Sorry I didn't notice I've exceed the 80 column. Here is the corrected post.
Some libraries fail when using Turkish locale. It is because of Turkish has 4 "I" (lowercase dotted i, lowercase dotless i, uppercase dotted i and uppercase dotless i). For details please look at [0] Some applications fails because of this (namely some bzr plugins which use email lib) and there is a potential every application which uses .upper()/.lower() can fail. Pardus [1], which is a distro from Turkey solves some of this issues by patching few libraries to use ascii upper/lower methods [2]. I guess these patches can be applied to Debian package without breaking anything else. The following is the list of related patches: http://svn.pardus.org.tr/pardus/2008/system/base/python/files/decimal-ascii.patch http://svn.pardus.org.tr/pardus/2008/system/base/python/files/python-2.5.1-codec-ascii-tolower.patch http://svn.pardus.org.tr/pardus/2008/system/base/python/files/trfix_emaillib.patch There is also related Python bug #1813 [3] which lists all the source files which use non-ascii lower/upper methods. That bug has also a patch [4], but since it is lower level stuff than Pardus' patches, it can break some stuff. That would be great if Turkish users can use Lenny's Python without a hassle. Thank you very much. [0] http://www.i18nguy.com/unicode/turkish-i18n.html [1] http://www.pardus.org.tr/eng/ [2] http://paketler.pardus.org.tr/2008/source/python.html [3] http://bugs.python.org/issue1813 [4] http://bugs.python.org/file9440/turklocale.patch -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
