On Feb 28, 2007, at 6:14 PM, Alan W. Irwin wrote: > Also, could you provide plparseopts in the thin interface? > Virtually every > language interface has plparseopts defined (normally with just one > argument). > That would have allowed me to write x10.adb completely in the thin > interface.
Before I started to make the change by adding plparseopts to the thin binding, just for kicks I tried to run an example using command line arguments. (I usually don't supply any in my normal testing, but I have tried it in the past to make sure that the command parsing was working, which it was.) So I've discovered that when using the 5.7.2 libraries, a bus error was generated with this call stack (ada_42 is the main program): #0 0x901316e0 in memcmp #1 0x907f280c in __CFInitialize #2 0x9083753c in __CFSetupFoundationBridging #3 0x929cc4f4 in _NSToDoAtProcessStart #4 0x9293ca60 in _NSInitializePlatform #5 0x90a40548 in call_class_loads #6 0x90a4045c in call_load_methods #7 0x90a3b2c8 in map_images #8 0x8fe0f3c0 in __dyld__ZN16ImageLoaderMachO14doNotificationE15dyld_image_modejPK15dyld_ image_info #9 0x8fe035a8 in __dyld__ZN4dyld12notifyAddingERSt6vectorIP11ImageLoaderSaIS2_EE #10 0x8fe0da94 in __dyld__ZN11ImageLoader4linkERKNS_11LinkContextENS_15BindingLazinessENS_ 18InitializerRunningEj #11 0x8fe04054 in __dyld__ZN4dyld4linkEP11ImageLoaderNS0_15BindingLazinessENS0_18Initializ erRunningE #12 0x8fe0992c in __dyld_dlopen #13 0x900308f4 in dlopen #14 0x96a472f8 in sys_dl_open #15 0x96a47d70 in tryall_dlopen #16 0x96a4b0e0 in try_dlopen #17 0x96a4b4c0 in lt_dlopenext #18 0x0016fd70 in plLoadDriver #19 0x0016eac8 in pllib_devinit #20 0x0016cf3c in c_plinit #21 0x00032e50 in plplot.initialize_plplot at plplot.adb:1615 #22 0x0001f3e4 in plplot.initialize_plotter at plplot.adb:34 #23 0x000420fc in testing_thick_binding.main_testing_thick_binding at testing_thick_binding.adb:74 #24 0x000490d0 in ada_42 at ada_42.adb:19 Using the same Ada sources but linking instead the old 5.5.3 libraries, there is no crash and the commands are parsed correctly. (One qualifier--to run the old libraries I have to edit the thin binding so that c_plmeridians and c_plmap either do or do not have the leading c_, but I hardly think this is the problem.) The last few calls in the call stack seem to be OS X calls, as NSxxx is a very common OS X symbol (stands for Next Step--why change it if it's not broken, right?) and CFxxx probably stands for Core Foundation, another OS X name convention. When there are no command line args, the program runs normally. Any ideas? Jerry ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
