[SourceJammer-devel] Re: SourceJammer and Encoding

[Marcelo S. <sal...@on...>]

   I've tried and find that the problem is a little worse than I 
imagined (or maybe it is not).

   You're right, ISO-8859-1 can't currupt win files: I had a win 
ISO-8859-1 win2K SJ server running for a long time, no probs. Thingy is, 
my Linux is UTF-8. There is this env var: "LANG=pt_BR.UTF-8" that seems 
to be sensitive for Tomcat. I didn't tell this earlier, but even my diff 
window on SJ was messed up, also for non-corrupted files on disk. 
Changing this var before running SJ, e.g.:

   > LANG="pt_BR.ISO-8859-1" sh sj_gui

seems to help a little. I bet if I set this var to ISO-8859-1 before 
running tomcat the problem will go away (for windows, that is). Then I 
will probably have to set 2 tomcat servers, one ISO-8559-1 and the other 
UTF-8. That's a lot of resources, so I thought of the alternative I've 
mentioned before.

    Aside of this complications, is there a way to use text files of 
many encodings on the same SJ archive/server? Suppose one have some 
webpage that has some .html for languages like Japanese, English, Arab, etc.
Would it be possible to store this files uncorrupted in SJ?

PS: Rob, if I were to implement this, would it be a hard thing? If you 
give me some directions, I might implement it, if needed.

===============================
   Marcelo Saldanha
   Ciência da Computação - UFMG


Robert MacGrogan wrote:
> Hi, Marcelo.
> 
> Glad to hear from you again. I recognized your name as soon as I saw it.
> 
> What you're saying sounds possible, but I'm not sure it's necessary. I've never noticed that using
> ISO-8859-1 corrupts Windows files. Can anyone out there tell me if I'm wrong about that?
> 
> Unfortunately, if you decide you want this change, you'll probably have to implement it yourself.
> I really can't make any promises at this point about when I'll (finally) get around to developing
> new SJ functionality.
> 
> --Rob
> 
> 
> 
> --- Marcelo Saldanha <sal...@on...> wrote:
> 
> 
>>   Hello, Robert!
>>
>>   It's been a while since my last e-mail. I'm the one who did that 
>>little translation tool, ages ago...
>>
>>   Now I'm back, and have a little question for you: Is it possible for 
>>SJ to use different encoding charsets for differente files? I had 
>>configured SJ server to store text files in UTF-8, since most of my 
>>files are on Linux, and are UTF-8. Now I need to store some Delphi files 
>>and it seems that these files are ISO-8859-1.
>>
>>   Well, my big problem is that this situation forces SJ to be mutually 
>>exclusive regarding the charsets. That is, I can either store using 
>>UTF-8 and have my Windows files corrupted or store using ISO-8859-1 and 
>>corrupt my Linux files. I thought of setting up two servers, but that 
>>would be too much of a hammer solution, and I'm trying to avoid that.
>>
>>   I can think of three possible solutions for this:
>>   1) SJ could use a specific encoding charset (utf8 or iso8859-1) for 
>>storing the text file, but also storing what encoding the user wants the 
>>  file to use. This way, during the Get SJ would convert the file 
>>between the storage and user charsets.
>>   2) SJ could store each file it its own encoding charset, eliminating 
>>the need of conversion.
>>   3) Combine the two above: SJ would store the file on its own 
>>encoding,  but also allow conversion (other users might want the file 
>>with a different encoding).
>>
>>   I hope you can help me find a solution for this problem.
>>
>>   Thanks!
>>
>>===============================
>>   Marcelo Saldanha
>>   Ciência da Computação - UFMG
>>
> 
> 
> 
> 
> 		
> __________________________________ 
> Yahoo! Mail - PC Magazine Editors' Choice 2005 
> http://mail.yahoo.com
>