I would suggest that you might be losing usefull information by ignoring comments :)
> -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On > Behalf Of John McKown > Sent: Sunday, September 07, 2008 1:00 PM > To: [email protected] > Subject: JCL parser status > > Well, on the off chance that anybody is interested, my creating a JCL > parser using Flex and Bison is coming along. My implementation is to read > a single JOB and create XML output. Parts of my code are "crufty" because > JCL is "crufty". But whoever designed the IF statement needs to be shot! > Almost every other JCL statement is in three parts: the label (jobname, > stepname, ddname, etc), the verb (JOB, EXEC, DD), then the operands for > that verb. The only exceptions are PEND, ELSE, ENDIF, and IF. Now, PEND, > ELSE, and ENDIF are easier because they don't have any operands, only > comments. But the IF statement! It has the keyword THEN in the area which > is the start of comments for all other JCL statements. And my lexer is > designed to automatically remove comments. So the THEN part of the IF > disappears. For now, I plan to allow the lexer to remove the THEN and > just assume it is there. If I can figure out a way to make the lexer > EASILY have a special case for the IF statement, then I'll test for it. > Since this particular implementation is to encode JCL into XML, then THEN > keyword is unneeded. It only exists on the IF statement and adds nothing > to the sematics. It is likely there just to make people comfortable. > > I vote for replacing JCL again. Preferrably with something that is > actually DESIGNED and INTEGRATED? The original OS/360 JCL may have been > designed well for the time. But it has grown old and crufty. And I haven't > looked as the speciality JCL statements yet (such as the PRINTDEV from > PSF?). > > -- > Q: What do theoretical physicists drink beer from? > A: An EIN stein. > > Maranatha! > John McKown > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO > Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
