[stringtemplate-interest] Translation using Recursive Templates

[Jones, Larry]

Oops. I think user error has betrayed me once again.

The following template does what I want (when I don't try to get the
first of the first of the rest)
max(args) ::= <<
<if(first(rest(rest(args))))>
MAX(<first(args)>, <max(rest(args))>)
<else>
MAX(<first(args)>, <first(rest(args))>)
<endif>
>>

Sorry for the noise.

-----Original Message-----
From: stringtemplate-interest-bounces at antlr.org
[mailto:stringtemplate-interest-bounces at antlr.org] On Behalf Of Jones,
Larry
Sent: Friday, August 29, 2008 3:07 PM
To: Terence Parr
Cc: stringtemplate-interest at antlr.org
Subject: Re: [stringtemplate-interest] Translation using Recursive
Templates


Ter,

I thought I had my problem licked, but I either:
- Iterate over all the children producing a linear list.
- Recurse one too many times.

For example, my source language has expressions like max(1, 2, 3, 4, 5).
I want to translate this to max(1, max(2, max(3, max(4, 5)))) because my
target language only takes two arguments for max().

I currently use the following template. A and B are a debugging device
to indicate which branch the template engine takes. The conditional is
attempting to determine if I have two arguments left.
max(args) ::= <<
<if(first(first(first(rest(args)))))>
A MAX(<first(args)>, <max(rest(args))>)
<else>
B MAX(<first(args)>, <first(rest(args))>)
<endif>
>>

The result of rendering the template is: 
a = A MAX(1, A MAX(2, A MAX(3, A MAX(4, B MAX(5, )))))

Notice that the "B" branch is taken when rest(args) is empty.

-----Original Message-----
From: Terence Parr [mailto:parrt at cs.usfca.edu] 
Sent: Friday, August 29, 2008 12:34 PM
To: Jones, Larry
Cc: stringtemplate-interest at antlr.org
Subject: Re: [stringtemplate-interest] Translation using Recursive
Templates



On Aug 27, 2008, at 7:58 PM, Jones, Larry wrote:

> I am translating one language to another. In the source language, I 
> have expressions such as (SRC A B C D) where SRC is a function ID
> and A..D are other expressions. This expression must translate to an  
> expression like (TGT A, (TGT B, (TGT C, D))) or (TGT(TGT(TGT A, B),  
> C, D))). (The value of the expression is independent of the order of  
> its arguments.) I have already parsed the source and created an AST.
>
> Because the output grammar is recursive, I attempted to recursively 
> invoke a template; however, I cannot determine how to terminate the
> recursion correctly. For example: tgt(args) ::=  
> <<tgt_fun(<first(args)>, <rest(args):tgt()>)>> produces incorrect  
> results.
>
> Any thoughts on the template I need to produce this translation?

Hi Larry,

Generally recursion in ST will terminate when there's no work to do.  
For example,

walk(t) ::= <<
<t.payload>
<t.children:{c | <c:walk()>}>
 >>

Here, if there are no children it does not recurse.
Ter

This e-mail and any attachments are intended only for use by the
addressee(s) named herein and may contain legally privileged and/or
confidential information. If you are not the intended recipient of this
e-mail, you are hereby notified any dissemination, distribution or
copying of this email, and any attachments thereto, is strictly
prohibited. If you receive this email in error please immediately notify
the sender and permanently delete the original copy and any copy of any
e-mail, and any printout thereof.
_______________________________________________
stringtemplate-interest mailing list stringtemplate-interest at antlr.org
http://www.antlr.org:8080/mailman/listinfo/stringtemplate-interest

This e-mail and any attachments are intended only for use by the addressee(s) named herein and may contain legally privileged and/or confidential information. If you are not the intended recipient of this e-mail, you are hereby notified any dissemination, distribution or copying of this email, and any attachments thereto, is strictly prohibited. If you receive this email in error please immediately notify the sender and permanently delete the original copy and any copy of any e-mail, and any printout thereof.