Charset-Error in Setup-Procedure

Next topic

Author	Message
Stefan Huegel	Friday 24 June 2005 7:27:30 am Hi there. I encounter a very strange phenomenon when trying to install EZP: in the last step before closing the installation process I am given the warning ... The database cannot be used, it uses the character set which is different from the requested charset iso-8859-1. The database (MySQL 4.1.8a) is UTF-8, however, and charset for the database cannot be changed (=> shared webspace). I changed the 'preferred and allowed charsets' for my German language isntallation by modifying the respective file share/locale folder - no way I can pass by that error message. Is there any way I can teach EZP to store information in UTF-8 in my database not ISO-8859-1? Any Feedback or help is highly appreciated. Thanx in advance! Stefan.
Birger Kollstrand	Sunday 26 June 2005 6:53:41 am Hi, THere is a good article about unicode (UTF-8) support in eZ. http://ez.no/ez_publish/documentation/configuration/configuration/language_and_charset/unicode_with_ez_publish That solved most of my problems. There are still issues though tha I have not been able to solve. One of them is that email adresses with other characters than the ascii is not accepted as user mail addresses. cu / birger....
Stefan Huegel	Monday 27 June 2005 2:27:12 am Hi Birger. First of all thanx for your reply. A had already been looking into the approach which you suggested. However it did not solve my problem. Still I am stuck at the point where the EZP isntallation process tries to write things into the DB. Then I am given the above mentioned error message. My DB-charset is 'latin1', collation is 'latin1_german2_ci'. As I am working with MySQL 4.1.8a there is no way I can set the charset to 'iso-8859-1'. Anyone any idea what I am doing wrong here? :o/ Stefan.
Birger Kollstrand	Monday 27 June 2005 11:07:12 am Hi, 'latin1' is the same as iso-8859-1. So if your DB is set to latin1/iso-8859-1 and you can't change it. Then all your setup of eZ should be like the default eZ setup which is iso-8859-1 if i understand the documentation correctly. http://www.cs.tut.fi/~jkorpela/latin1/ cu / birger....