hamdyz@prism.CS.ORST.EDU (Zikif Hamdy) (04/15/90)
From my previous posting asking for some information about MacYacc, language compiler, I get response from these generous souls Dennis Cohen from Claris Corp. Dave A. Fagan from Pacer Software, Inc. Pai Chou from U. of California, Berkeley Ralph Seguin from U of Michigan EECS Dept., Ann Arbor, MI Thank you very much!. Also I like to thank you everybody if your names are left out. Please accept my apology if this happens. Here is the summary (edited) In comp.sys.mac.programmer I asked: >Is anyone out there knows anything about MacYACC from Abraxas Software >Inc, Oregon? MacYACC is a compiler writing a new language. >It comes with a lexical analyzer and yacc compiler (UNIX compatible). >It works under MPW and LightSpeed C. At least the company claims that >way. Dennis Cohen wrote : I reviewed Abraxas "MacYacc 2.0" product for MacWorld. The review was scheduled for the May issue, but doesn't appear to be in it (they tell me they slipped it to June). It doesn't work under LightSpeed, it is an MPW tool and they ship a completely stripped-down MPW 2.02 Shell with the product. The examples with it are good, but don't think that it is as good as Earle Horton's bison port (which has the benefits of being free and coming with source). (later I asked...) >1. If it is not compatible with LightSpeed, is it possible to run >macyacc under MPW and then use the output files (yy.tab.c, etc) in >LS? Cohen : Yes, it can run under LS. >2. You mentioned about the program called bison, is it compatible >with THINK C 4.0 (plain, no oo modules is fine)? How about the >mpw-flex and mpw-yacc (?) that I can ftp from sumex-aim? Cohen : Yes, bison output is compatible with LSC. There is a LSC implementation of Bison (including source) about to appear for ftp from sumex-aim. >3. Can I use the lex and yacc under an UNIX machine and then use the >output files under THINK C? Cohen : Yes, if you're willing to do a little "massaging".i The manipulation of the large arrays was the sort of thing I was referencing -- I change them to dynamic allocation and wander through the code making the necessary adjustments. The LSC version should be available any day -- the moderators have received it, now they just have to verify it, prepare the abstract, and put it online. ------------------------------------------------------------------- Cohen's Disclaimer: Any opinions expressed above are _MINE_! ------------------------------------------------------------------- Dave A. Fagan wrote ... I've been using MacYacc for about 6 months. My biggest complaint is that it's not Macintosh software, it's hastily ported PC software. It's implemented as an MPW tool & has no interface to speak of. They could have at least put together a Commando interface on it. The manual is the same, all of the examples show the MS-DOS prompts & commands, the only thing they've done to change the manual for Macintosh is a global replace on PCYACC to MACYACC. The good news: it works. It produces a satisfactory parse engine which is all I need. It's diagnostic output is good. It lets you substitute your own parser skeleton - which you'll need to do if you want to handle large objects on the value stack and be friendly to Mac memory mangement. Pai Chou wrote ... I don't know if it's from Abraxas Software, but there is a public domain program by the same name. You should be able to get it from BMUG (Berkeley Macintosh User's Group) I think it's on programmer's disk #3. It is essentially the same as the UNIX yacc except it has the file finding dialog and a simple menu. It runs on 128K Mac's. The only problem with it is that it does not give you reduce-reduce error messages in words, but it puts a special character (generated using option-key) in front of the rule so I guess you can search for this special character. Other than that, it works great -- it successfully generates the parser for my Pascal compiler, which has more than 200 rules. ( --- repeated posting --- ) Ralph Sequin posted ... Well, I don't know about MacYACC, but you can get two PD programs that are equivalent to YACC. Firstly, there is GNU Bison. This can get obtained from just about any FTP site with GNU access (xanth.cs.odu.edu) Next, there is Berkeley YACC. Berkeley YACC is a PD version of YACC. The source is available on mrcnext.cso.uiuc.edu Check under the amiga/fish/ff299 directory. It is in a file called Yacc.zoo I am uncertain if you have Zoo for the Mac, so if you want me to convert it to another archive format, just send me some mail. --- End of Summary --- Note: I got MacYACC from Abraxas and... I am going to return the package back to Abraxas. In my opinion, Berkeley Yacc is much better (not to mention it includes the source codes). The manual from Abraxas is not very clear. The manual says one way and the technical support says the other way, but the support is correct. Who can afford using long distance just to run a sample/ tutorial program??? Can't stand those C>, /.../.../ stuff in the manual. So many READ ME files (almost every folder has one). That's why I bought a mac. Once again, I appreciate all help/answer from the people who answered my posting. Thank you to all of you. Zikif Hamdy -------------------------- ooo000ooo -------------------------------