Turkish language is probably a nightmare for programmers, especially for ones that something to do with internationalization. Why? Because Turkish alphabet has two characters: 'i' and 'ı' (i without dot); and their capitals 'İ' and 'I' respectively. Did you notice it?
It is unfortunate that in Turkish uppercase of 'i' is not 'I' (which is the case in English) And it gets worse as lowercase of 'I' is 'ı' not 'i'.
So, when you are programming you should always have this situation in mind especially if you are calling (most probably you are) toLowerCase, toUpperCase and equals in any programming language. Most languages have overridden methods which takes locale as parameters.
Especially, open source projects generally have such problems. I try to hunt these bugs and here is what I got for now:
- Eclipse Web Page Editor (Eclipse WTP Project):
Page Editor generates tags with 'ı'
https://bugs.eclipse.org/bugs/show_bug.cgi?id=225857 - Eclipse EMF:
Generator code from EMF model have some problems:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=220746 - uDig (User-friendly Desktop GIS):
Doesn't start in Turkish locale
http://jira.codehaus.org/browse/UDIG-1265 - SubSonic Tools (Data Access Layer creator for .NET)
My patch for their Turkish problem:
http://www.codeplex.com/subsonictools/SourceControl/PatchList.aspx
I will be hunting for more :)
No comments:
Post a Comment