GSOC 2011

[Luca Boasso]

Sorry for the late reply,
even tough I'm not an ANTLR expert, given my previous experience with the tool
and having read most of the official book, I could help more on this GSOC
idea.

I have looked at http://www.antlr.org/grammar/list and on google and I could not
find any recent ANTLR grammar for D.
In my opinion this project could be a good contribution for the D community
while being practical and useful in the short term.

The current ANTLR v3 is much more powerful than the previous version since it
can handle LL(*) grammars instead of the classic LL(k) one [More on LL(*) at:
http://www.antlr.org/papers/LL-star-PLDI11.pdf].

I had a quick look at the parse.c of dmd compiler. It's a recursive descent
parser but honestly I don't know the grammar enough to understand if
it is LL(k)
but definitively it could be parsed efficiently with the LL(*) parsing strategy.

I think that the essential aspects listed in the idea "ANTLR and Java based D
parser for IDE usage" could be pursued in the given 3 months time-line.

Could somebody tell me something more about the AST that should be generated as
output of the parser?


Luca Boasso


On 3/23/11, Bruno Medeiros <brunodomedeiros+spam at com.gmail> wrote:
> On 23/03/2011 14:59, spir wrote:
>> On 03/23/2011 02:40 PM, Bruno Medeiros wrote:
>>> The first two items (LexingandParsing,
>>> ANTLRandJavabasedDparserforIDEusage) are fairly concrete ideas.
>>> The last one, D Tools in D is far more general, and can involve a lot of
>>> different things
>>
>> It is instead more specific: "The aim is to boost the production of
>> fundamental tools for analysis of D source code in D: lexical,
>> syntactic, at best first steps of semantic analysis; producing code
>> representations as D data structures."
>>
>> Its usefulness /in the long term/ can be said more general in that it
>> would enable *further* creation of D tools in D (*), potentially useful
>> to all D programmers. Unlike eg a general-purpose parsing lib in D which
>> is specific to the parsing domain; on the other hand, a general-purpose
>> parsing lib does not parse only D code ;-).
>>
>> (*) See (again):
>> http://prowiki.org/wiki4d/wiki.cgi?GSOC_2011_Ideas#DtoolsinD
>>
>> Denis
>
> Yeah, I agree that D tools are (generally speaking) much more useful
> than a D parsing library. But that's not the point, the point is that
> with the limited manpower we have available (not just in terms of GSoC,
> but generally), it's very risky and unwise to embark on projects that
> then remain incomplete and never see the light of day as part of
> something useful.
> It's like comparing a bicycle to a car engine. Yeah, the car engine can
> allow to build a nice car but it's worthless on it's own and at least
> the bicycle can get somewhere faster. Even a skate or roller-blades are
> more useful.
> How much development has it even been in any D tools in D lately (in the
> last year or so)?
>
>
> --
> Bruno Medeiros - Software Engineer
>