[fixed] SET NAMES vs SET CHARACTER SET

Quote from: OpenGeek at Aug 17, 2008, 05:02 PM

I’ll try and take a look at it this week snop.  My apologies, I’ve been a bit swamped and had forgotten about the patch.

Great, I recommend you, after apply a patch look where all around he changed a code and on every place switch on a breakpoint. And debug every part of installation proccess, like this you best see what the patch doing.

My method when I have created a patch was that I have found every

mysql_query("SET CHARACTER SET {$database_connection_charset}");

in all MODx files and substitute this string with

mysql_query("{$database_connection_method} {$database_connection_charset}");

and I have adjusted installation proccess, so as he creating

$database_connection_method = ’SET CHARACTER SET’;

in a config.inc.php file in manager/includes/

Now I have updated patch, so as he was compatible with newset 096 branch and I have uploaded it to the CRUCIBLE under name set-character-set_set-names_0.9.6_branch_head revision_3934.tgz, so pls downloading this newest update.

snop, can you move the review you created for this out of draft status so we can take action on it? Also, if you provide a proper .patch file (unified diff) instead of a tar.gz, we can view the patch in context of the code directly in Crucible.

Also, please note, rather than editing all of the language files, the typical process is to only modify the english file and let the translators submit new files. They can tell how many entries are missing very easily.

ahaaa, I didn’t fully understand CRUCIBLE but I’m ok now.
It’s ready now and I have leaved out language files.

What happend, any reactions, any opened review

Quote from: snop at Aug 27, 2008, 11:14 AM

What happend, any reactions, any opened review

I will get to it; just been a busy month.

Quote from: OpenGeek at Aug 27, 2008, 08:45 PM

I will get to it; just been a busy month.

classic

I really need this feature, the other site with setup cp1250 global charset in mysql, but I need utf8, so I must do core hack :’( I can’t wait to this feature. Hurry up, need next release.

I need know If you plan implement my patch, because I don’t trust my core hack because it may have not work in the next release and what then ? If I will have tables in utf8 and databse will be set up to the e.g. cp1250 charset and in MODx will be still set SET CHARACTER SET ? In this case I will be in dead end

First of all, you still haven’t taken your review out of draft mode, so I can’t review it, nor does it show on my list of things to review. Please move it out of draft mode so we can take action on it.

Second, please do not tell me to hurry up; I appreciate that you want this implemented so as not to have to apply it to future releases, but this is a volunteer open source project and as there are very few volunteers committing code at the moment, you’re going to have to be patient until I get time to deal with it.

Third, yes, I do plan on implementing your patch, though I still don’t understand why you would have a database container with a collation different from your tables, or why you would want the mysql client to do charset conversion on this data to utf8. This kind of conversion is not lossless. Why not just store and present it in it’s native charset?

Quote from: OpenGeek at Sep 02, 2008, 11:50 AM

First of all, you still haven’t taken your review out of draft mode, so I can’t review it, nor does it show on my list of things to review. Please move it out of draft mode so we can take action on it.

So how can I my review out of the draft mode ? I don’t understand so pls, help me.

Quote from: OpenGeek at Sep 02, 2008, 11:50 AM

Second, please do not tell me to hurry up; I appreciate that you want this implemented so as not to have to apply it to future releases, but this is a volunteer open source project and as there are very few volunteers committing code at the moment, you’re going to have to be patient until I get time to deal with it.

Ok, sry If I tell something indecent, my english is bad.

Quote from: OpenGeek at Sep 02, 2008, 11:50 AM

Third, yes, I do plan on implementing your patch,


I must have absolutly certaintly, that It will be implemented. So now I know that will.

Quote from: OpenGeek at Sep 02, 2008, 11:50 AM

though I still don’t understand why you would have a database container with a collation different from your tables, or why you would want the mysql client to do charset conversion on this data to utf8. This kind of conversion is not lossless. Why not just store and present it in it’s native charset?

Because We want do ours sites in more languages and I don’t have absolutly certaintly that the conversion between utf8 and cp1250 will be without problems. e.g. japan or koreai or german, I didn’t test this conversions and if all character are converted without problems, but I don’t trust this conversions.