On Thu, 2011-02-10 at 10:11 +0100, Stadelmann Josef wrote: > And test and test and be aware that not every client can accept
:-) So far I was quite happy with it > per default message-response as large as 65536 bytes > (16 bit pointers eventually). I'll have to re-read your earlier post again (and search through the list), but generally now with libxml2 I don't have any general troubles with large data (if consider 900k as "large"). Right now I use this on Linux and Windows (both 32bit normally, few tests with 64bit). @Emanuele: Did libxml2 help in your case? Possibly it was a similar thing ... I still use 1.6. Rgds, tge > Also chunking and streaming > architectures/implemntation should ring alarm bells. > Josef > > -----Ursprüngliche Nachricht----- > Von: Emanuele Benedetti [mailto:be...@mclink.it] > Gesendet: Donnerstag, 10. Februar 2011 09:43 > An: Apache AXIS C User List > Betreff: Re: Guththila vs libxml2 > > In data 09 febbraio 2011 alle ore 20:00:19, Nandika Jayawardana > <jayaw...@gmail.com> ha scritto: > > > Guththila parser is based on pull model and should be slightly faster > > than > > libxml as explained by the article. In the last few years, Guththila > > became > > the default parser and hence undergone good amount of testing. Hence, you > > can use either parser for your usage. > > > Some times ago I had to recompile axis2 with libxml2 support rather than > Guththila as the latter silently discard messages bigger than 128K. I > don't know if the sistuation is changed with the last versions (I'm > speaking of a couple of years ago when was just out the axis2c version > 1.6). > > SO if you plan to receive messages bigger than that, make some tests. --------------------------------------------------------------------- To unsubscribe, e-mail: c-user-unsubscr...@axis.apache.org For additional commands, e-mail: c-user-h...@axis.apache.org
