Bug#317115: aptitude: some small bugs, 'search' and 'show' on cmdline, 'license' on curses.

2005-07-06 Thread Kenshi Muto
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Package: aptitude
Version: 0.3.3-1
Severity: normal
Tags: experimental

Hi,

UTF-8 support of new version 0.3.3 in experimental is very nice.
Eto-san and I checked it worked good shape with Japanese locale,
ja_JP.EUC-JP and ja_JP.UTF-8.

During test, we found some small problems.

1. `aptitude search' shows only first column (such as i,p,v), but
   misses package names and descriptions (any locales).
2. When we do `aptitude show' with LANG=ja_JP.EUC-JP or ja_JP.UTF-8,
   all of translations such as Package, Status are corrupted.
   I dumped them and found each first bytes of double-byte characters
   were removed.

'aptitude show cupsys' with LANG=ja_JP.EUC-JP in unstable:
0x: a5 d1 a5 c3 a5 b1 a1 bc - a5 b8 3a 20 63 75 70 73 ~: cups

'aptitude show cupsys' with LANG=ja_JP.EUC-JP in experimental:
0x: d1 c3 b1 fc b8 3a 20 63 - 75 70 73 79 73 0a b9 c6 ~~.: cupsys.~

   'a5 d1 a5 c3 a5 b1 a1 bc a5 b8' is correct Japanese translation of
   'Package' in EUC-JP encoding.

3. When we open 'license' on ncurses window, it shows too narrow
   window (any locales).

Thanks,
- -- 
Kenshi Muto
[EMAIL PROTECTED]
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 http://mailcrypt.sourceforge.net/

iEYEARECAAYFAkLLyyIACgkQQKW+7XLQPLGJxACglJkt+V6IwkAWSN+pxmBnfU8U
1v8An18nDL/LYjr1wRY2kCgE5vPXLYVV
=9l7o
-END PGP SIGNATURE-


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]



Bug#317115: aptitude: some small bugs, 'search' and 'show' on cmdline, 'license' on curses.

2005-07-06 Thread Daniel Burrows
On Wednesday 06 July 2005 08:14 am, Kenshi Muto wrote:
 1. `aptitude search' shows only first column (such as i,p,v), but
misses package names and descriptions (any locales).

  Yeah, I found that shortly after releasing this version and fixed it -- I 
forgot to transcode back to the native multibyte encoding before sending text 
to the terminal, so it was full of NUL characters and only the first 
character got printed.

 2. When we do `aptitude show' with LANG=ja_JP.EUC-JP or ja_JP.UTF-8,
all of translations such as Package, Status are corrupted.
I dumped them and found each first bytes of double-byte characters
were removed.

  Similar problem.  I've applied a fix to SVN.

 3. When we open 'license' on ncurses window, it shows too narrow
window (any locales).

  Seems to be due to the fact that I forgot to check whether characters were 
printable before passing them to wcwidth(), so I ended up with a negative 
width calculation.  (the GPL includes a form-feed character, which is 
unprintable)

  Daniel

-- 
/--- Daniel Burrows [EMAIL PROTECTED] --\
|  DROP THE SCYTHE AND TURN AROUND SLOWLY.  |
|-- Terry Pratchett, Reaper Man   |
\--- Be like the kid in the movie!  Play chess! -- http://www.uschess.org --/


pgplYq1wWw8Qm.pgp
Description: PGP signature