-
Notifications
You must be signed in to change notification settings - Fork 52
/
CHANGES.txt
130 lines (105 loc) · 4.14 KB
/
CHANGES.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
* 0.5.1 (yyyy.mm.dd)
-
* 0.5.0 (2023.07.01)
- #116: Add WordStopFilter and WordKeepFilter. (@mocobeta)
- #115: Run tests on Python 3.11 and 3.12 prerelease. (@mocobeta)
- #112: Add GitHub URL for PyPi. (@andriyor)
* 0.4.2 (2022.02.23)
- #90: Fix Analyzer usage. (@syou6162)
- #92: Add version info. (@norihitoishida)
- #95: Migrate to GitHub Actions. (@mocobeta)
- #101: Fix invalid import position of build.py. (@narupo)
- #105: Fix Non-deterministic behavior of tokenizer (@sueki1242)
- #107: Make system dictionary singleton to avoid 'Too much open files' error. (@mocobeta)
- #108: Bump minimum Python version to 3.7. (@mocobeta)
* 0.4.1 (2020.9.21)
- #87: Add progress indicator when compiling user dictionary (thanks @uezo)
- #88: Remove --mmap option from CLI tool
* 0.4.0 (2020.8.23)
- #72: Stop supporting Python 2.7.
- #73: Refactoring: Modernize source code.
- #81: Add profiling scripts (cProfile, tracemaloc).
- #82: Avoid frequent tuple unpacking.
- #83: Refactor lattice node and tokenizer token.
- #84: Improve mmap mode (reduce dictionary size, reduce memory consumption, and improve analysis performance).
- #85: Switch default mmap mode to True on 64bit architecture.
* 0.3.10 (2019.11.03)
- #69: Remove unnecessary sys.path change.
- #70: Remove python 3.3 and 3.4 support.
* 0.3.9 (2019.05.12)
- #63: Move "sysdic" directory to under janome package directory.
- #64: Embed dictionary index data to the source code.
* 0.3.8 (2019.04.03)
- Apply patch to mecab-ipadic for Japanese new era '令和'. #62
* 0.3.7 (2018.12.11)
- Visualize lattice graph by Graphviz. #2
- Janome CLI should support mmap option. #44
- Fix mmap file open mode. #45
- Add --version option to janome CLI. #51
(thanks @takahi-i)
- Support zip import and add the usage example. #54
(thanks @roy-freee)
- Use module-level named loggers. #55
- Add janome CLI for windows. #58
* 0.3.6 (2017.12.07)
- Optimize tokenize() method. #43
(thanks @nakagami)
- Trivial fixes in Tokenfilter. #46
- Fix list index out of range error when tokenizing large text #49
* 0.3.5 (2017.08.06)
- Optimize tokenize() method. #40
(thanks @nakagami)
- Set base_form attribute for unknown tokens. #41
- New TokenFilter to count frequenies of words. #42
* 0.3.4 (2017.07.29)
- Add analyzer framework. #37
* 0.3.3 (2017.07.23)
- Memory-mapped file support. #33
- Fix error on analyzing large text without panctuations. #34
(thanks @saito400)
- Refactor internal dictionary building to handle very large dictionary. #35
* 0.3.2 (2017.07.05)
- Fix install error on Windows. #32
(thanks @takeshi0406)
* 0.3.1 (2017.07.02)
- Add streaming mode. #29
- Add wakati-gaki mode. #30, #31
* 0.3.0 (2017.06.30)
- Limit lattice size to fix memory leak. #27
- Add API reference. #28
* 0.2.8 (2016.05.07)
- Fix nappropriate property names: infl_form, infl_type. #24
(thanks @ikawaha)
- Fix typo. #23
(thanks @kamatari)
* 0.2.7 (2016.03.05)
- Add kuromoji-like simplified user dictionary.
- (Python 2.7 only) Return unicode instead of str for properties such as part_of_speech, reading. #21
- Fix error with unknown characters. #22
* 0.2.6 (2015.10.26)
- Reduce memory usage for compiling system dictionary. #10
(thanks @hajime_nakagami)
- Add executable script 'janome'. (experimental)
- Reduce memory usage for creating Node and State objects by __slots__.
(thanks @atsuoishimoto)
- Minor refactoring.
* 0.2.5 (2015.05.11)
- Re-implement lru cache for Matcher.run()
- Optimize unknown words processing.
- Memory reduction for Lattice.
- Reduce dectionary data size.
* 0.2.4 (2015.05.03)
- Revert cache introduced in 0.2.3 (because there are troublesome bugs...)
* 0.2.3 (2015.05.03)
- Add lru words cache to improve speed of analysing
* 0.2.2 (2015.04.24)
- Fix bug in constructiong user dictionaries
* 0.2.0 (2015.04.24)
- Support Python 3x and 2.7 by the same codes. #9
(thanks @hajime_nakagami)
- Fix bug in transition costs calculation. #12
- reduce dictionary data size
* 0.1.4 (2015.04.11)
- reduce dictionary (fst) data file size
* 0.1.3 (2015.04.08)
- first release