Multibyte unicode characters about rumai HOT 7 CLOSED

I agree, and wmiir does not seem to have this problem:

wmiir xwrite /rbar/0-mpd "label ’"

from rumai.

Hi,

I can confirm this issue. After updating rbar field with an utf-8 character it crashes with the above message in wmiirc.log and also tags switching stops working. I cann't use localised date info or MPD titles/artist names with localised chars.

I suspect it's caused by hardcoded sizes for characters (only ASCII is handled) for communication stream in IXP Rumai module.

Is there a chance this bug/misfeature will be fixed soon ?

from rumai.

I'm pretty busy these days; feel free to contribute a patch. :)

from rumai.

I am using ruby 1.9.2p180 and my MPD status applet is able to write Unicode text to the bar label:

>> fs.rbar['0-mpd'].read
"colors #dcdcdc #343434 #444444\nlabel (mpd) \xC3\x98ystein Sev\xC3\xA5g: Consolation\n"

Notice that the UTF-8 characters are expressed as two bytes instead of a single \u{...} compound. Of course, this isn't a solution for the original issue, but it is somewhat puzzling.

>> puts "\xC3\x98ystein Sev\xC3\xA5g"
Øystein Sevåg
nil

>> puts "\xC3\x98ystein Sev\xC3\xA5g".encoding
UTF-8
nil

from rumai.

The problem was how Ruby 1.9 counts the length of a string: it counts characters, not bytes. I've got an ad-hoc solution working locally but my existing unit tests are failing. I'll push an update once I fix the failing tests.

from rumai.

Commit e7a9757 fixes this issue.

from rumai.

Finally fixed. Grreeaatt !
Thank goes to sunaku.

from rumai.