summaryrefslogtreecommitdiffstats
path: root/src/libstat
Commit message (Collapse)AuthorAgeFilesLines
...
* [Fix] Core: Fix address rotation bugVsevolod Stakhov2019-02-182-4/+4
| | | | | | | | | | | | | | | | | | | | | | Previously, upstream.get_addr function returned the new address of the upstream. Unfortunately, it was used for printing addresses. It caused the following situation: let's imagine we have A1 and A2 where A1 was initially selected. So the connection was performed to A1: Current addr Selected addr Connect+---------+ A2+------>A1 A1 | +-+Print failure<---+ A1+------>A2 A2 | +----+ +->Mark failure+-------->+ A2 | +----+ But the failure OP as well as log message told about `A2` where the real problem happened with `A1`. This commit adds distinguishing between getting the next and the current address of the upstream resolving this issue.
* [Minor] Stat: Fail-safety fix for object expansionVsevolod Stakhov2019-02-111-9/+3
|
* [Minor] Stat: Set more sane default prefix when users scripts are setVsevolod Stakhov2019-02-111-1/+1
|
* [Minor] Core: Fix flag definitionVsevolod Stakhov2019-01-221-3/+3
|
* [Minor] Slightly extend what we can treat as wordsVsevolod Stakhov2018-11-301-1/+1
|
* [Fix] Some fixes for raw partsVsevolod Stakhov2018-11-271-0/+1
|
* [Fix] Treat learning errors as non-fatalVsevolod Stakhov2018-11-271-4/+4
| | | | Issue: #2637
* [Minor] Fix for DSNVsevolod Stakhov2018-11-271-1/+1
|
* [Feature] Ignore bogus whitespaces in the wordsVsevolod Stakhov2018-11-262-1/+9
| | | | Issue: #2649
* [Project] Use URLs TLDs instead of !!EX!! in stat tokensVsevolod Stakhov2018-11-261-16/+39
|
* [Project] Use more generalised API to produce meta wordsVsevolod Stakhov2018-11-264-65/+90
|
* [Minor] Check language detector pointer before useVsevolod Stakhov2018-11-261-2/+2
|
* [Project] Finish basic tasks in new unicode projectVsevolod Stakhov2018-11-254-26/+43
|
* [Project] Rework parts conversion and serializationVsevolod Stakhov2018-11-251-8/+5
|
* [Project] Another try to normalize unicode properlyVsevolod Stakhov2018-11-253-113/+142
|
* [Project] Various unicode fixes in language detectorVsevolod Stakhov2018-11-251-3/+2
|
* [Project] Rework stemmingVsevolod Stakhov2018-11-244-9/+106
|
* [Project] Add function to normalize unicode on per words basisVsevolod Stakhov2018-11-243-2/+139
|
* [Project] Start words unicode structure reworkVsevolod Stakhov2018-11-242-56/+65
|
* [Minor] Do not use item in classifiers as it causes issuesVsevolod Stakhov2018-11-242-20/+0
| | | | Issue: #2636
* [Project] Migrate fuzzy redis backend to lua_redisVsevolod Stakhov2018-11-231-1/+3
|
* [Project] Use lua_redis for redis_cache as wellVsevolod Stakhov2018-11-231-109/+97
|
* [Project] Use lua_redis to configure servers in bayes Redis backendVsevolod Stakhov2018-11-231-107/+97
|
* [Minor] Fix some issues and add debuggingVsevolod Stakhov2018-11-161-5/+12
|
* [Minor] Allow to have bulk learn in the default configVsevolod Stakhov2018-11-161-0/+2
|
* [Project] Move all metatokens to lua_stat from CVsevolod Stakhov2018-11-162-126/+48
|
* [Project] Move stat tokens generation to LuaVsevolod Stakhov2018-11-162-0/+70
|
* [Rework] Improve bayes debug logging, remove unused stuffVsevolod Stakhov2018-11-166-153/+63
|
* [Minor] Add sanity checks for bayes probabilityVsevolod Stakhov2018-11-161-0/+13
|
* [Minor] Fix formulaVsevolod Stakhov2018-11-161-8/+8
|
* [Fix] Switch from chi-square to naive for large Fisher valueVsevolod Stakhov2018-11-151-2/+18
|
* [Minor] Check for empty bayes resultsVsevolod Stakhov2018-11-151-11/+15
|
* [Fix] Disable headernames in bayes temporarilyVsevolod Stakhov2018-11-151-0/+2
|
* [Feature] Skip stop words in statisticsVsevolod Stakhov2018-11-154-23/+35
|
* [Feature] Filter tokens in bayesVsevolod Stakhov2018-11-151-6/+20
|
* [Fix] Properly set learned countVsevolod Stakhov2018-11-151-0/+6
|
* [Fix] Rework bayes calculations...Vsevolod Stakhov2018-11-142-17/+14
|
* [Minor] Don't learn classifiers where expansion failsVsevolod Stakhov2018-11-101-3/+20
|
* [Minor] Move subject tokenisation to a separate routineVsevolod Stakhov2018-11-083-29/+70
| | | | Issue: #2623
* [Project] Another rework for symbols cacheVsevolod Stakhov2018-10-293-5/+5
|
* [Rework] Add more traces to async events, get rid of GQuarksVsevolod Stakhov2018-10-262-13/+14
|
* [Project] Adopt C modules and stuffVsevolod Stakhov2018-10-202-4/+30
|
* [Minor] Fix event pending checksVsevolod Stakhov2018-10-062-7/+7
|
* [CritFix] Fix words decay one more time (affects long messages)Vsevolod Stakhov2018-09-251-4/+8
|
* [Minor] More places where utext_close is requiredVsevolod Stakhov2018-09-141-0/+2
|
* [Fix] Fix more issues with watching of async eventsVsevolod Stakhov2018-09-122-8/+4
|
* [Fix] Fix words decay algorithmVsevolod Stakhov2018-09-111-1/+1
|
* [Fix] Fix crashes on task cleanupVsevolod Stakhov2018-09-102-4/+4
|
* [Minor] Properly set flag on text tokensVsevolod Stakhov2018-09-071-3/+4
|
* [Minor] Further fixes in tokenization algorithmVsevolod Stakhov2018-09-071-20/+28
|
generated by cgit v1.2.3 (git 2.39.1) at 2025-03-23 20:51:11 +0000