View Full Version : How to enter accentation?
15-05-04, 05:25 AM
I'm working on the italian locale where I have to put a lot of accented letters...
If I directly put the accented letter from my italian keyboard in the translation then I got strange chars on the rendering.
Any help would be appreciated.
15-05-04, 01:26 PM
as far as I know the letters you are entering are right, but you have to change the locale (language) of the page being displayed. That has to be a setting for all the pages I think ...
Instead of html pages beginning with
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
they should have [xml:lang="it"]
and instead of:
<meta http-equiv="Content-Type" content="text/html;charset=utf-8" />
the content should have [charset=someitaliandefaultcharset]
15-05-04, 01:33 PM
oh, ups, that is already a setting. edit your includes/config.php page and set the
$dPconfig['host_locale'] = "en";
setting to "it" - I hope that works.
16-05-04, 09:20 AM
I tried to change the config but same result.
Anyway I empirically have found a solution to the problem. :D
1) login as Administrator
2) set its locale with your target language
3) then do all the translation work and confirm
4) then if you want you can set different locale for the administrator
If you do the translation work from differente locale than your target all the locale strings are corrupted with strange character.
Help this could help other guys.
OK I found the problem.
edit this file :
//$locale_char_set = 'iso-8859-1';
$locale_char_set = 'utf-8';
$locale_char_set = 'iso-8859-1';
//$locale_char_set = 'utf-8';
and it works... why have they commented this line?
Is it the final solution? Can we change this safely?
Have you found an answer to your question (commented line)?
Anyway PoMM3, thanks a lot for the hack!
Speaking to ajdonnison - utf-8 was chosen because there were issues to do with saving and recalling accented characters to the database when using iso-8859-1. This recollection is decidedly hazy as this was quite sometime ago, it could also be that changes subsequent to that have sorted that out but he's really aren't sure off the top of his head and hasn't got time to investigate further at the moment. Somebody may have some light that they can shed on this.
No problem with iso-8859-1, what can of problem ajdonnison met?
18-06-04, 09:13 AM
Problems were that when you stored an accented character it would be saved as two bytes in the database, and when re-displayed would look like rubbish. As Karen said, this was some time ago and may have been solved by other changes to the translation system. YMMV.
So what's the final decision? I mean has source code been corrected on that point on main CVS branch or is it still a hack? If not how could I do it (i'm a sourceforge user)?
More generally I'm very willing to partipate to the development, how to start? how may I ask access to?
me too!!! I want to partipate!!!
I think that all the application should be UTF-8: hard-coded strings if any and also database strings. Unicode is really a better solution that all the various encodings needed for all earth languages.
The problem is that it seems that browsers (at least Konqueror under Linux) use their hosts locale settings for inputs instead of the encoding specified by the web page.
That should be changed: if a page is in UTF-8, the strings types in its entry fields should be given in UTF-8 to the server... Well, it's not a dotProject isssue but a browsers' one
vBulletin® v3.6.4, Copyright ©2000-2013, Jelsoft Enterprises Ltd.