axis-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Aleksander Slominski <as...@cs.indiana.edu>
Subject Re: UTF-8 char issue with 1.2RC2 ?
Date Wed, 02 Feb 2005 19:31:33 GMT
Mike Burati wrote:

> 
>  
>
>>i can confirm it that this is UTF8 related problem with the service
>>    
>>
>and not with AXIS.
>  
>
>>i have run the same test with XSUL SOAP stack using SDI and it failed
>>    
>>
>in my UTF8 decoder as well.
>  
>
the byte sequence is incorrect: 0xE0 0x20 ("a`" " ") - E0 indicates that 
it is start of two byte encoded sequence as high bit is set so following 
bitmask is expected: 1110xxxx 10xxxxxx 10xxxxxx but 0x20 is not 10xxxxxx 
so it is clearly incorrect UTF-8 (that explanation is not 100% guranteed 
as i tend to forget such stuff ...)

>Thanks, I was hoping it was the service as I didn't have time to debug
>what was going on deep down in AXIS (and then crimson/xerces) today :-).
>I had tried the service with SOAPScope and it didn't complain...
>
>Now that you clarified that it does fail with a UTF-8 issue with another
>(non-AXIS) toolkit, I double checked the simple SOAP envelope posting
>class (debug tool) I had written a couple years ago and sure enough I
>wasn't explicitly setting the encoding for the response stream, just for
>the outgoing soap envelope.  When I try to read the envelope using
>UTF-8, it fails for me too even with that simple output/intput stream
>URLConnection debugger.  If I tell it to use ISO-8859-1 for the response
>envelope, then it does decode that char correctly, so I'll assume the
>service provider is incorrectly encoding via ISO-8859-1 despite it
>claiming that it's encoded via UTF-8...
>  
>
i think you are right - what we see is Latin-1 output incorrectly 
described as UTF8 ...

>Thanks again - that saved me some time - I would've started debugging
>too low in the stack after running that first non-AXIS toolkit and
>seeing it work ok...  Your clarification that another toolkit failed
>kept me from digging the hole deeper when the problem was right at the
>surface :-)...
>  
>
i am glad to hear that ;-)

alek

-- 
The best way to predict the future is to invent it - Alan Kay


Mime
View raw message