Bug 821877

Hi, 

Some time ago, I want to fix the "unrecognizable character with Chinese encodings" problem for unzip.

So I adapted iconv patch for unzip from Archlinux and made a package.

patch: https://aur.archlinux.org/packages/unzip-iconv/?setlang=zh_CN

After some testing, I found my adapted patch was overriden by our own librcc patch(internally, they can be both applied, but librcc will take over everything about codepage, iconv has no chance), which means we have done things to resolve that problem.

But that librcc patch can't do anything with Chinese (And I can't apply my patch):

use the zip file I uploaded, try to unzip it.
 
I tested it with 12.2 and 12.3, with built unzip rpms from Archiving repo. (librcc version 0.2.10)

I built the source to a deb on Ubuntu 12.04, failed. (librcc version 0.2.9)

But some Arch users told me it displays good on their systems with the same source (librcc version 0.2.6) 

So the things we can do are:

fix the patch itself or librcc (we have to confirm with librcc upstream to see if CJK is included in the "many languages" mentioned in our librcc description)

if librcc can't, drop the patch and find another one supports CJK and European encodings well. because we can't fulfill some users' needs by compromising others'.

Actually community users have found one: libnatspec, from Altlinux. it seems to be a successor of the iconv patch (icon patch is named alt-iconv-*.patch). And it seems to be the best over the Internet.

I don't have the required knowledge to debug any deeper, so I need a developer to take a look at our librcc.

Greetings

Marguerite