aboutsummaryrefslogtreecommitdiffstats
path: root/native/jni/src/suggest/core/session (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Move dictionary code to top level dictionary dir.Keisuke Kuroyanagi2014-12-174-205/+4
| | | | | Bug: 18725954 Change-Id: Ia442ba4b5d84311057d83edf6e7aeb151d6a820b
* Create .cpp file for NgramContext.Keisuke Kuroyanagi2014-10-212-106/+138
| | | | | | Bug: 14425059 Change-Id: Ie950878817b9c80cc9c970e1a84880c9b9ab228a
* Rename prev_words_info.h to ngram_context.hKeisuke Kuroyanagi2014-10-092-1/+1
| | | | | Bug: 14425059 Change-Id: I0e906631ecad2361a8198b3f9e3394bb22c5bf83
* Rename PrevWordsInfo to NgramContext.Keisuke Kuroyanagi2014-10-093-19/+19
| | | | | Bug: 14425059 Change-Id: I30703fc80e9450d4e2dbfec965e7f9f4468f6a11
* Rename updateCounter to updateEntriesForWordWithNgramContext.Keisuke Kuroyanagi2014-10-091-0/+1
| | | | | Bug: 14425059 Change-Id: Id9b0dd7e32c711ed4292981517c3febd5fe9e897
* [ML13] Fix the locale passing in ProximityInfoJean Chalard2014-10-031-2/+6
| | | | | | | | | | | | | | | | | | | The locale is used to determine additional proximity characters. This is dependent on the dictionary language, but was passed as a function of the layout, which is wrong and would have given bad suggestions in multi-lingual mode. Ideally, additional proximity characters should be inserted in the dictionary header, but for now it's a rather simple change to get it from the dictionary's locale instead of the proximity info locale. Also, that allows us to remove completely the locale parameter from proximity info, which is a much needed change. This change has zero effect on unit tests and on regression tests. Bug: 11230254 Change-Id: If95157155db7dccd1f00b8ba55ccb3600283f9e4
* Implement updateCounter() by using existing entry adding methods.Keisuke Kuroyanagi2014-10-011-2/+12
| | | | | Bug: 14425059 Change-Id: I0b6cb80e1fb8f738e9c7d9e80fbc0c479546b879
* Use passed previous word count in PrevWordsInfo.Keisuke Kuroyanagi2014-09-143-22/+32
| | | | | | Bug: 14425059 Change-Id: I04007bdacf0176a05be7a27ef1c20c5b851d8bed
* Use WordIdArrayView for prevWordIds.Keisuke Kuroyanagi2014-09-112-6/+5
| | | | | Bug: 14425059 Change-Id: Ia84fb997d89564e60111b46ca83bbfa3b187f316
* Use word id for methods related to n-grams.Keisuke Kuroyanagi2014-09-033-23/+17
| | | | | | Bug: 14425059 Change-Id: I81e5d3793527776d3c9faa5594005ddbd4a71354
* Use IntArrayView for dictionary structure policy.Keisuke Kuroyanagi2014-08-271-13/+7
| | | | Change-Id: I0dc94908259d70d5085ff22abf422d90affb1452
* Remove PrevWordsInfo.getBigramsIteratorForPrediction().Keisuke Kuroyanagi2014-07-301-47/+0
| | | | | | | | | | | | | | | This changes the behavior of predictions for some rare cases. For example, let's consider the case of the prev word is "ABC", "ABC" and "abc" is in the dict, "ABC" doesn't have any bigrams, and "abc" has bigrams. Without this change, "abc"'s bigrams are used for prediction. But, with this change, "abc"'s bigrams are not used for prediction. This strategy is the same as the the current way of bigram probability computation during traversal (for suggestion). Bug: 14425059 Change-Id: I857ff39aa70aea265efa651451a63b759bb47d48
* Get bigrams iterator via dict structure policy.Keisuke Kuroyanagi2014-07-081-20/+18
| | | | | Bug: 14425059 Change-Id: Ie4c6afbb6a3b707b5aa8e9e4f941c9da6e6bcdbf
* Extend jni method to use multiple previous words.Keisuke Kuroyanagi2014-06-261-9/+38
| | | | | Bug: 14425059 Change-Id: I8150e8812ee32f8c713315a234b52e5b3a79d7a0
* Enable Beginning-of-Sentence prediction for contextual dict.Keisuke Kuroyanagi2014-06-121-0/+8
| | | | | | Bug: 14161647 Bug: 14119293 Change-Id: I0c00f13966db88e4de85e245e7bced43c9d474b2
* Add length check for attaching a BoS marker.Keisuke Kuroyanagi2014-05-281-1/+4
| | | | | Bug: 15297872 Change-Id: I7b0a27409b81b76561d19c6caa18a5011d3e90f6
* Support Beginning-of-Sentence in native codeKeisuke Kuroyanagi2014-05-231-16/+45
| | | | | Bug: 14119293 Change-Id: I0f382e33a19bf481823b23405d454de61ec835ff
* Use PrevWordsInfo to add/remove n(bi)-gram in native code.Keisuke Kuroyanagi2014-05-212-5/+33
| | | | | | Bug: 14119293 Bug: 14425059 Change-Id: I4b9a46bfd670b35195418eaee51456d44fb91b6d
* Support multiple previous words in DicNode.Keisuke Kuroyanagi2014-05-191-1/+1
| | | | | Bug: 14425059 Change-Id: Ib8682befe4d7d9fe5122eb538e7c804f75ded463
* Move prev word related logic to PrevWordsInfo.Keisuke Kuroyanagi2014-05-162-19/+55
| | | | | | Bug: 14119293 Bug: 14425059 Change-Id: I1bbd7ab4ace2c475f27bc468cb7b4d67e1ae2f9f
* Introduce MAX_PREV_WORD_COUNT_FOR_N_GRAM.Keisuke Kuroyanagi2014-05-153-19/+33
| | | | | Bug: 14425059 Change-Id: I587f90df026a14359d2ee452040bbfaf02c1ae51
* Refactoring: Move prev word information into PrevWordsInfo.Keisuke Kuroyanagi2014-05-153-6/+63
| | | | | | | Bug: 14119293 Bug: 14425059 Change-Id: I5a5f81c6b189e3ad1da093343a3121463f87c21c
* Cleanup DicTraverseSession.Keisuke Kuroyanagi2014-05-151-11/+0
| | | | Change-Id: I953665e1958e7214e9365e2bc9d3f3843016dc81
* s/stdint.h/cstdint/Ken Wakasa2014-04-071-1/+0
| | | | | | | | | Somehow, we were not able to use cstdint gcc target build. That's no longer the case with Clang. Removed unnecessary header inclusions too. Change-Id: Ic83a4adf696f1d5ec7a9809253f3c95804e622e1
* Remove mPartiallyCommited.Keisuke Kuroyanagi2014-03-282-5/+1
| | | | | | Bug: 13333066 Change-Id: I4e3aa31aa501866c23cc4b86093b866306161bfd
* Use nullptr instead of 0 for pointers.Keisuke Kuroyanagi2014-03-071-2/+2
| | | | Change-Id: Ic9425de07b896f940ca11c51968c637ebb897514
* Reset to 9bd6dac4708ad94fd0257c53e977df62b152e20cKen Wakasa2013-12-132-25/+10
| | | | | | The bulk merge from -bayo to klp-dev should not have been merged to master. Change-Id: I527a03a76f5247e4939a672f27c314dc11cbb854
* Output input index of second word's first char.Keisuke Kuroyanagi2013-09-171-1/+3
| | | | | | | | This is for partial commit. Bug: 9059617 Change-Id: I77a0245b63c29b7234fd9b06d362e05aebf30051
* Merge NOT_A_VALID_WORD_POS into NOT_A_DICT_POS.Keisuke Kuroyanagi2013-09-162-3/+3
| | | | Change-Id: I2ac3aa2a19d14e8c4159a59334e7816bd1409688
* Change cache capacity depending on the dictionary size.Keisuke Kuroyanagi2013-09-122-5/+15
| | | | | | Bug: 10699291 Change-Id: I7042d4c1307da2d991d4dd10d637f18026acb996
* Make terminal cache small.Keisuke Kuroyanagi2013-09-122-3/+4
| | | | | | | | | | | | | | | | | | | | Bug: 10699291 Before: (0) 2997.63 (0.98%) (1) 303760.77 (98.92%) (2) 274.94 (0.09%) (66) 0.38 (0.00%) Total 307062.27 (sum of others 307033.73) After: (0) 2953.22 (0.96%) (1) 304633.86 (98.93%) (2) 293.76 (0.10%) (66) 0.38 (0.00%) Total 307915.65 (sum of others 307881.22) Change-Id: I6905abeb590ef50d48ddfcd8ef8b5351af399150
* Move header reading methods to policyimpl.Keisuke Kuroyanagi2013-08-131-3/+2
| | | | | Bug: 6669677 Change-Id: I24cb32f28cf355485b431cc56d877000e3f946d4
* Use DictionaryStructureWithBufferPolicy for traversing.Keisuke Kuroyanagi2013-08-132-11/+10
| | | | | Bug: 6669677 Change-Id: If7bc038278b7217ed5ba4cb77b1238b2dad0001b
* Make DictionaryStructurePolicy have buffer info.Keisuke Kuroyanagi2013-08-092-3/+3
| | | | | | | | To support various format dictionary, extendable buffers or multiple files. Bug: 6669677 Change-Id: I203a5a4adc049a6322cfba4eacb4cb6715f8dfc2
* Replace NOT_VALID_WORD with NOT_A_VALID_WORD_POSKeisuke Kuroynagi2013-07-162-3/+3
| | | | | Bug: 6669677 Change-Id: If5fb58e8a98b04b767eaaf624cb6a40841b64f54
* Make bigram dictionary and traverse session use structure policy.Keisuke Kuroynagi2013-07-081-10/+7
| | | | | Bug: 6669677 Change-Id: I5a13290366afbb788da997b98d1c593963e92f3e
* Introduce BinaryDictionaryHeader to access binary dictionary header.Keisuke Kuroynagi2013-06-112-8/+3
| | | | | | | | BinaryDictionaryHeader abstracts header structure and access header information via only its instance. Bug: 6669677 Change-Id: Ib5ab0e8fef12265ddabc1d0165548b69953bba6a
* Purge DicTraverseWrapperSatoshi Kataoka2013-06-032-38/+19
| | | | | | bug: 8550444 Change-Id: Iad017e66ac579c6727b9f60ad9cda64e478200e5
* Use BinaryDictonaryInfo instead of raw pointers.Keisuke Kuroyanagi2013-06-022-6/+11
| | | | | | Bug: 6669677 Change-Id: I9792a872f1609de7c1ba0fc08d916047d6724c0b
* Move some dictionary related implementations to the dictionary directoryKen Wakasa2013-05-312-3/+3
| | | | Change-Id: I6fe71a540ad91ca46755be234abc8894242e2413
* Create a new directory for layout-related implementationsKen Wakasa2013-05-301-1/+1
| | | | Change-Id: Ia227bb296f20dab0dfc03dfcde3a06c624eca76c
* Make DicTraverseSession having suggest options.Keisuke Kuroyanagi2013-05-292-5/+11
| | | | Change-Id: I2b802c06634e5226c763fd838f3fc0918fc4a161
* Be careful about the dictionary size in detection methodsJean Chalard2013-05-081-1/+2
| | | | | Bug: 8857618 Change-Id: I29345ec96d53da601571ba73197a6485643a10a7
* Improve bigram frequency lookupTom Ouyang2013-04-262-5/+5
| | | | | | Bug: 8592527 Change-Id: I1908bcb552279b9acb140fe4d8d26b10ed9eda72
* Check bigrams for lowercased previous word if original is not foundTom Ouyang2013-04-221-1/+9
| | | | | Bug: 8592527 Change-Id: I70c4b81263d8b2cb7050364c124589f66900cfce
* Use additional multi-word cost per language (for Russian)Satoshi Kataoka2013-04-162-1/+14
| | | | | Bug: 7540133 Change-Id: I7eb7b8399746c15452ed2ed5069955e88fb546d3
* Enable touch coordinate correction for new algorithmSatoshi Kataoka2013-04-151-0/+4
| | | | | | Bug: 8505668 Change-Id: I07eb785c74c446777524104a3d2b61f0f591a498
* fix assertionSatoshi Kataoka2013-04-151-1/+1
| | | | Change-Id: I34ff470122aa0f34c66425d2eeae7590187d3074
* Add method to get dictionary flags from DicTraverseSessionTom Ouyang2013-04-082-1/+5
| | | | | | | | | This is needed for the traversal algorithm to know which digraphs (if any) are associated with the current dictionary. Bug: 8493920 Change-Id: I6ae088b55e50c0c2c7066d045b304c5068e23390
* Cleanup the include pathsKen Wakasa2013-04-082-3/+4
| | | | Change-Id: Ic93d39e230170cfc9f5be6d81d42006e1a61e347
generated by cgit v1.2.3-83-g751a (git 2.47.1) at 2025-08-02 09:26:12 +0000