[GCC-XML] Parsing a lot of files...

[Luis Paulo Torres]

Thanks for the help, guys!

I'm using QXmlStreamReader to process my outputs, and I tried to use
namespace to filter the classes and functions which I really want. I will
try use these ID/location informations to find the specific fields from my
classes. And after try it with Clang.



2013/10/7 Peter Hackett <peter.hackett at cal.berkeley.edu>

> The output of gccxml includes stuff like
>
> <File id="f765"
> name="/opt/icmanage/TOOLROOT/2.2/boost1_42/include/boost/memory_order.hpp"/>
>
> and other stuff like
>
>  <Enumeration id="_25085" name="memory_order" context="_2196"
> location="f765:41"
> ...
>
> (Note "f765")
>
> You could write a filter to grab only the stuff you care about.
>
> I user Qt's (http://qt-project.org)'s QDomDocument (xml reader/accessor)
> to process my gccxml output.
>
>
> On Mon, Oct 7, 2013 at 5:15 AM, Brad King <brad.king at kitware.com> wrote:
>
>> On 10/06/2013 07:57 AM, Luis Paulo Torres wrote:
>> > I just want informations from one file (my .cpp file).
>>
>> gccxml dumps the C++ interface of the entire translation unit by walking
>> the global namespace and dumping everything reachable from it.  Instead
>> you can restrict the starting point with -fxml-start=myname.  That will
>> reduce the output significantly but will still dump the dependencies
>> of your interfaces.  There is no way to restrict it further.
>>
>> You could also look at Clang which has a library for tool writing:
>>
>>  http://clang.llvm.org/docs/Tooling.html
>>
>> -Brad
>> _______________________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Follow this link to subscribe/unsubscribe:
>> http://www.gccxml.org/mailman/listinfo/gccxml
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/gccxml/attachments/20131007/c0d4161b/attachment.html>