So I've been brute forcing this at work and it looks like there may be some simplifications ... Compile the code and "hack the listing from the compile" - after all most COBOL compilations have a lots of useful information.
On Friday, May 9, 2014 12:56:12 PM UTC-5, clueless newbie wrote: > > Durand, > > Thanks for the link! > > COBOL programmers, in general, aren't adventurous -most will stick to the > standard that they're are familiar with. Large COBOL shops, on the other > hand, won't allow you to use an extension unless you can get written > sign-offs from each of your sixteen grandmothers --- too bad if you have > fewer! > > I've started by playing with amon's transforming > syntax.<http://longanswers.blogspot.com/2013/06/transforming-syntax.html>Once > I got a reworked version (I wanted the SLIF-BNF in the __DATA__ > section and the input to come from a file), I needed to pre-process the raw > cobol by stripping coomments and the sequence number and the program name > areas. So now I can get to the "guts" of the COBOL program. Simple regexes > allow me to split that into divisions. > > My plan is to begin with the IDENTIFICATION DIVISION and go (or fall down) > from there. > > On Friday, May 9, 2014 12:16:22 PM UTC-5, Durand Jean-Damien wrote: >> >> Beging a noob with COBOL I just wanna know if the language is perfectly >> standardized, or if programmers in it are tempted to use extensions that >> makes a general parser more difficult, like with C code using GNUCC only >> extensions. >> Side-effect of my question, is the COBOL source code you are targetting >> subject to some extensions, or is it writen if perfect (ANSI ?) COBOL. >> Thanks, >> >> Btw I found this link >> <http://www.cs.vu.nl/~x/coboldef/coboldef.html>interesting. >> >>> >>> -- You received this message because you are subscribed to the Google Groups "marpa parser" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
