Re: Compiler implementation language preference ?

On 2018-12-20, Martin Ward <martin@gkc.org.uk> wrote:
> On 19/12/18 19:54, Aaron Gray wrote:
>> Pity there are no real compiler-compilers anymore, hint-hint, I am
>> working on one to rule them all ;)
>>
>> [Please don't say you've invented another UNCOL. -John]
>
> I may be wrong but read Aaron's message to mean that he is
> working on a domain specific language for writing compilers,
> in the style of Language Oriented Programming,
> rather than a universal intermediate program representation language,
> which is what UNCOL attempted to be.
>
> The original paper on Language Oriented Programming:
>
> http://www.gkc.org.uk/martin/papers/middle-out-t.pdf


From that:


>> "In the case of the FermaT tool, the lowest level translator and
>> support library consists of 2–3,000 lines of LISP code. This
>> translates from low-level META WSL to LISP, all the rest of the system
>> is written in META WSL. To port the system to a new version of LISP, or
>> even to a new base language such as C, only requires rewriting the
>> lowest level translator: and this is a comparatively small task–in fact,
>> the first version of the translator was wr itten in less than three man
>> days. The FermaT system is currently being ported from a Unix environm
>> ent to a PC environment, using C rather than LISP as the implementation
>> language."


That language hopping aspect reminds me of MAL (Make A Lisp): a project
which bootstraps the same Lisp dialect (its own) in numerous different
languages (currently 74):


https://github.com/kanaka/mal

Return to the comp.compilers page.
Search the comp.compilers archives again.