blob: 284f068730ee2d8e2963b457e97c0f10db6dfe08 [file] [log] [blame]
Akron8dabdc12026-06-02 16:36:07 +020010.7.0 2026-06-08
Akronf3efc9e2026-06-03 10:46:37 +02002 - conllu2korapxml:
3 - Add automatic offset computation from 'text' comments
4 (implemented with AI assistance)
Akron8dabdc12026-06-02 16:36:07 +02005 - new option --text-sigle <template> to derive text sigle from newdoc id
6 (implemented with AI assistance)
Akron982dd1e2026-06-03 11:52:48 +02007 - Add --base-text option to generate data.xml with
8 reconstructed plain text from 'text' comment lines
9 (implemented with AI assistance)
Akronf3efc9e2026-06-03 10:46:37 +020010
Akron249fc832024-06-04 16:36:44 +0200110.6.3 2024-06-04
12 - Trim filenames to fix double space after filename metadata
13 - Set permissions of zip contents to 666
14 - Allow different foundries for morpho and dependency annotations
15 - Readme: Drop early stage warning
16 - Drop cherry-pick unfriendly test count prediction
17 - conllu2korapxml:
18 - escape &, <, >
19 - convert upos column to upos features
20
Marc Kupietz6dc97b72024-01-24 13:01:33 +0100210.6.2 2024-01-24
22 - Bump minimal perl version to 5.36 to improve unicode handling.
23 - korapxml2conllu
24 - Use implicit default utf8 encoding instead of explicit de/encodes. Speeds up processing by 10%.
25
Akrone24c9332023-03-22 10:59:33 +0100260.6.1 2023-03-22
27 - conllu2korapxml:
28 - Fix append for filehandle output.
29
Marc Kupietz66bb4952023-01-13 15:04:38 +0100300.6.0 2023-01-13
Marc Kupietz7e4cd6c2022-12-15 18:34:37 +010031 - korapxml2conllu:
Marc Kupietz534df182022-12-16 15:00:30 +010032 - the sigle-pattern option now affects the entire sigle
Marc Kupietz7e4cd6c2022-12-15 18:34:37 +010033 - handle docid attributes correctly if they are in a different line than their parent element <layer>
Akronf2b0bba2022-12-16 18:00:08 +010034 - Improve identification of offset errors
Marc Kupietz7e4cd6c2022-12-15 18:34:37 +010035
Marc Kupietz12f64e42022-09-29 08:58:16 +0200360.5.0 2022-09-29
37 - korapxml2conllu:
38 - --word2vec|lm-training-data option added to print word2vec input format
39 - --extract-metadata-regex added to extract some metadata values as context input for language model training
40 - by default sentence boundary information is now read from structure.xml files (use --s-bounds-from-morpho otherwise)
41 - use morpho.xml if present when run on base zips
42 - new option -c <columns>
43 - conllu2korapxml:
44 - ignore _-lemmas
45 - handle UDPipe comments
46 - ignore non-interpretable comments
47 - improve error handling for missing text ids and offsets
Marc Kupietzf1fdc192021-10-08 13:29:59 +020048
Marc Kupietza7d90c62021-07-31 23:48:13 +0200490.4.1 2021-07-31
50 - korapxml2conllu: fix patterns not extracted for last texts in archive
51
Marc Kupietz6beca9d2021-07-29 18:26:09 +0200520.4 2021-07-29
Marc Kupietzeb7d06a2021-03-19 16:29:16 +010053 - korapxml2conllu option -e <regex> added to extract element/attributes to comments
Marc Kupietz0ab8a2c2021-03-19 16:21:00 +010054
Marc Kupietz22858f82021-02-15 14:22:05 +0100550.3 2021-02-15
Marc Kupietz79ba1e52021-02-12 17:26:54 +010056 - Provide conllu2korapxml to convert from ConLL-U to KorAP-XML zip
57
Marc Kupietzb96c3862021-02-12 08:33:44 +0100580.2 2021-02-12
Marc Kupietzd8455832021-02-11 17:30:29 +010059 - Convert also KorAP-XML base zips
60
Marc Kupietz396b4d62021-02-12 08:29:35 +0100610.1 2020-09-23
62 - Initial release to GitHub.