[antlr-interest] How to look only for one rule?
Matthias Gutheil
matthias.gutheil at informatik.uni-mannheim.de
Sun Oct 22 04:11:27 PDT 2006
Many C files one by one...
-- Matthias
Loring Craymer schrieb:
> What are you trying to do, then? For a single file,
> building a custom tool is usually overkill. If you
> just want to do analysis of a single file, take a look
> at Columbus/CAN from
> <http://www.frontendart.com/>--it's a pretty nice tool
> and there is an evaluation period during which you can
> license it for free.
>
> --Loring
>
> --- Matthias Gutheil
> <matthias.gutheil at informatik.uni-mannheim.de> wrote:
>
>
>> Yes but the problem is, that C is context sensitive
>> and I only have one
>> c-file and not all the included stuff.
>> I had a try to change the C grammar, but thats a too
>> hard task for me at
>> the moment.
>>
>> -- Matthias
>>
>>
>> Loring Craymer schrieb:
>>
>>> Hmm. In cases where you have a complete grammar
>>>
>> (and
>>
>>> for C, the ANTLR 2 C grammar stuff done by Monty
>>>
>> and
>>
>>> others is definitely complete), I usually find it
>>> better to take that and recognize everything but
>>>
>> only
>>
>>> process what you need. I would suggest modifying
>>>
>> the
>>
>>> C grammar to build trees that only contain the
>>>
>> methods
>>
>>> of interest and none of the stuff that you want to
>>> ignore. That gets you to a solution fast, and
>>>
>> also
>>
>>> leaves room for extending the range of what you
>>>
>> are
>>
>>> interested in later.
>>>
>>> --Loring
>>>
>>>
>>> --- Terence Parr <parrt at cs.usfca.edu> wrote:
>>>
>>>
>>>
>>>> Hi. See the filter example, which does it all in
>>>> the lexer. Have a
>>>>
>>>> EVERYTHINGELSE : . ;
>>>>
>>>> rule at bottom to skip everything but your rules
>>>>
>> of
>>
>>>> interest.
>>>> Ter
>>>> On Oct 21, 2006, at 4:54 AM, Matthias Gutheil
>>>>
>> wrote:
>>
>>>>
>>>>
>>>>> Hello,
>>>>>
>>>>> I am writing a fuzzy C grammar and I am only
>>>>>
>>>>>
>>>> interesting in methods
>>>>
>>>>
>>>>> with parameters and return values.
>>>>>
>>>>> The method definition is not the problem, but I
>>>>>
>>>>>
>>>> want to skip
>>>>
>>>>
>>>>> anything else.
>>>>> For example something like (method | ~method)
>>>>>
>>>>>
>>>> doesn't work, cause
>>>>
>>>>
>>>>> so the first token af a real method (return type
>>>>>
>>>>>
>>>> or name) goes in
>>>>
>>>>
>>>>> ~method.
>>>>>
>>>>> Some tips?
>>>>>
>>>>> Matthias
>>>>>
>>>>>
>>>>
>>>>
>>> __________________________________________________
>>> Do You Yahoo!?
>>> Tired of spam? Yahoo! Mail has the best spam
>>>
>> protection around
>>
>>> http://mail.yahoo.com
>>>
>>>
>>
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
>
More information about the antlr-interest
mailing list