Using international/special characters breaks messages

[JeM7BsB]

Hi,

I have upgraded my forum (www.lowcarbforum.nl) to 2RC4 and now any message that contains an international/special character only shows up with the part till the position of such a character. Same thing happens in topic names. My forum language is set to Dutch utf8. Anyone has any idea what to check or configure?

Regards,

Eddy

[Norv]

Are your database tables columns UTF8? You can see the collation of the columns in PHPMyadmin, when looking at a table's structure (i.e. smf_messages). Please look at the columns, not the entire database or table.

In your file Settings.php, is there a line saying:

Code Select Expand


$db_character_set = 'utf8';


[JeM7BsB]

I imported an old database in to a clean install of 2RC4. Table yabbse_messages has UTF collation (utf_general_ci) in several columns/fields, for instance body, subject

I do have that line in my Settings.php

[Norv]

For what SMF version was the old database?

Did you have a backup of database structure and data, or only data?

What happens if you comment momentarily that line from Settings.php, i.e. replace it with:

Code Select Expand


// $db_character_set = 'utf8';


[JeM7BsB]

The old database was 2RC3

It was a complete database, so structure and data. In fact all users and topics/messages were transferred fine (except for the special character challenge)

Nothing changes when I comment the line

[Norv]

In phpMyAdmin, can you see a particular message (which appears "broken" on the forum) correctly displayed?

A link to your forum would be appreciated if possible.

[JeM7BsB]

The messages appear 'broken' in phpMyAdmin too. Just made a test message "qwertéüø' which shows up as 'qwert'

My forum is at http://www.lowcarbforum.nl

[Deaks]

JeM7BsB, does this issue still happen, as you not repsonded in 3 weeks im marking this as solved until you update us

[tycms]

I have exactly the same problem described in the first post of this subject. I previously started a new topic about this:
http://www.simplemachines.org/community/index.php?topic=415827

Following Norv's directions, I made further tests on this. Now, I suspect that this might be a bug in the update process of the smf_2-0-rc4_update package.


For anyone to easily replicate my results, I will describe what I have done step by step.

1) I started with a clean 2.0 RC3 install (using the smf_2-0-rc3_install.zip package). I selected the UTF-8 database option during the install.

2) I started a new topic and posted a new message containing non-ANSI characters.

3) I uploaded the smf_2-0-rc4_update.zip package and applied it to the RC3 install.

After the update, non-ANSI characters are corrupt (replaced with question marks).

[tycms]

Please also refer to the following bug report:

2.0 RC4 update bug?

[Illori]

please use a separate thread for your issue.

[tycms]

I already did that Illori. And I doubt this is "my issue". If this is a bug, it concerns the SMF development process and hence the whole community.

[JeM7BsB]

JeM7BsB, does this issue still happen, as you not repsonded in 3 weeks im marking this as solved until you update us

Well, as I didn't hear anything thing back I sort of solved it myself. It's possible to use special characters again on the forum, but all old messages are still 'broken'. Forum users seem to be ok with this