- Merged pr-103 AddWordEx
- Merged pr-96 Fix - memory allocted by cgo will not automaticlly free by gc.
- Merged pr-99 feature: add RemoveWord api #99
- upgrade cppjieba to v5.0.3
- merge pr-60
- remove bleve to pass travis-ci testing.
Notice:
- remove NewExtractor, and use its api in Jieba.
- upgrade cppjieba to v5.0.0
- fix issue 8
- add ExtractWithWeight api and fix issue 6
- fix issue 5
- upgrade cppjieba -> v4.8.1 and support (word,freq,tag) format of user dict
- add AddWord api
- upgrade cppjieba -> v4.8.0, and make CutForSearch api behaves the same as jieba by Python
- remove SetCutForSearchThreshold api
- upgrade cppjieba -> v4.7.0
- add new api: Tokenize for location information of words
- add new api: SetCutForSearchThreshold
- use Tokenize instead of Cut and SetCutForSearchThreshold(3) in gojieba/bleve
- NewJieba(...string) support variable arguments
- NewExtractor(...string) support variable arguments
- removed example/ and write Example in
*_test.go - add some kind of Benchmark, Testing
- fix bug in extractor.go: calling C.free manully to free momery allocated by C.CString
- supported bleve custom analyzer by using gojieba.
- Fix bug: calling C.free manully to free momery allocated by C.CString.
- Expose new api: Tag
Upgrade to fix potential trouble:
- limonp -> v0.6.0
- cppjieba -> v4.5.3
- Fix error in
go vet - Upgrade limonp to v0.5.4 and cppjieba to v4.5.0 to support more unicode character
Upgrade:
- cppjieba -> v4.4.1 to fix bug, see details in CppJieba ChangeLog v4.4.1
- 源码布局变动，增加 deps/ 管理外部依赖代码。
- 增加 Extractor 关键词抽取功能。
- Upgrade limonp to version v0.5.1
- Upgrade cppjieba to version v4.3.1
- 分词接口变动 New -> NewJieba
- 增加关键词抽取类 NewExtractor
- 修复多余日志输出的问题。
- 升级 cppjieba to v4.2.1
- 升级 cppjieba to v4.1.2
- 增加英文介绍
README_EN.md
- 完成基本分词功能：【全模式，精确模式，搜索引擎模式】