pyparsing-users

[Pyparsing] setResultsName: request for clarification

[stefaan.himpe <ste...@gm...>]

Hello,

I have difficulties in understanding how to use setResultsName 
correctly. The problem is illustrated in the code below:
In the first version, r.BODY is empty. In the second version, it 
contains what I expected it to contain. But I am not sure what the 
difference is between these two versions?

Thanks for any insights.

Best regards,
Stefaan.

---

import pyparsing as p

s = """
SECTION
{
   // {
   if (1) { dosomething; } else { if (0) then { a; } else { b } }
   printf ( "{" ); /* {
   {
   {
   */
   { nog; iets; }
}
"""

# VERSION 1
section_body = p.Combine(p.nestedExpr("{","}", 
ignoreExpr=p.quotedString|p.cppStyleComment|p.cStyleComment))
section_body.setParseAction(p.keepOriginalText)
section_body.setResultsName("BODY")
grammar = p.CaselessKeyword("SECTION").suppress() + section_body
r = grammar.parseString(s)
print r.BODY


# VERSION 2
section_body = p.Combine(p.nestedExpr("{","}", 
ignoreExpr=p.quotedString|p.cppStyleComment|p.cStyleComment))
section_body.setParseAction(p.keepOriginalText)
grammar = p.CaselessKeyword("SECTION").suppress() + \ 
section_body.setResultsName("BODY")
r = grammar.parseString(s)
print r.BODY

Re: [Pyparsing] setResultsName: request for clarification

[Ralph C. <ra...@in...>]

Hi Stefaan,

> I have difficulties in understanding how to use setResultsName
> correctly. The problem is illustrated in the code below: In the first
> version, r.BODY is empty. In the second version, it contains what I
> expected it to contain. But I am not sure what the difference is
> between these two versions?
> 
> # VERSION 1
> section_body = p.Combine(p.nestedExpr("{","}", 
> ignoreExpr=p.quotedString|p.cppStyleComment|p.cStyleComment))
> section_body.setParseAction(p.keepOriginalText)
> section_body.setResultsName("BODY")
> grammar = p.CaselessKeyword("SECTION").suppress() + section_body
> r = grammar.parseString(s)
> print r.BODY
> 
> # VERSION 2
> section_body = p.Combine(p.nestedExpr("{","}", 
> ignoreExpr=p.quotedString|p.cppStyleComment|p.cStyleComment))
> section_body.setParseAction(p.keepOriginalText)
> grammar = p.CaselessKeyword("SECTION").suppress() + \ 
>     section_body.setResultsName("BODY")
> r = grammar.parseString(s)
> print r.BODY

I would guess the intention is you can do

    address = ...
    command = address.setResultsName('from') + \
        address.setResultsName('to')

So your `version 1' is having little effect.

Cheers,


Ralph.

Re: [Pyparsing] setResultsName: request for clarification

[Stefaan H. <ste...@gm...>]

Hello Ralph,
> I would guess the intention is you can do
>
>     address = ...
>     command = address.setResultsName('from') + \
>         address.setResultsName('to')
>
> So your `version 1' is having little effect.
Thank you!
I hadn't looked at it this way.

What to me seemed like totally
inconsistent behaviour makes kind of sense now.

As I understand it now, giving a resultsname only
has an effect when using a parser as opposed to when defining one.

Thanks!
Stefaan.

Re: [Pyparsing] setResultsName: request for clarification

[Eike W. <eik...@gm...>]

On Thursday 01 May 2008 17:54, stefaan.himpe wrote:
> Hello,
>
> I have difficulties in understanding how to use setResultsName
> correctly. The problem is illustrated in the code below:
> In the first version, r.BODY is empty. In the second version, it
> contains what I expected it to contain. But I am not sure what the
> difference is between these two versions?

setResultsName creates a copy of the object. The original is left 
unchanged. 

>
> Thanks for any insights.
>
> Best regards,
> Stefaan.
>
> ---
>
> import pyparsing as p
>
> s = """
> SECTION
> {
>    // {
>    if (1) { dosomething; } else { if (0) then { a; } else { b } }
>    printf ( "{" ); /* {
>    {
>    {
>    */
>    { nog; iets; }
> }
> """
>
> # VERSION 1
> section_body = p.Combine(p.nestedExpr("{","}",
> ignoreExpr=p.quotedString|p.cppStyleComment|p.cStyleComment))
> section_body.setParseAction(p.keepOriginalText)

This statement has no effect. The parser with the name attached is 
forgotten.
> section_body.setResultsName("BODY")

> grammar = p.CaselessKeyword("SECTION").suppress() + section_body
> r = grammar.parseString(s)
> print r.BODY
>
>
> # VERSION 2
> section_body = p.Combine(p.nestedExpr("{","}",
> ignoreExpr=p.quotedString|p.cppStyleComment|p.cStyleComment))
> section_body.setParseAction(p.keepOriginalText)
> grammar = p.CaselessKeyword("SECTION").suppress() + \
> section_body.setResultsName("BODY")
> r = grammar.parseString(s)
> print r.BODY
>
Kind regards,
Eike.

Re: [Pyparsing] setResultsName: request for clarification

[stefaan.himpe <ste...@gm...>]

> setResultsName creates a copy of the object. The original is left 
> unchanged. 

Hello Eike,

Thanks a lot! Now I completely understand.

Best regards,
Stefaan.