pike.git/ lib/ modules/ Search.pmod/ Database.pmod/ MySQL.pike

Branch: Tag:

2022-09-14

2022-09-14 22:37:25 by Martin Nilsson <nilsson@fastmail.com>

A few more uses of default argument value.

2022-06-01

2022-06-01 23:38:49 by Martin Nilsson <nilsson@fastmail.com>

Added zero as return type on methods containing 'return 0'

2021-06-09

2021-06-09 13:26:05 by Chris Angelico <rosuav@gmail.com>

Add arg to some _sprintf callbacks to avoid spurious errors

2019-07-12

2019-07-12 14:01:04 by Marcus Comstedt <marcus@mc.pp.se>

Remove references to deprecated "master_sql"

2019-05-04

2019-05-04 09:12:19 by Arne Goedeke <el@laramies.com>

Merge remote-tracking branch 'origin/master' into new_utf8

2019-04-20

2019-04-20 11:03:22 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search: Protect lfuns.

2019-03-19

2019-03-19 12:33:55 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge commit '722771973bd' into patches/lyslyskom22891031

* commit '722771973bd': (6177 commits)
Verify that callablep responses are aligned with reality.
...

2019-03-14

2019-03-14 10:39:03 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge commit '2470270f500c728d10b8895314d8d8b07016e37b' into grubba/typechecker-automap

* commit '2470270f500c728d10b8895314d8d8b07016e37b': (18681 commits)
Removed the old typechecker.
...

2018-11-04

2018-11-04 16:11:11 by Arne Goedeke <el@laramies.com>

Merge remote-tracking branch 'origin/master' into new_utf8

2018-11-03

2018-11-03 14:21:37 by Marcus Comstedt <marcus@mc.pp.se>

Merge remote-tracking branch 'origin/8.1' into gobject-introspection

2018-02-15

2018-02-15 15:54:26 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge commit '75c9d1806f1a69ca21c27a2c2fe1b4a6ea38e77e' into patches/pike63

* commit '75c9d1806f1a69ca21c27a2c2fe1b4a6ea38e77e': (19587 commits)
...

2018-02-03

2018-02-03 09:48:55 by Peter Bortas <bortas@gmail.com>

Merge branch '8.1' into peter/travis

2018-01-17

2018-01-17 13:26:19 by Martin Karlgren <marty@roxen.com>

Search.MySQL: Fix splitting of blobs.

When an existing blob reached its size limit, the overflow blob(s)
got lost.

Potential fix for PIKE-52 (#8052).

2018-01-17 13:22:06 by Martin Karlgren <marty@roxen.com>

Search.MySQL: Fix splitting of blobs.

When an existing blob reached its size limit, the overflow blob(s)
got lost.

Potential fix for PIKE-52 (#8052).

2017-11-05

2017-11-05 15:53:18 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge branch 'grubba/rename_lfun_destroy' into 8.1

* grubba/rename_lfun_destroy:
Modules: Fixed logts of warnings.
Testsuite: Updated for LFUN::_destruct().
Compiler: Don't complain about LFUN::destroy() in compat mode.
Fix multiple warnings.
Runtime: LFUN::destroy() has been renamed to _destruct().
Compiler: Rename LFUN::destroy() to LFUN::_destruct().

2017-11-05 15:49:10 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Modules: Fixed logts of warnings.

More fall out from the LFUN::destroy ==> LFUN::_destruct rename.

2017-09-20

2017-09-20 12:10:23 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Added some optional consistency checks.

Defining SEARCH_DB_CONSISTENCY_CHECKS will now cause the module
to perform some extra SQL queries to make sure that the modified
entries are consistent.

2017-09-20 12:09:18 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Fixed truncation of word_hit.hits.

word_hit.hits should now lose its padding on the last write.

Also some related code cleanup.

2017-09-20 12:06:37 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Added some optional consistency checks.

Defining SEARCH_DB_CONSISTENCY_CHECKS will now cause the module
to perform some extra SQL queries to make sure that the modified
entries are consistent.

2017-09-20 12:03:02 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Fixed truncation of word_hit.hits.

word_hit.hits should now lose its padding on the last write.

Also some related code cleanup.

2017-09-19

2017-09-19 13:28:08 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Fixed bugs in store_to_db().

Work around for MySQL/MariaDB interpreting

INSERT('ABC', 3, 3, 'DEF')

as equivalent to just 'ABC' (and NOT 'ABCDEF').

Note that

INSERT('ABCX', 3, 3, 'DEF')

does result in 'ABCDEF'.

Also some code clean up to reduce the number of special cases.

2017-09-19 13:24:11 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Fixed bugs in store_to_db().

Work around for MySQL/MariaDB interpreting

INSERT('ABC', 3, 3, 'DEF')

as equivalent to just 'ABC' (and NOT 'ABCDEF').

Note that

INSERT('ABCX', 3, 3, 'DEF')

does result in 'ABCDEF'.

Also some code clean up to reduce the number of special cases.

2017-09-03

2017-09-03 10:46:36 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Updated type.

Fixes deprecation warning.

Fixes testsuite failures on platforms that have Mysql.

2017-08-05

2017-08-05 09:35:36 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Minor SQL query optimization.

Truncate the tables to clear them instead of removing the elements.

2017-06-09

2017-06-09 12:42:54 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Fixed race condition on update.

Data was retained between different critical zones.

2017-06-09 12:40:39 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Fixed race condition on update.

Data was retained between different critical zones.

2016-02-02

2016-02-02 14:05:30 by 0

Make glob expansion result stable and use a scoring strategy to prioritize
words when expansion result is capped (typically 100 in internal use).

2016-02-02 14:04:50 by 0

Make glob expansion result stable and use a scoring strategy to prioritize
words when expansion result is capped (typically 100 in internal use).

2015-10-11

2015-10-11 14:04:24 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge branch '8.1' into grubba/typechecker-mk3

* 8.1: (51 commits)
Compiler [Types]: Got rid of my_describe_type().
Compiler [Types]: Fixed wide strings in nested types.
Compiler [Types]: low_describe_type() now uses string_builder.
string_builder_sprintf: Fixed '+' for floats.
string_builder_sprintf: Added support for %T.
Calendar.mkrules: Improved backward compat.
Auth.OAuth.Client: Fixed AutoDoc markup issue.
New modules for communicating with different (RESTful) web apis (including OAuth and OAuth2 client modules) .
Documentation [%]: Added some clarifications and improved markup.
Inotify: Minor code cleanup. Potential fix for [CID 1325796].
Inotify: Removed some debug output.
Filesystem.Monitor [Inotify]: Improved support for blocking mode.
Inotify: Don't reference count immediate circular references.
Inotify: Removed some dead code.
System.Inotify, Filesystem.Monitor: Updates to use new Inotify API.
Inotify: Use fd_callback_boxes, internal event parsing.
Search.Database.MySQL: Censor the SQL-url.
Sql: Added censor_sql_url().
Search.Database.MySQL: Avoid long lived database connections.
Inotify: Don't throw errors in rm_watch() on invalid watch descriptor.
...

2015-10-07

2015-10-07 12:38:07 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Censor the SQL-url.

Also some minor code cleanup.

2015-10-07 12:37:47 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Avoid long lived database connections.

The automatic reconnection of Mysql connections has been disabled
by default for several years, so cached database connections may
very well become stale. The connection is now established when needed.

Fixes errors like "big_query(): Query ... failed (MySQL server has gone away)".

2015-10-07 12:34:20 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Censor the SQL-url.

Also some minor code cleanup.

2015-10-07 12:23:45 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Avoid long lived database connections.

The automatic reconnection of Mysql connections has been disabled
by default for several years, so cached database connections may
very well become stale. The connection is now established when needed.

Fixes errors like "big_query(): Query ... failed (MySQL server has gone away)".

2015-05-26

2015-05-26 10:33:42 by Martin Nilsson <nilsson@opera.com>

Removed trailing spaces.

2014-12-04

2014-12-04 19:27:14 by Martin Nilsson <nilsson@opera.com>

static -> protected

2014-12-04 19:26:20 by Martin Nilsson <nilsson@opera.com>

Non-zlib support.

2014-12-04 19:23:42 by 0

2014-12-04 19:23:42 by 0

2014-12-04 19:23:42 by 0

2014-12-04 19:23:42 by 0

2014-12-04 19:23:31 by 0

2014-12-04 19:23:09 by 0

2014-12-04 19:23:09 by 0

2014-09-04

2014-09-04 15:57:43 by Arne Goedeke <el@laramies.com>

Merge remote-tracking branch 'origin/8.0' into string_alloc

Conflicts:
src/stralloc.c

2014-08-29

2014-08-29 12:35:15 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Don't used DELAYED queries.

The DELAYED query modifier is:

* Not supported with InnoDB tables. This causes failures with
MySQL versions (eg 5.5) where InnoDB (and not MyISAM) is default.

* Ignored in MySQL 5.6.

* Deprecated in MySQL 5.6.6.

Fixes [bug 7255 (#7255)].

2014-08-29 12:34:41 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search.Database.MySQL: Don't used DELAYED queries.

The DELAYED query modifier is:

* Not supported with InnoDB tables. This causes failures with
MySQL versions (eg 5.5) where InnoDB (and not MyISAM) is default.

* Ignored in MySQL 5.6.

* Deprecated in MySQL 5.6.6.

Fixes [bug 7255 (#7255)].

2014-08-15

2014-08-15 13:49:53 by Martin Nilsson <nilsson@opera.com>

static -> protected

2014-05-16

2014-05-16 12:04:24 by Martin Nilsson <nilsson@opera.com>

Non-zlib support.

2014-01-05

2014-01-05 15:14:13 by Marcus Comstedt <marcus@mc.pp.se>

Merge branch '8.0' into gobject-introspection

2013-06-21

2013-06-21 09:18:55 by Arne Goedeke <el@laramies.com>

Merge remote-tracking branch 'origin/7.9' into pdf

2013-06-14

2013-06-14 09:47:28 by 0

The split_blobs() return value used a 0 for doc ID in the first tuple
which one of the call sites didn't expect (splitting blobs for words
not present in the database). We now always return a valid doc ID.

2013-06-14 09:47:28 by 0

Don't unlock and lock word_hit table every single iteration when syncing blobs.

2013-06-14 09:47:28 by 0

Introduce support for padded blobs. This is done to counter the performance
degradation seen when incrementally growing blobs which causes crazy record
fragmentation in MySQL's word_hit.MYD file.

The strategy is to grow blobs 25% when full, and otherwise write in place
and adjust counters to handle book-keeping of the padding. As a test case
I had a document containing ~1100 words (50% unique words) and indexed it
repeatedly to a database with 620.000 records; the old implementation
deteriorated from 270 ms -> 1100 ms per pass after 200 passes (although
OPTIMIZE TABLE would bring it back down again), while the new one is more
or less stable at 270-300 ms per pass throughout the whole test.

Padded blobs are not enabled by default unless higher layers of the
application indicates that it is ok. This is tested by resolving the
Search.Process.Compactor class since that code depdends on the blob format
and if an older compactor is present we cannot make the switch. (This would
preferably be solved by moving the compactor to this pmod.) If no compactor
is present the behavior (as well as the word_hit table definition) is
unchanged from earlier versions.

Note that if padding is activated in a multi-server setup with a shared
database all clients must be upgraded to the same implementation. The record
format is not backwards compatible since old blob readers will fail to skip
the padding, and old clients will write records with incorrect padding
counters.

2013-06-14 09:47:27 by 0

Add comment on restrictions on maximum blob size constant.

2013-06-14 09:33:11 by 0

Introduce support for padded blobs. This is done to counter the performance
degradation seen when incrementally growing blobs which causes crazy record
fragmentation in MySQL's word_hit.MYD file.

The strategy is to grow blobs 25% when full, and otherwise write in place
and adjust counters to handle book-keeping of the padding. As a test case
I had a document containing ~1100 words (50% unique words) and indexed it
repeatedly to a database with 620.000 records; the old implementation
deteriorated from 270 ms -> 1100 ms per pass after 200 passes (although
OPTIMIZE TABLE would bring it back down again), while the new one is more
or less stable at 270-300 ms per pass throughout the whole test.

Padded blobs are not enabled by default unless higher layers of the
application indicates that it is ok. This is tested by resolving the
Search.Process.Compactor class since that code depdends on the blob format
and if an older compactor is present we cannot make the switch. (This would
preferably be solved by moving the compactor to this pmod.) If no compactor
is present the behavior (as well as the word_hit table definition) is
unchanged from earlier versions.

Note that if padding is activated in a multi-server setup with a shared
database all clients must be upgraded to the same implementation. The record
format is not backwards compatible since old blob readers will fail to skip
the padding, and old clients will write records with incorrect padding
counters.

2013-06-14 09:33:11 by 0

Don't unlock and lock word_hit table every single iteration when syncing blobs.

2013-06-14 09:33:11 by 0

The split_blobs() return value used a 0 for doc ID in the first tuple
which one of the call sites didn't expect (splitting blobs for words
not present in the database). We now always return a valid doc ID.

2013-06-14 09:33:10 by 0

Add comment on restrictions on maximum blob size constant.

2013-06-11

2013-06-11 23:50:34 by Arne Goedeke <el@laramies.com>

Merge remote-tracking branch 'origin/7.9' into ba

2013-05-30

2013-05-30 11:47:59 by 0

Optimize query for word_hit blob length when there are several blobs for a
given word.

2013-05-30 11:47:16 by 0

Optimize query for word_hit blob length when there are several blobs for a
given word.

2013-02-06

2013-02-06 17:40:22 by 0

Add warning to catch unsynched data in memory before deallocating database
instance.

2013-02-05

2013-02-05 12:51:23 by 0

Indexing adds body and metadata fields in random order meaning that hit lists
aren't sorted numerically. This breaks subsequent phrase matching which assume
sorted hits. To correct this we sort blob data before writing it to the
database. Part of [bug 6679 (#6679)].

2013-02-05 12:48:23 by 0

Indexing adds body and metadata fields in random order meaning that hit lists
aren't sorted numerically. This breaks subsequent phrase matching which assume
sorted hits. To correct this we sort blob data before writing it to the
database. Part of [bug 6679 (#6679)].

2012-10-06

2012-10-06 11:38:03 by Marcus Comstedt <marcus@mc.pp.se>

Merge branch '7.9' into gobject-introspection

2012-07-01

2012-07-01 22:05:19 by Arne Goedeke <el@laramies.com>

Merge remote branch 'origin/7.9' into block_alloc

2012-06-20

2012-06-20 17:08:52 by Jonas Wallden <jonasw@roxen.com>

Clean up languageless document entries when indexing a multilanguage URI.
Part of [bug 6419 (#6419)].

2012-06-20 17:08:17 by Jonas Wallden <jonasw@roxen.com>

Clean up languageless document entries when indexing a multilanguage URI.
Part of [bug 6419 (#6419)].

2011-11-05

2011-11-05 15:02:44 by Martin Nilsson <nilsson@opera.com>

Removed $Id$.

2011-05-09

2011-05-09 00:23:37 by Martin Stjernholm <mast@lysator.liu.se>

Replaced the Roxen hack for publication date sorting with one that uses a
better field.

2011-05-09 00:23:28 by Martin Stjernholm <mast@lysator.liu.se>

Replaced the Roxen hack for publication date sorting with one that uses a
better field.

2011-02-01

2011-02-01 12:42:13 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search: Add an index on deleted_document.doc_id. Fixes [bug 5844 (#5844)].

2011-02-01 12:40:07 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Search: Add an index on deleted_document.doc_id. Fixes [bug 5844 (#5844)].

2010-10-26

2010-10-26 14:12:22 by Martin Stjernholm <mast@lysator.liu.se>

Reverted the move of Search.pmod from the Roxen module to Pike.

It stays in the Roxen Search module for Roxen <= 4.5, for compatibility. The
C-level part remains in Pike.

2010-10-26 10:03:44 by Martin Stjernholm <mast@lysator.liu.se>

More Roxen -> Pike adaptions in Search.pmod.

2010-10-26 09:52:45 by Martin Stjernholm <mast@lysator.liu.se>

More Roxen -> Pike adaptions in Search.pmod.

2010-10-06

2010-10-06 11:58:09 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge branch 'pike-modules/whitefish/4.5' into 7.4

2010-10-06 11:33:37 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge commit 'whitefish_split_4.5_5.0' into whitefish-fixup

This is in preparation to fixup Search.Process as well.

2010-10-06 10:14:50 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge remote branch 'pike-modules/whitefish/5.0' into 7.9

Updated Search.Utils and Search.Process to a non-roxen context.

Conflicts:
.gitattributes
.gitignore
lib/modules/.autodoc

2010-10-06 09:45:07 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Merge branch 'pike-modules/whitefish/5.0' into 7.8

Updated Search.Utils and Search.Process to a non-roxen context.

Conflicts:
.gitattributes
.gitignore
lib/modules/.autodoc

2010-01-20

2010-01-20 12:49:40 by Martin Jonsson <marty@roxen.com>

Add support for sorting on publication date (a.k.a visible from/external use) by specifying a sort="publdate" argument to the <search-results/> tag.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.92
Rev: lib/modules/Search.pmod/Query.pmod:1.37

2009-06-30

2009-06-30 12:35:33 by Fredrik Noring <noring@nocrew.org>

Added API function list_url_by_prefix(url_prefix, cb) to fetch indexed URL:s

- list_url_by_prefix(url_prefix, cb) is useful to implement e.g. a GC.

- For performance, coalesce several index content requests when possible.

- Added logging with timing values.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.10
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.91

2009-06-30 12:35:15 by Fredrik Noring <noring@nocrew.org>

Added API function list_url_by_prefix(url_prefix, cb) to fetch indexed URL:s

- list_url_by_prefix(url_prefix, cb) is useful to implement e.g. a GC.

- For performance, coalesce several index content requests when possible.

- Added logging with timing values.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.10
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.90

2009-06-26

2009-06-26 14:29:27 by Fredrik Noring <noring@nocrew.org>

Fixed leaked database URI entries when removing documents from index

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.8
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.90

2009-06-26 14:28:09 by Fredrik Noring <noring@nocrew.org>

Fixed leaked database URI entries when removing documents from index

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.8
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.89

2009-05-25

2009-05-25 18:26:52 by Martin Stjernholm <mast@lysator.liu.se>

Updated copyright notices.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.7
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.89
Rev: lib/modules/Search.pmod/Filter.pmod/Base.pike:1.18
Rev: lib/modules/Search.pmod/Filter.pmod/Charset.pmod:1.7
Rev: lib/modules/Search.pmod/Filter.pmod/HTML.pmod:1.44
Rev: lib/modules/Search.pmod/Filter.pmod/PDF.pmod:1.17
Rev: lib/modules/Search.pmod/Filter.pmod/PlainText.pmod:1.10
Rev: lib/modules/Search.pmod/Filter.pmod/Word.pmod:1.13
Rev: lib/modules/Search.pmod/Grammar.pmod/AbstractParser.pike:1.8
Rev: lib/modules/Search.pmod/Grammar.pmod/DefaultParser.pike:1.14
Rev: lib/modules/Search.pmod/Grammar.pmod/Lexer.pmod:1.9
Rev: lib/modules/Search.pmod/Grammar.pmod/debug.h:1.4
Rev: lib/modules/Search.pmod/Grammar.pmod/module.pmod:1.16
Rev: lib/modules/Search.pmod/Indexer.pmod:1.19
Rev: lib/modules/Search.pmod/Query.pmod:1.36
Rev: lib/modules/Search.pmod/RankingProfile.pike:1.21
Rev: lib/modules/Search.pmod/ResultSet.pike:1.3
Rev: lib/modules/Search.pmod/Utils.pmod:1.49
Rev: lib/modules/Search.pmod/module.pmod:1.18
Rev: lib/modules/Search.pmod/search.h:1.2

2009-05-25 12:41:23 by Martin Stjernholm <mast@lysator.liu.se>

Updated copyright notices.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.7
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.88
Rev: lib/modules/Search.pmod/Filter.pmod/Base.pike:1.18
Rev: lib/modules/Search.pmod/Filter.pmod/Charset.pmod:1.7
Rev: lib/modules/Search.pmod/Filter.pmod/HTML.pmod:1.44
Rev: lib/modules/Search.pmod/Filter.pmod/PDF.pmod:1.17
Rev: lib/modules/Search.pmod/Filter.pmod/PlainText.pmod:1.10
Rev: lib/modules/Search.pmod/Filter.pmod/Word.pmod:1.13
Rev: lib/modules/Search.pmod/Grammar.pmod/AbstractParser.pike:1.8
Rev: lib/modules/Search.pmod/Grammar.pmod/DefaultParser.pike:1.14
Rev: lib/modules/Search.pmod/Grammar.pmod/Lexer.pmod:1.9
Rev: lib/modules/Search.pmod/Grammar.pmod/debug.h:1.4
Rev: lib/modules/Search.pmod/Grammar.pmod/module.pmod:1.16
Rev: lib/modules/Search.pmod/Indexer.pmod:1.19
Rev: lib/modules/Search.pmod/Query.pmod:1.36
Rev: lib/modules/Search.pmod/RankingProfile.pike:1.20
Rev: lib/modules/Search.pmod/ResultSet.pike:1.3
Rev: lib/modules/Search.pmod/Utils.pmod:1.48
Rev: lib/modules/Search.pmod/module.pmod:1.18
Rev: lib/modules/Search.pmod/search.h:1.2

2008-08-15

2008-08-15 14:54:40 by Martin Stjernholm <mast@lysator.liu.se>

Moving to pike 7.8 and dropping compatibility with earlier pikes.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.88
Rev: lib/modules/Search.pmod/Queue.pmod/MySQL.pike:1.23

2008-03-26

2008-03-26 18:09:15 by Jonas Wallden <jonasw@roxen.com>

Implement negation of fields and globs by introducing a result set for all
documents in the database.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.87
Rev: lib/modules/Search.pmod/Grammar.pmod/module.pmod:1.13

2007-11-16

2007-11-16 10:12:03 by Marcus Wellhardh <wellhard@roxen.com>

Fix for [RT 10311] (partial). Added support for purge so documents will be removed from the index, i.e. add documents to the deleted_document table when a directory/file are purged. Also added a new button "Run Database Maintenance" in the "Database Profile/DB Settings" tab.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.6
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.86
Rev: lib/modules/Search.pmod/Queue.pmod/Base.pike:1.2
Rev: lib/modules/Search.pmod/Queue.pmod/MySQL.pike:1.21

2007-11-07

2007-11-07 16:50:15 by Marcus Wellhardh <wellhard@roxen.com>

Fixed [RT 10336]. Remove search words that will not fit into the database column size (64). Error symptom was: big_query(): Query failed (Duplicate entry ... for key 1).

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.85

2007-03-14

2007-03-14 18:03:01 by Anders Johansson <anders@roxen.com>

Fixed searching for words with for example the danish letter ae.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.79

2005-06-02

2005-06-02 16:27:22 by Martin Stjernholm <mast@lysator.liu.se>

Ensure that the tables don't remain locked if an error is thrown.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.84

2005-05-27

2005-05-27 11:39:52 by Martin Stjernholm <mast@lysator.liu.se>

Some pike 7.6 compat fixes.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.83
Rev: lib/modules/Search.pmod/Queue.pmod/MySQL.pike:1.20

2005-05-23

2005-05-23 15:18:13 by Anders Johansson <anders@roxen.com>

Lower performance impact on search queries while storing indexing results to DB. It was a problem for large databases where a single update query can take a number of seconds.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.82

2004-12-20

2004-12-20 12:04:49 by Anders Johansson <anders@roxen.com>

Fix bug causing frontends to try to write to a readonly search database. [bug 3830 (#3830)]

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.81

2004-12-20 12:04:49 by Anders Johansson <anders@roxen.com>

Fix bug causing frontends to try to write to a readonly search database. [bug 3830 (#3830)]

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.78

2004-08-24

2004-08-24 19:48:03 by Johan Schön <js@opera.com>

UTF8-encode words before storing them in MySQL, when not using the mergefile code path.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.80

2004-08-19

2004-08-19 11:35:21 by Fredrik Noring <noring@nocrew.org>

Potential fix for [bug 3772 (#3772)].

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.79

2004-08-18

2004-08-18 12:29:19 by Henrik Grubbström (Grubba) <grubba@grubba.org>

Now supports using the _WhiteFish.Blobs from 4.1.
Fixes [bug 3771 (#3771)].

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.77

2004-08-08

2004-08-08 14:22:53 by Johan Schön <js@opera.com>

Made the mergefile indexing method optional

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.78
Rev: lib/modules/Search.pmod/Indexer.pmod:1.18

2004-08-07

2004-08-07 15:27:00 by Johan Schön <js@opera.com>

Merge from external js/per/nilsson fork

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.5
Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.20(DEAD)
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.77
Rev: lib/modules/Search.pmod/Filter.pmod/Base.pike:1.16
Rev: lib/modules/Search.pmod/Filter.pmod/HTML.pmod:1.37
Rev: lib/modules/Search.pmod/Filter.pmod/Output.pike:1.1
Rev: lib/modules/Search.pmod/Filter.pmod/PDF.pmod:1.16
Rev: lib/modules/Search.pmod/Filter.pmod/PlainText.pmod:1.9
Rev: lib/modules/Search.pmod/Filter.pmod/Word.pmod:1.12
Rev: lib/modules/Search.pmod/Grammar.pmod/AbstractParser.pike:1.7
Rev: lib/modules/Search.pmod/Grammar.pmod/DefaultParser.pike:1.11
Rev: lib/modules/Search.pmod/Grammar.pmod/Lexer.pmod:1.7
Rev: lib/modules/Search.pmod/Grammar.pmod/module.pmod:1.12
Rev: lib/modules/Search.pmod/Indexer.pmod:1.17
Rev: lib/modules/Search.pmod/MergeFile.pike:1.1
Rev: lib/modules/Search.pmod/Process.pmod:1.12
Rev: lib/modules/Search.pmod/Query.pmod:1.26
Rev: src/modules/_WhiteFish/Makefile.in:1.7
Rev: src/modules/_WhiteFish/blob.c:1.34
Rev: src/modules/_WhiteFish/blobs.c:1.12
Rev: src/modules/_WhiteFish/buffer.c:1.13
Rev: src/modules/_WhiteFish/linkfarm.c:1.5
Rev: src/modules/_WhiteFish/resultset.c:1.24
Rev: src/modules/_WhiteFish/resultset.h:1.9
Rev: src/modules/_WhiteFish/whitefish.c:1.35

2004-03-02

2004-03-02 16:57:59 by Stefan Wallström <stewa@roxen.com>

Index path components 1 and 2 of the URI as separate data fields to make searching sub sections of a site easier.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.76

2004-03-02 16:57:50 by Stefan Wallström <stewa@roxen.com>

Index path components 1 and 2 of the URI as separate data fields to make searching sub sections of a site easier.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.76

2003-01-27

2003-01-27 15:10:10 by Mattias Andersson <matan@lysator.liu.se>

New features in 3.4, like renaming of profiles and indexing of any sitebuilder metadata, and updated version of wvWare.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.19
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.75
Rev: lib/modules/Search.pmod/Filter.pmod/PDF.pmod:1.13
Rev: lib/modules/Search.pmod/Filter.pmod/Word.pmod:1.10
Rev: lib/modules/Search.pmod/Filter.pmod/wvHtml.xml:1.3
Rev: lib/modules/Search.pmod/Process.pmod:1.10
Rev: lib/modules/Search.pmod/Utils.pmod:1.36

2002-11-26

2002-11-26 10:41:50 by Mattias Andersson <matan@lysator.liu.se>

Fixes problem with word break of search results when using globs.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.74

2002-05-15

2002-05-15 14:28:53 by Mattias Andersson <matan@lysator.liu.se>

Backport of glob search from 2.4.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.73
Rev: lib/modules/Search.pmod/Grammar.pmod/DefaultParser.pike:1.8
Rev: lib/modules/Search.pmod/Query.pmod:1.24

2002-03-12

2002-03-12 15:38:07 by Johan Schön <js@opera.com>

Added glob (* and ?) searches

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.73
Rev: lib/modules/Search.pmod/Grammar.pmod/DefaultParser.pike:1.10
Rev: lib/modules/Search.pmod/Query.pmod:1.24

2001-11-21

2001-11-21 14:47:33 by Johan Schön <js@opera.com>

Workaround for [Bug 2548 (#2548)].

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.72

2001-09-26

2001-09-26 09:17:35 by Anders Johansson <anders@roxen.com>

Fixed syntax error in query.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.71

2001-09-26 05:09:01 by Johan Schön <js@opera.com>

This table definition change speeds up the compacter a lot.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.70

2001-09-25

2001-09-25 22:02:38 by Johan Schön <js@opera.com>

Added a query-local blob cache with 10 entries, which is populated by reading 10 blob packets at once if possible. This reduces i/o time wait by as much as 50%

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.18
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.69
Rev: lib/modules/Search.pmod/Indexer.pmod:1.16
Rev: lib/modules/Search.pmod/Query.pmod:1.23

2001-08-31

2001-08-31 16:34:27 by Johan Schön <js@opera.com>

Changed inline doc to say that we use two letter ISO-639-1 language codes. Also changed database definition to allow 255 char language codes, to be on the safe side.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.17
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.68

2001-08-30

2001-08-30 20:36:16 by Johan Schön <js@opera.com>

Cleaned up debug message. Removed a catch from db->query to stop it from looping thousands of
times with an error each time.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.67

2001-08-20

2001-08-20 20:01:54 by Johan Schön <js@opera.com>

Added remove_metadata.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.16
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.66

2001-08-17

2001-08-17 22:56:42 by Johan Schön <js@opera.com>

Use Unicode Compose ("C") normalization on the words extracted. Also yet another bugfix for that unique word count method.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.65

2001-08-17 16:07:14 by Martin Nilsson <mani@lysator.liu.se>

Better approximation

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.64

2001-08-17 09:01:45 by Anders Johansson <anders@roxen.com>

Bugfixed the bugfix. :)

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.63

2001-08-16

2001-08-16 22:25:35 by Johan Schön <js@opera.com>

Bugfix :)

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.62

2001-08-16 22:07:16 by Johan Schön <js@opera.com>

Added statistics for the db profile info page

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.15
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.61

2001-08-16 21:27:49 by Martin Nilsson <mani@lysator.liu.se>

Readded debug, since it appears to need some debugging.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.60

2001-08-13

2001-08-13 19:51:32 by Martin Nilsson <mani@lysator.liu.se>

Removed some more debug

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.59

2001-08-09

2001-08-09 14:56:35 by Martin Nilsson <mani@lysator.liu.se>

Less debug

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.58

2001-08-09 11:39:10 by Johan Schön <js@opera.com>

Don't re-allocate already allocated field id's

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.57

2001-08-08

2001-08-08 11:00:08 by Per Hedbor <ph@opera.com>

Removed the queue from here, since it's elsewhere too

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.56

2001-08-08 08:51:52 by Anders Johansson <anders@roxen.com>

Return the correct type from get_special_metadata.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.55

2001-08-07

2001-08-07 15:33:07 by Johan Schön <js@opera.com>

Added get_special_metadata. (A better name wouldn't hurt.)

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.13
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.54

2001-08-07 15:17:32 by Johan Schön <js@opera.com>

AC filtering now works. (But only for URLs in the sitebuilder executing the query, so far.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.12
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.53

2001-08-01

2001-08-01 15:53:13 by Johan Schön <js@opera.com>

UTF-8 decode *all* metadata fields

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.52

2001-07-31

2001-07-31 15:29:47 by Johan Schön <js@opera.com>

Removed the unused anchor field

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.10
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.51
Rev: lib/modules/Search.pmod/Queue.pmod/MySQL.pike:1.2
Rev: lib/modules/Search.pmod/RankingProfile.pike:1.16
Rev: lib/modules/Search.pmod/Utils.pmod:1.19
Rev: src/modules/_WhiteFish/blob.c:1.26
Rev: src/modules/_WhiteFish/blob.h:1.13
Rev: src/modules/_WhiteFish/blobs.c:1.6
Rev: src/modules/_WhiteFish/whitefish.c:1.32

2001-07-16

2001-07-16 14:49:15 by Johan Schön <js@opera.com>

Added API method for retrieving dead documents.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.9
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.50

2001-07-04

2001-07-04 22:30:07 by Martin Nilsson <mani@lysator.liu.se>

Unique is for wimps.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.49

2001-07-04 20:46:32 by Martin Nilsson <mani@lysator.liu.se>

Fix

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.48

2001-07-04 20:42:21 by Martin Nilsson <mani@lysator.liu.se>

Index the words using strings rather than ints. Queries temporary disabled, since the do_query_foo() functions in _Whitefish need to be modified to use strings instead of ints.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.8
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.47
Rev: lib/modules/Search.pmod/Query.pmod:1.17
Rev: lib/modules/Search.pmod/Utils.pmod:1.8

2001-07-02

2001-07-02 13:06:53 by Johan Schön <js@opera.com>

Added database abstraction API docs, and cleaned up the MySQL db implementation accordingly.

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.7
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.46

2001-06-27

2001-06-27 06:08:59 by Johan Schön <js@opera.com>

Changed recreate_tables() to init_tables().

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.45

2001-06-26

2001-06-26 04:05:47 by Johan Schön <js@opera.com>

Changed 'field' table definition. Added 2*clear().

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.44

2001-06-26 01:55:32 by Johan Schön <js@opera.com>

Added force argument to add_uri. Blahonga.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.43

2001-06-23

2001-06-23 02:11:58 by Johan Schön <js@opera.com>

Removed debug

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.42

2001-06-23 02:09:20 by Johan Schön <js@opera.com>

Various bugfixes

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.41

2001-06-23 00:29:18 by Johan Schön <js@opera.com>

Typos

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.40

2001-06-23 00:25:50 by Johan Schön <js@opera.com>

Make sure the necessary fields are allocated before any function that uses them runs.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.39

2001-06-22

2001-06-22 23:55:40 by Johan Schön <js@opera.com>

Added a cache for list_fields()

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.38

2001-06-22 23:45:52 by Johan Schön <js@opera.com>

Added mapping(string:int) list_fields()

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.37

2001-06-22 23:28:45 by Johan Schön <js@opera.com>

Added a naive but hopefully working custom field id allocation algorithm

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.36

2001-06-22 23:05:24 by Johan Schön <js@opera.com>

Added a uri_md5 field in the queue table, and used it for all uri-based indexing. (URI's can sometimes be quite long, and the maximum index width in MySQL is 255 chars.)

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.35

2001-06-22 01:28:36 by Martin Nilsson <mani@lysator.liu.se>

Adding copyright message

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.6
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.34
Rev: lib/modules/Search.pmod/Filter.pmod/Base.pike:1.7
Rev: lib/modules/Search.pmod/Filter.pmod/Charset.pmod:1.4
Rev: lib/modules/Search.pmod/Filter.pmod/HTML.pmod:1.13
Rev: lib/modules/Search.pmod/Filter.pmod/PlainText.pmod:1.5
Rev: lib/modules/Search.pmod/Grammar.pmod/AbstractParser.pike:1.5
Rev: lib/modules/Search.pmod/Grammar.pmod/DefaultParser.pike:1.7
Rev: lib/modules/Search.pmod/Grammar.pmod/Lexer.pmod:1.5
Rev: lib/modules/Search.pmod/Grammar.pmod/debug.h:1.2
Rev: lib/modules/Search.pmod/Grammar.pmod/module.pmod:1.9
Rev: lib/modules/Search.pmod/Indexer.pmod:1.12
Rev: lib/modules/Search.pmod/Query.pmod:1.16
Rev: lib/modules/Search.pmod/RankingProfile.pike:1.13
Rev: lib/modules/Search.pmod/ResultSet.pike:1.2
Rev: lib/modules/Search.pmod/Utils.pmod:1.5
Rev: lib/modules/Search.pmod/module.pmod:1.15

2001-06-14

2001-06-14 13:32:30 by Johan Schön <js@opera.com>

Remove index.xml, index.html and index.htm from URLs before they are added to the queue. Also made the md5 checksum in the queue hex-ified

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.33

2001-06-11

2001-06-11 13:56:43 by Johan Schön <js@opera.com>

Catch any sql errors in sync()

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.32

2001-06-11 13:38:56 by Johan Schön <js@opera.com>

Typo

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.31

2001-06-11 13:38:11 by Johan Schön <js@opera.com>

Don't utf8-encode/decode the Gz:ed body string twice

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.30

2001-06-11 10:44:37 by Johan Schön <js@opera.com>

hash_word was used from elsewhere. Also fixed an off-by one indexing error.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.29

2001-06-11 09:59:36 by Per Hedbor <ph@opera.com>

Bugfix

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.28

2001-06-11 09:56:14 by Per Hedbor <ph@opera.com>

Optimized for speed

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.27

2001-06-10

2001-06-10 11:29:39 by Per Hedbor <ph@opera.com>

New Queue.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.26

2001-06-09

2001-06-09 23:21:51 by Johan Schön <js@opera.com>

Made md5 checksum hexified. Added Queue. Marked lots of symbols static.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.25

2001-06-05

2001-06-05 22:54:24 by Per Hedbor <ph@opera.com>

utf8-encode

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.24

2001-06-04

2001-06-04 13:43:24 by Johan Schön <js@opera.com>

Fixes a logic bug

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.23

2001-05-31

2001-05-31 06:08:28 by Johan Schön <js@opera.com>

Store and use document metadata. Also implemented db->remove_document()

Rev: lib/modules/Search.pmod/Database.pmod/Base.pmod:1.2
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.22
Rev: lib/modules/Search.pmod/Filter.pmod/Base.pike:1.5
Rev: lib/modules/Search.pmod/Grammar.pmod/Parser.pmod:1.2
Rev: lib/modules/Search.pmod/Indexer.pmod:1.6

2001-05-31 00:59:26 by Johan Schön <js@opera.com>

Cleaning up

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.21

2001-05-29

2001-05-29 12:25:28 by Johan Schön <js@opera.com>

Wrapper for the _WhiteFish.ResultSet class

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.20
Rev: lib/modules/Search.pmod/Filter.pmod/Base.pike:1.4
Rev: lib/modules/Search.pmod/ResultSet.pike:1.1

2001-05-26

2001-05-26 14:59:01 by Per Hedbor <ph@opera.com>

Better looking debug printouts and threaded sync()

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.19
Rev: lib/modules/Search.pmod/Filter.pmod/HTML.pmod:1.3
Rev: lib/modules/Search.pmod/Indexer.pmod:1.4

2001-05-26 14:11:35 by Per Hedbor <ph@opera.com>

Major speed optimization overhaul

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.18
Rev: lib/modules/Search.pmod/Filter.pmod/HTML.pmod:1.2
Rev: lib/modules/Search.pmod/Indexer.pmod:1.3
Rev: lib/modules/Search.pmod/Utils.pmod:1.2
Rev: src/modules/_WhiteFish/blob.c:1.20
Rev: src/modules/_WhiteFish/blob.h:1.10
Rev: src/modules/_WhiteFish/blobs.c:1.3

2001-05-25

2001-05-25 21:02:49 by Per Hedbor <ph@opera.com>

Optimized (hopefully) sync

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.17

2001-05-25 20:55:34 by Johan Schön <js@opera.com>

Optimization.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.16

2001-05-25 15:01:24 by Johan Schön <js@opera.com>

Now uses _WhiteFish.Blob instead.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.15

2001-05-17

2001-05-17 12:49:03 by Johan Schön <js@opera.com>

Redesigned again

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.14

2001-04-05

2001-04-05 10:26:27 by Johan Schön <js@opera.com>

Continued rewriting the SQL generating code

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.13

2001-03-28

2001-03-28 12:07:27 by Johan Schön <js@opera.com>

Halfway through rewriting the db layer to use temporary tables instead of one big join. It would be nice if MySQL could optimize joins a bit better...

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.12

2001-03-20

2001-03-20 04:55:39 by Johan Schön <js@opera.com>

Fixes some quoting problems in the sql queries.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.11

2001-03-19

2001-03-19 04:11:32 by Johan Schön <js@opera.com>

Generated SQL now more in sync with SQL table definitions.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.10

2001-03-19 03:51:31 by Johan Schön <js@opera.com>

Now compiles again

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.9

2001-03-18

2001-03-18 05:19:46 by Johan Schön <js@opera.com>

Many changes: adopting the insert methods to the new database design

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.8

2001-03-15

2001-03-15 21:34:01 by Johan Schön <js@opera.com>

Made Groupnode::build_sql() a lot less ugly. Added the leaf node Language.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.7

2001-03-15 16:03:41 by Johan Schön <js@opera.com>

Added code for generating a single SELECT statement for every possible query. This changes the database interface *ahem* somewhat.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.6

2001-01-05

2001-01-05 20:54:47 by Martin Nilsson <mani@lysator.liu.se>

More tables and no hex-version of MD5-id:s.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.5

2001-01-02

2001-01-02 12:49:18 by Johan Schön <js@opera.com>

More indices

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.4

2000-12-18

2000-12-18 22:46:16 by Martin Nilsson <mani@lysator.liu.se>

Changed API and return array(mapping) to queries, instead of mapping(string:Search.Document). Let MySQL do the trivial ranking stuff, like counting hits.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.3

2000-11-30

2000-11-30 16:19:15 by Johan Schön <js@opera.com>

More changes.

Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.2
Rev: lib/modules/Search.pmod/Filter.pmod/HTML.pike:1.2
Rev: lib/modules/Search.pmod/Filter.pmod/PlainText.pike:1.2
Rev: lib/modules/Search.pmod/Grammar.pmod/GoogleStyle.pike:1.2
Rev: lib/modules/Search.pmod/filters/text_html.pike:1.6(DEAD)
Rev: lib/modules/Search.pmod/filters/text_plain.pike:1.4(DEAD)
Rev: lib/modules/Search.pmod/module.pmod:1.8

2000-11-24

2000-11-24 00:24:16 by Johan Schön <js@opera.com>

Some restructuring :)

Rev: lib/modules/Search.pmod/Database.pmod/Base.pike:1.1
Rev: lib/modules/Search.pmod/Database.pmod/MySQL.pike:1.1
Rev: lib/modules/Search.pmod/Filter.pmod/Base.pike:1.1
Rev: lib/modules/Search.pmod/Filter.pmod/HTML.pike:1.1
Rev: lib/modules/Search.pmod/Filter.pmod/PlainText.pike:1.1
Rev: lib/modules/Search.pmod/Grammar.pmod/Base.pike:1.1
Rev: lib/modules/Search.pmod/Grammar.pmod/Foo.pike:1.1
Rev: lib/modules/Search.pmod/Grammar.pmod/GoogleStyle.pike:1.1
Rev: lib/modules/Search.pmod/Grammar.pmod/Naive.pike:1.1
Rev: lib/modules/Search.pmod/Grammar.pmod/Simple.pike:1.1
Rev: lib/modules/Search.pmod/module.pmod:1.7
Rev: lib/modules/Search.pmod/simple_sql_db.pike:1.8(DEAD)