Symphony.

I'm Chinese, Chinese characters were displayed as "???" in Symphony. Could this problem be fixed?

This is a foreign area to me (pun intended) but you may want to take a look at $settings['database']['character_encoding'] = 'utf8'; in /manifest/config.php.

It should work so long as your table character set and collation is something like utf8_unicode_ci and, as Lewis suggested, you make sure that /manifest/config.php has something like this:

$settings['database']['character_set'] = 'utf8';
$settings['database']['character_encoding'] = 'utf8';
$settings['database']['runtime_character_set_alter'] = '1';

for the database settings.

Thanks a lot to you all. It's fixed when I check the table character set and collation is utf8unicodeci. But the URL of the articles which have a chinese title will be blank! I hope the URL could be something like “http://www.aaa.com/articles/post-123/” when the title is chinese.

---PS. Hopes you to be able to understand me.

I think I should learn English well first. Sorry.

Your English is fine. I don't think it's possible to switch between filter methods in your data source. So, you would probably need to have all of your articles' URLs represented as article IDs.

Thank you again, let me have a try...

yes, i think article IDs will be your best solution. i was using the title of entries for urls, but i found that those IDs cut off around 50 characters. so anything that was different after 50 would just get cut off and mess up the filtering. therefore i had to change things to the IDs.

Lewis wrote:

you would probably need to have all of your articles' URLs represented as article IDs.

It isn't just Chinese that is affected in this way; so too is Hebrew with niqudim (vowel points). Here is an explanation of what I'm talking about, if you're interested.

I'm guessing it will be a while yet before scripts like this can be used in URLs.

David.

We have a website that uses a mixture of German and Chinese in an entry. The database table is encoded as UTF-8 with utf8_unicode_ci as collation. We are using the Typography with Markdown Extra formatter and for some reason were seeing question marks in the backend but properly encoded characters on the frontend, e. g. the following phrase

??????? Haben Sie Messer und Gabel?

will become

??????? Haben Sie Messer und Gabel?

in the backend and

你这有刀叉吗? Haben Sie Messer und Gabel?

on the frontend.

The problem begins when you try to edit the entry as it will be loaded with question marks instead of the proper Chinese characters.

Does anybody know how to solve this issue?

PS: It seems like this forum has some problems with Chinese as well. D’oh!

This is how the input looks like in the backend before saving:

Attachments:

sym_chinese.png

Using the entities instead of the actual characters in the backend is a useful workaround :)

Could it be a problem with the use of “normal” string functions (e.g. str_replace) rather than their multibyte string functions equivalents?

Well, but there shouldn’t be any replacements for the plain input value and the formatted_value in the database just works fine (saving the characters as entities).

@anyone: did you find a solution? I am working on a bilingual site, and everything has turned to question marks!

Maybe i can help you @dimlau, I'm chinese too. loves[at]kamwing.net