> Can you tell me what advantage(s) your version would have for Windows native > users?
I'm not sure Python integration into GDB is something that users are going to really notice. Mostly they're going to see their QStrings displayed nicely and the same for their STL containers and not think twice about how it's happening! Well it's like you said, binary vs source and also the usual msvc dll debuggability issues. My method compiles the Python project's sources (with a few patches, granted) using compilers I trust. There's always the test-suite if you want some extra re-assurance that things are working correctly. I'm not going to pretend that there aren't differences, but for the task of embedding into GDB (and any other task I've put them to) I've never had any problems, both with Qt Creator's pretty-printers, and the STL ones. I'm starting to repeat myself a bit, but in terms of stability, my NDK releases have had this GDB integration for ages now - the recent difference is that now it is cross compiled and updated to 2.7.3 and 3.3.0b2. The Windows version of the Necessitas Android Qt SDK has had > 13,000 downloads (and that was a few months ago - so likely nearer 20,000 now) and not a single issue has come up with the Python GDB integration. It just works. The patches have been tested and merged by Google too and down the line I'll be trying to get them into CPython. I guess in some ways it seems a bit like debating whether to take some msvc compiled expat libraries or to compile them from source. They'll both parse XML, but I know I'd rather build the software I release myself so I can investigate and fix it if it ever goes wrong. Cheers, Ray. On Sat, Sep 15, 2012 at 9:55 PM, John E. / TDM <[email protected]> wrote: > On 9/15/2012 1:30 AM, Ray Donnelly wrote: >> Hi John, >> >> Not sure you'll be interested but I build my own version of Python >> into some Android NDKs, and this Python is cross compilable/compiled. >> >From a Linux box, 12 variants can be built (Windows, Mac, Linux) * >> (x86, x86_64) * (2.7.3, 3.3.0b1). Native shouldn't be much of a >> stretch too but it's not something I've tried. Here are the some 2.7.3 >> Windows builds I made. > > Thanks for the input. Ultimately it seems to me that Ruben's method will > prove the most stable, as it compiles directly against the Python > project's own released binaries. Can you tell me what advantage(s) your > version would have for Windows native users? > > -John E. / TDM > > ------------------------------------------------------------------------------ > How fast is your code? > 3 out of 4 devs don\\\'t know how their code performs in production. > Find out how slow your code is with AppDynamics Lite. > http://ad.doubleclick.net/clk;262219672;13503038;z? > http://info.appdynamics.com/FreeJavaPerformanceDownload.html > _______________________________________________ > Mingw-w64-public mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/mingw-w64-public ------------------------------------------------------------------------------ How fast is your code? 3 out of 4 devs don\\\'t know how their code performs in production. Find out how slow your code is with AppDynamics Lite. http://ad.doubleclick.net/clk;262219672;13503038;z? http://info.appdynamics.com/FreeJavaPerformanceDownload.html _______________________________________________ Mingw-w64-public mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mingw-w64-public
