i<pInfo->nConstraint; i++){
|
||||
if( pInfo->aConstraint[i].usable
|
||||
&& pInfo->aConstraint[i].iColumn==0
|
||||
&& pInfo->aConstraint[i].op==SQLITE_INDEX_CONSTRAINT_EQ
|
||||
){
|
||||
pInfo->idxNum = 1;
|
||||
pInfo->aConstraintUsage[i].argvIndex = 1;
|
||||
pInfo->aConstraintUsage[i].omit = 1;
|
||||
pInfo->estimatedCost = 1;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
}
|
||||
|
||||
pInfo->idxNum = 0;
|
||||
assert( pInfo->estimatedCost>1000000.0 );
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xOpen - Open a cursor.
|
||||
*/
|
||||
static int fts3tokOpenMethod(sqlite3_vtab *pVTab, sqlite3_vtab_cursor **ppCsr){
|
||||
Fts3tokCursor *pCsr;
|
||||
UNUSED_PARAMETER(pVTab);
|
||||
|
||||
pCsr = (Fts3tokCursor *)sqlite3_malloc(sizeof(Fts3tokCursor));
|
||||
if( pCsr==0 ){
|
||||
return SQLITE_NOMEM;
|
||||
}
|
||||
memset(pCsr, 0, sizeof(Fts3tokCursor));
|
||||
|
||||
*ppCsr = (sqlite3_vtab_cursor *)pCsr;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Reset the tokenizer cursor passed as the only argument. As if it had
|
||||
** just been returned by fts3tokOpenMethod().
|
||||
*/
|
||||
static void fts3tokResetCursor(Fts3tokCursor *pCsr){
|
||||
if( pCsr->pCsr ){
|
||||
Fts3tokTable *pTab = (Fts3tokTable *)(pCsr->base.pVtab);
|
||||
pTab->pMod->xClose(pCsr->pCsr);
|
||||
pCsr->pCsr = 0;
|
||||
}
|
||||
sqlite3_free(pCsr->zInput);
|
||||
pCsr->zInput = 0;
|
||||
pCsr->zToken = 0;
|
||||
pCsr->nToken = 0;
|
||||
pCsr->iStart = 0;
|
||||
pCsr->iEnd = 0;
|
||||
pCsr->iPos = 0;
|
||||
pCsr->iRowid = 0;
|
||||
}
|
||||
|
||||
/*
|
||||
** xClose - Close a cursor.
|
||||
*/
|
||||
static int fts3tokCloseMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3tokCursor *pCsr = (Fts3tokCursor *)pCursor;
|
||||
|
||||
fts3tokResetCursor(pCsr);
|
||||
sqlite3_free(pCsr);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xNext - Advance the cursor to the next row, if any.
|
||||
*/
|
||||
static int fts3tokNextMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3tokCursor *pCsr = (Fts3tokCursor *)pCursor;
|
||||
Fts3tokTable *pTab = (Fts3tokTable *)(pCursor->pVtab);
|
||||
int rc; /* Return code */
|
||||
|
||||
pCsr->iRowid++;
|
||||
rc = pTab->pMod->xNext(pCsr->pCsr,
|
||||
&pCsr->zToken, &pCsr->nToken,
|
||||
&pCsr->iStart, &pCsr->iEnd, &pCsr->iPos
|
||||
);
|
||||
|
||||
if( rc!=SQLITE_OK ){
|
||||
fts3tokResetCursor(pCsr);
|
||||
if( rc==SQLITE_DONE ) rc = SQLITE_OK;
|
||||
}
|
||||
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** xFilter - Initialize a cursor to point at the start of its data.
|
||||
*/
|
||||
static int fts3tokFilterMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* The cursor used for this query */
|
||||
int idxNum, /* Strategy index */
|
||||
const char *idxStr, /* Unused */
|
||||
int nVal, /* Number of elements in apVal */
|
||||
sqlite3_value **apVal /* Arguments for the indexing scheme */
|
||||
){
|
||||
int rc = SQLITE_ERROR;
|
||||
Fts3tokCursor *pCsr = (Fts3tokCursor *)pCursor;
|
||||
Fts3tokTable *pTab = (Fts3tokTable *)(pCursor->pVtab);
|
||||
UNUSED_PARAMETER(idxStr);
|
||||
UNUSED_PARAMETER(nVal);
|
||||
|
||||
fts3tokResetCursor(pCsr);
|
||||
if( idxNum==1 ){
|
||||
const char *zByte = (const char *)sqlite3_value_text(apVal[0]);
|
||||
int nByte = sqlite3_value_bytes(apVal[0]);
|
||||
pCsr->zInput = sqlite3_malloc(nByte+1);
|
||||
if( pCsr->zInput==0 ){
|
||||
rc = SQLITE_NOMEM;
|
||||
}else{
|
||||
memcpy(pCsr->zInput, zByte, nByte);
|
||||
pCsr->zInput[nByte] = 0;
|
||||
rc = pTab->pMod->xOpen(pTab->pTok, pCsr->zInput, nByte, &pCsr->pCsr);
|
||||
if( rc==SQLITE_OK ){
|
||||
pCsr->pCsr->pTokenizer = pTab->pTok;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if( rc!=SQLITE_OK ) return rc;
|
||||
return fts3tokNextMethod(pCursor);
|
||||
}
|
||||
|
||||
/*
|
||||
** xEof - Return true if the cursor is at EOF, or false otherwise.
|
||||
*/
|
||||
static int fts3tokEofMethod(sqlite3_vtab_cursor *pCursor){
|
||||
Fts3tokCursor *pCsr = (Fts3tokCursor *)pCursor;
|
||||
return (pCsr->zToken==0);
|
||||
}
|
||||
|
||||
/*
|
||||
** xColumn - Return a column value.
|
||||
*/
|
||||
static int fts3tokColumnMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* Cursor to retrieve value from */
|
||||
sqlite3_context *pCtx, /* Context for sqlite3_result_xxx() calls */
|
||||
int iCol /* Index of column to read value from */
|
||||
){
|
||||
Fts3tokCursor *pCsr = (Fts3tokCursor *)pCursor;
|
||||
|
||||
/* CREATE TABLE x(input, token, start, end, position) */
|
||||
switch( iCol ){
|
||||
case 0:
|
||||
sqlite3_result_text(pCtx, pCsr->zInput, -1, SQLITE_TRANSIENT);
|
||||
break;
|
||||
case 1:
|
||||
sqlite3_result_text(pCtx, pCsr->zToken, pCsr->nToken, SQLITE_TRANSIENT);
|
||||
break;
|
||||
case 2:
|
||||
sqlite3_result_int(pCtx, pCsr->iStart);
|
||||
break;
|
||||
case 3:
|
||||
sqlite3_result_int(pCtx, pCsr->iEnd);
|
||||
break;
|
||||
default:
|
||||
assert( iCol==4 );
|
||||
sqlite3_result_int(pCtx, pCsr->iPos);
|
||||
break;
|
||||
}
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** xRowid - Return the current rowid for the cursor.
|
||||
*/
|
||||
static int fts3tokRowidMethod(
|
||||
sqlite3_vtab_cursor *pCursor, /* Cursor to retrieve value from */
|
||||
sqlite_int64 *pRowid /* OUT: Rowid value */
|
||||
){
|
||||
Fts3tokCursor *pCsr = (Fts3tokCursor *)pCursor;
|
||||
*pRowid = (sqlite3_int64)pCsr->iRowid;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Register the fts3tok module with database connection db. Return SQLITE_OK
|
||||
** if successful or an error code if sqlite3_create_module() fails.
|
||||
*/
|
||||
int sqlite3Fts3InitTok(sqlite3 *db, Fts3Hash *pHash){
|
||||
static const sqlite3_module fts3tok_module = {
|
||||
0, /* iVersion */
|
||||
fts3tokConnectMethod, /* xCreate */
|
||||
fts3tokConnectMethod, /* xConnect */
|
||||
fts3tokBestIndexMethod, /* xBestIndex */
|
||||
fts3tokDisconnectMethod, /* xDisconnect */
|
||||
fts3tokDisconnectMethod, /* xDestroy */
|
||||
fts3tokOpenMethod, /* xOpen */
|
||||
fts3tokCloseMethod, /* xClose */
|
||||
fts3tokFilterMethod, /* xFilter */
|
||||
fts3tokNextMethod, /* xNext */
|
||||
fts3tokEofMethod, /* xEof */
|
||||
fts3tokColumnMethod, /* xColumn */
|
||||
fts3tokRowidMethod, /* xRowid */
|
||||
0, /* xUpdate */
|
||||
0, /* xBegin */
|
||||
0, /* xSync */
|
||||
0, /* xCommit */
|
||||
0, /* xRollback */
|
||||
0, /* xFindFunction */
|
||||
0, /* xRename */
|
||||
0, /* xSavepoint */
|
||||
0, /* xRelease */
|
||||
0 /* xRollbackTo */
|
||||
};
|
||||
int rc; /* Return code */
|
||||
|
||||
rc = sqlite3_create_module(db, "fts3tokenize", &fts3tok_module, (void*)pHash);
|
||||
return rc;
|
||||
}
|
||||
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
@ -0,0 +1,507 @@
|
||||
/*
|
||||
** 2007 June 22
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
** This is part of an SQLite module implementing full-text search.
|
||||
** This particular file implements the generic tokenizer interface.
|
||||
*/
|
||||
|
||||
/*
|
||||
** The code in this file is only compiled if:
|
||||
**
|
||||
** * The FTS3 module is being built as an extension
|
||||
** (in which case SQLITE_CORE is not defined), or
|
||||
**
|
||||
** * The FTS3 module is being built into the core of
|
||||
** SQLite (in which case SQLITE_ENABLE_FTS3 is defined).
|
||||
*/
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
|
||||
#include <assert.h>
|
||||
#include <string.h>
|
||||
|
||||
/*
|
||||
** Implementation of the SQL scalar function for accessing the underlying
|
||||
** hash table. This function may be called as follows:
|
||||
**
|
||||
** SELECT <function-name>(<key-name>);
|
||||
** SELECT <function-name>(<key-name>, <pointer>);
|
||||
**
|
||||
** where <function-name> is the name passed as the second argument
|
||||
** to the sqlite3Fts3InitHashTable() function (e.g. 'fts3_tokenizer').
|
||||
**
|
||||
** If the <pointer> argument is specified, it must be a blob value
|
||||
** containing a pointer to be stored as the hash data corresponding
|
||||
** to the string <key-name>. If <pointer> is not specified, then
|
||||
** the string <key-name> must already exist in the has table. Otherwise,
|
||||
** an error is returned.
|
||||
**
|
||||
** Whether or not the <pointer> argument is specified, the value returned
|
||||
** is a blob containing the pointer stored as the hash data corresponding
|
||||
** to string <key-name> (after the hash-table is updated, if applicable).
|
||||
*/
|
||||
static void scalarFunc(
|
||||
sqlite3_context *context,
|
||||
int argc,
|
||||
sqlite3_value **argv
|
||||
){
|
||||
Fts3Hash *pHash;
|
||||
void *pPtr = 0;
|
||||
const unsigned char *zName;
|
||||
int nName;
|
||||
|
||||
assert( argc==1 || argc==2 );
|
||||
|
||||
pHash = (Fts3Hash *)sqlite3_user_data(context);
|
||||
|
||||
zName = sqlite3_value_text(argv[0]);
|
||||
nName = sqlite3_value_bytes(argv[0])+1;
|
||||
|
||||
if( argc==2 ){
|
||||
#ifdef SQLITE_ENABLE_FTS3_TOKENIZER
|
||||
void *pOld;
|
||||
int n = sqlite3_value_bytes(argv[1]);
|
||||
if( zName==0 || n!=sizeof(pPtr) ){
|
||||
sqlite3_result_error(context, "argument type mismatch", -1);
|
||||
return;
|
||||
}
|
||||
pPtr = *(void **)sqlite3_value_blob(argv[1]);
|
||||
pOld = sqlite3Fts3HashInsert(pHash, (void *)zName, nName, pPtr);
|
||||
if( pOld==pPtr ){
|
||||
sqlite3_result_error(context, "out of memory", -1);
|
||||
return;
|
||||
}
|
||||
#else
|
||||
sqlite3_result_error(context, "fts3tokenize: "
|
||||
"disabled - rebuild with -DSQLITE_ENABLE_FTS3_TOKENIZER", -1
|
||||
);
|
||||
return;
|
||||
#endif /* SQLITE_ENABLE_FTS3_TOKENIZER */
|
||||
}else
|
||||
{
|
||||
if( zName ){
|
||||
pPtr = sqlite3Fts3HashFind(pHash, zName, nName);
|
||||
}
|
||||
if( !pPtr ){
|
||||
char *zErr = sqlite3_mprintf("unknown tokenizer: %s", zName);
|
||||
sqlite3_result_error(context, zErr, -1);
|
||||
sqlite3_free(zErr);
|
||||
return;
|
||||
}
|
||||
}
|
||||
|
||||
sqlite3_result_blob(context, (void *)&pPtr, sizeof(pPtr), SQLITE_TRANSIENT);
|
||||
}
|
||||
|
||||
int sqlite3Fts3IsIdChar(char c){
|
||||
static const char isFtsIdChar[] = {
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 0x */
|
||||
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 1x */
|
||||
0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 2x */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, /* 3x */
|
||||
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 4x */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, /* 5x */
|
||||
0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, /* 6x */
|
||||
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, /* 7x */
|
||||
};
|
||||
return (c&0x80 || isFtsIdChar[(int)(c)]);
|
||||
}
|
||||
|
||||
const char *sqlite3Fts3NextToken(const char *zStr, int *pn){
|
||||
const char *z1;
|
||||
const char *z2 = 0;
|
||||
|
||||
/* Find the start of the next token. */
|
||||
z1 = zStr;
|
||||
while( z2==0 ){
|
||||
char c = *z1;
|
||||
switch( c ){
|
||||
case '\0': return 0; /* No more tokens here */
|
||||
case '\'':
|
||||
case '"':
|
||||
case '`': {
|
||||
z2 = z1;
|
||||
while( *++z2 && (*z2!=c || *++z2==c) );
|
||||
break;
|
||||
}
|
||||
case '[':
|
||||
z2 = &z1[1];
|
||||
while( *z2 && z2[0]!=']' ) z2++;
|
||||
if( *z2 ) z2++;
|
||||
break;
|
||||
|
||||
default:
|
||||
if( sqlite3Fts3IsIdChar(*z1) ){
|
||||
z2 = &z1[1];
|
||||
while( sqlite3Fts3IsIdChar(*z2) ) z2++;
|
||||
}else{
|
||||
z1++;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
*pn = (int)(z2-z1);
|
||||
return z1;
|
||||
}
|
||||
|
||||
int sqlite3Fts3InitTokenizer(
|
||||
Fts3Hash *pHash, /* Tokenizer hash table */
|
||||
const char *zArg, /* Tokenizer name */
|
||||
sqlite3_tokenizer **ppTok, /* OUT: Tokenizer (if applicable) */
|
||||
char **pzErr /* OUT: Set to malloced error message */
|
||||
){
|
||||
int rc;
|
||||
char *z = (char *)zArg;
|
||||
int n = 0;
|
||||
char *zCopy;
|
||||
char *zEnd; /* Pointer to nul-term of zCopy */
|
||||
sqlite3_tokenizer_module *m;
|
||||
|
||||
zCopy = sqlite3_mprintf("%s", zArg);
|
||||
if( !zCopy ) return SQLITE_NOMEM;
|
||||
zEnd = &zCopy[strlen(zCopy)];
|
||||
|
||||
z = (char *)sqlite3Fts3NextToken(zCopy, &n);
|
||||
if( z==0 ){
|
||||
assert( n==0 );
|
||||
z = zCopy;
|
||||
}
|
||||
z[n] = '\0';
|
||||
sqlite3Fts3Dequote(z);
|
||||
|
||||
m = (sqlite3_tokenizer_module *)sqlite3Fts3HashFind(pHash,z,(int)strlen(z)+1);
|
||||
if( !m ){
|
||||
sqlite3Fts3ErrMsg(pzErr, "unknown tokenizer: %s", z);
|
||||
rc = SQLITE_ERROR;
|
||||
}else{
|
||||
char const **aArg = 0;
|
||||
int iArg = 0;
|
||||
z = &z[n+1];
|
||||
while( z<zEnd && (NULL!=(z = (char *)sqlite3Fts3NextToken(z, &n))) ){
|
||||
int nNew = sizeof(char *)*(iArg+1);
|
||||
char const **aNew = (const char **)sqlite3_realloc((void *)aArg, nNew);
|
||||
if( !aNew ){
|
||||
sqlite3_free(zCopy);
|
||||
sqlite3_free((void *)aArg);
|
||||
return SQLITE_NOMEM;
|
||||
}
|
||||
aArg = aNew;
|
||||
aArg[iArg++] = z;
|
||||
z[n] = '\0';
|
||||
sqlite3Fts3Dequote(z);
|
||||
z = &z[n+1];
|
||||
}
|
||||
rc = m->xCreate(iArg, aArg, ppTok);
|
||||
assert( rc!=SQLITE_OK || *ppTok );
|
||||
if( rc!=SQLITE_OK ){
|
||||
sqlite3Fts3ErrMsg(pzErr, "unknown tokenizer");
|
||||
}else{
|
||||
(*ppTok)->pModule = m;
|
||||
}
|
||||
sqlite3_free((void *)aArg);
|
||||
}
|
||||
|
||||
sqlite3_free(zCopy);
|
||||
return rc;
|
||||
}
|
||||
|
||||
|
||||
#ifdef SQLITE_TEST
|
||||
|
||||
#include <tcl.h>
|
||||
#include <string.h>
|
||||
|
||||
/*
|
||||
** Implementation of a special SQL scalar function for testing tokenizers
|
||||
** designed to be used in concert with the Tcl testing framework. This
|
||||
** function must be called with two or more arguments:
|
||||
**
|
||||
** SELECT <function-name>(<key-name>, ..., <input-string>);
|
||||
**
|
||||
** where <function-name> is the name passed as the second argument
|
||||
** to the sqlite3Fts3InitHashTable() function (e.g. 'fts3_tokenizer')
|
||||
** concatenated with the string '_test' (e.g. 'fts3_tokenizer_test').
|
||||
**
|
||||
** The return value is a string that may be interpreted as a Tcl
|
||||
** list. For each token in the <input-string>, three elements are
|
||||
** added to the returned list. The first is the token position, the
|
||||
** second is the token text (folded, stemmed, etc.) and the third is the
|
||||
** substring of <input-string> associated with the token. For example,
|
||||
** using the built-in "simple" tokenizer:
|
||||
**
|
||||
** SELECT fts_tokenizer_test('simple', 'I don't see how');
|
||||
**
|
||||
** will return the string:
|
||||
**
|
||||
** "{0 i I 1 dont don't 2 see see 3 how how}"
|
||||
**
|
||||
*/
|
||||
static void testFunc(
|
||||
sqlite3_context *context,
|
||||
int argc,
|
||||
sqlite3_value **argv
|
||||
){
|
||||
Fts3Hash *pHash;
|
||||
sqlite3_tokenizer_module *p;
|
||||
sqlite3_tokenizer *pTokenizer = 0;
|
||||
sqlite3_tokenizer_cursor *pCsr = 0;
|
||||
|
||||
const char *zErr = 0;
|
||||
|
||||
const char *zName;
|
||||
int nName;
|
||||
const char *zInput;
|
||||
int nInput;
|
||||
|
||||
const char *azArg[64];
|
||||
|
||||
const char *zToken;
|
||||
int nToken = 0;
|
||||
int iStart = 0;
|
||||
int iEnd = 0;
|
||||
int iPos = 0;
|
||||
int i;
|
||||
|
||||
Tcl_Obj *pRet;
|
||||
|
||||
if( argc<2 ){
|
||||
sqlite3_result_error(context, "insufficient arguments", -1);
|
||||
return;
|
||||
}
|
||||
|
||||
nName = sqlite3_value_bytes(argv[0]);
|
||||
zName = (const char *)sqlite3_value_text(argv[0]);
|
||||
nInput = sqlite3_value_bytes(argv[argc-1]);
|
||||
zInput = (const char *)sqlite3_value_text(argv[argc-1]);
|
||||
|
||||
pHash = (Fts3Hash *)sqlite3_user_data(context);
|
||||
p = (sqlite3_tokenizer_module *)sqlite3Fts3HashFind(pHash, zName, nName+1);
|
||||
|
||||
if( !p ){
|
||||
char *zErr2 = sqlite3_mprintf("unknown tokenizer: %s", zName);
|
||||
sqlite3_result_error(context, zErr2, -1);
|
||||
sqlite3_free(zErr2);
|
||||
return;
|
||||
}
|
||||
|
||||
pRet = Tcl_NewObj();
|
||||
Tcl_IncrRefCount(pRet);
|
||||
|
||||
for(i=1; i<argc-1; i++){
|
||||
azArg[i-1] = (const char *)sqlite3_value_text(argv[i]);
|
||||
}
|
||||
|
||||
if( SQLITE_OK!=p->xCreate(argc-2, azArg, &pTokenizer) ){
|
||||
zErr = "error in xCreate()";
|
||||
goto finish;
|
||||
}
|
||||
pTokenizer->pModule = p;
|
||||
if( sqlite3Fts3OpenTokenizer(pTokenizer, 0, zInput, nInput, &pCsr) ){
|
||||
zErr = "error in xOpen()";
|
||||
goto finish;
|
||||
}
|
||||
|
||||
while( SQLITE_OK==p->xNext(pCsr, &zToken, &nToken, &iStart, &iEnd, &iPos) ){
|
||||
Tcl_ListObjAppendElement(0, pRet, Tcl_NewIntObj(iPos));
|
||||
Tcl_ListObjAppendElement(0, pRet, Tcl_NewStringObj(zToken, nToken));
|
||||
zToken = &zInput[iStart];
|
||||
nToken = iEnd-iStart;
|
||||
Tcl_ListObjAppendElement(0, pRet, Tcl_NewStringObj(zToken, nToken));
|
||||
}
|
||||
|
||||
if( SQLITE_OK!=p->xClose(pCsr) ){
|
||||
zErr = "error in xClose()";
|
||||
goto finish;
|
||||
}
|
||||
if( SQLITE_OK!=p->xDestroy(pTokenizer) ){
|
||||
zErr = "error in xDestroy()";
|
||||
goto finish;
|
||||
}
|
||||
|
||||
finish:
|
||||
if( zErr ){
|
||||
sqlite3_result_error(context, zErr, -1);
|
||||
}else{
|
||||
sqlite3_result_text(context, Tcl_GetString(pRet), -1, SQLITE_TRANSIENT);
|
||||
}
|
||||
Tcl_DecrRefCount(pRet);
|
||||
}
|
||||
|
||||
#ifdef SQLITE_ENABLE_FTS3_TOKENIZER
|
||||
static
|
||||
int registerTokenizer(
|
||||
sqlite3 *db,
|
||||
char *zName,
|
||||
const sqlite3_tokenizer_module *p
|
||||
){
|
||||
int rc;
|
||||
sqlite3_stmt *pStmt;
|
||||
const char zSql[] = "SELECT fts3_tokenizer(?, ?)";
|
||||
|
||||
rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, 0);
|
||||
if( rc!=SQLITE_OK ){
|
||||
return rc;
|
||||
}
|
||||
|
||||
sqlite3_bind_text(pStmt, 1, zName, -1, SQLITE_STATIC);
|
||||
sqlite3_bind_blob(pStmt, 2, &p, sizeof(p), SQLITE_STATIC);
|
||||
sqlite3_step(pStmt);
|
||||
|
||||
return sqlite3_finalize(pStmt);
|
||||
}
|
||||
#endif /* SQLITE_ENABLE_FTS3_TOKENIZER */
|
||||
|
||||
|
||||
static
|
||||
int queryTokenizer(
|
||||
sqlite3 *db,
|
||||
char *zName,
|
||||
const sqlite3_tokenizer_module **pp
|
||||
){
|
||||
int rc;
|
||||
sqlite3_stmt *pStmt;
|
||||
const char zSql[] = "SELECT fts3_tokenizer(?)";
|
||||
|
||||
*pp = 0;
|
||||
rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, 0);
|
||||
if( rc!=SQLITE_OK ){
|
||||
return rc;
|
||||
}
|
||||
|
||||
sqlite3_bind_text(pStmt, 1, zName, -1, SQLITE_STATIC);
|
||||
if( SQLITE_ROW==sqlite3_step(pStmt) ){
|
||||
if( sqlite3_column_type(pStmt, 0)==SQLITE_BLOB ){
|
||||
memcpy((void *)pp, sqlite3_column_blob(pStmt, 0), sizeof(*pp));
|
||||
}
|
||||
}
|
||||
|
||||
return sqlite3_finalize(pStmt);
|
||||
}
|
||||
|
||||
void sqlite3Fts3SimpleTokenizerModule(sqlite3_tokenizer_module const**ppModule);
|
||||
|
||||
/*
|
||||
** Implementation of the scalar function fts3_tokenizer_internal_test().
|
||||
** This function is used for testing only, it is not included in the
|
||||
** build unless SQLITE_TEST is defined.
|
||||
**
|
||||
** The purpose of this is to test that the fts3_tokenizer() function
|
||||
** can be used as designed by the C-code in the queryTokenizer and
|
||||
** registerTokenizer() functions above. These two functions are repeated
|
||||
** in the README.tokenizer file as an example, so it is important to
|
||||
** test them.
|
||||
**
|
||||
** To run the tests, evaluate the fts3_tokenizer_internal_test() scalar
|
||||
** function with no arguments. An assert() will fail if a problem is
|
||||
** detected. i.e.:
|
||||
**
|
||||
** SELECT fts3_tokenizer_internal_test();
|
||||
**
|
||||
*/
|
||||
static void intTestFunc(
|
||||
sqlite3_context *context,
|
||||
int argc,
|
||||
sqlite3_value **argv
|
||||
){
|
||||
int rc;
|
||||
const sqlite3_tokenizer_module *p1;
|
||||
const sqlite3_tokenizer_module *p2;
|
||||
sqlite3 *db = (sqlite3 *)sqlite3_user_data(context);
|
||||
|
||||
UNUSED_PARAMETER(argc);
|
||||
UNUSED_PARAMETER(argv);
|
||||
|
||||
/* Test the query function */
|
||||
sqlite3Fts3SimpleTokenizerModule(&p1);
|
||||
rc = queryTokenizer(db, "simple", &p2);
|
||||
assert( rc==SQLITE_OK );
|
||||
assert( p1==p2 );
|
||||
rc = queryTokenizer(db, "nosuchtokenizer", &p2);
|
||||
assert( rc==SQLITE_ERROR );
|
||||
assert( p2==0 );
|
||||
assert( 0==strcmp(sqlite3_errmsg(db), "unknown tokenizer: nosuchtokenizer") );
|
||||
|
||||
/* Test the storage function */
|
||||
#ifdef SQLITE_ENABLE_FTS3_TOKENIZER
|
||||
rc = registerTokenizer(db, "nosuchtokenizer", p1);
|
||||
assert( rc==SQLITE_OK );
|
||||
rc = queryTokenizer(db, "nosuchtokenizer", &p2);
|
||||
assert( rc==SQLITE_OK );
|
||||
assert( p2==p1 );
|
||||
#endif
|
||||
|
||||
sqlite3_result_text(context, "ok", -1, SQLITE_STATIC);
|
||||
}
|
||||
|
||||
#endif
|
||||
|
||||
/*
|
||||
** Set up SQL objects in database db used to access the contents of
|
||||
** the hash table pointed to by argument pHash. The hash table must
|
||||
** been initialized to use string keys, and to take a private copy
|
||||
** of the key when a value is inserted. i.e. by a call similar to:
|
||||
**
|
||||
** sqlite3Fts3HashInit(pHash, FTS3_HASH_STRING, 1);
|
||||
**
|
||||
** This function adds a scalar function (see header comment above
|
||||
** scalarFunc() in this file for details) and, if ENABLE_TABLE is
|
||||
** defined at compilation time, a temporary virtual table (see header
|
||||
** comment above struct HashTableVtab) to the database schema. Both
|
||||
** provide read/write access to the contents of *pHash.
|
||||
**
|
||||
** The third argument to this function, zName, is used as the name
|
||||
** of both the scalar and, if created, the virtual table.
|
||||
*/
|
||||
int sqlite3Fts3InitHashTable(
|
||||
sqlite3 *db,
|
||||
Fts3Hash *pHash,
|
||||
const char *zName
|
||||
){
|
||||
int rc = SQLITE_OK;
|
||||
void *p = (void *)pHash;
|
||||
const int any = SQLITE_ANY;
|
||||
|
||||
#ifdef SQLITE_TEST
|
||||
char *zTest = 0;
|
||||
char *zTest2 = 0;
|
||||
void *pdb = (void *)db;
|
||||
zTest = sqlite3_mprintf("%s_test", zName);
|
||||
zTest2 = sqlite3_mprintf("%s_internal_test", zName);
|
||||
if( !zTest || !zTest2 ){
|
||||
rc = SQLITE_NOMEM;
|
||||
}
|
||||
#endif
|
||||
|
||||
if( SQLITE_OK==rc ){
|
||||
rc = sqlite3_create_function(db, zName, 1, any, p, scalarFunc, 0, 0);
|
||||
}
|
||||
if( SQLITE_OK==rc ){
|
||||
rc = sqlite3_create_function(db, zName, 2, any, p, scalarFunc, 0, 0);
|
||||
}
|
||||
#ifdef SQLITE_TEST
|
||||
if( SQLITE_OK==rc ){
|
||||
rc = sqlite3_create_function(db, zTest, -1, any, p, testFunc, 0, 0);
|
||||
}
|
||||
if( SQLITE_OK==rc ){
|
||||
rc = sqlite3_create_function(db, zTest2, 0, any, pdb, intTestFunc, 0, 0);
|
||||
}
|
||||
#endif
|
||||
|
||||
#ifdef SQLITE_TEST
|
||||
sqlite3_free(zTest);
|
||||
sqlite3_free(zTest2);
|
||||
#endif
|
||||
|
||||
return rc;
|
||||
}
|
||||
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
@ -0,0 +1,161 @@
|
||||
/*
|
||||
** 2006 July 10
|
||||
**
|
||||
** The author disclaims copyright to this source code.
|
||||
**
|
||||
*************************************************************************
|
||||
** Defines the interface to tokenizers used by fulltext-search. There
|
||||
** are three basic components:
|
||||
**
|
||||
** sqlite3_tokenizer_module is a singleton defining the tokenizer
|
||||
** interface functions. This is essentially the class structure for
|
||||
** tokenizers.
|
||||
**
|
||||
** sqlite3_tokenizer is used to define a particular tokenizer, perhaps
|
||||
** including customization information defined at creation time.
|
||||
**
|
||||
** sqlite3_tokenizer_cursor is generated by a tokenizer to generate
|
||||
** tokens from a particular input.
|
||||
*/
|
||||
#ifndef _FTS3_TOKENIZER_H_
|
||||
#define _FTS3_TOKENIZER_H_
|
||||
|
||||
/* TODO(shess) Only used for SQLITE_OK and SQLITE_DONE at this time.
|
||||
** If tokenizers are to be allowed to call sqlite3_*() functions, then
|
||||
** we will need a way to register the API consistently.
|
||||
*/
|
||||
#include "sqlite3.h"
|
||||
|
||||
/*
|
||||
** Structures used by the tokenizer interface. When a new tokenizer
|
||||
** implementation is registered, the caller provides a pointer to
|
||||
** an sqlite3_tokenizer_module containing pointers to the callback
|
||||
** functions that make up an implementation.
|
||||
**
|
||||
** When an fts3 table is created, it passes any arguments passed to
|
||||
** the tokenizer clause of the CREATE VIRTUAL TABLE statement to the
|
||||
** sqlite3_tokenizer_module.xCreate() function of the requested tokenizer
|
||||
** implementation. The xCreate() function in turn returns an
|
||||
** sqlite3_tokenizer structure representing the specific tokenizer to
|
||||
** be used for the fts3 table (customized by the tokenizer clause arguments).
|
||||
**
|
||||
** To tokenize an input buffer, the sqlite3_tokenizer_module.xOpen()
|
||||
** method is called. It returns an sqlite3_tokenizer_cursor object
|
||||
** that may be used to tokenize a specific input buffer based on
|
||||
** the tokenization rules supplied by a specific sqlite3_tokenizer
|
||||
** object.
|
||||
*/
|
||||
typedef struct sqlite3_tokenizer_module sqlite3_tokenizer_module;
|
||||
typedef struct sqlite3_tokenizer sqlite3_tokenizer;
|
||||
typedef struct sqlite3_tokenizer_cursor sqlite3_tokenizer_cursor;
|
||||
|
||||
struct sqlite3_tokenizer_module {
|
||||
|
||||
/*
|
||||
** Structure version. Should always be set to 0 or 1.
|
||||
*/
|
||||
int iVersion;
|
||||
|
||||
/*
|
||||
** Create a new tokenizer. The values in the argv[] array are the
|
||||
** arguments passed to the "tokenizer" clause of the CREATE VIRTUAL
|
||||
** TABLE statement that created the fts3 table. For example, if
|
||||
** the following SQL is executed:
|
||||
**
|
||||
** CREATE .. USING fts3( ... , tokenizer <tokenizer-name> arg1 arg2)
|
||||
**
|
||||
** then argc is set to 2, and the argv[] array contains pointers
|
||||
** to the strings "arg1" and "arg2".
|
||||
**
|
||||
** This method should return either SQLITE_OK (0), or an SQLite error
|
||||
** code. If SQLITE_OK is returned, then *ppTokenizer should be set
|
||||
** to point at the newly created tokenizer structure. The generic
|
||||
** sqlite3_tokenizer.pModule variable should not be initialized by
|
||||
** this callback. The caller will do so.
|
||||
*/
|
||||
int (*xCreate)(
|
||||
int argc, /* Size of argv array */
|
||||
const char *const*argv, /* Tokenizer argument strings */
|
||||
sqlite3_tokenizer **ppTokenizer /* OUT: Created tokenizer */
|
||||
);
|
||||
|
||||
/*
|
||||
** Destroy an existing tokenizer. The fts3 module calls this method
|
||||
** exactly once for each successful call to xCreate().
|
||||
*/
|
||||
int (*xDestroy)(sqlite3_tokenizer *pTokenizer);
|
||||
|
||||
/*
|
||||
** Create a tokenizer cursor to tokenize an input buffer. The caller
|
||||
** is responsible for ensuring that the input buffer remains valid
|
||||
** until the cursor is closed (using the xClose() method).
|
||||
*/
|
||||
int (*xOpen)(
|
||||
sqlite3_tokenizer *pTokenizer, /* Tokenizer object */
|
||||
const char *pInput, int nBytes, /* Input buffer */
|
||||
sqlite3_tokenizer_cursor **ppCursor /* OUT: Created tokenizer cursor */
|
||||
);
|
||||
|
||||
/*
|
||||
** Destroy an existing tokenizer cursor. The fts3 module calls this
|
||||
** method exactly once for each successful call to xOpen().
|
||||
*/
|
||||
int (*xClose)(sqlite3_tokenizer_cursor *pCursor);
|
||||
|
||||
/*
|
||||
** Retrieve the next token from the tokenizer cursor pCursor. This
|
||||
** method should either return SQLITE_OK and set the values of the
|
||||
** "OUT" variables identified below, or SQLITE_DONE to indicate that
|
||||
** the end of the buffer has been reached, or an SQLite error code.
|
||||
**
|
||||
** *ppToken should be set to point at a buffer containing the
|
||||
** normalized version of the token (i.e. after any case-folding and/or
|
||||
** stemming has been performed). *pnBytes should be set to the length
|
||||
** of this buffer in bytes. The input text that generated the token is
|
||||
** identified by the byte offsets returned in *piStartOffset and
|
||||
** *piEndOffset. *piStartOffset should be set to the index of the first
|
||||
** byte of the token in the input buffer. *piEndOffset should be set
|
||||
** to the index of the first byte just past the end of the token in
|
||||
** the input buffer.
|
||||
**
|
||||
** The buffer *ppToken is set to point at is managed by the tokenizer
|
||||
** implementation. It is only required to be valid until the next call
|
||||
** to xNext() or xClose().
|
||||
*/
|
||||
/* TODO(shess) current implementation requires pInput to be
|
||||
** nul-terminated. This should either be fixed, or pInput/nBytes
|
||||
** should be converted to zInput.
|
||||
*/
|
||||
int (*xNext)(
|
||||
sqlite3_tokenizer_cursor *pCursor, /* Tokenizer cursor */
|
||||
const char **ppToken, int *pnBytes, /* OUT: Normalized text for token */
|
||||
int *piStartOffset, /* OUT: Byte offset of token in input buffer */
|
||||
int *piEndOffset, /* OUT: Byte offset of end of token in input buffer */
|
||||
int *piPosition /* OUT: Number of tokens returned before this one */
|
||||
);
|
||||
|
||||
/***********************************************************************
|
||||
** Methods below this point are only available if iVersion>=1.
|
||||
*/
|
||||
|
||||
/*
|
||||
** Configure the language id of a tokenizer cursor.
|
||||
*/
|
||||
int (*xLanguageid)(sqlite3_tokenizer_cursor *pCsr, int iLangid);
|
||||
};
|
||||
|
||||
struct sqlite3_tokenizer {
|
||||
const sqlite3_tokenizer_module *pModule; /* The module for this tokenizer */
|
||||
/* Tokenizer implementations will typically add additional fields */
|
||||
};
|
||||
|
||||
struct sqlite3_tokenizer_cursor {
|
||||
sqlite3_tokenizer *pTokenizer; /* Tokenizer for this cursor. */
|
||||
/* Tokenizer implementations will typically add additional fields */
|
||||
};
|
||||
|
||||
int fts3_global_term_cnt(int iTerm, int iCol);
|
||||
int fts3_term_cnt(int iTerm, int iCol);
|
||||
|
||||
|
||||
#endif /* _FTS3_TOKENIZER_H_ */
|
||||
@ -0,0 +1,234 @@
|
||||
/*
|
||||
** 2006 Oct 10
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
** Implementation of the "simple" full-text-search tokenizer.
|
||||
*/
|
||||
|
||||
/*
|
||||
** The code in this file is only compiled if:
|
||||
**
|
||||
** * The FTS3 module is being built as an extension
|
||||
** (in which case SQLITE_CORE is not defined), or
|
||||
**
|
||||
** * The FTS3 module is being built into the core of
|
||||
** SQLite (in which case SQLITE_ENABLE_FTS3 is defined).
|
||||
*/
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
|
||||
#include <assert.h>
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
|
||||
#include "fts3_tokenizer.h"
|
||||
|
||||
typedef struct simple_tokenizer {
|
||||
sqlite3_tokenizer base;
|
||||
char delim[128]; /* flag ASCII delimiters */
|
||||
} simple_tokenizer;
|
||||
|
||||
typedef struct simple_tokenizer_cursor {
|
||||
sqlite3_tokenizer_cursor base;
|
||||
const char *pInput; /* input we are tokenizing */
|
||||
int nBytes; /* size of the input */
|
||||
int iOffset; /* current position in pInput */
|
||||
int iToken; /* index of next token to be returned */
|
||||
char *pToken; /* storage for current token */
|
||||
int nTokenAllocated; /* space allocated to zToken buffer */
|
||||
} simple_tokenizer_cursor;
|
||||
|
||||
|
||||
static int simpleDelim(simple_tokenizer *t, unsigned char c){
|
||||
return c<0x80 && t->delim[c];
|
||||
}
|
||||
static int fts3_isalnum(int x){
|
||||
return (x>='0' && x<='9') || (x>='A' && x<='Z') || (x>='a' && x<='z');
|
||||
}
|
||||
|
||||
/*
|
||||
** Create a new tokenizer instance.
|
||||
*/
|
||||
static int simpleCreate(
|
||||
int argc, const char * const *argv,
|
||||
sqlite3_tokenizer **ppTokenizer
|
||||
){
|
||||
simple_tokenizer *t;
|
||||
|
||||
t = (simple_tokenizer *) sqlite3_malloc(sizeof(*t));
|
||||
if( t==NULL ) return SQLITE_NOMEM;
|
||||
memset(t, 0, sizeof(*t));
|
||||
|
||||
/* TODO(shess) Delimiters need to remain the same from run to run,
|
||||
** else we need to reindex. One solution would be a meta-table to
|
||||
** track such information in the database, then we'd only want this
|
||||
** information on the initial create.
|
||||
*/
|
||||
if( argc>1 ){
|
||||
int i, n = (int)strlen(argv[1]);
|
||||
for(i=0; i<n; i++){
|
||||
unsigned char ch = argv[1][i];
|
||||
/* We explicitly don't support UTF-8 delimiters for now. */
|
||||
if( ch>=0x80 ){
|
||||
sqlite3_free(t);
|
||||
return SQLITE_ERROR;
|
||||
}
|
||||
t->delim[ch] = 1;
|
||||
}
|
||||
} else {
|
||||
/* Mark non-alphanumeric ASCII characters as delimiters */
|
||||
int i;
|
||||
for(i=1; i<0x80; i++){
|
||||
t->delim[i] = !fts3_isalnum(i) ? -1 : 0;
|
||||
}
|
||||
}
|
||||
|
||||
*ppTokenizer = &t->base;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Destroy a tokenizer
|
||||
*/
|
||||
static int simpleDestroy(sqlite3_tokenizer *pTokenizer){
|
||||
sqlite3_free(pTokenizer);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Prepare to begin tokenizing a particular string. The input
|
||||
** string to be tokenized is pInput[0..nBytes-1]. A cursor
|
||||
** used to incrementally tokenize this string is returned in
|
||||
** *ppCursor.
|
||||
*/
|
||||
static int simpleOpen(
|
||||
sqlite3_tokenizer *pTokenizer, /* The tokenizer */
|
||||
const char *pInput, int nBytes, /* String to be tokenized */
|
||||
sqlite3_tokenizer_cursor **ppCursor /* OUT: Tokenization cursor */
|
||||
){
|
||||
simple_tokenizer_cursor *c;
|
||||
|
||||
UNUSED_PARAMETER(pTokenizer);
|
||||
|
||||
c = (simple_tokenizer_cursor *) sqlite3_malloc(sizeof(*c));
|
||||
if( c==NULL ) return SQLITE_NOMEM;
|
||||
|
||||
c->pInput = pInput;
|
||||
if( pInput==0 ){
|
||||
c->nBytes = 0;
|
||||
}else if( nBytes<0 ){
|
||||
c->nBytes = (int)strlen(pInput);
|
||||
}else{
|
||||
c->nBytes = nBytes;
|
||||
}
|
||||
c->iOffset = 0; /* start tokenizing at the beginning */
|
||||
c->iToken = 0;
|
||||
c->pToken = NULL; /* no space allocated, yet. */
|
||||
c->nTokenAllocated = 0;
|
||||
|
||||
*ppCursor = &c->base;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Close a tokenization cursor previously opened by a call to
|
||||
** simpleOpen() above.
|
||||
*/
|
||||
static int simpleClose(sqlite3_tokenizer_cursor *pCursor){
|
||||
simple_tokenizer_cursor *c = (simple_tokenizer_cursor *) pCursor;
|
||||
sqlite3_free(c->pToken);
|
||||
sqlite3_free(c);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Extract the next token from a tokenization cursor. The cursor must
|
||||
** have been opened by a prior call to simpleOpen().
|
||||
*/
|
||||
static int simpleNext(
|
||||
sqlite3_tokenizer_cursor *pCursor, /* Cursor returned by simpleOpen */
|
||||
const char **ppToken, /* OUT: *ppToken is the token text */
|
||||
int *pnBytes, /* OUT: Number of bytes in token */
|
||||
int *piStartOffset, /* OUT: Starting offset of token */
|
||||
int *piEndOffset, /* OUT: Ending offset of token */
|
||||
int *piPosition /* OUT: Position integer of token */
|
||||
){
|
||||
simple_tokenizer_cursor *c = (simple_tokenizer_cursor *) pCursor;
|
||||
simple_tokenizer *t = (simple_tokenizer *) pCursor->pTokenizer;
|
||||
unsigned char *p = (unsigned char *)c->pInput;
|
||||
|
||||
while( c->iOffset<c->nBytes ){
|
||||
int iStartOffset;
|
||||
|
||||
/* Scan past delimiter characters */
|
||||
while( c->iOffset<c->nBytes && simpleDelim(t, p[c->iOffset]) ){
|
||||
c->iOffset++;
|
||||
}
|
||||
|
||||
/* Count non-delimiter characters. */
|
||||
iStartOffset = c->iOffset;
|
||||
while( c->iOffset<c->nBytes && !simpleDelim(t, p[c->iOffset]) ){
|
||||
c->iOffset++;
|
||||
}
|
||||
|
||||
if( c->iOffset>iStartOffset ){
|
||||
int i, n = c->iOffset-iStartOffset;
|
||||
if( n>c->nTokenAllocated ){
|
||||
char *pNew;
|
||||
c->nTokenAllocated = n+20;
|
||||
pNew = sqlite3_realloc(c->pToken, c->nTokenAllocated);
|
||||
if( !pNew ) return SQLITE_NOMEM;
|
||||
c->pToken = pNew;
|
||||
}
|
||||
for(i=0; i<n; i++){
|
||||
/* TODO(shess) This needs expansion to handle UTF-8
|
||||
** case-insensitivity.
|
||||
*/
|
||||
unsigned char ch = p[iStartOffset+i];
|
||||
c->pToken[i] = (char)((ch>='A' && ch<='Z') ? ch-'A'+'a' : ch);
|
||||
}
|
||||
*ppToken = c->pToken;
|
||||
*pnBytes = n;
|
||||
*piStartOffset = iStartOffset;
|
||||
*piEndOffset = c->iOffset;
|
||||
*piPosition = c->iToken++;
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
}
|
||||
return SQLITE_DONE;
|
||||
}
|
||||
|
||||
/*
|
||||
** The set of routines that implement the simple tokenizer
|
||||
*/
|
||||
static const sqlite3_tokenizer_module simpleTokenizerModule = {
|
||||
0,
|
||||
simpleCreate,
|
||||
simpleDestroy,
|
||||
simpleOpen,
|
||||
simpleClose,
|
||||
simpleNext,
|
||||
0,
|
||||
};
|
||||
|
||||
/*
|
||||
** Allocate a new simple tokenizer. Return a pointer to the new
|
||||
** tokenizer in *ppModule
|
||||
*/
|
||||
void sqlite3Fts3SimpleTokenizerModule(
|
||||
sqlite3_tokenizer_module const**ppModule
|
||||
){
|
||||
*ppModule = &simpleTokenizerModule;
|
||||
}
|
||||
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
@ -0,0 +1,393 @@
|
||||
/*
|
||||
** 2012 May 24
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
**
|
||||
** Implementation of the "unicode" full-text-search tokenizer.
|
||||
*/
|
||||
|
||||
#ifndef SQLITE_DISABLE_FTS3_UNICODE
|
||||
|
||||
#include "fts3Int.h"
|
||||
#if !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3)
|
||||
|
||||
#include <assert.h>
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
|
||||
#include "fts3_tokenizer.h"
|
||||
|
||||
/*
|
||||
** The following two macros - READ_UTF8 and WRITE_UTF8 - have been copied
|
||||
** from the sqlite3 source file utf.c. If this file is compiled as part
|
||||
** of the amalgamation, they are not required.
|
||||
*/
|
||||
#ifndef SQLITE_AMALGAMATION
|
||||
|
||||
static const unsigned char sqlite3Utf8Trans1[] = {
|
||||
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
|
||||
0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
|
||||
0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
|
||||
0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
|
||||
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
|
||||
0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
|
||||
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
|
||||
0x00, 0x01, 0x02, 0x03, 0x00, 0x01, 0x00, 0x00,
|
||||
};
|
||||
|
||||
#define READ_UTF8(zIn, zTerm, c) \
|
||||
c = *(zIn++); \
|
||||
if( c>=0xc0 ){ \
|
||||
c = sqlite3Utf8Trans1[c-0xc0]; \
|
||||
while( zIn!=zTerm && (*zIn & 0xc0)==0x80 ){ \
|
||||
c = (c<<6) + (0x3f & *(zIn++)); \
|
||||
} \
|
||||
if( c<0x80 \
|
||||
|| (c&0xFFFFF800)==0xD800 \
|
||||
|| (c&0xFFFFFFFE)==0xFFFE ){ c = 0xFFFD; } \
|
||||
}
|
||||
|
||||
#define WRITE_UTF8(zOut, c) { \
|
||||
if( c<0x00080 ){ \
|
||||
*zOut++ = (u8)(c&0xFF); \
|
||||
} \
|
||||
else if( c<0x00800 ){ \
|
||||
*zOut++ = 0xC0 + (u8)((c>>6)&0x1F); \
|
||||
*zOut++ = 0x80 + (u8)(c & 0x3F); \
|
||||
} \
|
||||
else if( c<0x10000 ){ \
|
||||
*zOut++ = 0xE0 + (u8)((c>>12)&0x0F); \
|
||||
*zOut++ = 0x80 + (u8)((c>>6) & 0x3F); \
|
||||
*zOut++ = 0x80 + (u8)(c & 0x3F); \
|
||||
}else{ \
|
||||
*zOut++ = 0xF0 + (u8)((c>>18) & 0x07); \
|
||||
*zOut++ = 0x80 + (u8)((c>>12) & 0x3F); \
|
||||
*zOut++ = 0x80 + (u8)((c>>6) & 0x3F); \
|
||||
*zOut++ = 0x80 + (u8)(c & 0x3F); \
|
||||
} \
|
||||
}
|
||||
|
||||
#endif /* ifndef SQLITE_AMALGAMATION */
|
||||
|
||||
typedef struct unicode_tokenizer unicode_tokenizer;
|
||||
typedef struct unicode_cursor unicode_cursor;
|
||||
|
||||
struct unicode_tokenizer {
|
||||
sqlite3_tokenizer base;
|
||||
int bRemoveDiacritic;
|
||||
int nException;
|
||||
int *aiException;
|
||||
};
|
||||
|
||||
struct unicode_cursor {
|
||||
sqlite3_tokenizer_cursor base;
|
||||
const unsigned char *aInput; /* Input text being tokenized */
|
||||
int nInput; /* Size of aInput[] in bytes */
|
||||
int iOff; /* Current offset within aInput[] */
|
||||
int iToken; /* Index of next token to be returned */
|
||||
char *zToken; /* storage for current token */
|
||||
int nAlloc; /* space allocated at zToken */
|
||||
};
|
||||
|
||||
|
||||
/*
|
||||
** Destroy a tokenizer allocated by unicodeCreate().
|
||||
*/
|
||||
static int unicodeDestroy(sqlite3_tokenizer *pTokenizer){
|
||||
if( pTokenizer ){
|
||||
unicode_tokenizer *p = (unicode_tokenizer *)pTokenizer;
|
||||
sqlite3_free(p->aiException);
|
||||
sqlite3_free(p);
|
||||
}
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** As part of a tokenchars= or separators= option, the CREATE VIRTUAL TABLE
|
||||
** statement has specified that the tokenizer for this table shall consider
|
||||
** all characters in string zIn/nIn to be separators (if bAlnum==0) or
|
||||
** token characters (if bAlnum==1).
|
||||
**
|
||||
** For each codepoint in the zIn/nIn string, this function checks if the
|
||||
** sqlite3FtsUnicodeIsalnum() function already returns the desired result.
|
||||
** If so, no action is taken. Otherwise, the codepoint is added to the
|
||||
** unicode_tokenizer.aiException[] array. For the purposes of tokenization,
|
||||
** the return value of sqlite3FtsUnicodeIsalnum() is inverted for all
|
||||
** codepoints in the aiException[] array.
|
||||
**
|
||||
** If a standalone diacritic mark (one that sqlite3FtsUnicodeIsdiacritic()
|
||||
** identifies as a diacritic) occurs in the zIn/nIn string it is ignored.
|
||||
** It is not possible to change the behavior of the tokenizer with respect
|
||||
** to these codepoints.
|
||||
*/
|
||||
static int unicodeAddExceptions(
|
||||
unicode_tokenizer *p, /* Tokenizer to add exceptions to */
|
||||
int bAlnum, /* Replace Isalnum() return value with this */
|
||||
const char *zIn, /* Array of characters to make exceptions */
|
||||
int nIn /* Length of z in bytes */
|
||||
){
|
||||
const unsigned char *z = (const unsigned char *)zIn;
|
||||
const unsigned char *zTerm = &z[nIn];
|
||||
int iCode;
|
||||
int nEntry = 0;
|
||||
|
||||
assert( bAlnum==0 || bAlnum==1 );
|
||||
|
||||
while( z<zTerm ){
|
||||
READ_UTF8(z, zTerm, iCode);
|
||||
assert( (sqlite3FtsUnicodeIsalnum(iCode) & 0xFFFFFFFE)==0 );
|
||||
if( sqlite3FtsUnicodeIsalnum(iCode)!=bAlnum
|
||||
&& sqlite3FtsUnicodeIsdiacritic(iCode)==0
|
||||
){
|
||||
nEntry++;
|
||||
}
|
||||
}
|
||||
|
||||
if( nEntry ){
|
||||
int *aNew; /* New aiException[] array */
|
||||
int nNew; /* Number of valid entries in array aNew[] */
|
||||
|
||||
aNew = sqlite3_realloc(p->aiException, (p->nException+nEntry)*sizeof(int));
|
||||
if( aNew==0 ) return SQLITE_NOMEM;
|
||||
nNew = p->nException;
|
||||
|
||||
z = (const unsigned char *)zIn;
|
||||
while( z<zTerm ){
|
||||
READ_UTF8(z, zTerm, iCode);
|
||||
if( sqlite3FtsUnicodeIsalnum(iCode)!=bAlnum
|
||||
&& sqlite3FtsUnicodeIsdiacritic(iCode)==0
|
||||
){
|
||||
int i, j;
|
||||
for(i=0; i<nNew && aNew[i]<iCode; i++);
|
||||
for(j=nNew; j>i; j--) aNew[j] = aNew[j-1];
|
||||
aNew[i] = iCode;
|
||||
nNew++;
|
||||
}
|
||||
}
|
||||
p->aiException = aNew;
|
||||
p->nException = nNew;
|
||||
}
|
||||
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Return true if the p->aiException[] array contains the value iCode.
|
||||
*/
|
||||
static int unicodeIsException(unicode_tokenizer *p, int iCode){
|
||||
if( p->nException>0 ){
|
||||
int *a = p->aiException;
|
||||
int iLo = 0;
|
||||
int iHi = p->nException-1;
|
||||
|
||||
while( iHi>=iLo ){
|
||||
int iTest = (iHi + iLo) / 2;
|
||||
if( iCode==a[iTest] ){
|
||||
return 1;
|
||||
}else if( iCode>a[iTest] ){
|
||||
iLo = iTest+1;
|
||||
}else{
|
||||
iHi = iTest-1;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
/*
|
||||
** Return true if, for the purposes of tokenization, codepoint iCode is
|
||||
** considered a token character (not a separator).
|
||||
*/
|
||||
static int unicodeIsAlnum(unicode_tokenizer *p, int iCode){
|
||||
assert( (sqlite3FtsUnicodeIsalnum(iCode) & 0xFFFFFFFE)==0 );
|
||||
return sqlite3FtsUnicodeIsalnum(iCode) ^ unicodeIsException(p, iCode);
|
||||
}
|
||||
|
||||
/*
|
||||
** Create a new tokenizer instance.
|
||||
*/
|
||||
static int unicodeCreate(
|
||||
int nArg, /* Size of array argv[] */
|
||||
const char * const *azArg, /* Tokenizer creation arguments */
|
||||
sqlite3_tokenizer **pp /* OUT: New tokenizer handle */
|
||||
){
|
||||
unicode_tokenizer *pNew; /* New tokenizer object */
|
||||
int i;
|
||||
int rc = SQLITE_OK;
|
||||
|
||||
pNew = (unicode_tokenizer *) sqlite3_malloc(sizeof(unicode_tokenizer));
|
||||
if( pNew==NULL ) return SQLITE_NOMEM;
|
||||
memset(pNew, 0, sizeof(unicode_tokenizer));
|
||||
pNew->bRemoveDiacritic = 1;
|
||||
|
||||
for(i=0; rc==SQLITE_OK && i<nArg; i++){
|
||||
const char *z = azArg[i];
|
||||
int n = (int)strlen(z);
|
||||
|
||||
if( n==19 && memcmp("remove_diacritics=1", z, 19)==0 ){
|
||||
pNew->bRemoveDiacritic = 1;
|
||||
}
|
||||
else if( n==19 && memcmp("remove_diacritics=0", z, 19)==0 ){
|
||||
pNew->bRemoveDiacritic = 0;
|
||||
}
|
||||
else if( n>=11 && memcmp("tokenchars=", z, 11)==0 ){
|
||||
rc = unicodeAddExceptions(pNew, 1, &z[11], n-11);
|
||||
}
|
||||
else if( n>=11 && memcmp("separators=", z, 11)==0 ){
|
||||
rc = unicodeAddExceptions(pNew, 0, &z[11], n-11);
|
||||
}
|
||||
else{
|
||||
/* Unrecognized argument */
|
||||
rc = SQLITE_ERROR;
|
||||
}
|
||||
}
|
||||
|
||||
if( rc!=SQLITE_OK ){
|
||||
unicodeDestroy((sqlite3_tokenizer *)pNew);
|
||||
pNew = 0;
|
||||
}
|
||||
*pp = (sqlite3_tokenizer *)pNew;
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** Prepare to begin tokenizing a particular string. The input
|
||||
** string to be tokenized is pInput[0..nBytes-1]. A cursor
|
||||
** used to incrementally tokenize this string is returned in
|
||||
** *ppCursor.
|
||||
*/
|
||||
static int unicodeOpen(
|
||||
sqlite3_tokenizer *p, /* The tokenizer */
|
||||
const char *aInput, /* Input string */
|
||||
int nInput, /* Size of string aInput in bytes */
|
||||
sqlite3_tokenizer_cursor **pp /* OUT: New cursor object */
|
||||
){
|
||||
unicode_cursor *pCsr;
|
||||
|
||||
pCsr = (unicode_cursor *)sqlite3_malloc(sizeof(unicode_cursor));
|
||||
if( pCsr==0 ){
|
||||
return SQLITE_NOMEM;
|
||||
}
|
||||
memset(pCsr, 0, sizeof(unicode_cursor));
|
||||
|
||||
pCsr->aInput = (const unsigned char *)aInput;
|
||||
if( aInput==0 ){
|
||||
pCsr->nInput = 0;
|
||||
}else if( nInput<0 ){
|
||||
pCsr->nInput = (int)strlen(aInput);
|
||||
}else{
|
||||
pCsr->nInput = nInput;
|
||||
}
|
||||
|
||||
*pp = &pCsr->base;
|
||||
UNUSED_PARAMETER(p);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Close a tokenization cursor previously opened by a call to
|
||||
** simpleOpen() above.
|
||||
*/
|
||||
static int unicodeClose(sqlite3_tokenizer_cursor *pCursor){
|
||||
unicode_cursor *pCsr = (unicode_cursor *) pCursor;
|
||||
sqlite3_free(pCsr->zToken);
|
||||
sqlite3_free(pCsr);
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Extract the next token from a tokenization cursor. The cursor must
|
||||
** have been opened by a prior call to simpleOpen().
|
||||
*/
|
||||
static int unicodeNext(
|
||||
sqlite3_tokenizer_cursor *pC, /* Cursor returned by simpleOpen */
|
||||
const char **paToken, /* OUT: Token text */
|
||||
int *pnToken, /* OUT: Number of bytes at *paToken */
|
||||
int *piStart, /* OUT: Starting offset of token */
|
||||
int *piEnd, /* OUT: Ending offset of token */
|
||||
int *piPos /* OUT: Position integer of token */
|
||||
){
|
||||
unicode_cursor *pCsr = (unicode_cursor *)pC;
|
||||
unicode_tokenizer *p = ((unicode_tokenizer *)pCsr->base.pTokenizer);
|
||||
int iCode = 0;
|
||||
char *zOut;
|
||||
const unsigned char *z = &pCsr->aInput[pCsr->iOff];
|
||||
const unsigned char *zStart = z;
|
||||
const unsigned char *zEnd;
|
||||
const unsigned char *zTerm = &pCsr->aInput[pCsr->nInput];
|
||||
|
||||
/* Scan past any delimiter characters before the start of the next token.
|
||||
** Return SQLITE_DONE early if this takes us all the way to the end of
|
||||
** the input. */
|
||||
while( z<zTerm ){
|
||||
READ_UTF8(z, zTerm, iCode);
|
||||
if( unicodeIsAlnum(p, iCode) ) break;
|
||||
zStart = z;
|
||||
}
|
||||
if( zStart>=zTerm ) return SQLITE_DONE;
|
||||
|
||||
zOut = pCsr->zToken;
|
||||
do {
|
||||
int iOut;
|
||||
|
||||
/* Grow the output buffer if required. */
|
||||
if( (zOut-pCsr->zToken)>=(pCsr->nAlloc-4) ){
|
||||
char *zNew = sqlite3_realloc(pCsr->zToken, pCsr->nAlloc+64);
|
||||
if( !zNew ) return SQLITE_NOMEM;
|
||||
zOut = &zNew[zOut - pCsr->zToken];
|
||||
pCsr->zToken = zNew;
|
||||
pCsr->nAlloc += 64;
|
||||
}
|
||||
|
||||
/* Write the folded case of the last character read to the output */
|
||||
zEnd = z;
|
||||
iOut = sqlite3FtsUnicodeFold(iCode, p->bRemoveDiacritic);
|
||||
if( iOut ){
|
||||
WRITE_UTF8(zOut, iOut);
|
||||
}
|
||||
|
||||
/* If the cursor is not at EOF, read the next character */
|
||||
if( z>=zTerm ) break;
|
||||
READ_UTF8(z, zTerm, iCode);
|
||||
}while( unicodeIsAlnum(p, iCode)
|
||||
|| sqlite3FtsUnicodeIsdiacritic(iCode)
|
||||
);
|
||||
|
||||
/* Set the output variables and return. */
|
||||
pCsr->iOff = (int)(z - pCsr->aInput);
|
||||
*paToken = pCsr->zToken;
|
||||
*pnToken = (int)(zOut - pCsr->zToken);
|
||||
*piStart = (int)(zStart - pCsr->aInput);
|
||||
*piEnd = (int)(zEnd - pCsr->aInput);
|
||||
*piPos = pCsr->iToken++;
|
||||
return SQLITE_OK;
|
||||
}
|
||||
|
||||
/*
|
||||
** Set *ppModule to a pointer to the sqlite3_tokenizer_module
|
||||
** structure for the unicode tokenizer.
|
||||
*/
|
||||
void sqlite3Fts3UnicodeTokenizer(sqlite3_tokenizer_module const **ppModule){
|
||||
static const sqlite3_tokenizer_module module = {
|
||||
0,
|
||||
unicodeCreate,
|
||||
unicodeDestroy,
|
||||
unicodeOpen,
|
||||
unicodeClose,
|
||||
unicodeNext,
|
||||
0,
|
||||
};
|
||||
*ppModule = &module;
|
||||
}
|
||||
|
||||
#endif /* !defined(SQLITE_CORE) || defined(SQLITE_ENABLE_FTS3) */
|
||||
#endif /* ifndef SQLITE_DISABLE_FTS3_UNICODE */
|
||||
@ -0,0 +1,365 @@
|
||||
/*
|
||||
** 2012 May 25
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
*/
|
||||
|
||||
/*
|
||||
** DO NOT EDIT THIS MACHINE GENERATED FILE.
|
||||
*/
|
||||
|
||||
#ifndef SQLITE_DISABLE_FTS3_UNICODE
|
||||
#if defined(SQLITE_ENABLE_FTS3) || defined(SQLITE_ENABLE_FTS4)
|
||||
|
||||
#include <assert.h>
|
||||
|
||||
/*
|
||||
** Return true if the argument corresponds to a unicode codepoint
|
||||
** classified as either a letter or a number. Otherwise false.
|
||||
**
|
||||
** The results are undefined if the value passed to this function
|
||||
** is less than zero.
|
||||
*/
|
||||
int sqlite3FtsUnicodeIsalnum(int c){
|
||||
/* Each unsigned integer in the following array corresponds to a contiguous
|
||||
** range of unicode codepoints that are not either letters or numbers (i.e.
|
||||
** codepoints for which this function should return 0).
|
||||
**
|
||||
** The most significant 22 bits in each 32-bit value contain the first
|
||||
** codepoint in the range. The least significant 10 bits are used to store
|
||||
** the size of the range (always at least 1). In other words, the value
|
||||
** ((C<<22) + N) represents a range of N codepoints starting with codepoint
|
||||
** C. It is not possible to represent a range larger than 1023 codepoints
|
||||
** using this format.
|
||||
*/
|
||||
static const unsigned int aEntry[] = {
|
||||
0x00000030, 0x0000E807, 0x00016C06, 0x0001EC2F, 0x0002AC07,
|
||||
0x0002D001, 0x0002D803, 0x0002EC01, 0x0002FC01, 0x00035C01,
|
||||
0x0003DC01, 0x000B0804, 0x000B480E, 0x000B9407, 0x000BB401,
|
||||
0x000BBC81, 0x000DD401, 0x000DF801, 0x000E1002, 0x000E1C01,
|
||||
0x000FD801, 0x00120808, 0x00156806, 0x00162402, 0x00163C01,
|
||||
0x00164437, 0x0017CC02, 0x00180005, 0x00181816, 0x00187802,
|
||||
0x00192C15, 0x0019A804, 0x0019C001, 0x001B5001, 0x001B580F,
|
||||
0x001B9C07, 0x001BF402, 0x001C000E, 0x001C3C01, 0x001C4401,
|
||||
0x001CC01B, 0x001E980B, 0x001FAC09, 0x001FD804, 0x00205804,
|
||||
0x00206C09, 0x00209403, 0x0020A405, 0x0020C00F, 0x00216403,
|
||||
0x00217801, 0x0023901B, 0x00240004, 0x0024E803, 0x0024F812,
|
||||
0x00254407, 0x00258804, 0x0025C001, 0x00260403, 0x0026F001,
|
||||
0x0026F807, 0x00271C02, 0x00272C03, 0x00275C01, 0x00278802,
|
||||
0x0027C802, 0x0027E802, 0x00280403, 0x0028F001, 0x0028F805,
|
||||
0x00291C02, 0x00292C03, 0x00294401, 0x0029C002, 0x0029D401,
|
||||
0x002A0403, 0x002AF001, 0x002AF808, 0x002B1C03, 0x002B2C03,
|
||||
0x002B8802, 0x002BC002, 0x002C0403, 0x002CF001, 0x002CF807,
|
||||
0x002D1C02, 0x002D2C03, 0x002D5802, 0x002D8802, 0x002DC001,
|
||||
0x002E0801, 0x002EF805, 0x002F1803, 0x002F2804, 0x002F5C01,
|
||||
0x002FCC08, 0x00300403, 0x0030F807, 0x00311803, 0x00312804,
|
||||
0x00315402, 0x00318802, 0x0031FC01, 0x00320802, 0x0032F001,
|
||||
0x0032F807, 0x00331803, 0x00332804, 0x00335402, 0x00338802,
|
||||
0x00340802, 0x0034F807, 0x00351803, 0x00352804, 0x00355C01,
|
||||
0x00358802, 0x0035E401, 0x00360802, 0x00372801, 0x00373C06,
|
||||
0x00375801, 0x00376008, 0x0037C803, 0x0038C401, 0x0038D007,
|
||||
0x0038FC01, 0x00391C09, 0x00396802, 0x003AC401, 0x003AD006,
|
||||
0x003AEC02, 0x003B2006, 0x003C041F, 0x003CD00C, 0x003DC417,
|
||||
0x003E340B, 0x003E6424, 0x003EF80F, 0x003F380D, 0x0040AC14,
|
||||
0x00412806, 0x00415804, 0x00417803, 0x00418803, 0x00419C07,
|
||||
0x0041C404, 0x0042080C, 0x00423C01, 0x00426806, 0x0043EC01,
|
||||
0x004D740C, 0x004E400A, 0x00500001, 0x0059B402, 0x005A0001,
|
||||
0x005A6C02, 0x005BAC03, 0x005C4803, 0x005CC805, 0x005D4802,
|
||||
0x005DC802, 0x005ED023, 0x005F6004, 0x005F7401, 0x0060000F,
|
||||
0x0062A401, 0x0064800C, 0x0064C00C, 0x00650001, 0x00651002,
|
||||
0x0066C011, 0x00672002, 0x00677822, 0x00685C05, 0x00687802,
|
||||
0x0069540A, 0x0069801D, 0x0069FC01, 0x006A8007, 0x006AA006,
|
||||
0x006C0005, 0x006CD011, 0x006D6823, 0x006E0003, 0x006E840D,
|
||||
0x006F980E, 0x006FF004, 0x00709014, 0x0070EC05, 0x0071F802,
|
||||
0x00730008, 0x00734019, 0x0073B401, 0x0073C803, 0x00770027,
|
||||
0x0077F004, 0x007EF401, 0x007EFC03, 0x007F3403, 0x007F7403,
|
||||
0x007FB403, 0x007FF402, 0x00800065, 0x0081A806, 0x0081E805,
|
||||
0x00822805, 0x0082801A, 0x00834021, 0x00840002, 0x00840C04,
|
||||
0x00842002, 0x00845001, 0x00845803, 0x00847806, 0x00849401,
|
||||
0x00849C01, 0x0084A401, 0x0084B801, 0x0084E802, 0x00850005,
|
||||
0x00852804, 0x00853C01, 0x00864264, 0x00900027, 0x0091000B,
|
||||
0x0092704E, 0x00940200, 0x009C0475, 0x009E53B9, 0x00AD400A,
|
||||
0x00B39406, 0x00B3BC03, 0x00B3E404, 0x00B3F802, 0x00B5C001,
|
||||
0x00B5FC01, 0x00B7804F, 0x00B8C00C, 0x00BA001A, 0x00BA6C59,
|
||||
0x00BC00D6, 0x00BFC00C, 0x00C00005, 0x00C02019, 0x00C0A807,
|
||||
0x00C0D802, 0x00C0F403, 0x00C26404, 0x00C28001, 0x00C3EC01,
|
||||
0x00C64002, 0x00C6580A, 0x00C70024, 0x00C8001F, 0x00C8A81E,
|
||||
0x00C94001, 0x00C98020, 0x00CA2827, 0x00CB003F, 0x00CC0100,
|
||||
0x01370040, 0x02924037, 0x0293F802, 0x02983403, 0x0299BC10,
|
||||
0x029A7C01, 0x029BC008, 0x029C0017, 0x029C8002, 0x029E2402,
|
||||
0x02A00801, 0x02A01801, 0x02A02C01, 0x02A08C09, 0x02A0D804,
|
||||
0x02A1D004, 0x02A20002, 0x02A2D011, 0x02A33802, 0x02A38012,
|
||||
0x02A3E003, 0x02A4980A, 0x02A51C0D, 0x02A57C01, 0x02A60004,
|
||||
0x02A6CC1B, 0x02A77802, 0x02A8A40E, 0x02A90C01, 0x02A93002,
|
||||
0x02A97004, 0x02A9DC03, 0x02A9EC01, 0x02AAC001, 0x02AAC803,
|
||||
0x02AADC02, 0x02AAF802, 0x02AB0401, 0x02AB7802, 0x02ABAC07,
|
||||
0x02ABD402, 0x02AF8C0B, 0x03600001, 0x036DFC02, 0x036FFC02,
|
||||
0x037FFC01, 0x03EC7801, 0x03ECA401, 0x03EEC810, 0x03F4F802,
|
||||
0x03F7F002, 0x03F8001A, 0x03F88007, 0x03F8C023, 0x03F95013,
|
||||
0x03F9A004, 0x03FBFC01, 0x03FC040F, 0x03FC6807, 0x03FCEC06,
|
||||
0x03FD6C0B, 0x03FF8007, 0x03FFA007, 0x03FFE405, 0x04040003,
|
||||
0x0404DC09, 0x0405E411, 0x0406400C, 0x0407402E, 0x040E7C01,
|
||||
0x040F4001, 0x04215C01, 0x04247C01, 0x0424FC01, 0x04280403,
|
||||
0x04281402, 0x04283004, 0x0428E003, 0x0428FC01, 0x04294009,
|
||||
0x0429FC01, 0x042CE407, 0x04400003, 0x0440E016, 0x04420003,
|
||||
0x0442C012, 0x04440003, 0x04449C0E, 0x04450004, 0x04460003,
|
||||
0x0446CC0E, 0x04471404, 0x045AAC0D, 0x0491C004, 0x05BD442E,
|
||||
0x05BE3C04, 0x074000F6, 0x07440027, 0x0744A4B5, 0x07480046,
|
||||
0x074C0057, 0x075B0401, 0x075B6C01, 0x075BEC01, 0x075C5401,
|
||||
0x075CD401, 0x075D3C01, 0x075DBC01, 0x075E2401, 0x075EA401,
|
||||
0x075F0C01, 0x07BBC002, 0x07C0002C, 0x07C0C064, 0x07C2800F,
|
||||
0x07C2C40E, 0x07C3040F, 0x07C3440F, 0x07C4401F, 0x07C4C03C,
|
||||
0x07C5C02B, 0x07C7981D, 0x07C8402B, 0x07C90009, 0x07C94002,
|
||||
0x07CC0021, 0x07CCC006, 0x07CCDC46, 0x07CE0014, 0x07CE8025,
|
||||
0x07CF1805, 0x07CF8011, 0x07D0003F, 0x07D10001, 0x07D108B6,
|
||||
0x07D3E404, 0x07D4003E, 0x07D50004, 0x07D54018, 0x07D7EC46,
|
||||
0x07D9140B, 0x07DA0046, 0x07DC0074, 0x38000401, 0x38008060,
|
||||
0x380400F0,
|
||||
};
|
||||
static const unsigned int aAscii[4] = {
|
||||
0xFFFFFFFF, 0xFC00FFFF, 0xF8000001, 0xF8000001,
|
||||
};
|
||||
|
||||
if( c<128 ){
|
||||
return ( (aAscii[c >> 5] & (1 << (c & 0x001F)))==0 );
|
||||
}else if( c<(1<<22) ){
|
||||
unsigned int key = (((unsigned int)c)<<10) | 0x000003FF;
|
||||
int iRes = 0;
|
||||
int iHi = sizeof(aEntry)/sizeof(aEntry[0]) - 1;
|
||||
int iLo = 0;
|
||||
while( iHi>=iLo ){
|
||||
int iTest = (iHi + iLo) / 2;
|
||||
if( key >= aEntry[iTest] ){
|
||||
iRes = iTest;
|
||||
iLo = iTest+1;
|
||||
}else{
|
||||
iHi = iTest-1;
|
||||
}
|
||||
}
|
||||
assert( aEntry[0]<key );
|
||||
assert( key>=aEntry[iRes] );
|
||||
return (((unsigned int)c) >= ((aEntry[iRes]>>10) + (aEntry[iRes]&0x3FF)));
|
||||
}
|
||||
return 1;
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** If the argument is a codepoint corresponding to a lowercase letter
|
||||
** in the ASCII range with a diacritic added, return the codepoint
|
||||
** of the ASCII letter only. For example, if passed 235 - "LATIN
|
||||
** SMALL LETTER E WITH DIAERESIS" - return 65 ("LATIN SMALL LETTER
|
||||
** E"). The resuls of passing a codepoint that corresponds to an
|
||||
** uppercase letter are undefined.
|
||||
*/
|
||||
static int remove_diacritic(int c){
|
||||
unsigned short aDia[] = {
|
||||
0, 1797, 1848, 1859, 1891, 1928, 1940, 1995,
|
||||
2024, 2040, 2060, 2110, 2168, 2206, 2264, 2286,
|
||||
2344, 2383, 2472, 2488, 2516, 2596, 2668, 2732,
|
||||
2782, 2842, 2894, 2954, 2984, 3000, 3028, 3336,
|
||||
3456, 3696, 3712, 3728, 3744, 3896, 3912, 3928,
|
||||
3968, 4008, 4040, 4106, 4138, 4170, 4202, 4234,
|
||||
4266, 4296, 4312, 4344, 4408, 4424, 4472, 4504,
|
||||
6148, 6198, 6264, 6280, 6360, 6429, 6505, 6529,
|
||||
61448, 61468, 61534, 61592, 61642, 61688, 61704, 61726,
|
||||
61784, 61800, 61836, 61880, 61914, 61948, 61998, 62122,
|
||||
62154, 62200, 62218, 62302, 62364, 62442, 62478, 62536,
|
||||
62554, 62584, 62604, 62640, 62648, 62656, 62664, 62730,
|
||||
62924, 63050, 63082, 63274, 63390,
|
||||
};
|
||||
char aChar[] = {
|
||||
'\0', 'a', 'c', 'e', 'i', 'n', 'o', 'u', 'y', 'y', 'a', 'c',
|
||||
'd', 'e', 'e', 'g', 'h', 'i', 'j', 'k', 'l', 'n', 'o', 'r',
|
||||
's', 't', 'u', 'u', 'w', 'y', 'z', 'o', 'u', 'a', 'i', 'o',
|
||||
'u', 'g', 'k', 'o', 'j', 'g', 'n', 'a', 'e', 'i', 'o', 'r',
|
||||
'u', 's', 't', 'h', 'a', 'e', 'o', 'y', '\0', '\0', '\0', '\0',
|
||||
'\0', '\0', '\0', '\0', 'a', 'b', 'd', 'd', 'e', 'f', 'g', 'h',
|
||||
'h', 'i', 'k', 'l', 'l', 'm', 'n', 'p', 'r', 'r', 's', 't',
|
||||
'u', 'v', 'w', 'w', 'x', 'y', 'z', 'h', 't', 'w', 'y', 'a',
|
||||
'e', 'i', 'o', 'u', 'y',
|
||||
};
|
||||
|
||||
unsigned int key = (((unsigned int)c)<<3) | 0x00000007;
|
||||
int iRes = 0;
|
||||
int iHi = sizeof(aDia)/sizeof(aDia[0]) - 1;
|
||||
int iLo = 0;
|
||||
while( iHi>=iLo ){
|
||||
int iTest = (iHi + iLo) / 2;
|
||||
if( key >= aDia[iTest] ){
|
||||
iRes = iTest;
|
||||
iLo = iTest+1;
|
||||
}else{
|
||||
iHi = iTest-1;
|
||||
}
|
||||
}
|
||||
assert( key>=aDia[iRes] );
|
||||
return ((c > (aDia[iRes]>>3) + (aDia[iRes]&0x07)) ? c : (int)aChar[iRes]);
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Return true if the argument interpreted as a unicode codepoint
|
||||
** is a diacritical modifier character.
|
||||
*/
|
||||
int sqlite3FtsUnicodeIsdiacritic(int c){
|
||||
unsigned int mask0 = 0x08029FDF;
|
||||
unsigned int mask1 = 0x000361F8;
|
||||
if( c<768 || c>817 ) return 0;
|
||||
return (c < 768+32) ?
|
||||
(mask0 & (1 << (c-768))) :
|
||||
(mask1 & (1 << (c-768-32)));
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Interpret the argument as a unicode codepoint. If the codepoint
|
||||
** is an upper case character that has a lower case equivalent,
|
||||
** return the codepoint corresponding to the lower case version.
|
||||
** Otherwise, return a copy of the argument.
|
||||
**
|
||||
** The results are undefined if the value passed to this function
|
||||
** is less than zero.
|
||||
*/
|
||||
int sqlite3FtsUnicodeFold(int c, int bRemoveDiacritic){
|
||||
/* Each entry in the following array defines a rule for folding a range
|
||||
** of codepoints to lower case. The rule applies to a range of nRange
|
||||
** codepoints starting at codepoint iCode.
|
||||
**
|
||||
** If the least significant bit in flags is clear, then the rule applies
|
||||
** to all nRange codepoints (i.e. all nRange codepoints are upper case and
|
||||
** need to be folded). Or, if it is set, then the rule only applies to
|
||||
** every second codepoint in the range, starting with codepoint C.
|
||||
**
|
||||
** The 7 most significant bits in flags are an index into the aiOff[]
|
||||
** array. If a specific codepoint C does require folding, then its lower
|
||||
** case equivalent is ((C + aiOff[flags>>1]) & 0xFFFF).
|
||||
**
|
||||
** The contents of this array are generated by parsing the CaseFolding.txt
|
||||
** file distributed as part of the "Unicode Character Database". See
|
||||
** http://www.unicode.org for details.
|
||||
*/
|
||||
static const struct TableEntry {
|
||||
unsigned short iCode;
|
||||
unsigned char flags;
|
||||
unsigned char nRange;
|
||||
} aEntry[] = {
|
||||
{65, 14, 26}, {181, 64, 1}, {192, 14, 23},
|
||||
{216, 14, 7}, {256, 1, 48}, {306, 1, 6},
|
||||
{313, 1, 16}, {330, 1, 46}, {376, 116, 1},
|
||||
{377, 1, 6}, {383, 104, 1}, {385, 50, 1},
|
||||
{386, 1, 4}, {390, 44, 1}, {391, 0, 1},
|
||||
{393, 42, 2}, {395, 0, 1}, {398, 32, 1},
|
||||
{399, 38, 1}, {400, 40, 1}, {401, 0, 1},
|
||||
{403, 42, 1}, {404, 46, 1}, {406, 52, 1},
|
||||
{407, 48, 1}, {408, 0, 1}, {412, 52, 1},
|
||||
{413, 54, 1}, {415, 56, 1}, {416, 1, 6},
|
||||
{422, 60, 1}, {423, 0, 1}, {425, 60, 1},
|
||||
{428, 0, 1}, {430, 60, 1}, {431, 0, 1},
|
||||
{433, 58, 2}, {435, 1, 4}, {439, 62, 1},
|
||||
{440, 0, 1}, {444, 0, 1}, {452, 2, 1},
|
||||
{453, 0, 1}, {455, 2, 1}, {456, 0, 1},
|
||||
{458, 2, 1}, {459, 1, 18}, {478, 1, 18},
|
||||
{497, 2, 1}, {498, 1, 4}, {502, 122, 1},
|
||||
{503, 134, 1}, {504, 1, 40}, {544, 110, 1},
|
||||
{546, 1, 18}, {570, 70, 1}, {571, 0, 1},
|
||||
{573, 108, 1}, {574, 68, 1}, {577, 0, 1},
|
||||
{579, 106, 1}, {580, 28, 1}, {581, 30, 1},
|
||||
{582, 1, 10}, {837, 36, 1}, {880, 1, 4},
|
||||
{886, 0, 1}, {902, 18, 1}, {904, 16, 3},
|
||||
{908, 26, 1}, {910, 24, 2}, {913, 14, 17},
|
||||
{931, 14, 9}, {962, 0, 1}, {975, 4, 1},
|
||||
{976, 140, 1}, {977, 142, 1}, {981, 146, 1},
|
||||
{982, 144, 1}, {984, 1, 24}, {1008, 136, 1},
|
||||
{1009, 138, 1}, {1012, 130, 1}, {1013, 128, 1},
|
||||
{1015, 0, 1}, {1017, 152, 1}, {1018, 0, 1},
|
||||
{1021, 110, 3}, {1024, 34, 16}, {1040, 14, 32},
|
||||
{1120, 1, 34}, {1162, 1, 54}, {1216, 6, 1},
|
||||
{1217, 1, 14}, {1232, 1, 88}, {1329, 22, 38},
|
||||
{4256, 66, 38}, {4295, 66, 1}, {4301, 66, 1},
|
||||
{7680, 1, 150}, {7835, 132, 1}, {7838, 96, 1},
|
||||
{7840, 1, 96}, {7944, 150, 8}, {7960, 150, 6},
|
||||
{7976, 150, 8}, {7992, 150, 8}, {8008, 150, 6},
|
||||
{8025, 151, 8}, {8040, 150, 8}, {8072, 150, 8},
|
||||
{8088, 150, 8}, {8104, 150, 8}, {8120, 150, 2},
|
||||
{8122, 126, 2}, {8124, 148, 1}, {8126, 100, 1},
|
||||
{8136, 124, 4}, {8140, 148, 1}, {8152, 150, 2},
|
||||
{8154, 120, 2}, {8168, 150, 2}, {8170, 118, 2},
|
||||
{8172, 152, 1}, {8184, 112, 2}, {8186, 114, 2},
|
||||
{8188, 148, 1}, {8486, 98, 1}, {8490, 92, 1},
|
||||
{8491, 94, 1}, {8498, 12, 1}, {8544, 8, 16},
|
||||
{8579, 0, 1}, {9398, 10, 26}, {11264, 22, 47},
|
||||
{11360, 0, 1}, {11362, 88, 1}, {11363, 102, 1},
|
||||
{11364, 90, 1}, {11367, 1, 6}, {11373, 84, 1},
|
||||
{11374, 86, 1}, {11375, 80, 1}, {11376, 82, 1},
|
||||
{11378, 0, 1}, {11381, 0, 1}, {11390, 78, 2},
|
||||
{11392, 1, 100}, {11499, 1, 4}, {11506, 0, 1},
|
||||
{42560, 1, 46}, {42624, 1, 24}, {42786, 1, 14},
|
||||
{42802, 1, 62}, {42873, 1, 4}, {42877, 76, 1},
|
||||
{42878, 1, 10}, {42891, 0, 1}, {42893, 74, 1},
|
||||
{42896, 1, 4}, {42912, 1, 10}, {42922, 72, 1},
|
||||
{65313, 14, 26},
|
||||
};
|
||||
static const unsigned short aiOff[] = {
|
||||
1, 2, 8, 15, 16, 26, 28, 32,
|
||||
37, 38, 40, 48, 63, 64, 69, 71,
|
||||
79, 80, 116, 202, 203, 205, 206, 207,
|
||||
209, 210, 211, 213, 214, 217, 218, 219,
|
||||
775, 7264, 10792, 10795, 23228, 23256, 30204, 54721,
|
||||
54753, 54754, 54756, 54787, 54793, 54809, 57153, 57274,
|
||||
57921, 58019, 58363, 61722, 65268, 65341, 65373, 65406,
|
||||
65408, 65410, 65415, 65424, 65436, 65439, 65450, 65462,
|
||||
65472, 65476, 65478, 65480, 65482, 65488, 65506, 65511,
|
||||
65514, 65521, 65527, 65528, 65529,
|
||||
};
|
||||
|
||||
int ret = c;
|
||||
|
||||
assert( c>=0 );
|
||||
assert( sizeof(unsigned short)==2 && sizeof(unsigned char)==1 );
|
||||
|
||||
if( c<128 ){
|
||||
if( c>='A' && c<='Z' ) ret = c + ('a' - 'A');
|
||||
}else if( c<65536 ){
|
||||
int iHi = sizeof(aEntry)/sizeof(aEntry[0]) - 1;
|
||||
int iLo = 0;
|
||||
int iRes = -1;
|
||||
|
||||
while( iHi>=iLo ){
|
||||
int iTest = (iHi + iLo) / 2;
|
||||
int cmp = (c - aEntry[iTest].iCode);
|
||||
if( cmp>=0 ){
|
||||
iRes = iTest;
|
||||
iLo = iTest+1;
|
||||
}else{
|
||||
iHi = iTest-1;
|
||||
}
|
||||
}
|
||||
assert( iRes<0 || c>=aEntry[iRes].iCode );
|
||||
|
||||
if( iRes>=0 ){
|
||||
const struct TableEntry *p = &aEntry[iRes];
|
||||
if( c<(p->iCode + p->nRange) && 0==(0x01 & p->flags & (p->iCode ^ c)) ){
|
||||
ret = (c + (aiOff[p->flags>>1])) & 0x0000FFFF;
|
||||
assert( ret>0 );
|
||||
}
|
||||
}
|
||||
|
||||
if( bRemoveDiacritic ) ret = remove_diacritic(ret);
|
||||
}
|
||||
|
||||
else if( c>=66560 && c<66600 ){
|
||||
ret = c + 40;
|
||||
}
|
||||
|
||||
return ret;
|
||||
}
|
||||
#endif /* defined(SQLITE_ENABLE_FTS3) || defined(SQLITE_ENABLE_FTS4) */
|
||||
#endif /* !defined(SQLITE_DISABLE_FTS3_UNICODE) */
|
||||
5676
query_classifier/qc_sqlite/sqlite-src-3110100/ext/fts3/fts3_write.c
Normal file
5676
query_classifier/qc_sqlite/sqlite-src-3110100/ext/fts3/fts3_write.c
Normal file
File diff suppressed because it is too large
Load Diff
@ -0,0 +1,122 @@
|
||||
|
||||
|
||||
#--------------------------------------------------------------------------
|
||||
# This script contains several sub-programs used to test FTS3/FTS4
|
||||
# performance. It does not run the queries directly, but generates SQL
|
||||
# scripts that can be run using the shell tool.
|
||||
#
|
||||
# The following cases are tested:
|
||||
#
|
||||
# 1. Inserting documents into an FTS3 table.
|
||||
# 2. Optimizing an FTS3 table (i.e. "INSERT INTO t1 VALUES('optimize')").
|
||||
# 3. Deleting documents from an FTS3 table.
|
||||
# 4. Querying FTS3 tables.
|
||||
#
|
||||
|
||||
# Number of tokens in vocabulary. And number of tokens in each document.
|
||||
#
|
||||
set VOCAB_SIZE 2000
|
||||
set DOC_SIZE 100
|
||||
|
||||
set NUM_INSERTS 100000
|
||||
set NUM_SELECTS 1000
|
||||
|
||||
# Force everything in this script to be deterministic.
|
||||
#
|
||||
expr {srand(0)}
|
||||
|
||||
proc usage {} {
|
||||
puts stderr "Usage: $::argv0 <rows> <selects>"
|
||||
exit -1
|
||||
}
|
||||
|
||||
proc sql {sql} {
|
||||
puts $::fd $sql
|
||||
}
|
||||
|
||||
|
||||
# Return a list of $nWord randomly generated tokens each between 2 and 10
|
||||
# characters in length.
|
||||
#
|
||||
proc build_vocab {nWord} {
|
||||
set ret [list]
|
||||
set chars [list a b c d e f g h i j k l m n o p q r s t u v w x y z]
|
||||
for {set i 0} {$i<$nWord} {incr i} {
|
||||
set len [expr {int((rand()*9.0)+2)}]
|
||||
set term ""
|
||||
for {set j 0} {$j<$len} {incr j} {
|
||||
append term [lindex $chars [expr {int(rand()*[llength $chars])}]]
|
||||
}
|
||||
lappend ret $term
|
||||
}
|
||||
set ret
|
||||
}
|
||||
|
||||
proc select_term {} {
|
||||
set n [llength $::vocab]
|
||||
set t [expr int(rand()*$n*3)]
|
||||
if {$t>=2*$n} { set t [expr {($t-2*$n)/100}] }
|
||||
if {$t>=$n} { set t [expr {($t-$n)/10}] }
|
||||
lindex $::vocab $t
|
||||
}
|
||||
|
||||
proc select_doc {nTerm} {
|
||||
set ret [list]
|
||||
for {set i 0} {$i<$nTerm} {incr i} {
|
||||
lappend ret [select_term]
|
||||
}
|
||||
set ret
|
||||
}
|
||||
|
||||
proc test_1 {nInsert} {
|
||||
sql "PRAGMA synchronous = OFF;"
|
||||
sql "DROP TABLE IF EXISTS t1;"
|
||||
sql "CREATE VIRTUAL TABLE t1 USING fts4;"
|
||||
for {set i 0} {$i < $nInsert} {incr i} {
|
||||
set doc [select_doc $::DOC_SIZE]
|
||||
sql "INSERT INTO t1 VALUES('$doc');"
|
||||
}
|
||||
}
|
||||
|
||||
proc test_2 {} {
|
||||
sql "INSERT INTO t1(t1) VALUES('optimize');"
|
||||
}
|
||||
|
||||
proc test_3 {nSelect} {
|
||||
for {set i 0} {$i < $nSelect} {incr i} {
|
||||
sql "SELECT count(*) FROM t1 WHERE t1 MATCH '[select_term]';"
|
||||
}
|
||||
}
|
||||
|
||||
proc test_4 {nSelect} {
|
||||
for {set i 0} {$i < $nSelect} {incr i} {
|
||||
sql "SELECT count(*) FROM t1 WHERE t1 MATCH '[select_term] [select_term]';"
|
||||
}
|
||||
}
|
||||
|
||||
if {[llength $argv]!=0} usage
|
||||
|
||||
set ::vocab [build_vocab $::VOCAB_SIZE]
|
||||
|
||||
set ::fd [open fts3speed_insert.sql w]
|
||||
test_1 $NUM_INSERTS
|
||||
close $::fd
|
||||
|
||||
set ::fd [open fts3speed_select.sql w]
|
||||
test_3 $NUM_SELECTS
|
||||
close $::fd
|
||||
|
||||
set ::fd [open fts3speed_select2.sql w]
|
||||
test_4 $NUM_SELECTS
|
||||
close $::fd
|
||||
|
||||
set ::fd [open fts3speed_optimize.sql w]
|
||||
test_2
|
||||
close $::fd
|
||||
|
||||
puts "Success. Created files:"
|
||||
puts " fts3speed_insert.sql"
|
||||
puts " fts3speed_select.sql"
|
||||
puts " fts3speed_select2.sql"
|
||||
puts " fts3speed_optimize.sql"
|
||||
|
||||
@ -0,0 +1,115 @@
|
||||
#!/usr/bin/tclsh
|
||||
#
|
||||
# This script builds a single C code file holding all of FTS3 code.
|
||||
# The name of the output file is fts3amal.c. To build this file,
|
||||
# first do:
|
||||
#
|
||||
# make target_source
|
||||
#
|
||||
# The make target above moves all of the source code files into
|
||||
# a subdirectory named "tsrc". (This script expects to find the files
|
||||
# there and will not work if they are not found.)
|
||||
#
|
||||
# After the "tsrc" directory has been created and populated, run
|
||||
# this script:
|
||||
#
|
||||
# tclsh mkfts3amal.tcl
|
||||
#
|
||||
# The amalgamated FTS3 code will be written into fts3amal.c
|
||||
#
|
||||
|
||||
# Open the output file and write a header comment at the beginning
|
||||
# of the file.
|
||||
#
|
||||
set out [open fts3amal.c w]
|
||||
set today [clock format [clock seconds] -format "%Y-%m-%d %H:%M:%S UTC" -gmt 1]
|
||||
puts $out [subst \
|
||||
{/******************************************************************************
|
||||
** This file is an amalgamation of separate C source files from the SQLite
|
||||
** Full Text Search extension 2 (fts3). By combining all the individual C
|
||||
** code files into this single large file, the entire code can be compiled
|
||||
** as a one translation unit. This allows many compilers to do optimizations
|
||||
** that would not be possible if the files were compiled separately. It also
|
||||
** makes the code easier to import into other projects.
|
||||
**
|
||||
** This amalgamation was generated on $today.
|
||||
*/}]
|
||||
|
||||
# These are the header files used by FTS3. The first time any of these
|
||||
# files are seen in a #include statement in the C code, include the complete
|
||||
# text of the file in-line. The file only needs to be included once.
|
||||
#
|
||||
foreach hdr {
|
||||
fts3.h
|
||||
fts3_hash.h
|
||||
fts3_tokenizer.h
|
||||
sqlite3.h
|
||||
sqlite3ext.h
|
||||
} {
|
||||
set available_hdr($hdr) 1
|
||||
}
|
||||
|
||||
# 78 stars used for comment formatting.
|
||||
set s78 \
|
||||
{*****************************************************************************}
|
||||
|
||||
# Insert a comment into the code
|
||||
#
|
||||
proc section_comment {text} {
|
||||
global out s78
|
||||
set n [string length $text]
|
||||
set nstar [expr {60 - $n}]
|
||||
set stars [string range $s78 0 $nstar]
|
||||
puts $out "/************** $text $stars/"
|
||||
}
|
||||
|
||||
# Read the source file named $filename and write it into the
|
||||
# sqlite3.c output file. If any #include statements are seen,
|
||||
# process them approprately.
|
||||
#
|
||||
proc copy_file {filename} {
|
||||
global seen_hdr available_hdr out
|
||||
set tail [file tail $filename]
|
||||
section_comment "Begin file $tail"
|
||||
set in [open $filename r]
|
||||
while {![eof $in]} {
|
||||
set line [gets $in]
|
||||
if {[regexp {^#\s*include\s+["<]([^">]+)[">]} $line all hdr]} {
|
||||
if {[info exists available_hdr($hdr)]} {
|
||||
if {$available_hdr($hdr)} {
|
||||
section_comment "Include $hdr in the middle of $tail"
|
||||
copy_file tsrc/$hdr
|
||||
section_comment "Continuing where we left off in $tail"
|
||||
}
|
||||
} elseif {![info exists seen_hdr($hdr)]} {
|
||||
set seen_hdr($hdr) 1
|
||||
puts $out $line
|
||||
}
|
||||
} elseif {[regexp {^#ifdef __cplusplus} $line]} {
|
||||
puts $out "#if 0"
|
||||
} elseif {[regexp {^#line} $line]} {
|
||||
# Skip #line directives.
|
||||
} else {
|
||||
puts $out $line
|
||||
}
|
||||
}
|
||||
close $in
|
||||
section_comment "End of $tail"
|
||||
}
|
||||
|
||||
|
||||
# Process the source files. Process files containing commonly
|
||||
# used subroutines first in order to help the compiler find
|
||||
# inlining opportunities.
|
||||
#
|
||||
foreach file {
|
||||
fts3.c
|
||||
fts3_hash.c
|
||||
fts3_porter.c
|
||||
fts3_tokenizer.c
|
||||
fts3_tokenizer1.c
|
||||
} {
|
||||
copy_file tsrc/$file
|
||||
}
|
||||
|
||||
close $out
|
||||
@ -0,0 +1,875 @@
|
||||
/*
|
||||
** This program is a debugging and analysis utility that displays
|
||||
** information about an FTS3 or FTS4 index.
|
||||
**
|
||||
** Link this program against the SQLite3 amalgamation with the
|
||||
** SQLITE_ENABLE_FTS4 compile-time option. Then run it as:
|
||||
**
|
||||
** fts3view DATABASE
|
||||
**
|
||||
** to get a list of all FTS3/4 tables in DATABASE, or do
|
||||
**
|
||||
** fts3view DATABASE TABLE COMMAND ....
|
||||
**
|
||||
** to see various aspects of the TABLE table. Type fts3view with no
|
||||
** arguments for a list of available COMMANDs.
|
||||
*/
|
||||
#include <stdio.h>
|
||||
#include <stdarg.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
#include <ctype.h>
|
||||
#include "sqlite3.h"
|
||||
|
||||
/*
|
||||
** Extra command-line arguments:
|
||||
*/
|
||||
int nExtra;
|
||||
char **azExtra;
|
||||
|
||||
/*
|
||||
** Look for a command-line argument.
|
||||
*/
|
||||
const char *findOption(const char *zName, int hasArg, const char *zDefault){
|
||||
int i;
|
||||
const char *zResult = zDefault;
|
||||
for(i=0; i<nExtra; i++){
|
||||
const char *z = azExtra[i];
|
||||
while( z[0]=='-' ) z++;
|
||||
if( strcmp(z, zName)==0 ){
|
||||
int j = 1;
|
||||
if( hasArg==0 || i==nExtra-1 ) j = 0;
|
||||
zResult = azExtra[i+j];
|
||||
while( i+j<nExtra ){
|
||||
azExtra[i] = azExtra[i+j+1];
|
||||
i++;
|
||||
}
|
||||
break;
|
||||
}
|
||||
}
|
||||
return zResult;
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Prepare an SQL query
|
||||
*/
|
||||
static sqlite3_stmt *prepare(sqlite3 *db, const char *zFormat, ...){
|
||||
va_list ap;
|
||||
char *zSql;
|
||||
sqlite3_stmt *pStmt;
|
||||
int rc;
|
||||
|
||||
va_start(ap, zFormat);
|
||||
zSql = sqlite3_vmprintf(zFormat, ap);
|
||||
va_end(ap);
|
||||
rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, 0);
|
||||
if( rc ){
|
||||
fprintf(stderr, "Error: %s\nSQL: %s\n", sqlite3_errmsg(db), zSql);
|
||||
exit(1);
|
||||
}
|
||||
sqlite3_free(zSql);
|
||||
return pStmt;
|
||||
}
|
||||
|
||||
/*
|
||||
** Run an SQL statement
|
||||
*/
|
||||
static int runSql(sqlite3 *db, const char *zFormat, ...){
|
||||
va_list ap;
|
||||
char *zSql;
|
||||
int rc;
|
||||
|
||||
va_start(ap, zFormat);
|
||||
zSql = sqlite3_vmprintf(zFormat, ap);
|
||||
rc = sqlite3_exec(db, zSql, 0, 0, 0);
|
||||
va_end(ap);
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** Show the table schema
|
||||
*/
|
||||
static void showSchema(sqlite3 *db, const char *zTab){
|
||||
sqlite3_stmt *pStmt;
|
||||
pStmt = prepare(db,
|
||||
"SELECT sql FROM sqlite_master"
|
||||
" WHERE name LIKE '%q%%'"
|
||||
" ORDER BY 1",
|
||||
zTab);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
printf("%s;\n", sqlite3_column_text(pStmt, 0));
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
pStmt = prepare(db, "PRAGMA page_size");
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
printf("PRAGMA page_size=%s;\n", sqlite3_column_text(pStmt, 0));
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
pStmt = prepare(db, "PRAGMA journal_mode");
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
printf("PRAGMA journal_mode=%s;\n", sqlite3_column_text(pStmt, 0));
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
pStmt = prepare(db, "PRAGMA auto_vacuum");
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
const char *zType = "???";
|
||||
switch( sqlite3_column_int(pStmt, 0) ){
|
||||
case 0: zType = "OFF"; break;
|
||||
case 1: zType = "FULL"; break;
|
||||
case 2: zType = "INCREMENTAL"; break;
|
||||
}
|
||||
printf("PRAGMA auto_vacuum=%s;\n", zType);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
pStmt = prepare(db, "PRAGMA encoding");
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
printf("PRAGMA encoding=%s;\n", sqlite3_column_text(pStmt, 0));
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
}
|
||||
|
||||
/*
|
||||
** Read a 64-bit variable-length integer from memory starting at p[0].
|
||||
** Return the number of bytes read, or 0 on error.
|
||||
** The value is stored in *v.
|
||||
*/
|
||||
int getVarint(const unsigned char *p, sqlite_int64 *v){
|
||||
const unsigned char *q = p;
|
||||
sqlite_uint64 x = 0, y = 1;
|
||||
while( (*q&0x80)==0x80 && q-(unsigned char *)p<9 ){
|
||||
x += y * (*q++ & 0x7f);
|
||||
y <<= 7;
|
||||
}
|
||||
x += y * (*q++);
|
||||
*v = (sqlite_int64) x;
|
||||
return (int) (q - (unsigned char *)p);
|
||||
}
|
||||
|
||||
|
||||
/* Show the content of the %_stat table
|
||||
*/
|
||||
static void showStat(sqlite3 *db, const char *zTab){
|
||||
sqlite3_stmt *pStmt;
|
||||
pStmt = prepare(db, "SELECT id, value FROM '%q_stat'", zTab);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
printf("stat[%d] =", sqlite3_column_int(pStmt, 0));
|
||||
switch( sqlite3_column_type(pStmt, 1) ){
|
||||
case SQLITE_INTEGER: {
|
||||
printf(" %d\n", sqlite3_column_int(pStmt, 1));
|
||||
break;
|
||||
}
|
||||
case SQLITE_BLOB: {
|
||||
unsigned char *x = (unsigned char*)sqlite3_column_blob(pStmt, 1);
|
||||
int len = sqlite3_column_bytes(pStmt, 1);
|
||||
int i = 0;
|
||||
sqlite3_int64 v;
|
||||
while( i<len ){
|
||||
i += getVarint(x, &v);
|
||||
printf(" %lld", v);
|
||||
}
|
||||
printf("\n");
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
}
|
||||
|
||||
/*
|
||||
** Report on the vocabulary. This creates an fts4aux table with a random
|
||||
** name, but deletes it in the end.
|
||||
*/
|
||||
static void showVocabulary(sqlite3 *db, const char *zTab){
|
||||
char *zAux;
|
||||
sqlite3_uint64 r;
|
||||
sqlite3_stmt *pStmt;
|
||||
int nDoc = 0;
|
||||
int nToken = 0;
|
||||
int nOccurrence = 0;
|
||||
int nTop;
|
||||
int n, i;
|
||||
|
||||
sqlite3_randomness(sizeof(r), &r);
|
||||
zAux = sqlite3_mprintf("viewer_%llx", zTab, r);
|
||||
runSql(db, "BEGIN");
|
||||
pStmt = prepare(db, "SELECT count(*) FROM %Q", zTab);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
nDoc = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
printf("Number of documents...................... %9d\n", nDoc);
|
||||
|
||||
runSql(db, "CREATE VIRTUAL TABLE %s USING fts4aux(%Q)", zAux, zTab);
|
||||
pStmt = prepare(db,
|
||||
"SELECT count(*), sum(occurrences) FROM %s WHERE col='*'",
|
||||
zAux);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
nToken = sqlite3_column_int(pStmt, 0);
|
||||
nOccurrence = sqlite3_column_int(pStmt, 1);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
printf("Total tokens in all documents............ %9d\n", nOccurrence);
|
||||
printf("Total number of distinct tokens.......... %9d\n", nToken);
|
||||
if( nToken==0 ) goto end_vocab;
|
||||
|
||||
n = 0;
|
||||
pStmt = prepare(db, "SELECT count(*) FROM %s"
|
||||
" WHERE col='*' AND occurrences==1", zAux);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
n = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
printf("Tokens used exactly once................. %9d %5.2f%%\n",
|
||||
n, n*100.0/nToken);
|
||||
|
||||
n = 0;
|
||||
pStmt = prepare(db, "SELECT count(*) FROM %s"
|
||||
" WHERE col='*' AND documents==1", zAux);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
n = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
printf("Tokens used in only one document......... %9d %5.2f%%\n",
|
||||
n, n*100.0/nToken);
|
||||
|
||||
if( nDoc>=2000 ){
|
||||
n = 0;
|
||||
pStmt = prepare(db, "SELECT count(*) FROM %s"
|
||||
" WHERE col='*' AND occurrences<=%d", zAux, nDoc/1000);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
n = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
printf("Tokens used in 0.1%% or less of docs...... %9d %5.2f%%\n",
|
||||
n, n*100.0/nToken);
|
||||
}
|
||||
|
||||
if( nDoc>=200 ){
|
||||
n = 0;
|
||||
pStmt = prepare(db, "SELECT count(*) FROM %s"
|
||||
" WHERE col='*' AND occurrences<=%d", zAux, nDoc/100);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
n = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
printf("Tokens used in 1%% or less of docs........ %9d %5.2f%%\n",
|
||||
n, n*100.0/nToken);
|
||||
}
|
||||
|
||||
nTop = atoi(findOption("top", 1, "25"));
|
||||
printf("The %d most common tokens:\n", nTop);
|
||||
pStmt = prepare(db,
|
||||
"SELECT term, documents FROM %s"
|
||||
" WHERE col='*'"
|
||||
" ORDER BY documents DESC, term"
|
||||
" LIMIT %d", zAux, nTop);
|
||||
i = 0;
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
i++;
|
||||
n = sqlite3_column_int(pStmt, 1);
|
||||
printf(" %2d. %-30s %9d docs %5.2f%%\n", i,
|
||||
sqlite3_column_text(pStmt, 0), n, n*100.0/nDoc);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
|
||||
end_vocab:
|
||||
runSql(db, "ROLLBACK");
|
||||
sqlite3_free(zAux);
|
||||
}
|
||||
|
||||
/*
|
||||
** Report on the number and sizes of segments
|
||||
*/
|
||||
static void showSegmentStats(sqlite3 *db, const char *zTab){
|
||||
sqlite3_stmt *pStmt;
|
||||
int nSeg = 0;
|
||||
sqlite3_int64 szSeg = 0, mxSeg = 0;
|
||||
int nIdx = 0;
|
||||
sqlite3_int64 szIdx = 0, mxIdx = 0;
|
||||
int nRoot = 0;
|
||||
sqlite3_int64 szRoot = 0, mxRoot = 0;
|
||||
sqlite3_int64 mx;
|
||||
int nLeaf;
|
||||
int n;
|
||||
int pgsz;
|
||||
int mxLevel;
|
||||
int i;
|
||||
|
||||
pStmt = prepare(db,
|
||||
"SELECT count(*), sum(length(block)), max(length(block))"
|
||||
" FROM '%q_segments'",
|
||||
zTab);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
nSeg = sqlite3_column_int(pStmt, 0);
|
||||
szSeg = sqlite3_column_int64(pStmt, 1);
|
||||
mxSeg = sqlite3_column_int64(pStmt, 2);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
pStmt = prepare(db,
|
||||
"SELECT count(*), sum(length(block)), max(length(block))"
|
||||
" FROM '%q_segments' a JOIN '%q_segdir' b"
|
||||
" WHERE a.blockid BETWEEN b.leaves_end_block+1 AND b.end_block",
|
||||
zTab, zTab);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
nIdx = sqlite3_column_int(pStmt, 0);
|
||||
szIdx = sqlite3_column_int64(pStmt, 1);
|
||||
mxIdx = sqlite3_column_int64(pStmt, 2);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
pStmt = prepare(db,
|
||||
"SELECT count(*), sum(length(root)), max(length(root))"
|
||||
" FROM '%q_segdir'",
|
||||
zTab);
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
nRoot = sqlite3_column_int(pStmt, 0);
|
||||
szRoot = sqlite3_column_int64(pStmt, 1);
|
||||
mxRoot = sqlite3_column_int64(pStmt, 2);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
|
||||
printf("Number of segments....................... %9d\n", nSeg+nRoot);
|
||||
printf("Number of leaf segments.................. %9d\n", nSeg-nIdx);
|
||||
printf("Number of index segments................. %9d\n", nIdx);
|
||||
printf("Number of root segments.................. %9d\n", nRoot);
|
||||
printf("Total size of all segments............... %9lld\n", szSeg+szRoot);
|
||||
printf("Total size of all leaf segments.......... %9lld\n", szSeg-szIdx);
|
||||
printf("Total size of all index segments......... %9lld\n", szIdx);
|
||||
printf("Total size of all root segments.......... %9lld\n", szRoot);
|
||||
if( nSeg>0 ){
|
||||
printf("Average size of all segments............. %11.1f\n",
|
||||
(double)(szSeg+szRoot)/(double)(nSeg+nRoot));
|
||||
printf("Average size of leaf segments............ %11.1f\n",
|
||||
(double)(szSeg-szIdx)/(double)(nSeg-nIdx));
|
||||
}
|
||||
if( nIdx>0 ){
|
||||
printf("Average size of index segments........... %11.1f\n",
|
||||
(double)szIdx/(double)nIdx);
|
||||
}
|
||||
if( nRoot>0 ){
|
||||
printf("Average size of root segments............ %11.1f\n",
|
||||
(double)szRoot/(double)nRoot);
|
||||
}
|
||||
mx = mxSeg;
|
||||
if( mx<mxRoot ) mx = mxRoot;
|
||||
printf("Maximum segment size..................... %9lld\n", mx);
|
||||
printf("Maximum index segment size............... %9lld\n", mxIdx);
|
||||
printf("Maximum root segment size................ %9lld\n", mxRoot);
|
||||
|
||||
pStmt = prepare(db, "PRAGMA page_size");
|
||||
pgsz = 1024;
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
pgsz = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
printf("Database page size....................... %9d\n", pgsz);
|
||||
pStmt = prepare(db,
|
||||
"SELECT count(*)"
|
||||
" FROM '%q_segments' a JOIN '%q_segdir' b"
|
||||
" WHERE a.blockid BETWEEN b.start_block AND b.leaves_end_block"
|
||||
" AND length(a.block)>%d",
|
||||
zTab, zTab, pgsz-45);
|
||||
n = 0;
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
n = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
nLeaf = nSeg - nIdx;
|
||||
printf("Leaf segments larger than %5d bytes.... %9d %5.2f%%\n",
|
||||
pgsz-45, n, nLeaf>0 ? n*100.0/nLeaf : 0.0);
|
||||
|
||||
pStmt = prepare(db, "SELECT max(level%%1024) FROM '%q_segdir'", zTab);
|
||||
mxLevel = 0;
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
mxLevel = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
|
||||
for(i=0; i<=mxLevel; i++){
|
||||
pStmt = prepare(db,
|
||||
"SELECT count(*), sum(len), avg(len), max(len), sum(len>%d),"
|
||||
" count(distinct idx)"
|
||||
" FROM (SELECT length(a.block) AS len, idx"
|
||||
" FROM '%q_segments' a JOIN '%q_segdir' b"
|
||||
" WHERE (a.blockid BETWEEN b.start_block"
|
||||
" AND b.leaves_end_block)"
|
||||
" AND (b.level%%1024)==%d)",
|
||||
pgsz-45, zTab, zTab, i);
|
||||
if( sqlite3_step(pStmt)==SQLITE_ROW
|
||||
&& (nLeaf = sqlite3_column_int(pStmt, 0))>0
|
||||
){
|
||||
nIdx = sqlite3_column_int(pStmt, 5);
|
||||
sqlite3_int64 sz;
|
||||
printf("For level %d:\n", i);
|
||||
printf(" Number of indexes...................... %9d\n", nIdx);
|
||||
printf(" Number of leaf segments................ %9d\n", nLeaf);
|
||||
if( nIdx>1 ){
|
||||
printf(" Average leaf segments per index........ %11.1f\n",
|
||||
(double)nLeaf/(double)nIdx);
|
||||
}
|
||||
printf(" Total size of all leaf segments........ %9lld\n",
|
||||
(sz = sqlite3_column_int64(pStmt, 1)));
|
||||
printf(" Average size of leaf segments.......... %11.1f\n",
|
||||
sqlite3_column_double(pStmt, 2));
|
||||
if( nIdx>1 ){
|
||||
printf(" Average leaf segment size per index.... %11.1f\n",
|
||||
(double)sz/(double)nIdx);
|
||||
}
|
||||
printf(" Maximum leaf segment size.............. %9lld\n",
|
||||
sqlite3_column_int64(pStmt, 3));
|
||||
n = sqlite3_column_int(pStmt, 4);
|
||||
printf(" Leaf segments larger than %5d bytes.. %9d %5.2f%%\n",
|
||||
pgsz-45, n, n*100.0/nLeaf);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
** Print a single "tree" line of the segdir map output.
|
||||
*/
|
||||
static void printTreeLine(sqlite3_int64 iLower, sqlite3_int64 iUpper){
|
||||
printf(" tree %9lld", iLower);
|
||||
if( iUpper>iLower ){
|
||||
printf(" thru %9lld (%lld blocks)", iUpper, iUpper-iLower+1);
|
||||
}
|
||||
printf("\n");
|
||||
}
|
||||
|
||||
/*
|
||||
** Check to see if the block of a %_segments entry is NULL.
|
||||
*/
|
||||
static int isNullSegment(sqlite3 *db, const char *zTab, sqlite3_int64 iBlockId){
|
||||
sqlite3_stmt *pStmt;
|
||||
int rc = 1;
|
||||
|
||||
pStmt = prepare(db, "SELECT block IS NULL FROM '%q_segments'"
|
||||
" WHERE blockid=%lld", zTab, iBlockId);
|
||||
if( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
rc = sqlite3_column_int(pStmt, 0);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
return rc;
|
||||
}
|
||||
|
||||
/*
|
||||
** Show a map of segments derived from the %_segdir table.
|
||||
*/
|
||||
static void showSegdirMap(sqlite3 *db, const char *zTab){
|
||||
int mxIndex, iIndex;
|
||||
sqlite3_stmt *pStmt = 0;
|
||||
sqlite3_stmt *pStmt2 = 0;
|
||||
int prevLevel;
|
||||
|
||||
pStmt = prepare(db, "SELECT max(level/1024) FROM '%q_segdir'", zTab);
|
||||
if( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
mxIndex = sqlite3_column_int(pStmt, 0);
|
||||
}else{
|
||||
mxIndex = 0;
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
|
||||
printf("Number of inverted indices............... %3d\n", mxIndex+1);
|
||||
pStmt = prepare(db,
|
||||
"SELECT level, idx, start_block, leaves_end_block, end_block, rowid"
|
||||
" FROM '%q_segdir'"
|
||||
" WHERE level/1024==?"
|
||||
" ORDER BY level DESC, idx",
|
||||
zTab);
|
||||
pStmt2 = prepare(db,
|
||||
"SELECT blockid FROM '%q_segments'"
|
||||
" WHERE blockid BETWEEN ? AND ? ORDER BY blockid",
|
||||
zTab);
|
||||
for(iIndex=0; iIndex<=mxIndex; iIndex++){
|
||||
if( mxIndex>0 ){
|
||||
printf("**************************** Index %d "
|
||||
"****************************\n", iIndex);
|
||||
}
|
||||
sqlite3_bind_int(pStmt, 1, iIndex);
|
||||
prevLevel = -1;
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
int iLevel = sqlite3_column_int(pStmt, 0)%1024;
|
||||
int iIdx = sqlite3_column_int(pStmt, 1);
|
||||
sqlite3_int64 iStart = sqlite3_column_int64(pStmt, 2);
|
||||
sqlite3_int64 iLEnd = sqlite3_column_int64(pStmt, 3);
|
||||
sqlite3_int64 iEnd = sqlite3_column_int64(pStmt, 4);
|
||||
char rtag[20];
|
||||
if( iLevel!=prevLevel ){
|
||||
printf("level %2d idx %2d", iLevel, iIdx);
|
||||
prevLevel = iLevel;
|
||||
}else{
|
||||
printf(" idx %2d", iIdx);
|
||||
}
|
||||
sqlite3_snprintf(sizeof(rtag), rtag, "r%lld",
|
||||
sqlite3_column_int64(pStmt,5));
|
||||
printf(" root %9s\n", rtag);
|
||||
if( iLEnd>iStart ){
|
||||
sqlite3_int64 iLower, iPrev = 0, iX;
|
||||
if( iLEnd+1<=iEnd ){
|
||||
sqlite3_bind_int64(pStmt2, 1, iLEnd+1);
|
||||
sqlite3_bind_int64(pStmt2, 2, iEnd);
|
||||
iLower = -1;
|
||||
while( sqlite3_step(pStmt2)==SQLITE_ROW ){
|
||||
iX = sqlite3_column_int64(pStmt2, 0);
|
||||
if( iLower<0 ){
|
||||
iLower = iPrev = iX;
|
||||
}else if( iX==iPrev+1 ){
|
||||
iPrev = iX;
|
||||
}else{
|
||||
printTreeLine(iLower, iPrev);
|
||||
iLower = iPrev = iX;
|
||||
}
|
||||
}
|
||||
sqlite3_reset(pStmt2);
|
||||
if( iLower>=0 ){
|
||||
if( iLower==iPrev && iLower==iEnd
|
||||
&& isNullSegment(db,zTab,iLower)
|
||||
){
|
||||
printf(" null %9lld\n", iLower);
|
||||
}else{
|
||||
printTreeLine(iLower, iPrev);
|
||||
}
|
||||
}
|
||||
}
|
||||
printf(" leaves %9lld thru %9lld (%lld blocks)\n",
|
||||
iStart, iLEnd, iLEnd - iStart + 1);
|
||||
}
|
||||
}
|
||||
sqlite3_reset(pStmt);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
sqlite3_finalize(pStmt2);
|
||||
}
|
||||
|
||||
/*
|
||||
** Decode a single segment block and display the results on stdout.
|
||||
*/
|
||||
static void decodeSegment(
|
||||
const unsigned char *aData, /* Content to print */
|
||||
int nData /* Number of bytes of content */
|
||||
){
|
||||
sqlite3_int64 iChild = 0;
|
||||
sqlite3_int64 iPrefix;
|
||||
sqlite3_int64 nTerm;
|
||||
sqlite3_int64 n;
|
||||
sqlite3_int64 iDocsz;
|
||||
int iHeight;
|
||||
sqlite3_int64 i = 0;
|
||||
int cnt = 0;
|
||||
char zTerm[1000];
|
||||
|
||||
i += getVarint(aData, &n);
|
||||
iHeight = (int)n;
|
||||
printf("height: %d\n", iHeight);
|
||||
if( iHeight>0 ){
|
||||
i += getVarint(aData+i, &iChild);
|
||||
printf("left-child: %lld\n", iChild);
|
||||
}
|
||||
while( i<nData ){
|
||||
if( (cnt++)>0 ){
|
||||
i += getVarint(aData+i, &iPrefix);
|
||||
}else{
|
||||
iPrefix = 0;
|
||||
}
|
||||
i += getVarint(aData+i, &nTerm);
|
||||
if( iPrefix+nTerm+1 >= sizeof(zTerm) ){
|
||||
fprintf(stderr, "term to long\n");
|
||||
exit(1);
|
||||
}
|
||||
memcpy(zTerm+iPrefix, aData+i, (size_t)nTerm);
|
||||
zTerm[iPrefix+nTerm] = 0;
|
||||
i += nTerm;
|
||||
if( iHeight==0 ){
|
||||
i += getVarint(aData+i, &iDocsz);
|
||||
printf("term: %-25s doclist %7lld bytes offset %lld\n", zTerm, iDocsz, i);
|
||||
i += iDocsz;
|
||||
}else{
|
||||
printf("term: %-25s child %lld\n", zTerm, ++iChild);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Print a a blob as hex and ascii.
|
||||
*/
|
||||
static void printBlob(
|
||||
const unsigned char *aData, /* Content to print */
|
||||
int nData /* Number of bytes of content */
|
||||
){
|
||||
int i, j;
|
||||
const char *zOfstFmt;
|
||||
const int perLine = 16;
|
||||
|
||||
if( (nData&~0xfff)==0 ){
|
||||
zOfstFmt = " %03x: ";
|
||||
}else if( (nData&~0xffff)==0 ){
|
||||
zOfstFmt = " %04x: ";
|
||||
}else if( (nData&~0xfffff)==0 ){
|
||||
zOfstFmt = " %05x: ";
|
||||
}else if( (nData&~0xffffff)==0 ){
|
||||
zOfstFmt = " %06x: ";
|
||||
}else{
|
||||
zOfstFmt = " %08x: ";
|
||||
}
|
||||
|
||||
for(i=0; i<nData; i += perLine){
|
||||
fprintf(stdout, zOfstFmt, i);
|
||||
for(j=0; j<perLine; j++){
|
||||
if( i+j>nData ){
|
||||
fprintf(stdout, " ");
|
||||
}else{
|
||||
fprintf(stdout,"%02x ", aData[i+j]);
|
||||
}
|
||||
}
|
||||
for(j=0; j<perLine; j++){
|
||||
if( i+j>nData ){
|
||||
fprintf(stdout, " ");
|
||||
}else{
|
||||
fprintf(stdout,"%c", isprint(aData[i+j]) ? aData[i+j] : '.');
|
||||
}
|
||||
}
|
||||
fprintf(stdout,"\n");
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
** Convert text to a 64-bit integer
|
||||
*/
|
||||
static sqlite3_int64 atoi64(const char *z){
|
||||
sqlite3_int64 v = 0;
|
||||
while( z[0]>='0' && z[0]<='9' ){
|
||||
v = v*10 + z[0] - '0';
|
||||
z++;
|
||||
}
|
||||
return v;
|
||||
}
|
||||
|
||||
/*
|
||||
** Return a prepared statement which, when stepped, will return in its
|
||||
** first column the blob associated with segment zId. If zId begins with
|
||||
** 'r' then it is a rowid of a %_segdir entry. Otherwise it is a
|
||||
** %_segment entry.
|
||||
*/
|
||||
static sqlite3_stmt *prepareToGetSegment(
|
||||
sqlite3 *db, /* The database */
|
||||
const char *zTab, /* The FTS3/4 table name */
|
||||
const char *zId /* ID of the segment to open */
|
||||
){
|
||||
sqlite3_stmt *pStmt;
|
||||
if( zId[0]=='r' ){
|
||||
pStmt = prepare(db, "SELECT root FROM '%q_segdir' WHERE rowid=%lld",
|
||||
zTab, atoi64(zId+1));
|
||||
}else{
|
||||
pStmt = prepare(db, "SELECT block FROM '%q_segments' WHERE blockid=%lld",
|
||||
zTab, atoi64(zId));
|
||||
}
|
||||
return pStmt;
|
||||
}
|
||||
|
||||
/*
|
||||
** Print the content of a segment or of the root of a segdir. The segment
|
||||
** or root is identified by azExtra[0]. If the first character of azExtra[0]
|
||||
** is 'r' then the remainder is the integer rowid of the %_segdir entry.
|
||||
** If the first character of azExtra[0] is not 'r' then, then all of
|
||||
** azExtra[0] is an integer which is the block number.
|
||||
**
|
||||
** If the --raw option is present in azExtra, then a hex dump is provided.
|
||||
** Otherwise a decoding is shown.
|
||||
*/
|
||||
static void showSegment(sqlite3 *db, const char *zTab){
|
||||
const unsigned char *aData;
|
||||
int nData;
|
||||
sqlite3_stmt *pStmt;
|
||||
|
||||
pStmt = prepareToGetSegment(db, zTab, azExtra[0]);
|
||||
if( sqlite3_step(pStmt)!=SQLITE_ROW ){
|
||||
sqlite3_finalize(pStmt);
|
||||
return;
|
||||
}
|
||||
nData = sqlite3_column_bytes(pStmt, 0);
|
||||
aData = sqlite3_column_blob(pStmt, 0);
|
||||
printf("Segment %s of size %d bytes:\n", azExtra[0], nData);
|
||||
if( findOption("raw", 0, 0)!=0 ){
|
||||
printBlob(aData, nData);
|
||||
}else{
|
||||
decodeSegment(aData, nData);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
}
|
||||
|
||||
/*
|
||||
** Decode a single doclist and display the results on stdout.
|
||||
*/
|
||||
static void decodeDoclist(
|
||||
const unsigned char *aData, /* Content to print */
|
||||
int nData /* Number of bytes of content */
|
||||
){
|
||||
sqlite3_int64 iPrevDocid = 0;
|
||||
sqlite3_int64 iDocid;
|
||||
sqlite3_int64 iPos;
|
||||
sqlite3_int64 iPrevPos = 0;
|
||||
sqlite3_int64 iCol;
|
||||
int i = 0;
|
||||
|
||||
while( i<nData ){
|
||||
i += getVarint(aData+i, &iDocid);
|
||||
printf("docid %lld col0", iDocid+iPrevDocid);
|
||||
iPrevDocid += iDocid;
|
||||
iPrevPos = 0;
|
||||
while( 1 ){
|
||||
i += getVarint(aData+i, &iPos);
|
||||
if( iPos==1 ){
|
||||
i += getVarint(aData+i, &iCol);
|
||||
printf(" col%lld", iCol);
|
||||
iPrevPos = 0;
|
||||
}else if( iPos==0 ){
|
||||
printf("\n");
|
||||
break;
|
||||
}else{
|
||||
iPrevPos += iPos - 2;
|
||||
printf(" %lld", iPrevPos);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
/*
|
||||
** Print the content of a doclist. The segment or segdir-root is
|
||||
** identified by azExtra[0]. If the first character of azExtra[0]
|
||||
** is 'r' then the remainder is the integer rowid of the %_segdir entry.
|
||||
** If the first character of azExtra[0] is not 'r' then, then all of
|
||||
** azExtra[0] is an integer which is the block number. The offset
|
||||
** into the segment is identified by azExtra[1]. The size of the doclist
|
||||
** is azExtra[2].
|
||||
**
|
||||
** If the --raw option is present in azExtra, then a hex dump is provided.
|
||||
** Otherwise a decoding is shown.
|
||||
*/
|
||||
static void showDoclist(sqlite3 *db, const char *zTab){
|
||||
const unsigned char *aData;
|
||||
sqlite3_int64 offset;
|
||||
int nData;
|
||||
sqlite3_stmt *pStmt;
|
||||
|
||||
offset = atoi64(azExtra[1]);
|
||||
nData = atoi(azExtra[2]);
|
||||
pStmt = prepareToGetSegment(db, zTab, azExtra[0]);
|
||||
if( sqlite3_step(pStmt)!=SQLITE_ROW ){
|
||||
sqlite3_finalize(pStmt);
|
||||
return;
|
||||
}
|
||||
aData = sqlite3_column_blob(pStmt, 0);
|
||||
printf("Doclist at %s offset %lld of size %d bytes:\n",
|
||||
azExtra[0], offset, nData);
|
||||
if( findOption("raw", 0, 0)!=0 ){
|
||||
printBlob(aData+offset, nData);
|
||||
}else{
|
||||
decodeDoclist(aData+offset, nData);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
}
|
||||
|
||||
/*
|
||||
** Show the top N largest segments
|
||||
*/
|
||||
static void listBigSegments(sqlite3 *db, const char *zTab){
|
||||
int nTop, i;
|
||||
sqlite3_stmt *pStmt;
|
||||
sqlite3_int64 sz;
|
||||
sqlite3_int64 id;
|
||||
|
||||
nTop = atoi(findOption("top", 1, "25"));
|
||||
printf("The %d largest segments:\n", nTop);
|
||||
pStmt = prepare(db,
|
||||
"SELECT blockid, length(block) AS len FROM '%q_segments'"
|
||||
" ORDER BY 2 DESC, 1"
|
||||
" LIMIT %d", zTab, nTop);
|
||||
i = 0;
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
i++;
|
||||
id = sqlite3_column_int64(pStmt, 0);
|
||||
sz = sqlite3_column_int64(pStmt, 1);
|
||||
printf(" %2d. %9lld size %lld\n", i, id, sz);
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
}
|
||||
|
||||
|
||||
|
||||
static void usage(const char *argv0){
|
||||
fprintf(stderr, "Usage: %s DATABASE\n"
|
||||
" or: %s DATABASE FTS3TABLE ARGS...\n", argv0, argv0);
|
||||
fprintf(stderr,
|
||||
"ARGS:\n"
|
||||
" big-segments [--top N] show the largest segments\n"
|
||||
" doclist BLOCKID OFFSET SIZE [--raw] Decode a doclist\n"
|
||||
" schema FTS table schema\n"
|
||||
" segdir directory of segments\n"
|
||||
" segment BLOCKID [--raw] content of a segment\n"
|
||||
" segment-stats info on segment sizes\n"
|
||||
" stat the %%_stat table\n"
|
||||
" vocabulary [--top N] document vocabulary\n"
|
||||
);
|
||||
exit(1);
|
||||
}
|
||||
|
||||
int main(int argc, char **argv){
|
||||
sqlite3 *db;
|
||||
int rc;
|
||||
const char *zTab;
|
||||
const char *zCmd;
|
||||
|
||||
if( argc<2 ) usage(argv[0]);
|
||||
rc = sqlite3_open(argv[1], &db);
|
||||
if( rc ){
|
||||
fprintf(stderr, "Cannot open %s\n", argv[1]);
|
||||
exit(1);
|
||||
}
|
||||
if( argc==2 ){
|
||||
sqlite3_stmt *pStmt;
|
||||
int cnt = 0;
|
||||
pStmt = prepare(db, "SELECT b.sql"
|
||||
" FROM sqlite_master a, sqlite_master b"
|
||||
" WHERE a.name GLOB '*_segdir'"
|
||||
" AND b.name=substr(a.name,1,length(a.name)-7)"
|
||||
" ORDER BY 1");
|
||||
while( sqlite3_step(pStmt)==SQLITE_ROW ){
|
||||
cnt++;
|
||||
printf("%s;\n", sqlite3_column_text(pStmt, 0));
|
||||
}
|
||||
sqlite3_finalize(pStmt);
|
||||
if( cnt==0 ){
|
||||
printf("/* No FTS3/4 tables found in database %s */\n", argv[1]);
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
if( argc<4 ) usage(argv[0]);
|
||||
zTab = argv[2];
|
||||
zCmd = argv[3];
|
||||
nExtra = argc-4;
|
||||
azExtra = argv+4;
|
||||
if( strcmp(zCmd,"big-segments")==0 ){
|
||||
listBigSegments(db, zTab);
|
||||
}else if( strcmp(zCmd,"doclist")==0 ){
|
||||
if( argc<7 ) usage(argv[0]);
|
||||
showDoclist(db, zTab);
|
||||
}else if( strcmp(zCmd,"schema")==0 ){
|
||||
showSchema(db, zTab);
|
||||
}else if( strcmp(zCmd,"segdir")==0 ){
|
||||
showSegdirMap(db, zTab);
|
||||
}else if( strcmp(zCmd,"segment")==0 ){
|
||||
if( argc<5 ) usage(argv[0]);
|
||||
showSegment(db, zTab);
|
||||
}else if( strcmp(zCmd,"segment-stats")==0 ){
|
||||
showSegmentStats(db, zTab);
|
||||
}else if( strcmp(zCmd,"stat")==0 ){
|
||||
showStat(db, zTab);
|
||||
}else if( strcmp(zCmd,"vocabulary")==0 ){
|
||||
showVocabulary(db, zTab);
|
||||
}else{
|
||||
usage(argv[0]);
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
File diff suppressed because it is too large
Load Diff
File diff suppressed because it is too large
Load Diff
@ -0,0 +1,694 @@
|
||||
|
||||
source [file join [file dirname [info script]] parseunicode.tcl]
|
||||
|
||||
proc print_rd {map} {
|
||||
global tl_lookup_table
|
||||
set aChar [list]
|
||||
set lRange [list]
|
||||
|
||||
set nRange 1
|
||||
set iFirst [lindex $map 0 0]
|
||||
set cPrev [lindex $map 0 1]
|
||||
|
||||
foreach m [lrange $map 1 end] {
|
||||
foreach {i c} $m {}
|
||||
|
||||
if {$cPrev == $c} {
|
||||
for {set j [expr $iFirst+$nRange]} {$j<$i} {incr j} {
|
||||
if {[info exists tl_lookup_table($j)]==0} break
|
||||
}
|
||||
|
||||
if {$j==$i} {
|
||||
set nNew [expr {(1 + $i - $iFirst)}]
|
||||
if {$nNew<=8} {
|
||||
set nRange $nNew
|
||||
continue
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
lappend lRange [list $iFirst $nRange]
|
||||
lappend aChar $cPrev
|
||||
|
||||
set iFirst $i
|
||||
set cPrev $c
|
||||
set nRange 1
|
||||
}
|
||||
lappend lRange [list $iFirst $nRange]
|
||||
lappend aChar $cPrev
|
||||
|
||||
puts "/*"
|
||||
puts "** If the argument is a codepoint corresponding to a lowercase letter"
|
||||
puts "** in the ASCII range with a diacritic added, return the codepoint"
|
||||
puts "** of the ASCII letter only. For example, if passed 235 - \"LATIN"
|
||||
puts "** SMALL LETTER E WITH DIAERESIS\" - return 65 (\"LATIN SMALL LETTER"
|
||||
puts "** E\"). The resuls of passing a codepoint that corresponds to an"
|
||||
puts "** uppercase letter are undefined."
|
||||
puts "*/"
|
||||
puts "static int ${::remove_diacritic}(int c)\{"
|
||||
puts " unsigned short aDia\[\] = \{"
|
||||
puts -nonewline " 0, "
|
||||
set i 1
|
||||
foreach r $lRange {
|
||||
foreach {iCode nRange} $r {}
|
||||
if {($i % 8)==0} {puts "" ; puts -nonewline " " }
|
||||
incr i
|
||||
|
||||
puts -nonewline [format "%5d" [expr ($iCode<<3) + $nRange-1]]
|
||||
puts -nonewline ", "
|
||||
}
|
||||
puts ""
|
||||
puts " \};"
|
||||
puts " char aChar\[\] = \{"
|
||||
puts -nonewline " '\\0', "
|
||||
set i 1
|
||||
foreach c $aChar {
|
||||
set str "'$c', "
|
||||
if {$c == ""} { set str "'\\0', " }
|
||||
|
||||
if {($i % 12)==0} {puts "" ; puts -nonewline " " }
|
||||
incr i
|
||||
puts -nonewline "$str"
|
||||
}
|
||||
puts ""
|
||||
puts " \};"
|
||||
puts {
|
||||
unsigned int key = (((unsigned int)c)<<3) | 0x00000007;
|
||||
int iRes = 0;
|
||||
int iHi = sizeof(aDia)/sizeof(aDia[0]) - 1;
|
||||
int iLo = 0;
|
||||
while( iHi>=iLo ){
|
||||
int iTest = (iHi + iLo) / 2;
|
||||
if( key >= aDia[iTest] ){
|
||||
iRes = iTest;
|
||||
iLo = iTest+1;
|
||||
}else{
|
||||
iHi = iTest-1;
|
||||
}
|
||||
}
|
||||
assert( key>=aDia[iRes] );
|
||||
return ((c > (aDia[iRes]>>3) + (aDia[iRes]&0x07)) ? c : (int)aChar[iRes]);}
|
||||
puts "\}"
|
||||
}
|
||||
|
||||
proc print_isdiacritic {zFunc map} {
|
||||
|
||||
set lCode [list]
|
||||
foreach m $map {
|
||||
foreach {code char} $m {}
|
||||
if {$code && $char == ""} { lappend lCode $code }
|
||||
}
|
||||
set lCode [lsort -integer $lCode]
|
||||
set iFirst [lindex $lCode 0]
|
||||
set iLast [lindex $lCode end]
|
||||
|
||||
set i1 0
|
||||
set i2 0
|
||||
|
||||
foreach c $lCode {
|
||||
set i [expr $c - $iFirst]
|
||||
if {$i < 32} {
|
||||
set i1 [expr {$i1 | (1<<$i)}]
|
||||
} else {
|
||||
set i2 [expr {$i2 | (1<<($i-32))}]
|
||||
}
|
||||
}
|
||||
|
||||
puts "/*"
|
||||
puts "** Return true if the argument interpreted as a unicode codepoint"
|
||||
puts "** is a diacritical modifier character."
|
||||
puts "*/"
|
||||
puts "int ${zFunc}\(int c)\{"
|
||||
puts " unsigned int mask0 = [format "0x%08X" $i1];"
|
||||
puts " unsigned int mask1 = [format "0x%08X" $i2];"
|
||||
|
||||
puts " if( c<$iFirst || c>$iLast ) return 0;"
|
||||
puts " return (c < $iFirst+32) ?"
|
||||
puts " (mask0 & (1 << (c-$iFirst))) :"
|
||||
puts " (mask1 & (1 << (c-$iFirst-32)));"
|
||||
puts "\}"
|
||||
}
|
||||
|
||||
|
||||
#-------------------------------------------------------------------------
|
||||
|
||||
proc an_load_separator_ranges {} {
|
||||
global unicodedata.txt
|
||||
set lSep [an_load_unicodedata_text ${unicodedata.txt}]
|
||||
unset -nocomplain iFirst
|
||||
unset -nocomplain nRange
|
||||
set lRange [list]
|
||||
foreach sep $lSep {
|
||||
if {0==[info exists iFirst]} {
|
||||
set iFirst $sep
|
||||
set nRange 1
|
||||
} elseif { $sep == ($iFirst+$nRange) } {
|
||||
incr nRange
|
||||
} else {
|
||||
lappend lRange [list $iFirst $nRange]
|
||||
set iFirst $sep
|
||||
set nRange 1
|
||||
}
|
||||
}
|
||||
lappend lRange [list $iFirst $nRange]
|
||||
set lRange
|
||||
}
|
||||
|
||||
proc an_print_range_array {lRange} {
|
||||
set iFirstMax 0
|
||||
set nRangeMax 0
|
||||
foreach range $lRange {
|
||||
foreach {iFirst nRange} $range {}
|
||||
if {$iFirst > $iFirstMax} {set iFirstMax $iFirst}
|
||||
if {$nRange > $nRangeMax} {set nRangeMax $nRange}
|
||||
}
|
||||
if {$iFirstMax >= (1<<22)} {error "first-max is too large for format"}
|
||||
if {$nRangeMax >= (1<<10)} {error "range-max is too large for format"}
|
||||
|
||||
puts -nonewline " "
|
||||
puts [string trim {
|
||||
/* Each unsigned integer in the following array corresponds to a contiguous
|
||||
** range of unicode codepoints that are not either letters or numbers (i.e.
|
||||
** codepoints for which this function should return 0).
|
||||
**
|
||||
** The most significant 22 bits in each 32-bit value contain the first
|
||||
** codepoint in the range. The least significant 10 bits are used to store
|
||||
** the size of the range (always at least 1). In other words, the value
|
||||
** ((C<<22) + N) represents a range of N codepoints starting with codepoint
|
||||
** C. It is not possible to represent a range larger than 1023 codepoints
|
||||
** using this format.
|
||||
*/
|
||||
}]
|
||||
puts -nonewline " static const unsigned int aEntry\[\] = \{"
|
||||
set i 0
|
||||
foreach range $lRange {
|
||||
foreach {iFirst nRange} $range {}
|
||||
set u32 [format "0x%08X" [expr ($iFirst<<10) + $nRange]]
|
||||
|
||||
if {($i % 5)==0} {puts "" ; puts -nonewline " "}
|
||||
puts -nonewline " $u32,"
|
||||
incr i
|
||||
}
|
||||
puts ""
|
||||
puts " \};"
|
||||
}
|
||||
|
||||
proc an_print_ascii_bitmap {lRange} {
|
||||
foreach range $lRange {
|
||||
foreach {iFirst nRange} $range {}
|
||||
for {set i $iFirst} {$i < ($iFirst+$nRange)} {incr i} {
|
||||
if {$i<=127} { set a($i) 1 }
|
||||
}
|
||||
}
|
||||
|
||||
set aAscii [list 0 0 0 0]
|
||||
foreach key [array names a] {
|
||||
set idx [expr $key >> 5]
|
||||
lset aAscii $idx [expr [lindex $aAscii $idx] | (1 << ($key&0x001F))]
|
||||
}
|
||||
|
||||
puts " static const unsigned int aAscii\[4\] = \{"
|
||||
puts -nonewline " "
|
||||
foreach v $aAscii { puts -nonewline [format " 0x%08X," $v] }
|
||||
puts ""
|
||||
puts " \};"
|
||||
}
|
||||
|
||||
proc print_isalnum {zFunc lRange} {
|
||||
puts "/*"
|
||||
puts "** Return true if the argument corresponds to a unicode codepoint"
|
||||
puts "** classified as either a letter or a number. Otherwise false."
|
||||
puts "**"
|
||||
puts "** The results are undefined if the value passed to this function"
|
||||
puts "** is less than zero."
|
||||
puts "*/"
|
||||
puts "int ${zFunc}\(int c)\{"
|
||||
an_print_range_array $lRange
|
||||
an_print_ascii_bitmap $lRange
|
||||
puts {
|
||||
if( (unsigned int)c<128 ){
|
||||
return ( (aAscii[c >> 5] & (1 << (c & 0x001F)))==0 );
|
||||
}else if( (unsigned int)c<(1<<22) ){
|
||||
unsigned int key = (((unsigned int)c)<<10) | 0x000003FF;
|
||||
int iRes = 0;
|
||||
int iHi = sizeof(aEntry)/sizeof(aEntry[0]) - 1;
|
||||
int iLo = 0;
|
||||
while( iHi>=iLo ){
|
||||
int iTest = (iHi + iLo) / 2;
|
||||
if( key >= aEntry[iTest] ){
|
||||
iRes = iTest;
|
||||
iLo = iTest+1;
|
||||
}else{
|
||||
iHi = iTest-1;
|
||||
}
|
||||
}
|
||||
assert( aEntry[0]<key );
|
||||
assert( key>=aEntry[iRes] );
|
||||
return (((unsigned int)c) >= ((aEntry[iRes]>>10) + (aEntry[iRes]&0x3FF)));
|
||||
}
|
||||
return 1;}
|
||||
puts "\}"
|
||||
}
|
||||
|
||||
proc print_test_isalnum {zFunc lRange} {
|
||||
foreach range $lRange {
|
||||
foreach {iFirst nRange} $range {}
|
||||
for {set i $iFirst} {$i < ($iFirst+$nRange)} {incr i} { set a($i) 1 }
|
||||
}
|
||||
|
||||
puts "static int isalnum_test(int *piCode)\{"
|
||||
puts -nonewline " unsigned char aAlnum\[\] = \{"
|
||||
for {set i 0} {$i < 70000} {incr i} {
|
||||
if {($i % 32)==0} { puts "" ; puts -nonewline " " }
|
||||
set bFlag [expr ![info exists a($i)]]
|
||||
puts -nonewline "${bFlag},"
|
||||
}
|
||||
puts ""
|
||||
puts " \};"
|
||||
|
||||
puts -nonewline " int aLargeSep\[\] = \{"
|
||||
set i 0
|
||||
foreach iSep [lsort -integer [array names a]] {
|
||||
if {$iSep<70000} continue
|
||||
if {($i % 8)==0} { puts "" ; puts -nonewline " " }
|
||||
puts -nonewline " $iSep,"
|
||||
incr i
|
||||
}
|
||||
puts ""
|
||||
puts " \};"
|
||||
puts -nonewline " int aLargeOther\[\] = \{"
|
||||
set i 0
|
||||
foreach iSep [lsort -integer [array names a]] {
|
||||
if {$iSep<70000} continue
|
||||
if {[info exists a([expr $iSep-1])]==0} {
|
||||
if {($i % 8)==0} { puts "" ; puts -nonewline " " }
|
||||
puts -nonewline " [expr $iSep-1],"
|
||||
incr i
|
||||
}
|
||||
if {[info exists a([expr $iSep+1])]==0} {
|
||||
if {($i % 8)==0} { puts "" ; puts -nonewline " " }
|
||||
puts -nonewline " [expr $iSep+1],"
|
||||
incr i
|
||||
}
|
||||
}
|
||||
puts ""
|
||||
puts " \};"
|
||||
|
||||
puts [subst -nocommands {
|
||||
int i;
|
||||
for(i=0; i<sizeof(aAlnum)/sizeof(aAlnum[0]); i++){
|
||||
if( ${zFunc}(i)!=aAlnum[i] ){
|
||||
*piCode = i;
|
||||
return 1;
|
||||
}
|
||||
}
|
||||
for(i=0; i<sizeof(aLargeSep)/sizeof(aLargeSep[0]); i++){
|
||||
if( ${zFunc}(aLargeSep[i])!=0 ){
|
||||
*piCode = aLargeSep[i];
|
||||
return 1;
|
||||
}
|
||||
}
|
||||
for(i=0; i<sizeof(aLargeOther)/sizeof(aLargeOther[0]); i++){
|
||||
if( ${zFunc}(aLargeOther[i])!=1 ){
|
||||
*piCode = aLargeOther[i];
|
||||
return 1;
|
||||
}
|
||||
}
|
||||
}]
|
||||
puts " return 0;"
|
||||
puts "\}"
|
||||
}
|
||||
|
||||
#-------------------------------------------------------------------------
|
||||
|
||||
proc tl_create_records {} {
|
||||
global tl_lookup_table
|
||||
|
||||
set iFirst ""
|
||||
set nOff 0
|
||||
set nRange 0
|
||||
set nIncr 0
|
||||
|
||||
set lRecord [list]
|
||||
foreach code [lsort -integer [array names tl_lookup_table]] {
|
||||
set mapping $tl_lookup_table($code)
|
||||
if {$iFirst == ""} {
|
||||
set iFirst $code
|
||||
set nOff [expr $mapping - $code]
|
||||
set nRange 1
|
||||
set nIncr 1
|
||||
} else {
|
||||
set diff [expr $code - ($iFirst + ($nIncr * ($nRange - 1)))]
|
||||
if { $nRange==1 && ($diff==1 || $diff==2) } {
|
||||
set nIncr $diff
|
||||
}
|
||||
|
||||
if {$diff != $nIncr || ($mapping - $code)!=$nOff} {
|
||||
if { $nRange==1 } {set nIncr 1}
|
||||
lappend lRecord [list $iFirst $nIncr $nRange $nOff]
|
||||
set iFirst $code
|
||||
set nOff [expr $mapping - $code]
|
||||
set nRange 1
|
||||
set nIncr 1
|
||||
} else {
|
||||
incr nRange
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
lappend lRecord [list $iFirst $nIncr $nRange $nOff]
|
||||
|
||||
set lRecord
|
||||
}
|
||||
|
||||
proc tl_print_table_header {} {
|
||||
puts -nonewline " "
|
||||
puts [string trim {
|
||||
/* Each entry in the following array defines a rule for folding a range
|
||||
** of codepoints to lower case. The rule applies to a range of nRange
|
||||
** codepoints starting at codepoint iCode.
|
||||
**
|
||||
** If the least significant bit in flags is clear, then the rule applies
|
||||
** to all nRange codepoints (i.e. all nRange codepoints are upper case and
|
||||
** need to be folded). Or, if it is set, then the rule only applies to
|
||||
** every second codepoint in the range, starting with codepoint C.
|
||||
**
|
||||
** The 7 most significant bits in flags are an index into the aiOff[]
|
||||
** array. If a specific codepoint C does require folding, then its lower
|
||||
** case equivalent is ((C + aiOff[flags>>1]) & 0xFFFF).
|
||||
**
|
||||
** The contents of this array are generated by parsing the CaseFolding.txt
|
||||
** file distributed as part of the "Unicode Character Database". See
|
||||
** http://www.unicode.org for details.
|
||||
*/
|
||||
}]
|
||||
puts " static const struct TableEntry \{"
|
||||
puts " unsigned short iCode;"
|
||||
puts " unsigned char flags;"
|
||||
puts " unsigned char nRange;"
|
||||
puts " \} aEntry\[\] = \{"
|
||||
}
|
||||
|
||||
proc tl_print_table_entry {togglevar entry liOff} {
|
||||
upvar $togglevar t
|
||||
foreach {iFirst nIncr nRange nOff} $entry {}
|
||||
|
||||
if {$iFirst > (1<<16)} { return 1 }
|
||||
|
||||
if {[info exists t]==0} {set t 0}
|
||||
if {$t==0} { puts -nonewline " " }
|
||||
|
||||
set flags 0
|
||||
if {$nIncr==2} { set flags 1 ; set nRange [expr $nRange * 2]}
|
||||
if {$nOff<0} { incr nOff [expr (1<<16)] }
|
||||
|
||||
set idx [lsearch $liOff $nOff]
|
||||
if {$idx<0} {error "malfunction generating aiOff"}
|
||||
set flags [expr $flags + $idx*2]
|
||||
|
||||
set txt "{$iFirst, $flags, $nRange},"
|
||||
if {$t==2} {
|
||||
puts $txt
|
||||
} else {
|
||||
puts -nonewline [format "% -23s" $txt]
|
||||
}
|
||||
set t [expr ($t+1)%3]
|
||||
|
||||
return 0
|
||||
}
|
||||
|
||||
proc tl_print_table_footer {togglevar} {
|
||||
upvar $togglevar t
|
||||
if {$t!=0} {puts ""}
|
||||
puts " \};"
|
||||
}
|
||||
|
||||
proc tl_print_if_entry {entry} {
|
||||
foreach {iFirst nIncr nRange nOff} $entry {}
|
||||
if {$nIncr==2} {error "tl_print_if_entry needs improvement!"}
|
||||
|
||||
puts " else if( c>=$iFirst && c<[expr $iFirst+$nRange] )\{"
|
||||
puts " ret = c + $nOff;"
|
||||
puts " \}"
|
||||
}
|
||||
|
||||
proc tl_generate_ioff_table {lRecord} {
|
||||
foreach entry $lRecord {
|
||||
foreach {iFirst nIncr nRange iOff} $entry {}
|
||||
if {$iOff<0} { incr iOff [expr (1<<16)] }
|
||||
if {[info exists a($iOff)]} continue
|
||||
set a($iOff) 1
|
||||
}
|
||||
|
||||
set liOff [lsort -integer [array names a]]
|
||||
if {[llength $liOff]>128} { error "Too many distinct ioffs" }
|
||||
return $liOff
|
||||
}
|
||||
|
||||
proc tl_print_ioff_table {liOff} {
|
||||
puts -nonewline " static const unsigned short aiOff\[\] = \{"
|
||||
set i 0
|
||||
foreach off $liOff {
|
||||
if {($i % 8)==0} {puts "" ; puts -nonewline " "}
|
||||
puts -nonewline [format "% -7s" "$off,"]
|
||||
incr i
|
||||
}
|
||||
puts ""
|
||||
puts " \};"
|
||||
|
||||
}
|
||||
|
||||
proc print_fold {zFunc} {
|
||||
|
||||
set lRecord [tl_create_records]
|
||||
|
||||
set lHigh [list]
|
||||
puts "/*"
|
||||
puts "** Interpret the argument as a unicode codepoint. If the codepoint"
|
||||
puts "** is an upper case character that has a lower case equivalent,"
|
||||
puts "** return the codepoint corresponding to the lower case version."
|
||||
puts "** Otherwise, return a copy of the argument."
|
||||
puts "**"
|
||||
puts "** The results are undefined if the value passed to this function"
|
||||
puts "** is less than zero."
|
||||
puts "*/"
|
||||
puts "int ${zFunc}\(int c, int bRemoveDiacritic)\{"
|
||||
|
||||
set liOff [tl_generate_ioff_table $lRecord]
|
||||
tl_print_table_header
|
||||
foreach entry $lRecord {
|
||||
if {[tl_print_table_entry toggle $entry $liOff]} {
|
||||
lappend lHigh $entry
|
||||
}
|
||||
}
|
||||
tl_print_table_footer toggle
|
||||
tl_print_ioff_table $liOff
|
||||
|
||||
puts [subst -nocommands {
|
||||
int ret = c;
|
||||
|
||||
assert( sizeof(unsigned short)==2 && sizeof(unsigned char)==1 );
|
||||
|
||||
if( c<128 ){
|
||||
if( c>='A' && c<='Z' ) ret = c + ('a' - 'A');
|
||||
}else if( c<65536 ){
|
||||
const struct TableEntry *p;
|
||||
int iHi = sizeof(aEntry)/sizeof(aEntry[0]) - 1;
|
||||
int iLo = 0;
|
||||
int iRes = -1;
|
||||
|
||||
assert( c>aEntry[0].iCode );
|
||||
while( iHi>=iLo ){
|
||||
int iTest = (iHi + iLo) / 2;
|
||||
int cmp = (c - aEntry[iTest].iCode);
|
||||
if( cmp>=0 ){
|
||||
iRes = iTest;
|
||||
iLo = iTest+1;
|
||||
}else{
|
||||
iHi = iTest-1;
|
||||
}
|
||||
}
|
||||
|
||||
assert( iRes>=0 && c>=aEntry[iRes].iCode );
|
||||
p = &aEntry[iRes];
|
||||
if( c<(p->iCode + p->nRange) && 0==(0x01 & p->flags & (p->iCode ^ c)) ){
|
||||
ret = (c + (aiOff[p->flags>>1])) & 0x0000FFFF;
|
||||
assert( ret>0 );
|
||||
}
|
||||
|
||||
if( bRemoveDiacritic ) ret = ${::remove_diacritic}(ret);
|
||||
}
|
||||
}]
|
||||
|
||||
foreach entry $lHigh {
|
||||
tl_print_if_entry $entry
|
||||
}
|
||||
|
||||
puts ""
|
||||
puts " return ret;"
|
||||
puts "\}"
|
||||
}
|
||||
|
||||
proc print_fold_test {zFunc mappings} {
|
||||
global tl_lookup_table
|
||||
|
||||
foreach m $mappings {
|
||||
set c [lindex $m 1]
|
||||
if {$c == ""} {
|
||||
set extra([lindex $m 0]) 0
|
||||
} else {
|
||||
scan $c %c i
|
||||
set extra([lindex $m 0]) $i
|
||||
}
|
||||
}
|
||||
|
||||
puts "static int fold_test(int *piCode)\{"
|
||||
puts -nonewline " static int aLookup\[\] = \{"
|
||||
for {set i 0} {$i < 70000} {incr i} {
|
||||
|
||||
set expected $i
|
||||
catch { set expected $tl_lookup_table($i) }
|
||||
set expected2 $expected
|
||||
catch { set expected2 $extra($expected2) }
|
||||
|
||||
if {($i % 4)==0} { puts "" ; puts -nonewline " " }
|
||||
puts -nonewline "$expected, $expected2, "
|
||||
}
|
||||
puts " \};"
|
||||
puts " int i;"
|
||||
puts " for(i=0; i<sizeof(aLookup)/sizeof(aLookup\[0\]); i++)\{"
|
||||
puts " int iCode = (i/2);"
|
||||
puts " int bFlag = i & 0x0001;"
|
||||
puts " if( ${zFunc}\(iCode, bFlag)!=aLookup\[i\] )\{"
|
||||
puts " *piCode = iCode;"
|
||||
puts " return 1;"
|
||||
puts " \}"
|
||||
puts " \}"
|
||||
puts " return 0;"
|
||||
puts "\}"
|
||||
}
|
||||
|
||||
|
||||
proc print_fileheader {} {
|
||||
puts [string trim {
|
||||
/*
|
||||
** 2012 May 25
|
||||
**
|
||||
** The author disclaims copyright to this source code. In place of
|
||||
** a legal notice, here is a blessing:
|
||||
**
|
||||
** May you do good and not evil.
|
||||
** May you find forgiveness for yourself and forgive others.
|
||||
** May you share freely, never taking more than you give.
|
||||
**
|
||||
******************************************************************************
|
||||
*/
|
||||
|
||||
/*
|
||||
** DO NOT EDIT THIS MACHINE GENERATED FILE.
|
||||
*/
|
||||
}]
|
||||
puts ""
|
||||
if {$::generate_fts5_code} {
|
||||
# no-op
|
||||
} else {
|
||||
puts "#ifndef SQLITE_DISABLE_FTS3_UNICODE"
|
||||
puts "#if defined(SQLITE_ENABLE_FTS3) || defined(SQLITE_ENABLE_FTS4)"
|
||||
}
|
||||
puts ""
|
||||
puts "#include <assert.h>"
|
||||
puts ""
|
||||
}
|
||||
|
||||
proc print_test_main {} {
|
||||
puts ""
|
||||
puts "#include <stdio.h>"
|
||||
puts ""
|
||||
puts "int main(int argc, char **argv)\{"
|
||||
puts " int r1, r2;"
|
||||
puts " int code;"
|
||||
puts " r1 = isalnum_test(&code);"
|
||||
puts " if( r1 ) printf(\"isalnum(): Problem with code %d\\n\",code);"
|
||||
puts " else printf(\"isalnum(): test passed\\n\");"
|
||||
puts " r2 = fold_test(&code);"
|
||||
puts " if( r2 ) printf(\"fold(): Problem with code %d\\n\",code);"
|
||||
puts " else printf(\"fold(): test passed\\n\");"
|
||||
puts " return (r1 || r2);"
|
||||
puts "\}"
|
||||
}
|
||||
|
||||
# Proces the command line arguments. Exit early if they are not to
|
||||
# our liking.
|
||||
#
|
||||
proc usage {} {
|
||||
puts -nonewline stderr "Usage: $::argv0 ?-test? ?-fts5? "
|
||||
puts stderr "<CaseFolding.txt file> <UnicodeData.txt file>"
|
||||
exit 1
|
||||
}
|
||||
if {[llength $argv]<2} usage
|
||||
set unicodedata.txt [lindex $argv end]
|
||||
set casefolding.txt [lindex $argv end-1]
|
||||
|
||||
set remove_diacritic remove_diacritic
|
||||
set generate_test_code 0
|
||||
set generate_fts5_code 0
|
||||
set function_prefix "sqlite3Fts"
|
||||
for {set i 0} {$i < [llength $argv]-2} {incr i} {
|
||||
switch -- [lindex $argv $i] {
|
||||
-test {
|
||||
set generate_test_code 1
|
||||
}
|
||||
-fts5 {
|
||||
set function_prefix sqlite3Fts5
|
||||
set generate_fts5_code 1
|
||||
set remove_diacritic fts5_remove_diacritic
|
||||
}
|
||||
default {
|
||||
usage
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
print_fileheader
|
||||
|
||||
# Print the isalnum() function to stdout.
|
||||
#
|
||||
set lRange [an_load_separator_ranges]
|
||||
print_isalnum ${function_prefix}UnicodeIsalnum $lRange
|
||||
|
||||
# Leave a gap between the two generated C functions.
|
||||
#
|
||||
puts ""
|
||||
puts ""
|
||||
|
||||
# Load the fold data. This is used by the [rd_XXX] commands
|
||||
# as well as [print_fold].
|
||||
tl_load_casefolding_txt ${casefolding.txt}
|
||||
|
||||
set mappings [rd_load_unicodedata_text ${unicodedata.txt}]
|
||||
print_rd $mappings
|
||||
puts ""
|
||||
puts ""
|
||||
print_isdiacritic ${function_prefix}UnicodeIsdiacritic $mappings
|
||||
puts ""
|
||||
puts ""
|
||||
|
||||
# Print the fold() function to stdout.
|
||||
#
|
||||
print_fold ${function_prefix}UnicodeFold
|
||||
|
||||
# Print the test routines and main() function to stdout, if -test
|
||||
# was specified.
|
||||
#
|
||||
if {$::generate_test_code} {
|
||||
print_test_isalnum ${function_prefix}UnicodeIsalnum $lRange
|
||||
print_fold_test ${function_prefix}UnicodeFold $mappings
|
||||
print_test_main
|
||||
}
|
||||
|
||||
if {$generate_fts5_code} {
|
||||
# no-op
|
||||
} else {
|
||||
puts "#endif /* defined(SQLITE_ENABLE_FTS3) || defined(SQLITE_ENABLE_FTS4) */"
|
||||
puts "#endif /* !defined(SQLITE_DISABLE_FTS3_UNICODE) */"
|
||||
}
|
||||
@ -0,0 +1,146 @@
|
||||
|
||||
#--------------------------------------------------------------------------
|
||||
# Parameter $zName must be a path to the file UnicodeData.txt. This command
|
||||
# reads the file and returns a list of mappings required to remove all
|
||||
# diacritical marks from a unicode string. Each mapping is itself a list
|
||||
# consisting of two elements - the unicode codepoint and the single ASCII
|
||||
# character that it should be replaced with, or an empty string if the
|
||||
# codepoint should simply be removed from the input. Examples:
|
||||
#
|
||||
# { 224 a } (replace codepoint 224 to "a")
|
||||
# { 769 "" } (remove codepoint 769 from input)
|
||||
#
|
||||
# Mappings are only returned for non-upper case codepoints. It is assumed
|
||||
# that the input has already been folded to lower case.
|
||||
#
|
||||
proc rd_load_unicodedata_text {zName} {
|
||||
global tl_lookup_table
|
||||
|
||||
set fd [open $zName]
|
||||
set lField {
|
||||
code
|
||||
character_name
|
||||
general_category
|
||||
canonical_combining_classes
|
||||
bidirectional_category
|
||||
character_decomposition_mapping
|
||||
decimal_digit_value
|
||||
digit_value
|
||||
numeric_value
|
||||
mirrored
|
||||
unicode_1_name
|
||||
iso10646_comment_field
|
||||
uppercase_mapping
|
||||
lowercase_mapping
|
||||
titlecase_mapping
|
||||
}
|
||||
set lRet [list]
|
||||
|
||||
while { ![eof $fd] } {
|
||||
set line [gets $fd]
|
||||
if {$line == ""} continue
|
||||
|
||||
set fields [split $line ";"]
|
||||
if {[llength $fields] != [llength $lField]} { error "parse error: $line" }
|
||||
foreach $lField $fields {}
|
||||
if { [llength $character_decomposition_mapping]!=2
|
||||
|| [string is xdigit [lindex $character_decomposition_mapping 0]]==0
|
||||
} {
|
||||
continue
|
||||
}
|
||||
|
||||
set iCode [expr "0x$code"]
|
||||
set iAscii [expr "0x[lindex $character_decomposition_mapping 0]"]
|
||||
set iDia [expr "0x[lindex $character_decomposition_mapping 1]"]
|
||||
|
||||
if {[info exists tl_lookup_table($iCode)]} continue
|
||||
|
||||
if { ($iAscii >= 97 && $iAscii <= 122)
|
||||
|| ($iAscii >= 65 && $iAscii <= 90)
|
||||
} {
|
||||
lappend lRet [list $iCode [string tolower [format %c $iAscii]]]
|
||||
set dia($iDia) 1
|
||||
}
|
||||
}
|
||||
|
||||
foreach d [array names dia] {
|
||||
lappend lRet [list $d ""]
|
||||
}
|
||||
set lRet [lsort -integer -index 0 $lRet]
|
||||
|
||||
close $fd
|
||||
set lRet
|
||||
}
|
||||
|
||||
#-------------------------------------------------------------------------
|
||||
# Parameter $zName must be a path to the file UnicodeData.txt. This command
|
||||
# reads the file and returns a list of codepoints (integers). The list
|
||||
# contains all codepoints in the UnicodeData.txt assigned to any "General
|
||||
# Category" that is not a "Letter" or "Number".
|
||||
#
|
||||
proc an_load_unicodedata_text {zName} {
|
||||
set fd [open $zName]
|
||||
set lField {
|
||||
code
|
||||
character_name
|
||||
general_category
|
||||
canonical_combining_classes
|
||||
bidirectional_category
|
||||
character_decomposition_mapping
|
||||
decimal_digit_value
|
||||
digit_value
|
||||
numeric_value
|
||||
mirrored
|
||||
unicode_1_name
|
||||
iso10646_comment_field
|
||||
uppercase_mapping
|
||||
lowercase_mapping
|
||||
titlecase_mapping
|
||||
}
|
||||
set lRet [list]
|
||||
|
||||
while { ![eof $fd] } {
|
||||
set line [gets $fd]
|
||||
if {$line == ""} continue
|
||||
|
||||
set fields [split $line ";"]
|
||||
if {[llength $fields] != [llength $lField]} { error "parse error: $line" }
|
||||
foreach $lField $fields {}
|
||||
|
||||
set iCode [expr "0x$code"]
|
||||
set bAlnum [expr {
|
||||
[lsearch {L N} [string range $general_category 0 0]] >= 0
|
||||
|| $general_category=="Co"
|
||||
}]
|
||||
|
||||
if { !$bAlnum } { lappend lRet $iCode }
|
||||
}
|
||||
|
||||
close $fd
|
||||
set lRet
|
||||
}
|
||||
|
||||
proc tl_load_casefolding_txt {zName} {
|
||||
global tl_lookup_table
|
||||
|
||||
set fd [open $zName]
|
||||
while { ![eof $fd] } {
|
||||
set line [gets $fd]
|
||||
if {[string range $line 0 0] == "#"} continue
|
||||
if {$line == ""} continue
|
||||
|
||||
foreach x {a b c d} {unset -nocomplain $x}
|
||||
foreach {a b c d} [split $line ";"] {}
|
||||
|
||||
set a2 [list]
|
||||
set c2 [list]
|
||||
foreach elem $a { lappend a2 [expr "0x[string trim $elem]"] }
|
||||
foreach elem $c { lappend c2 [expr "0x[string trim $elem]"] }
|
||||
set b [string trim $b]
|
||||
set d [string trim $d]
|
||||
|
||||
if {$b=="C" || $b=="S"} { set tl_lookup_table($a2) $c2 }
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user