- Fields in CTEs are marked as being used in subselects.
- In qc_mysqlembedded all selects must be walked if CTE is present.
- In qc_sqlite unions need special handling.
The default sql mode must now be provided explicitly when the query
classifier is setup. This is in preparation for "sql_mode" becoming
a global configuration parameter of MaxScale.
The sql mode is now a property of the info object used for storing
parsing related information. It is initialized with the value of the
sql mode qc_sqlite was initialized with.
This will be further changed so that the mode can be adjusted at
runtime so that the sql mode can be the property of a session. That
is, once set it will affect all future parsing for that session.
In PL/SQL BEGIN starts a block (i.e. not a transaction). Whenever we
see that, we assume it is such a block, consume all input and set the
type to QUERY_TYPE_WRITE to ensure it goes to master.
Both 10.3 and Oracle support sequence pseudo colums and corresponding
functions. Getting the next number in the sequence is in both cases
obtained using nextval/nextval() but the current number is in Oracle
obtained using currval/currval() and in 10.3 using lastval/lastval().
These fields/functions are now ignored, in the sense that they will
not show up in the field/function infos. However, they will cause the
type mask of the statement to contain the bit QUERY_TYPE_WRITE so that
statements accessing the sequence will always be sent to the master.
Oracle SQL accepts statements like
PREPARE stmt FROM 'INSERT INTO t1 VALUES (:a,:b)';
PREPARE stmt FROM 'INSERT INTO t1 VALUES (:"a",:"b")';
That is, the variable may be surrounded by quotes.
The embedded parser of 10.3 parses things slightly differently than
how the embedded parser of earlier releases does. When comparing the
output of different query classifiers you need to be able to specify
with what qc_sqlite should be compatible with.
EXPLAIN statements are no longer parsed completely as doing so makes
it hard to modify the grammar for the needs or Oracle SQL.
Consequently, for an EXPLAIN statement you now bascially only get the
type and the operation (the newly added QUERY_OP_EXPLAIN and QUERY_OP_SHOW).
The other information is not interesting and is related to
information_schema and similar tables.
Changed the query operation enum to contain implicit enum values instead
of providing. The operation was never used as a bitmask so it is pointless
to declare them as such.
Added the EXECUTE type to the enum and used it in qc_sqlite and
qc_mysqlembedded.
By default, only the essentials - the type and the operation - of
a statement will be collected and only if fields, tables, functions
and databases are explicitly asked for, will they be collected.
However, a statement will be parsed at most twice; if parsing is
needed a second time then all information will be collected.
If it is known that some particular information is needed, then
qc_parse() can be called explicitly to ensure it is collected
at first parsing.
It is now possible to specify what information the caller is interested
in. With this the cost for collecting information during the query parsing
that nobody is interested in can be avoided.
