Vanilla 1 is no longer supported or maintained. If you need a copy, you can get it here.
HackerOne users: Testing against this community violates our program's Terms of Service and will result in your bounty being denied.

Bug: Mutibyte charactors encoding problem in mySQL database version 4.1 and later .

edited September 2006 in Vanilla 1.0 Help
I thank Lossumo give me this pretty tools ! It is very good for my persional site . But when I use it and try to backup my database , I meet a problem . Since mySQL ver 4.1 update , it add a new feature called muti language support , this make PHP data connection changed in mutibyte languages. If don't change them , it will insert wrong data in database when you use Chinese or other mutibyte language . You can see right charactors in database . But we cannot backup database via phpMyAdmin 's SQL dump file , the SQL dump is wrong , and when we try to restore them to database , over half of charactors will be changed and unreadable . This problem is found in all encoder of Chinese language , mySQL version 4.1 and later . Cause we can't see right data when this problem fixed ,we must first get right data from database via some tools or PHP script , phpMyAdmin or other database tools like mySQL front will never get the right data from wrong database , just php programs what using same way to access database with Vanilla conrrent version . this tools must get data from database and make a right SQL dump and supporting all encoding . like utf8 and gb2312 in Chinese etc . A problem detail and fix solution advice is on my blog ( . It is not whole solution , I think it must get mySQL database server version first , then deside if use the SET NAMES query . and it must get the encoding setting from vanilla setting to instead value "utf8" in my code . I have hacked my Vanilla myself . But I ' m a html coder and flash coder , I can't fix all of it . So my code now is just a 'hack' , not a fix . Wish the fix soon !


  • Options
    MarkMark Vanilla Staff
    I had actually already implemented this in Vanilla 1.0.1, but then it caused a bunch of problems with data entered before it was implemented - that data came out without the correct encoding and looked all garbled.

    I will be adding it back in for the next revision, but I need to come up with a way for people to either (a) turn it off or (b) convert their existing data to a compatible format after the change.
  • Options
    i had the same problem with you,icebird i move my forum to other hosting, but after i restore my data,i found that half of charactors unreadable :(
  • Options
    @mark: Did you see other posts I made on this subject, here and on Vanilla Dev, giving a method to build an upgrader from latin1 encoding to utf8?
    I suppose so since I posted in 3 place, at least, but this forum being so fast growing, I'm not sure you read any.
This discussion has been closed.