Unicode doesn't work with emoticons

[m_gol]

When I try to make my own set of emoticons, I cannot put tekst written in Unicode (with ą, ę, ł etc.) in emoticons.py. I moved to Jabber from Polish IM named Gadu-Gadu and I think many people doing the same would like to have their old, GG-style emoticons in Gajim; but they can't do this, because some emots used in GG uses Unicode characters (for example, <głaszcze> should return one of emoticons).

[asterix]

see also #3327

[asterix]

adding

# coding=utf-8

at the begining of the file fixes the problem (see http://www.python.org/dev/peps/pep-0263/)

I add that line to existing pack in our sources

[asterix]

(In [8554]) add encoding to emoticons files. fixes #3227

[roidelapluie]

(In [8606]) Add encoding to emoticons files. Fix #3227.

[m_gol]

I don't know how to do it - I added a following line:

'<głaszcze>': 'coolglasses.png',

to my emoticons.py and it didn't work; when I wrote <glaszcze> instead of <głaszcze>, it worked. I tried to add the information about UTF-8 coding at the beginning of the emoticons.py file and it didn't change anything, again.

I have "static" emoticon set enabled.

[asterix]

handle utf-8 emoticons

[asterix]

could you test the attached patch ?

[m_gol]

How can I apply it? I can see a gajim.py file neither in my src Gajim directory nor anywhere else inside the Gajim folder.

[asterix]

it is in src folder. But maybe you're under winows ?

[m_gol]

Yes, I'm under Windows, temporarily.

[asterix]

ok so forget it. There are many things to install to use vn version under windows. I tested under lin and tried with chars like é and that works. So I'll commit it.

[asterix]

hmm there is a problem with the IGNORECASE of re module:

re.search('é', 'qwért', re.IGNORECASE)

that works

now if I do the same with 'qwÉrt' instead of 'qwért', that doesn't work.

so if you write ":é" as an emoticone in emoticons.py, only :É will work.

[m_gol]

"so if you write ":é" as an emoticone in emoticons.py, only :É will work." So couldn't You just put the to-be-emoticon text through some procedure, to translate it from ":é" to ":É"? Then emots would fit chosen pattern (after some changes), wouldn't they?

[asterix]

it's what we do, when we load emoticons, we upper() them. and when we compare, we use the re.IGNORECASE argument, so it should match, but it doesn't.

I just posted to comp.lang.python ... I'll see

[asterix]

(In [9053]) handle unicode chars in emoticons. fixes #3227

[asterix]

(In [9054]) handle unicode chars in emoticons. fixes #3227