Except for the XS part, the COBOL to Perl "translator" would Perl.

On Wednesday, May 7, 2014 5:06:08 PM UTC-5, Michael Roberts wrote:
>
> So? You'll get Perl out no matter what you write the hypothetical 
> translator in. Maybe I'm missing the point.
>
> On Thu, 08 May 2014 00:04:25 +0200, clueless newbie 
> <[email protected]<javascript:>> 
> wrote:
>
> Plus all but the XS is Perl.
>
> On Wednesday, May 7, 2014 4:57:08 PM UTC-5, Michael Roberts wrote:
>>
>> You could definitely steal their code generation as a start, though. (I 
>> have to admit Marpa would be much cooler.)
>>
>> On Wed, 07 May 2014 23:55:41 +0200, clueless newbie <[email protected]> 
>> wrote:
>>
>> I think the Marpa approach has more appeal than bootstrapping from 
>> open-cobol. 
>>
>>
>> On Wednesday, May 7, 2014 4:41:50 PM UTC-5, Michael Roberts wrote:
>>>
>>> Here you go: 
>>> http://sourceforge.net/projects/open-cobol/<http://www.google.com/url?q=http%3A%2F%2Fsourceforge.net%2Fprojects%2Fopen-cobol%2F&sa=D&sntz=1&usg=AFQjCNG9VFPZlfAdI3h5LP-3zk993L-aYw>
>>>  GnuCOBOL actually works by translating COBOL into intermediate C. It would 
>>> be relatively easy to convert that into intermediate Perl instead of C. 
>>> You're done. Sure, it would be crappy Perl, but it would be Perl.
>>>
>>> On Wed, 07 May 2014 23:36:39 +0200, clueless newbie <[email protected]> 
>>> wrote:
>>>
>>> I can only judge management's goals by management's selection of 
>>> resources it is willing to dedicate to the project.
>>>
>>> Few Perl programmers with only a year under their belt can code stuff 
>>> worthy of a Burke or a Conway.
>>>
>>>
>>> On Wednesday, May 7, 2014 4:14:25 PM UTC-5, Aristotle Pagaltzis wrote:
>>>>
>>>> * clueless newbie <[email protected]> [2014-05-07 22:40]: 
>>>> > In earlier days the same thing was said of compilers. That a good 
>>>> > assembler programmer could always outcode the compiler. NOTE the 
>>>> > keyword good! The problem is lack of good programmers! 
>>>>
>>>> I’m afraid you picked the wrong metaphor. 
>>>>
>>>> 1. The goals are fundamentally different. 
>>>>
>>>>    Nobody uses a compiler to do a one-time switch from a C codebase to 
>>>>    an assembler codebase so that they can hire assembler programmers to 
>>>>    maintain it. In fact no one even cares what the output of the 
>>>>    compiler looks like specifically, just that it accurately reflects 
>>>>    the semantics of the code in the source language and runs as fast as 
>>>>    possible. 
>>>>
>>>>    So what is asked of a compiler is much less demanding than the kind 
>>>>    of project you have been tasked with. 
>>>>
>>>> 2. Much more fundamentally, assembler offers fewer abstractions than 
>>>> any 
>>>>    non-esoteric language. In fact, every real compiler always 
>>>> translates 
>>>>    in the direction of fewer abstractions. 
>>>>
>>>>    But Perl has a lot *more* abstractions than COBOL. 
>>>>
>>>>    You are really trying to go in the other direction – i.e. to write 
>>>>    a decompiler, essentially. Those do exist, but they are very 
>>>> limited. 
>>>>    All the useful ones work only because they assume they are looking 
>>>> at 
>>>>    the output of some (quite particular) compiler, and they translate 
>>>>    that back to the original language. There is no decompiler that can 
>>>>    take arbitrary assembler and spit readable and maintainable C out 
>>>> the 
>>>>    other end. 
>>>>
>>>> If the ultimate goal of your project really is just accurate mechanical 
>>>> translation, of the sort a compiler does, then yes absolutely, that can 
>>>> be achieved. 
>>>>
>>>> If the goal is to get readable maintainable Perl out of the exercise, … 
>>>>
>>>> Regards, 
>>>> -- 
>>>> Aristotle Pagaltzis // <http://plasmasturm.org/> 
>>>>
>>> -- 
>>> You received this message because you are subscribed to a topic in the 
>>> Google Groups "marpa parser" group.
>>> To unsubscribe from this topic, visit 
>>> https://groups.google.com/d/topic/marpa-parser/OBh6tMOCaPE/unsubscribe.
>>> To unsubscribe from this group and all its topics, send an email to 
>>> [email protected].
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>>
>>>
>>>
>>> -- 
>> You received this message because you are subscribed to a topic in the 
>> Google Groups "marpa parser" group.
>> To unsubscribe from this topic, visit 
>> https://groups.google.com/d/topic/marpa-parser/OBh6tMOCaPE/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to 
>> [email protected].
>> For more options, visit https://groups.google.com/d/optout.
>>
>>
>>
>>
>> -- 
> You received this message because you are subscribed to a topic in the 
> Google Groups "marpa parser" group.
> To unsubscribe from this topic, visit 
> https://groups.google.com/d/topic/marpa-parser/OBh6tMOCaPE/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to 
> [email protected] <javascript:>.
> For more options, visit https://groups.google.com/d/optout.
>
>
>
>
>

-- 
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.

Reply via email to