Code::Blocks Forums
User forums => Help => Topic started by: hongxing on October 26, 2005, 02:40:39 pm
-
I want to use C::B in windows and linux, the problem is my app use xerces-c as xml parsing.
I linux, I can using GNU make to compile xerces-c, when coming to Windows, I have to use mingw32.
my question is how to compile xerces-c using mingw32 in windows ? has anyone do that?
-
my question is how to compile xerces-c using mingw32 in windows ? has anyone do that?
I have and it's not trivial...
-
I'd use TinyXML, but if you need a good parser, why not use libxml2?
-
It is not a new project, I am now porting it to windows 2000
====================
tinyXML can not support Unicode, so I can not use characters like Chinese, Korean.
============================
libxml2 is simple and high performance ? can it support Chinese/Korean characters? if is , why C::B not using it but tinyXML?
===============================
can libxml2 be compiled under mingw32?
-
tinyXML is used in C::B because of it's simplicity and little overhead, it's not even a library, in fact it's only 3 .h/.cpp straight files.
As for other XML libraries, it depends on your needings, but you might want to look here:
http://xmlbench.sourceforge.net/results/features200402/index.html
http://xmlbench.sourceforge.net/results/benchmark200402/index.html
Xerces doesn't seems to be bad, but the binary size is HUGE and it's rather slow.
If I have to choose, I would choose libxml2 as rickg22 suggests, because it's well-featured and with good performance and library size overall.
-
If I recall correctly, libxml2 also needs libiconv for the handling of unicode. But then again, i never used libxml2. I just know PHP uses it.
-
Update: Check out this thread at the mingw mailing list.
http://sourceforge.net/mailarchive/message.php?msg_id=1225103
Hmmm... it appears Xerces-C can be compiled with GCC. But I'm not sure... are you sure there's not a makefile around?
-
tinyXML can not support Unicode, so I can not use characters like Chinese, Korean.
Not true. tinyXML has been supporting UTF-8 for one and a half years.