[antlr-interest] Re: SQL grammar, help on ambig warns

[lgcraymer]

--- In antlr-interest at y..., Ruslan Zasukhin <sunshine at p...> wrote:
> on 8/26/02 7:59, lgcraymer at lgc at m... wrote:
> 
> > Ruslan--
> > 
> > One likely scenario is that somewhere in your grammar you have a 
rule
> > which follows character_string_type with "LPAREN" and ANTLR can't 
tell
> > whether to assign an LPAREN to that or to "character_string_type" 
in
> > the optional subrule.  You may need a syntactic predicate to
> > disambiguate ...
> 
> As I understand using of syntax predicate is THE ONLY way to run 
around
> this?
> 
> I just do all my best to keep grammar k=1 and avoid syntax 
predicates to get

The k value affects ANTLR grammar analysis much more than it affects 
generated code.  Setting k=50, for example, can cause the analysis to 
take "forever" to generate code from an input grammar, but the 
generated code does not change (if k is already large enough to remove 
ambiguities).  In your case, k=2 may be a good choice:  "LPAREN 
unsigned_integer" may not be ambiguous.

> it as fat as possible. I need this SQL parser for real-life DBMS.
> 
> If speak about syntax predicate...
> How it must looks to still have optional branch.
> I think about adding empty alternative.
> Is this correct? 

Yes, in fact that is what ANTLR does internally.

> 
> character_string_type
>      :    ( "character" | "char" )
>              ( "varying" LPAREN! unsigned_integer RPAREN!
>              |     
>                 (( LPAREN! unsigned_integer RPAREN! ) =>
>                      ( LPAREN! unsigned_integer RPAREN! )
>                 | /* nothing */
>                 )
>             )      
> 
> Just for interest. I assume that it is possible redo some grammar 
rules so
> it will be possible to avoid syntax predicate, yes? Because for YACC 
it is
> possible.

Usually, although it can be hard to figure out what changes will work. 
 That's one of the reasons that I'd like to have an ANTLR IDE with 
refactoring browser/editor included ...

--Loring

> 
> 
> >> --------------------------------------------------------------
> >> character_string_type
> >>     :    ( "character" | "char" )
> >>             (    "varying" LPAREN! unsigned_integer RPAREN!
> >>             |     ( LPAREN! unsigned_integer RPAREN! )?     
<<<<<<
> > here
> >>             )
> >>     |    ("varchar" | "string")
> >>             (     LPAREN! unsigned_integer
> >>                      (COMMA! character_string_literal)? RPAREN! )
> >>     ;    
> >> --------------------------------------------------------------
> 
> -- 
> Best regards,
> Ruslan Zasukhin      [ I feel the need...the need for speed ]
> -------------------------------------------------------------
> e-mail: ruslan at p...
> web: http://www.paradigmasoft.com
> 
> To subscribe to the Valentina mail list
> send a letter to valentina-on at l...
> -------------------------------------------------------------


 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/