Bugzilla – Bug 104867
GPM fails to insert non-ASCII characters properly
Last modified: 2008-01-08 10:23:10 UTC
When trying to copy & paste non-ASCII characters with GPM on the Linux console (framebuffer) into Bash, those characters do not appear (e.g "möglich" will appear as "mglich"). Furthermore Bash gets confused: when trying to insert an ASCII character after the 'm', some "magic" character appears, and when moving the cursor along the word, characters appear and disappear, and Bash's readline gets out of sync regarding display and internal character position. GPM should be able to paster non_ACSII characters as well.
I'll look at it.
I'll try to debug this today...
Created attachment 47148 [details] patch to allow copy and paste of latin1 1. The actual copy-and-paste is done by the kernel, so this is not a GPM bug. 2. Unfortunatelly, it's IMHO impossible to reconstruct the UTF-8 text when copying from the console -- this patch just enables it for Latin1 (Umlauts for instance) and replaces other characters with a question mark to avoid the "magic" character from appearing. I'm not sure whether this doesn't break other languages, where it might have worked before -- someone would have to check, I'm not a kernel hacker.
It should be fixed in kernel -> reassign to the kernel developers.
Ok, I'll take a look at what we can do with this. Jiri, would you be willing to take a look into the console code again?
I think it is actually possible to reverse-map the unicode->glyph mapping; the inverse_translations table just needs to be initialized accordingly. I'll have a look at it next week.
Any progress? ;)
Submited a patch to the kernel CVS, should be in the next beta.
fixed in 10.1 beta 7
In openSUSE 10.2: I tried to copy & paste some german umlauts (ä, ö) with GPM in Linux console. These characters didn't copy.
Ales, does this still happen on 11.0? I can't reproduce it.
It works in 11Alpha0 :-)
Thanks, closing as FIXED