MXS-1070
Now both qc_mysqlembedded and qc_sqlite return the same stuff
for the same statement, and both include also operators in
addition to pure functions. Whether that is the right approach,
is still subject to debate.
However, if we want to make it possible to disable e.g. the
use of concat as in "select concat(a) from t", where a is a string,
to prevent the bypassing of the masking filter, then conceptually
it should be possible to prevent "select a+0 from t", where a is an
int, as well.
The setting up and the initialization of the query classifier has
now been separated. The gateway explicitly sets up the query
classifier (i.e. chooses which one to use and what arguments to
provide), but the actual initialization is performed as part of
the general module initialization.
The MXS_MODULDE object now contains optinal pointers for functions
to be called att process and thread startup and shutdown. Since the
functions were added to the end, strictly speaking, all structures
would not have needed to have been modified, but better to be
explicit. In a subsequent change, these will be called.
C++ does not support flexible arrays, so for the time being C++
modules are restricted to 10 parameters. Better approach is to
factor out the parameters to a separate array and then just store
a pointer to that array in MXS_MODULE.
The MODULE_INFO is now the main object which is used by modules to convey
information to the MaxScale core. The MXS_MODULE name is more apt as it
now contains the actual module definition.
The old MODULES structure was moved into load_utils.c as an internal
implementation and was renamed so that it is not confused with the new
MODULE structure.
The modules are now declared with a common macro. This allows future
additions to the module loading process while also making the loaded
symbol name a constant.
This allows modules to only expose one entry point with a consistent
signature. In the future, this could be used to implement declarations of
module parameters.
qc_get_table_names() and qc_get_database_names() now always set
the number items, also in case of error. That is, it is now always
safe to start iterating using the number of items without first
checking the returned pointer.
When a backslash is encountered, the backslash should not be
copied but only the character after that.
For the sake of completeness, a few more characters would have
to be handled explicitly, but as the content of a string will not
affect the statement's classification there is not much point in
doing that.
Sqlite3 performs some lazy initialization, during which it internally
parses some SQL statements of its own. Earlier there was detection code
for noticing that, but it was costly and errorprone.
Now, sqlite3 is forced to perform the initialization at startup so that
we no longer need any detection code.
Together with the field names, now qc_get_field_info also returns
field usage information, that is, in what context a field is used.
This allows, for instance, the cache to take action if a a particular
field is selected (SELECT a FROM ...), but not if it is used in a
GROUP BY clause (...GROUP BY a).
This caused a significant modifications of qc_mysqlembedded that
earlier did not walk the parse-tree, but instead looped over of a
list of st_select_lex instances that, the name notwithstanding,
also contain information about other things but SELECTs. The former
approach lost all contextual information, so it was not possible
to know where a particular field was used.
Now the parse tree is walked, which means that the contextual
information is known, and thus the field usage can be updated.
Function is no longer used and it was quite unoptimal, so now
removed.
qc_get_prepare_name, qc_get_prepare_operation and qc_get_field_info
that were missing from qc_dummy added at the same time.
A HAVING clause can only refer to names that already have been mentioned
or if "SELECT *" is used. Either way, the HAVING names need not be separately
collected.
We now collect more information about a particular field and
then, if necessary, copy the data over into affected_fields if
someone is interested in that.
The comparison program is extended as well, but qc_get_fields_infos()
is not tested, because qc_mysqlembedded does not implement this yet.
This function returns more detailed information about the fields
of a statement. Supersedes qc_get_affected_fields() that will
be deprecated and removed.
Note that this function now introduced new kind of behaviour; the
returned data belongs to the GWBUF and remains valid for as long as
the GWBUF is alive. That means that unnecessary copying need not
be done.
The operation of the statement to be prepared is no longer
reported as the operation of the PREPARE statement.
Instead, when the type of the statement is
QUERY_TYPE_PREPARE_NAMED_STMT, the operation can be obtained
using qc_get_prepare_operation().
The qc_mysqlembedded implementation will be provided in a
subsequent commit.
The operation of a PREPARE statement will be that of the preparable
statement. That will make it possible to know whether an EXECUTEd
prepared statement e.g. is a SELECT or an UPDATE.
When a DESCRIBE <table> or a SHOW COLUMNS IN <table> query is done, the
actual query is performed on tables in the information_schema
database. This might be what actually happens on the backend server but
this information is not really useful when we need to know which database
the query targets.
By passing the actual table names instead of the underlying table names,
the schemarouter is able to detect where these statements should be
routed.
Now more information about a transaction start is provided. When
a transaction start statement is parsed, the type of the statement
with be QUERY_TYPE_BEGIN_TRX anded with QUERY_TYPE_READ or
QUERY_TYPE_WRITE if the transaction was explicitly started as READ
ONLY or READ WRITE.
Now also BEGIN WORK and [COMMIT|ROLLBACK] WORK are recognized.
"AND CHAIN" will still need to be recognized.
The general purpose stuff in skygw_utils.h was moved to utils.h
and the corresponding implementation from skygw_utils.cc to utils.c.
Includes updated accordingly.
Skygw_utils.h is now only used by log_manager and by mlist, which
is only used by log_manager. Consequently, skygw_utils.h was moved
to server/maxscale.
Utils.h needs a separate overhaul.
- Headers now to be included as <maxscale/xyz.h>
- First step, no cleanup of headers has been made. Only moving
from one place to another + necessary modifications.
Previously a query like e.g.
delete t11.*, t12.* from t11,t12 where t11.a = t12.a;
would cause a leak. The code for freeing that memory was present
but commented out because it used to cause a crash.
Now no crash appears, so it would seem that the crash was caused
by something else that no longer is present.
973b983 Merge branch 'release-2.0.0' into develop
255dd23 Make spinlock functions take const argument
6e23bab Fix bitmask reallocation
338c189 Rename and clean up slavelag filter
3ea8f28 Fix possible NULL pointer dereference
bfe6738 MXS-830: Add module information to logged messages
1fad962 Fix strncat usage
d38997a Adjust log throttling policy
0be4e4b Add hashtable_item_strcasecmp
726100e Take hashtable convenience functions into use
5e7744a Fix typo in maxadmin.md
c5778c8 Merge branch 'release-2.0.0' into develop
b5762af Move from tmpnam to mkstemp
d6f2c71 Add convenience functions to hashtable
359058a MXS-825: Add support for --execdir
636347c Enable runtime reconfiguration of log throttling
ef9fba9 Improve log throttling documentation
aef917a Implement log throttling
e3a5349 Remove shardrouter.c
8051e80 Remove custom qc_sqlite allocation functions
fd34d60 Initial implementation of the learning firewall
a8752a8 Removed "filestem option" from example
1ef2519 Removed "filestem option" from example
0815cc8 Cleanup spinlock.h
ab4dc99 Clean up hashtable.h
ef2c078 Add prototypes for hashtable copy and free functions
fb5cfaf Add 'log_throttling' configuration entry
300d823 Add proper prototypes for hashtable hash and cmp functions
1c649aa qc_mysqlembedded: Include skygw_...h without path.
d276160 Add missing RPM scripts
e70e644 Fix HTTPAuth installation
1b2b389 Combine utils into the server directory
3ff9913 Add missing utils headers to devel package
407efb2 Fix minor packaging problems
99aa6ad Split MaxScale into core, experimental and devel packages
1290386 Merge branch 'develop' of ssh://github.com/mariadb-corporation/maxscale-new into develop
e59f148 Make scripts POSIX sh compatible
7319266 Fixed SHOW SLAVE STATUS in bonlog router
f8d760a Update Binlogrouter.md
0a904ed Update Replication-Proxy-Binlog-Router-Tutorial.md
75d4202 Update Replication-Proxy-Binlog-Router-Tutorial.md
b8651fc Add missing newline in listmanager.h
c7ad047 Add note about user data caches to release notes
70ccc2b Merge branch 'release-2.0.0' into develop
575d1b6 Mistake - dummy session needs list markers set.
8364508 Merge branch 'develop' into binlog_server_semisync
868b902 Update MaxScale limitations
2c8b327 Store listener caches in separate directories
6e183ec Create unique user data caches for each listeners
f643685 Don't free orphaned tee filter sessions
4179afa Allow binlogrouter to be used without a listener
7ad79af Add function for freeing a listener
677a0a2 Move authentication data from services to listeners
4f12af7 Merge remote-tracking branch 'origin/MXS-677' into develop
1419b81 Semi-Sync support to binlog server: code review updtate
0ea0f01 Semi-Sync support to binlog server: added missing routine
4aad909 Semi-Sync support to binlog server
b824e1e Add authenticator support to httpd.c
705a688 Change tabs to spaces
d0c419e Change method of adding list fields to e.g. DCB
25504fc Document the changed routing priority of hints
41666d1 Remove use_ssl_if_enabled global option
a3584e9 Make routing hints have highest priority
34a1d24 Updated document with new binlog router option
01eedc5 Updated documentation with SSL usage
8a4c0f6 Update Replication-Proxy-Binlog-Router-Tutorial.md
4e374aa Update Replication-Proxy-Binlog-Router-Tutorial.md
f3f3c57 Update Replication-Proxy-Binlog-Router-Tutorial.md
617b79f Binlog Server: error messages typo fix
fa8dfae Binlog Server: error messages review
1b8819c Fix freeing of schemarouter session memory
07f49e1 MXS-788: new code review fix
1fd3b09 MXS-788: show services now displays SSL info
6ca2584 MXS-788 code review fix
ae6a7d0 MXS-788 code review
43d3474 Master server SSL connection
90b2377 Use correct variable in listmanager pre-allocation
9a5b238 Fix listmanager pre-allocation
9c78625 Fix a memory leak when backend authentication fails
e59a966 Fix hang in list_find_free
ff30223 Fix freeing of shared data in schemarouter
fc8f9d3 Add missing include in luafilter
ecf7f53 Add missing NULL value to filter parameter array
636d849 Update memory allocation approach
f0d1d38 Add new allocation functions
97d00a0 Fix writing of uninitialized data to logs
e72c9b2 Merge branch 'release-2.0.0' into develop
cf2b712 Merge branch 'release-2.0.0' into develop
8917c5c Change the logic behind valid list entry checks
c10deff Improve documentation about version_string
f59f1f7 Merge branch 'develop' of ssh://github.com/mariadb-corporation/maxscale-new into develop
c88edb3 Backend authentication failure improvement
abd5bee Revert "Backend authentication failure improvement"
5bb3107 Backend authentication failure improvement
b7f434a Add new allocation functions
3f022fa Fix stupid mistake
99c4317 Merge remote-tracking branch 'origin/MXS-677' into develop
3c1ded6 Added connection/authentication failure error reporting in SHOW SLAVE STATUS
0a60f7b Tidy up and deal with review points.
ba103ff blr_slave.c: Update strncpy usage
467331e blr_master.c: Strncpy usage updates
d2b7c0c Merge remote-tracking branch 'origin/develop-nullauth-merge' into develop
5a8c1d0 qc: Measure execution time at the right place.
bccdb93 Merge branch 'NullAuthDeny' into develop
2e6511c Add 5.5.5 prefix to all version strings that lack it
314655a Improve DCB and session initialization and list handling
e1c43f0 MXS-655: Make MaxScale logging logrotate(8) compatible
ce36afd MXS-626: Don't log a header unless maxlog enabled
dcd47a7 blr_file.c: Replace uses of strncpy
6b8f576 bls_slave.c: Replace strncpy with memcpy
68a0039 Add list preallocation, tidy up, simplify init.
cb37d1b Fix copyright etc headers.
11a400d Tidy; comment; fix bad copies and mistakes.
7e36ec4 Add list manager files.
c4794e3 Initial code for list manager.
1b42e25 Merge remote-tracking branch 'origin/MXS-765' into develop
d50f617 Fix problems, extend tests, respond to review.
dcb4a91 Filter test folder removed
0b60dbe Add a couple of comments.
83cdba0 Fix overwriting problem.
ba5d353 Fix overwriting problem.
53671cb Small fixes in response to review.
173d049 blr.c: Review strncpy usage
4ff6ef2 binlog_common.c: Replace strncpy with memcpy
f238e03 maxbinlogcheck.s: Replace strncpy
9807f8d harness: Replace unnecessary use of strncpy
8c7fe6a avro: Modify strncpy usage
9b8008e Small improvements.
b7f784f Fix mistakes in testqueuemanager.c
cc26962 Restore missing poll.c code; add testqueuemanager.c.
2e91806 Format the filter harness
22059e6 Initial implementation connection queueing.
c604dc2 readwritesplit.c: Improve COM_INIT_DB handling
454d920 schemarouter.c: Replace strncpy with strcpy
8e85d66 sharding_common.c: Too long a database name handled explicitly
77f4446 Astyle schemarouter
491f7c2 maxinfo.c: Replace strncpy with memcpy
6b98105 maxinfo: Reformat with astyle
c1dbf08 Handle oversize user and database names
5fa4a0f Merge branch 'develop' of ssh://github.com/mariadb-corporation/maxscale-new into develop
706963b BLR_DBUSERS_TAIL new var in blr.h
d75b9af Tweak comments, remove trailing blanks.
ab2400a Optimise statistics gathering by inline & simpler fns.
fb59ddc Remove unnecessary strncpy/strncat usage in Binlog Server
bdcd551 resultset.c: Change strncpy to memcpy
c6b1c5e Reject rather than cut too long a path
6d8f112 Remove unnecessary strncpy/strncat usage
18bf5ed Remove unnecessary strncpy usage
dc0e2db Make maxpasswd more userfriendly
c9c8695 Fix calculation of padded_len in encryptPassword
2cfd2c6 dbusers.c: Check strncpy usage
7ab9342 Make more thorough checks in secrets_readKeys
be7d593 Format cli.c debugcli.c testroute.c webserver.c
1ee5efb config.c: Check usage of strncpy
3043b12 gq_utils.c: Unnecessary use of strncpy removed
77874ac Add help to maxkeys
38392a3 Update secrets_writeKeys documentation
2d1325c Make SSL optional in MaxScale's own communication
bda00da Fix avro build failures
b2cb31a Add more OOM macros
41ccf17 Fix strdup usage
a48f732 Fix realloc calls
20771f6 Add forgotten extern "C" block
8faf35a Add maxscale allocation functions
bb47890 Add macros for OOM logging
afea388 Fix silly mistakes.
6dafd22 Make deny default for null auth; move code from common to auth.
