Hawkin
/
arangodb
mirror of https://gitee.com/bigwinds/arangodb
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
21,662 Commits 358 Branches 495 Tags 994 MiB
Commit Graph

181 Commits

Author SHA1 Message Date
Jan Steemann ffab0ef47d renamed file 2015-06-18 17:29:49 +02:00
Michael Hackstein 99afc6ec87 CXX ExampleMatcher learned about internal attributes. This makes all AQL-Graph tests pass again ;) 2015-05-30 00:00:49 -07:00
Jan Steemann 8126781ca7 fixed crash 2015-05-29 11:52:07 +02:00
Jan Steemann 4cd37f8d55 fix MacOS compile errors 2015-05-29 10:01:19 +02:00
Jan Steemann 88198767dd logging 2015-05-29 09:49:22 +02:00
Jan Steemann 4dde6e49cd goodbye barriers, welcome ditches 2015-05-28 10:23:30 +02:00
Jan Steemann 8d29b600d3 fixed Windows compile errors 2015-05-26 18:40:27 +02:00
Jan Steemann 4b67c7a735 several fixes for indexes 2015-05-26 13:35:11 +02:00
Jan Steemann 0d6bd8876c fixed wrong memory zone 2015-05-26 09:41:10 +02:00
Jan Steemann a9f2769ffa now compiles and links 2015-05-23 04:31:13 +02:00
Jan Steemann 9a1b3d0b52 in the middle of changes - need to go home now 2015-05-22 22:15:26 +02:00
Jan Steemann a5b6f00b7f moved indexes to cxx 2015-05-22 20:15:14 +02:00
Jan Steemann bc76cf5079 use std::vector 2015-05-22 12:58:40 +02:00
Jan Steemann 3033846791 catch exception when filling indexes 2015-05-22 09:29:56 +02:00
Michael Hackstein 86c7f87dd3 Merge branch 'cppTrav2' into devel 2015-05-19 16:17:27 -07:00
Michael Hackstein 5360172a7b The ExampleMatcher is now able to use a v8 json object as example for matching in the constructor 2015-05-15 09:56:16 -07:00
Michael Hackstein 7ba7482d58 Implemented an ExampleMatcher class to be used for path filtering. Is now used in byExample queries as well. API not fixed yet. 2015-05-14 13:29:42 -07:00
Jan Steemann f8a4df2ad4 Merge branch 'edge-index-resize' of https://github.com/arangodb/arangodb into devel 
Conflicts:
	lib/Basics/associative-multi.cpp
 2015-05-13 12:52:38 +02:00
Willi Goesgens 421d9fbb9e Add performance logging for index building. 2015-05-13 10:49:38 +02:00
Jan Steemann a007a867ee "privatized" TRI_vector_t internals. 
This allows changing the internals of TRI_vector_t structs in order to make the struct smaller.
On 64 bits, the size of TRI_vector_t is reduced from 48 bytes to 28 bytes.
TRI_json_t does benefit from this, as its biggest component is a TRI_vector_t.
 2015-05-05 10:31:02 +02:00
Max Neunhoeffer 65c4c5629d Highly experimental aggregation experiment. Do not use. 2015-04-30 00:44:39 -07:00
Max Neunhoeffer 1bea0bede9 Finish new ReadWriteLockCPP11 for collections. 2015-04-29 21:03:05 -07:00
Jan Steemann 629883041c added asserts 2015-04-21 12:08:09 +02:00
Jan Steemann 61499432b7 do not remove constant sorts when creating plans, but during optimization phase 
less utf8 comparisons
 2015-04-20 11:32:29 +02:00
Jan Steemann d8597d9662 fixed edge index lookups in cluster 2015-04-13 14:36:16 +02:00
Frank Celler 4c5d57f1b8 merged Exceptions.h and Exception.h 2015-03-23 14:05:19 +01:00
Jan Steemann 7dc3fc3d5c safely create collections and databases 2015-03-19 15:56:12 +01:00
Jan Steemann 6360c798e7 added option `--database.ignore-logfile-errors` 
Conflicts:
	arangod/RestServer/ArangoServer.cpp
 2015-03-18 17:46:11 +01:00
Jan Steemann b5fb457d7b ensure there is a datafile info entry for each datafile of a collection 2015-03-18 17:44:38 +01:00
Jan Steemann 659ef9e6af made _uncollectedLogfileEntries an atomic variable 2015-03-18 17:39:36 +01:00
Jan Steemann 7f121fc352 use barrier 2015-03-04 11:35:32 +01:00
Jan Steemann bc4ad22ab6 parallel secondary index building when loading collections 2015-03-03 10:52:51 +01:00
Jan Steemann 1461d75ed4 improve collection loading time 
This patch reduces collection loading time by preallocating enough space in primary index ahead of time.
When a collection is closed, the number of documents in the collection will be stored in the collection's JSON info file.
This value is used to determine the initial size for the primary index when the collection is loaded next time.

Datafile iteration has also been made slightly faster.

The above changes will have a significant benefit when the collection's datafiles are already in the OS buffer cache, and when there are no secondary indexes.
Loading datafiles from disk or building secondary indexes may be more time-consuming than the improvements reapable by this patch, but the patch shouldn't hurt anyway.
 2015-02-26 16:35:51 +01:00
Jan Steemann f29591135a adjusted geo index API 2015-02-18 14:59:46 +01:00
Jan Steemann a26bd2b223 auto-migrate old hash indexes 2015-02-10 11:34:51 +01:00
Jan Steemann 82c7522d4e test fixes 2015-02-09 20:08:06 +01:00
Jan Steemann 003d7e9112 sparse indexes, initial commit, untested 2015-02-07 03:07:41 +01:00
Jan Steemann 70fba85a67 enforce an error code 2015-01-22 15:20:57 +01:00
Jan Steemann eac7ce5d13 removed unused define 2015-01-16 21:58:13 +01:00
Jan Steemann 2d7a7b87aa move variable declarations to usage 2015-01-13 20:46:33 +01:00
Jan Steemann f3b142849f fixed leaks during reverts 2015-01-03 01:28:23 +01:00
Max Neunhoeffer 05c069f361 Another fix for the distributed locking problem in cluster AQL. 2014-12-23 14:35:19 +01:00
Max Neunhoeffer 1782039138 Fix cluster-wide locking for AQL. 2014-12-22 16:21:45 +01:00
Max Neunhoeffer e50a705d9a Squashed commit of the following: organise locking in distributed AQL 
Locking is now done in an extra round after the query is fully
instanciated in the cluster. All participating shards are locked
in alphabetical order of their shard ID (local collection name).
For this to work there is a new action in the RestAqlHandler plus a
mechanism to prevent the usual locking from happening: Each thread has a
thread local static class variable of
  triagens::arango::Transaction::_makeNolockHeaders
which is of type std::unordered_set<std::string>*.
Whenever this is not equal to nullptr and a local collection name is
stored in there, no locking or unlocking takes place. This information
is forwarded by the X-Arango-Nolock HTTP header, whenever an HTTP
request is sent via ClusterComm to a shard.
 2014-12-22 14:40:22 +01:00
Jan Steemann 20e7fe7a7a the great rename: array => object, list => array 2014-12-18 21:07:06 +01:00
Jan Steemann db1ddb880b wait for sync thread outside of lock 2014-12-17 14:21:24 +01:00
Jan Steemann d1bf47a14a fixed issue when rolling back remove operations that could not remove from indexes 2014-11-01 18:03:57 +01:00
Jan Steemann 999d8ffd1d small optimizations 2014-10-08 16:10:27 +02:00
Jan Steemann d5fb057603 less locking 2014-09-24 13:32:39 +02:00
Jan Steemann 1fff0ecac2 optimization for recovery of collections that are going to be dropped anyway 2014-09-18 18:20:56 +02:00