Andrew, Glad you asked. Surely I need a faq to answer those...
Pointers are not handled. Or they will be mishandled. I may leave a choice to convert the notations this.that and this->that as this[that] I've seen object.rect->topleft and this gets more confusing... so object[rect,topleft] is the answer but it's gets weirder the more code you see ;) Pointers can be much more complex than just simple assignment of variable addresses. Even a for loop can hard to parse for(mystruct* s->t=0; (x ? i,z)< thisfcuntion();++i) // etc... is a beast to translate... But I got some ways of doing some of these cases. Recognizing how a for loop is incremented alone means different strategies for repeat statements. Like basic's goto statements what do you do? Now, pointers... How could I translate a register barrel shifter for example? in a code like this #define SIZEOF(thisstruct) int *b=thisstruct->nextpointer where you move from struct element to another using the pointer is best done with a simple thisstruct[b,element] Some C++ or Java oriented constructs/logic must be done by hand. As I said it does the tedious translation and parsing the job. you did see the screen shots right? > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] Behalf Of > [EMAIL PROTECTED] > Sent: Saturday, August 07, 2004 14:22 > To: [EMAIL PROTECTED] > Subject: Re: translater tool, whoa > > > > > I wrote a transcript translator engine for c, pascal, basic > > which also handles parts of fortran, php, flash, java and js. > > > > -- Please send me more info on it. Btw, how did you handle the > pointer stuff > in those languages??? > > Andrew > > _______________________________________________ > use-revolution mailing list > [EMAIL PROTECTED] > http://lists.runrev.com/mailman/listinfo/use-revolution _______________________________________________ use-revolution mailing list [EMAIL PROTECTED] http://lists.runrev.com/mailman/listinfo/use-revolution
