Menu

#431 LilyPad for Windows' menus should be unicode

Verified
nobody
None
Defect
2015-03-02
2007-08-30
Anonymous
No

Originally created by: *anonymous

Originally created by: gpermus@gmail.com

(screenshot attached)

For example, `File' is showed as "-t-@-C--(F)" (left most item on menu).
It should be shown  as U+30d5 U+30a1 U+30a4 U+30eb
(Fu A-Small I Ru, in Katakana) and "(F)".

$ perl -e 'print "\x30\xd5\x30\xa1\x30\xa4\x30\xeb"' | iconv -f ucs-2 -t
sjis |
hexdump -C
00000000  83 74 83 40 83 43 83 8b                           |.t.@.C..|

On Japanese Windows, strings should be written in CP932 (or Unicode),
but lilypond does not.

1 Attachments

Discussion

  • Google Importer

    Google Importer - 2010-07-10

    Originally posted by: pnorcks@gmail.com

    This issue should be fixed with the next LilyPad release, though I don't know how to test on Windows with a Japanese locale...

    Owner: pnorcks
    Status: Started

     
  • Google Importer

    Google Importer - 2010-08-13

    Originally posted by: pnorcks@gmail.com

    (No comment was entered for this change.)

    Labels: -Priority-Low Priority-Postponed
    Owner: ---
    Status: Accepted

     
  • Google Importer

    Google Importer - 2011-10-30

    Originally posted by: pkx1...@gmail.com

    Also reported by a user using Chinese localization of Windows 7

    http://lists.gnu.org/archive/html/bug-lilypond/2011-10/msg00620.html

    Summary: LilyPad for Windows' menus should be unicode
    Labels: -Priority-Postponed

     
  • Google Importer

    Google Importer - 2014-10-16

    Originally posted by: truer...@gmail.com

    lilypad in lilypond-2.19.15-1.mingw.exe is still broken Japanese strings.
    Probably it is broken also in the languages which are non-ASCII (non ISO-8859-1).
    This is caused by the bug of binutils 2.19.
    Therefore, it is solved if you use binutils 2.24.
    And, iconv is also required.

    If lilypad is built in the Windows environment, windres (included in binutils 2.19) generate the correct binary.
    In this case, windres uses Windows API for codepage conversion.

    However, if in the Linux environment (non-Windows environment), it generate the incorrect binary for non-ASCII (non ISO-8859-1) resources.
    In this case, windres can't use Windows API and should use iconv.

    binutils 2.19 has the bug that is about how to use iconv.
    This bug is fixed by the following commits.

    http://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commitdiff;h=6f4c2146e75b03cec4668959b212259f3342b586
    http://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commitdiff;h=7e93ea4b52b60c00219bc5e1d84576774f67e98b

     
  • Google Importer

    Google Importer - 2014-10-16

    Originally posted by: truer...@gmail.com

    This issue is not solved even if binutils 2.24 is used without iconv.
    I think that iconv can be forced with like the following patch.

    --- a/binutils/winduni.c    2014-10-13 01:25:53 +0900
    +++ b/binutils/winduni.c    2014-10-13 01:33:13 +0900
    @@ -44,6 +44,10 @@

    #if HAVE_ICONV
    #include <iconv.h>
    +#else
    +#if !defined (_WIN32) && !defined (__CYGWIN__)
    +#error iconv is not found.
    +#endif
    #endif

    static rc_uint_type wind_WideCharToMultiByte (rc_uint_type, const unichar *, char *, rc_uint_type);

     
  • Google Importer

    Google Importer - 2015-03-02

    Originally posted by: truer...@gmail.com

    This issue is fixed by lilypond-2.19.16-1.mingw.exe.
    Because GUB's binutils has been updated to 2.25.

     
  • Google Importer

    Google Importer - 2015-03-02

    Originally posted by: PhilEHol...@googlemail.com

    (No comment was entered for this change.)

    Status: Fixed_2_19_16

     
  • Google Importer

    Google Importer - 2015-03-02

    Originally posted by: PhilEHol...@googlemail.com

    Taking a liberty by marking as verified a bug I marked as fixed, but I would argue that Masamichi claimed it fixed...

    Status: Verified

     
MongoDB Logo MongoDB