Hi
You can answer just to Phpmyadmin-devel, I'm subcribed to it, so if you cc me I'll get answer twice.
I think the second one would be better. PMA_mysql_query(something) is definetly easier to read and maintain than the first method.
Ok I will do this change
Definetly splitting the patch up will get code accepted and tested faster.
Yes it's sure. I just didn't split it till now, because I just kept one modified copy of PMA and generated patch from it. So till there didn't anything happen about applying I did it this way.
How big are they? They might belong in a new file instead. Possibly libraries/langconv.lib.php3
About 80 lines of functions, 20 lines of comments and 20 lines of code to load php extensions.
- added lang/{czech,english}-utf8.inc.php3
Could you write some documentation on how people should write UTF-8 translations?
I would prefer if they could be automatically generated, which should be possible. In patch there are also sync_english.sh and sync_czech.sh which do this for czech/english. But this could be done automatically for all languages. It's just converting chatset and adding one line about allowing recoding:
iconv -f iso8859-2 -t utf-8 english.inc.php3| sed -e 's/iso-8859-1/utf-8/' -e '/$charset/a\ $allow_recoding = TRUE;' > english-utf8.inc.php3
How well tested is this patch in terms of browsers? Is there any danger of
it breaking older or uncommon systems. One testcase I would like to see is using PMA from a web-enabled cell phone.
I tested it in all current versions of browsers (Mozilla, Opera, MSIE, links, lynx) under Linux and Widnows + some older versions (MSIE 4, Netscape 4.7x). Unicode is correctly supported by all of them. I didn't test it on any special systems like PDA or cell phones, but there still remains possibility to work as non patched version works - you can disable it in config or you can use language files which don't allow it.
Michal Cihar
On Fri, 7 Jun 2002, Michal Cihar wrote:
You can answer just to Phpmyadmin-devel, I'm subcribed to it, so if you cc me I'll get answer twice.
My bad, I normally do only reply to the list.
Definetly splitting the patch up will get code accepted and tested faster.
Yes it's sure. I just didn't split it till now, because I just kept one modified copy of PMA and generated patch from it. So till there didn't anything happen about applying I did it this way.
Ah, ok.
How big are they? They might belong in a new file instead. Possibly libraries/langconv.lib.php3
About 80 lines of functions, 20 lines of comments and 20 lines of code to load php extensions.
I would say that this probably would be better off in a new file then, for clarity's sake mostly.
My DB-Config code will be going into these two files: libraries/db_config.lib.php3 (approx 200 lines) libraries/xpath/ (approx 4700 lines)
iconv -f iso8859-2 -t utf-8 english.inc.php3 | sed -e 's/iso-8859-1/utf-8/' -e '/$charset/a\ $allow_recoding = TRUE;' > english-utf8.inc.php3
DO document this somewhere in PMA. That conversion function would be useful. Possibly have some of this in scripts/ like my scripts/convertcfg.pl. Could we get some more input on the creation of scripts for things like this?
But my original question was intended in a different sense. I had meant that some documentation should exist on how to write a UTF-8 translation from scratch for a language that has no normal form. Elvish dialects and Klingon from the far reachs of Unicode come to mind. Some people I know actually speak elvish in everyday life, instead of other languages, as for a few of them, they started out having elvish or klingon as their only common language for communication.
I tested it in all current versions of browsers (Mozilla, Opera, MSIE, links, lynx) under Linux and Widnows + some older versions (MSIE 4, Netscape 4.7x). Unicode is correctly supported by all of them. I didn't test it on any special systems like PDA or cell phones, but there still remains possibility to work as non patched version works - you can disable it in config or you can use language files which don't allow it.
Ok, that is good enough for the moment, I don't know if PDA/cell phones support unicode even. Just that I had an emergency once, and had to fix something in my mySQL DB from PMA over a web-enabled PDA.
-
Michal Cihar -
Robin Johnson