1
0
Fork 0
Commit Graph

126 Commits

Author SHA1 Message Date
Jan Steemann 92abde4932 more tests for indexes to overcome getIndexes() mysteries (related to issue #366) 2013-01-23 10:43:20 +01:00
Jan Steemann 346e364767 issue #366 2013-01-23 10:28:00 +01:00
Jan Steemann 452a266475 fixed memory error 2013-01-18 17:56:38 +01:00
Jan Steemann 9a003247d0 removed the _bidirectional attribute for edges
This will be reimplemented later on collection, and not on individual edges level.
Lucas & the others were sure directed & undirected edges should not be mixed in the same collection
2013-01-18 17:02:30 +01:00
Jan Steemann bbade8c0b8 removed the non-working indexSubstrings property for fulltext indexing 2013-01-17 15:58:28 +01:00
Jan Steemann 7a6db16241 added db._version() 2013-01-15 16:44:48 +01:00
Jan Steemann 2e5f65a2ad speed up insertion into edges index 2013-01-15 15:13:34 +01:00
Jan Steemann 4818f22b21 fixes 2013-01-14 02:58:19 +01:00
Jan Steemann 459f935d00 intptr_t to uintptr_t conversion 2013-01-02 15:00:59 +01:00
Jan Steemann 33ea87fccb cppcheck 2012-12-19 09:20:29 +01:00
Jan Steemann 669fe4275a cppcheck 2012-12-18 16:44:06 +01:00
Jan Steemann 6c0b840db0 clang fixes 2012-12-18 15:28:28 +01:00
Jan Steemann 96637435ab fulltext: more query operators, more tests, more controlled memory usage and statistics 2012-12-18 14:56:06 +01:00
Jan Steemann 0374ed417e fixed & reactivated compaction 2012-12-15 02:17:37 +01:00
Jan Steemann b8ca0b2d4d fulltext index massive rewrite 2012-12-15 01:01:25 +01:00
Jan Steemann b5d0457f3a fixed some issues reported by cppcheck 2012-12-12 18:42:11 +01:00
Jan Steemann 691ec4d536 more verbosity on failed fulltext insertion 2012-12-10 15:50:49 +01:00
Jan Steemann fb4afd0353 fulltext index 2012-12-10 12:16:48 +01:00
Jan Steemann 7c5d36bf0f adjusted default sizes 2012-12-07 18:04:29 +01:00
Jan Steemann cbaa57a1bd cleanup of fulltext indexes, resizing etc. 2012-12-07 18:00:01 +01:00
Jan Steemann 7261465139 Merge branch 'devel' of https://github.com/triAGENS/ArangoDB into devel 2012-12-07 13:51:31 +01:00
Jan Steemann c49ce0a52f fulltext resizing, still incomplete 2012-12-07 13:51:19 +01:00
Jan Steemann 63b4c94b0f fulltext resizing, halfway 2012-12-07 13:50:29 +01:00
a-brandt c4cc85ab9f memory bugfix 2012-12-07 13:47:02 +01:00
a-brandt 1794ea49fa added new function for getting words 2012-12-07 13:09:16 +01:00
Jan Steemann f89475df47 some cleanup 2012-12-07 09:53:00 +01:00
Jan Steemann 0b001b7a87 fixed memory leaks, added newest code by Richard 2012-12-06 16:43:33 +01:00
Jan Steemann 38809d4612 added callbacks for index cleanup 2012-12-06 15:46:11 +01:00
Jan Steemann 69986a5de0 fixed a segfault when searching for empty words in fulltext 2012-12-06 09:28:02 +01:00
Jan Steemann 21fa3ed616 small stuff, oom handling etc. 2012-12-04 21:21:26 +01:00
Jan Steemann 1811ce9186 fixed some issues found by cppcheck 2012-12-04 18:21:29 +01:00
Jan Steemann 5742ab6fba fixed unicode lowercasing 2012-12-04 11:12:38 +01:00
Jan Steemann fcc880e021 use TRI_Allocate 2012-12-03 17:54:55 +01:00
Jan Steemann d8506d7a13 implemented updating of documents 2012-12-03 17:32:04 +01:00
Jan Steemann 1130df4330 utf8 lower-casing and normalisation for fulltext strings 2012-12-03 17:14:54 +01:00
Jan Steemann 9b44e4aa45 added very limited & alpha fulltext query capability
db.collection.FULLTEXT() allows to issue queries using a fulltext index.
Needs to be passed a comma-separated query string with the search words.
Search words can be prefixed with either:
* prefix: does a prefix search
* complete: does a complete-match search
* substring: does a substring search

This is still very very alpha. It crashes and produces wrong results in some cases.
2012-12-02 03:24:20 +01:00
Jan Steemann d2aef2dcdc add words to fulltext index
this parses the fulltext-indexed attributes of documents when there's a fulltext index, and adds the individual words to the index.
As the fulltext index is case-sensitive, all words are added to the index in lower case.
The text tokenisation implementation is still very naive and currently works properly only for character ranges [a-z] and [A-Z].
Unicode words are also supported, but they are not normalised nor lower-cased yet. Additionally, unicode punctuation characters are not excluded and will also be added to the index.
Updating documents that are fulltext-indexed currently does not work.
2012-12-02 00:55:59 +01:00
Jan Steemann f517aa0059 ensureFulltextIndex 2012-11-30 17:53:02 +01:00
Jan Steemann 19fb9d0aea added stub functions for fulltext indexes 2012-11-30 14:50:02 +01:00
Jan Steemann f84207e368 removed redundant declaration of edges index. edges index is now a regular index 2012-11-27 14:14:31 +01:00
Jan Steemann eac11fb793 issue #293 2012-11-24 03:16:23 +01:00
a-brandt 71d3d54de1 collection info struct changed 2012-11-15 11:24:35 +01:00
Jan Steemann d1bc1513aa moved creation of primary index into index.c 2012-10-31 13:55:57 +01:00
Jan Steemann 21735bb267 minor changes 2012-10-29 15:34:28 +01:00
Frank Celler 3aebed3020 Merge branch '1.1' of github.com:triAGENS/ArangoDB into 1.1
Conflicts:
	Documentation/Makefile.files
2012-10-17 18:03:46 +02:00
Frank Celler 1efad1d9c7 fixed file permissions 2012-10-17 18:01:57 +02:00
Jan Steemann c5996869b6 hash index fixes 2012-10-16 17:43:22 +02:00
Jan Steemann 0beeeb6c2a reduce element size in hash index 2012-10-16 15:15:31 +02:00
Jan Steemann eba0456c05 added some memory checks 2012-10-11 15:49:20 +02:00
Jan Steemann a4b979222f fixed memory zone issues, memleaks 2012-10-10 19:05:16 +02:00
Jan Steemann 6847caac80 changed misleading member name (secondaryIndexes => allIndexes) 2012-10-10 11:24:36 +02:00
Oreste Panaia 2a71a3c371 minor changes for windows compatability 2012-09-28 17:34:22 +08:00
Oreste Panaia f5316acb9d Merge branch 'devel' of https://github.com/triAGENS/ArangoDB into devel
Conflicts:
	arangod/Ahuacatl/ahuacatl-grammar.c
	arangod/V8Server/v8-vocbase.cpp
	arangod/VocBase/document-collection.c
	arangod/VocBase/document-collection.h
2012-09-17 22:40:12 +08:00
Oreste Panaia c7e5eae52d Added additional error numbers for bitarray index creation 2012-09-17 19:57:10 +08:00
Jan Steemann 1ef037ce8c renamed indexes to secondaryIndexes 2012-09-17 09:54:12 +02:00
Jan Steemann 1b1d7ba877 renamed simple_collection to document_collection 2012-09-14 14:46:37 +02:00
Jan Steemann 9bc99b7c9d renamed document collection to primary collection 2012-09-14 10:19:41 +02:00
Jan Steemann cfe44acda4 moved variable to top-level scope. otherwise bad(TM) things may happen 2012-09-11 10:45:47 +02:00
Oreste Panaia f93eef34d7 git
Merge branch 'devel' of https://github.com/triAGENS/ArangoDB into devel

Conflicts:
	lib/ShapedJson/json-shaper.c
2012-09-10 20:53:41 +08:00
Oreste Panaia 8847d87580 added limited AQL support for bit indexes 2012-09-10 20:42:14 +08:00
Jan Steemann febd5bfa8e removed member TRI_doc_mptr_t->_document 2012-09-04 13:20:01 +02:00
Oreste Panaia d632fb820b Merge branch 'devel' of https://github.com/triAGENS/ArangoDB into devel 2012-09-03 17:09:00 +08:00
Oreste Panaia b7d239f629 draft index interface 2012-09-03 17:05:39 +08:00
Frank Celler 8b34bbb9de fixed permission 2012-08-31 08:59:59 +02:00
Jan Steemann b8e7d5634d issue #153: edge collection should be a flag for a collection 2012-08-28 10:48:24 +02:00
Oreste Panaia 03cce62c3a bitarray index now support explicit values 2012-08-02 15:20:43 +08:00
Frank Celler f126016484 added ExtractShapedJsonVocShaper 2012-07-16 15:42:41 +02:00
Frank Celler 4c517c6d3c removed exec flag 2012-07-10 10:46:20 +02:00
Oreste Panaia c72f44c949 Bitarray indexes update 2012-07-05 13:59:26 +08:00
Oreste Panaia 4ea30529ba Merge branch 'devel' of https://github.com/triAGENS/ArangoDB into devel 2012-06-29 14:28:10 +08:00
Oreste Panaia 2e3b6531c8 Some fixes for bitarray indexes 2012-06-29 14:27:30 +08:00
Jan Steemann c4db88bdea fixed doxygen errors 2012-06-27 13:22:40 +02:00
Oreste Panaia a9f9c3af55 trying to upload bitarrays 2012-06-27 14:21:21 +08:00
Oreste Panaia 20bbdab435 Bitarray index first draft 2012-06-26 20:05:49 +08:00
Oreste Panaia 9949c9e75c renamed sl-operator.c and sl-operator.h to index-operator.c and index-operator.h respectively 2012-06-14 10:04:41 +08:00
Frank Celler d2c758d663 the great rename 2012-06-08 15:01:25 +02:00