From: Dave J. <dav...@go...> - 2009-11-30 04:57:43
|
Hi , Russ wrote: > ...... Could you make the classes available for that world? There are no Java classes for this, since the EAI library and the main programs are both written in Tcl, however AFAIK someone (I think it was Doug) a Python version working (using a wrapper around Sarah's client library). He would have a version of my EAI test program in Python, and a copy of the WRL file. The Python script should be very easy to translate to another language, but since the test program can be reduced to a 'two-liner' here is the Tcl code. Essential syntax: OO: Class newObjectName (instantiation) OO: Object/Class method thing thing thing the {}'s are string delimiters . . # Connect to the browser . . # Create an instance of a Node object Node labelTemp # retrieve the 'message' node , which is then used to inform the user of the test progress labelTemp getNode {message} # getNode is a method in Node, not a method in Browser. I guess in Java you would say something like: Declare a object called labelTemp of class Node Call the browser's getNode method with 'message' as a string parameter and put the value into labelTemp Dave 2009/11/30 Russ Kinter <py...@ve...>: > > >> -----Original Message----- >> From: fre...@cr... [mailto:fre...@cr...] On Behalf Of >> Dave Joubert >> Sent: Sunday, November 29, 2009 5:41 PM >> To: FreeWRL X3D/VRML Viewer >> Subject: [FreeWRL] More problem in the EAI >> >> Hi, >> >> I have been spending a lot of time fighting with the telecomms company >> (they broke my wires twice in one week!!), and getting my 64bit >> environment functional. >> >> I have now got around to running my first test on the EAI, but using >> my Tcl based library on the client side, and the new (1.22.5 , 64bit) >> FreeWRL crashes on a test it used to pass. >> >> Just the critical portion of valgrind is shown: if the previous stuff >> is important, I will send it. >> >> EAISERVER: no client yet >> readEAIsocket--, retval 1 >> read in from socket 11 bytes, max 8192 bfct 0 cmd <1A message >> > >> handle_EAI-- Data is :1A message >> : >> EAI_parse_commands:start of while loop, strlen 11 str :1A message >> : >> EAI - seq number 1 >> EAI command GETNODE (A) strlen 10 >> >> ... 1 GETNODE message >> EAI_GetNode - getting message >> ==28396== Invalid write of size 4 >> ==28396== at 0x4E7B20D: lexer_fromString (CParseLexer.c:163) >> ==28396== by 0x4EC4644: DEFNameIndex (X3DParser.c:273) >> ==28396== by 0x4F2C4AA: EAI_GetNode (EAIHelpers.c:261) >> ==28396== by 0x4F29898: handleGETNODE (EAIEventsIn.c:656) >> ==28396== by 0x4F2AB13: EAI_parse_commands (EAIEventsIn.c:200) >> ==28396== by 0x4F2CE63: handle_EAI (EAIServ.c:346) >> ==28396== by 0x4E6F58C: EventLoop (MainLoop.c:584) >> ==28396== by 0x4E7005D: displayThread (MainLoop.c:1281) >> ==28396== by 0x5C2DA03: start_thread (pthread_create.c:300) >> ==28396== by 0x91247BC: clone (clone.S:112) >> ==28396== Address 0x90 is not stack'd, malloc'd or (recently) free'd >> ==28396== >> ==28396== >> ==28396== HEAP SUMMARY: >> ==28396== in use at exit: 962,873 bytes in 4,532 blocks >> ==28396== total heap usage: 13,458 allocs, 8,926 frees, 4,124,780 >> bytes allocated >> ==28396== >> ==28396== LEAK SUMMARY: >> ==28396== definitely lost: 1,828 bytes in 70 blocks >> ==28396== indirectly lost: 4,866 bytes in 115 blocks >> ==28396== possibly lost: 279,588 bytes in 276 blocks >> ==28396== still reachable: 676,591 bytes in 4,071 blocks >> ==28396== suppressed: 0 bytes in 0 blocks >> ==28396== Rerun with --leak-check=full to see details of leaked memory >> >> The WRL file, which does nothing until the remote EAI app starts >> talking to it, has not changed, and is attached. >> >> Dave > [Russ Kinter] I seem to remember that three years ago FreeWRL's EAI did not > work on a 64bit -well it got the browser, but crashed out right when trying > to get an MFNode eventOut and if you can't get children_changed value you > might as well not have EAI at all. Dave, Could you make the classes > available for that world? > > > _______________________________________________ > FreeWRL mailing list > Fr...@cr... > |