[PATCHv6] Update the i18n, UTF-8, and translit_* files to Unicode 8.0.0
Pravin Satpute
psatpute@redhat.com
Thu Jul 23 13:08:00 GMT 2015
>----- Original Message -----
>From: "Mike FABIAN" <mfabian@redhat.com>
>To: "GLIBC Devel" <libc-alpha@sourceware.org>
>Cc: "Marko Myllynen" <myllynen@redhat.com>, "Pravin Satpute" <psatpute@redhat.com>, "Jens Petersen" <petersen@redhat.com>
>Sent: Monday, June 29, 2015 6:39:55 PM
>Subject: [PATCHv6] Update the i18n, UTF-8, and translit_* files to Unicode 8.0.0
>
>
>Therefore, I changed gen_translit_combining a bit to make exceptions for
>these negated mathematical symbols.
>
>New patch set attached.
I reviewed all the patches. Following are comments.
0001-Remove-duplicate-transliterations-for-U-0152-and-U-0.patch
- This patch is self explanatory removes duplicates. I checked for other duplicates there are none now.
0002-Addition-and-fixes-for-translit_neutral.patch
- Most of the combinations are from http://unicode.org/cldr/trac/browser/trunk/common/transforms/Latin-ASCII.xml
- And othre combinations as suggested by Marko on bugzilla #16061
0003-Update-the-translit-files-to-Unicode-7.0.0.patch
Resolves bug 16061
- This files added auto generation scripts for
gen_translit_circle.py, gen_translit_cjk_compat.p, gen_translit_combining.py, gen_translit_compat.py, gen_translit_font.py and gen_translit_fraction.py
I found minor pylint improvemetns with pylint-1.4.3. Fixed in git. Only remaining one is as follows.
*_compat.py, *_cjk_compat.py, *_combining R:112, 0: Too many branches (16/12) (too-many-branches). As we ignored for earlier UnicodegWe can ignore this since complexity of script required this.
Two suggestions:
Non blocker 1. I think we can get rid of functions read_input_file, output_head and output_tail and can just add one function for format_out_file.
Non blocker 2. In long term may be we can have single script for generating translit but it can affect understanding.
0004-Add-transliteration-rules-for-da-nb-nn-and-sv-locale.patch
Fixes long standing issue #89. Looks fine to me.
0005-Unicode-8.0.0-update.patch
Patch is clean. Only 19 characters got changed in i18n those are NEW TAI LUE due to changes in UnicodeData.txt.
Change report available
1. https://raw.githubusercontent.com/pravins/glibc-i18n/master/unicode8-0/UTF-8-report
2. https://raw.githubusercontent.com/pravins/glibc-i18n/master/unicode8-0/UTF-8-report
Regards,
Pravin Satpute
More information about the Libc-alpha
mailing list