summaryrefslogtreecommitdiffstats
path: root/container-search
Commit message (Collapse)AuthorAgeFilesLines
* Handle that normalizing leads to an empty queryJon Bratseth2017-08-291-7/+8
|
* Make assignment of sessionid independent of wether getSessionId(true) has ↵Henning Baldersheim2017-08-282-4/+8
| | | | | | been called prior to clone(). Now it will behave as if getSessionId() has NOT been called, which is what is considered current behavior by 99% of usage patterns.
* Restore old beahvior on clone.Henning Baldersheim2017-08-282-10/+19
| | | | | - If getSessionId(true) has been called prior to clone, allcones will share sessionid with parent. - If not, they will all get their own.
* remove unused codeHenning Baldersheim2017-08-281-1/+0
|
* Remove unused code.Henning Baldersheim2017-08-281-8/+0
|
* Add info about enabled caches and their keys.Henning Baldersheim2017-08-283-0/+58
|
* Generate an explicit timeout messageJon Bratseth2017-08-212-1/+3
|
* Add copyright headerJon Bratseth2017-08-181-0/+1
|
* wantsRPCSummarFill -> wantsRPCSummaryFillHenning Baldersheim2017-08-172-3/+3
|
* Merge pull request #3136 from vespa-engine/balder/use-rank-profile-in-sessionidHenning Baldersheim2017-08-179-48/+71
|\ | | | | Add rankprofile to the base sessionId.
| * Remove unused imports in my open files.Henning Baldersheim2017-08-175-6/+0
| |
| * Update comments and imports.Henning Baldersheim2017-08-173-7/+8
| |
| * Use your own rankprofile, and avoid breaking public API.Henning Baldersheim2017-08-178-37/+60
| |
| * Add rankprofile to the base sessionId.Henning Baldersheim2017-08-168-21/+26
| |
* | Return an error when ranking.queryCache is not enabled when it needs to be.Henning Baldersheim2017-08-163-29/+74
|/
* Also propagate rank profile.Henning Baldersheim2017-08-161-2/+5
|
* Send the sessionId down if present.Henning Baldersheim2017-08-151-3/+8
|
* Remove reference to yca proxyJon Bratseth2017-08-081-1/+1
|
* Remove unnecessary instances of the character sequence 'yca'Jon Bratseth2017-08-084-47/+47
|
* Remove unnecessary catchJon Bratseth2017-08-021-1/+1
|
* Avoid timing sensitivityJon Bratseth2017-08-022-6/+7
|
* Cleanup pom.xmlBjørn Christian Seime2017-07-241-14/+0
|
* Remove building of old osgi bundlesBjørn Christian Seime2017-07-2427-646/+0
| | | | These bundles are not used in any unit or system test.
* Only create timeout error when it is actually neededBjørn Christian Seime2017-07-121-6/+7
|
* Even if we claim efficiency by lazy setting, we actually must do it eventually.Henning Baldersheim2017-06-221-1/+2
| | | | | If not there is no point in having a placeholder for it. Not doing so makes sorting/compaing/hashing extremely expensive.
* Avoid the implicit sorting that deepIterator does as a sideeffect.Henning Baldersheim2017-06-221-2/+2
| | | | It is not worthwhile. Especially when you do not have a document id.
* use underVespaHomeArne H Juul2017-06-192-2/+3
|
* VESPA_HOME should not end with / anymoreArne H Juul2017-06-171-2/+0
| | | | | * best practice for HOME-like env vars is to be just the name of the directory without trailing slash.
* Update copyright headersJon Bratseth2017-06-141282-1271/+1288
|
* Revert "Update copyright headers"Jon Bratseth2017-06-141281-1281/+1264
|
* Update copyright headersJon Bratseth2017-06-141281-1264/+1281
|
* Merge pull request #2771 from yahoo/bratseth/remove-unnecessary-filesTor Egge2017-06-141-5/+0
|\ | | | | Remove unnecessary files
| * Remove unnecessary filesJon Bratseth2017-06-141-5/+0
| |
* | Remove carriage returnJon Bratseth2017-06-143-3/+3
|/
* Revert "Revert "Encode as UTF-8""Jon Bratseth2017-06-141-1/+1
|
* Revert "Encode as UTF-8"Tor Egge2017-06-141-1/+1
|
* Encode as UTF-8Jon Bratseth2017-06-141-1/+1
|
* Revert "Copyright header"Jon Bratseth2017-06-131283-1291/+1279
|
* Copyright headerJon Bratseth2017-06-131283-1279/+1291
|
* Nonfunctional changesJon Bratseth2017-06-091-1/+1
|
* Nonfunctional changesJon Bratseth2017-06-087-37/+111
|
* use standard formatting for Override annotationArne H Juul2017-05-3052-121/+242
|
* Add more testing of stddev aggregatorBjørn Christian Seime2017-05-292-0/+3
|
* Use either logger if-guard or lambda, not both.Tor Brede Vekterli2017-05-241-4/+1
|
* Also dump visitor session trace on failure, not just on successTor Brede Vekterli2017-05-241-3/+4
|
* Merge pull request #2506 from yahoo/arnej/remove-extra-gitignoreJon Bratseth2017-05-1920-87/+21
|\ | | | | Arnej/remove extra gitignore
| * remove old unused ignoresArne Juul2017-05-1918-41/+0
| |
| * simplify .gitignoreArne Juul2017-05-191-44/+21
| |
| * remove extra .gitignoreArne Juul2017-05-191-2/+0
| |
* | Change NFKC normalization default to falseJon Bratseth2017-05-192-3/+6
| | | | | | | | | | | | | | | | | | Currently NFKC normalization - always happens for tokenized fields regardless of this setting (query and index) - never happens for non-tokenized fields on the indexing side Because of this, OFF is a better default as it does not change anything for tokenized fields but makes things consistent by default for non-tokenized fields.