1. Add GA to test BreakIterator under LSTM configuration (remove Thai
and Burmese dictionary and include Thai and Burmese LSTM)
2. Add LSTMDataName for the purpose of testing.
3. Add file base test code to test BreakIterator match results from test
file generated by pythong code in
https://github.com/unicode-org/lstm_word_segmentation/blob/master/segment_text.py
4. Fix a LSTMBreakEngine::divideUpDictionaryRange bug when the return value
should only contains the number of words found when the passed in foundBreaks
already contains some data.
5. Change the cintltest TestSwapData from testing thaidict to laodict so
it will not break while we filter out thaidict under the LSTM
configuration.
Also remove the use of the unsafe double-checked lock idiom in the same
function, SimpleDateFormat::tzFormat(). Synchronization now always uses a
mutex, which is slower, but in the context of format or parse operations,
shouldn't be significant.
Added synchronization to one more unsafe direct reference to a const
SimpleDateFormat::fTimeZoneFormat. In the assignment operator.
clang 11 throws an error when using undefined variables starting with
TARGET_OS_, and TARGET_OS_SIMULATOR is missing in earlier versions
of macOS, so the code would not compile without an additional guard.
See also ICU-13440.
clang 11 throws an error when using undefined variables starting with
TARGET_OS_, and TARGET_OS_SIMULATOR is missing in earlier versions
of macOS, so the code would not compile without an additional guard.
See also ICU-13440.
- improve checks for CLDR resolutions
- update batch fetch from 50 to 100
- do not show no-commit tickets under illegal-state
- add navigation
- use numeric IDs for status, resolution, types with constants
(S_, R_, I_)
fix bugs from CLDR-14224
- do not show not-in-query issues as open
- cache single-issue queries
- give information on each problem as to what it is
- put 'review' in a separate section (not counted in Problems)
- more detailed fix info under each issue
- fix section headings, always show component, add nocopyright option
- sort open issues by componnent
- sort the 'review' list
ICU-20444:
- exclude (via git-cherry) any commits already cherry picked to the branch.
- used a workaround to https://github.com/gitpython-developers/GitPython/issues/846
that seemed to work well,
- warn about issues with excluded commits
CLDR-14524:
- Exclude "Obsolete" from requiring commits
evaluates to the empty string and the result files are missing the
perf test name as a result. E.g. '_Mar_16_155820.html' instead of
'unisetperf_Mar_16_155820.html'.
Not obvious why the regex was even created since it doesn't seem
necessary because the name of the caller doesn't require additional
parsing. The rpaticular code is from 2002, maybe changes in Perl
account for the issue(?)
