«

Hi,

I am new to Parse::RecDescent and I came across this problem of the
parser not being able to ignore the newline character. Is there a way
I can ask the parser to ignore all newlines?

Regards
John


### parser_test.pl ###
eval '(exit $?0)' && eval 'exec perl -w -S $0 ${1+"$@"}' && eval 'exec
perl -w -S $0 $argv:q' if 0;
#
use strict;
use diagnostics;
use Parse::RecDescent;

my $grammar = q {

start: identifier(s)

identifier : /\S+/
{ print $item[0]."\n"; }
};
my $parser = Parse::RecDescent->new($grammar);

open(IN,"data.txt") or die "Cannot open data.txt";
# slurp all the lines
my @lines = <IN>;
defined $parser->start(@lines) or die "Didn't match anything";
### end of parser

# data.txt
Head node1
Tail node2

On 2010-04-24 01:51, <> wrote:
> I am new to Parse::RecDescent and I came across this problem of the
> parser not being able to ignore the newline character. Is there a way
> I can ask the parser to ignore all newlines?
[...]
> open(IN,"data.txt") or die "Cannot open data.txt";
> # slurp all the lines
> my @lines = <IN>;
> defined $parser->start(@lines) or die "Didn't match anything";

You are calling $parser->start with two arguments here: "Head node1\n"
and "Tail node2\n".

AFAICS this isn't allowed. (But the error message is strange - I guess
Parse::RecDescent uses a second parameter for internal and undocumented
purposes.

So I think your question is really: How can I read in an entire file all
at once?

This is a FAQ. read

perldoc -q 'entire file'

hp

Thanks for the pointer to the perldoc section! I tried this suggestion
in that section and I did not get the newline problem:

$var = do { local $/; <INPUT> };

My updated prog is:
my $grammar = q {

start: identifier(s)

identifier : /\S+/
{ print $item[1]."\n"; }
};
my $parser = Parse::RecDescent->new($grammar);

open(IN,"data.txt") or die "Cannot open data.txt";
# slurp all the lines
my $var = do { local $/; <IN> };
#my @lines = <IN>;
defined $parser->start($var) or die "Didn't match anything";

It now prints:
Head
node1
Tail
node2

Like I wanted it to print.

Regards
John

>>>>> "fc" == freesoft12@gmail com <> writes:

fc> Thanks for the pointer to the perldoc section! I tried this suggestion
fc> in that section and I did not get the newline problem:

fc> $var = do { local $/; <INPUT> };

use File::Slurp. cleaner and faster. especially since you are using a
slower parser.

uri

--
Uri Guttman ------ -------- http://www.sysarch.com --
----- Perl Code Review , Architecture, Development, Training, Support ------
--------- Gourmet Hot Cocoa Mix ---- http://bestfriendscocoa.com ---------

On Apr 24, 8:53*am, "freesof...@gmail.com" <freesof...@gmail.com>
wrote:
> Thanks for the pointer to the perldoc section! I tried this suggestion
> in that section and I did not get the newline problem:
>
> $var = do { local $/; <INPUT> };
>
> My updated prog is:
> my $grammar = q {
>
> start: identifier(s)
>
> identifier : /\S+/
> * * * * * * { print $item[1]."\n"; }};

>
> my $parser = Parse::RecDescent->new($grammar);
>
> open(IN,"data.txt") or die "Cannot open data.txt";
> # slurp all the lines
> my $var = do { local $/; <IN> };
> #my @lines = <IN>;
> defined $parser->start($var) or die "Didn't match anything";
>
> It now prints:
> Head
> node1
> Tail
> node2
>

And you don't really need to slurp:

while (<IN>)
{
defined $parser->start($_)
or die "Didn't match anything";
}

--
Charles DeRykus

On 2010-04-24 18:45, C.DeRykus <> wrote:
> On Apr 24, 8:53*am, "freesof...@gmail.com" <freesof...@gmail.com>
> wrote:
>> Thanks for the pointer to the perldoc section! I tried this suggestion
>> in that section and I did not get the newline problem:
[...]
>> my $parser = Parse::RecDescent->new($grammar);
>>
>> open(IN,"data.txt") or die "Cannot open data.txt";
>> my $var = do { local $/; <IN> };
>> defined $parser->start($var) or die "Didn't match anything";
[...]
> And you don't really need to slurp:
>
> while (<IN>)
> {
> defined $parser->start($_)
> or die "Didn't match anything";
> }

That parses every line separately which is in general not the same as
parsing a whole file.

hp

On Apr 24, 12:04*pm, "Peter J. Holzer" <hjp-usen...@hjp.at> wrote:
> On 2010-04-24 18:45, C.DeRykus <dery...@gmail.com> wrote:
>
> [...]
> > And you don't really need to slurp:
>
> > while (<IN>)
> > {
> > * * * * defined $parser->start($_)
> > * * * * * * * or die "Didn't match anything";
> > }
>
> That parses every line separately which is in general not the same as
> parsing a whole file.
>

Right, a simple example like the OP's, doesn't but even
a slightly more complex grammar could. For instance, if
the grammar required a blank line too:


start: identifier(s) blank

identifier : /\S+/
{ print $item[1],"\n"; }

blank: /^$/
{ print "blank...\n"; }


Then, you'd need to slurp... I think.

--
Charles DeRykus

I can't slurp as my orig grammar has a hierarchy to it (meaning that
the identifiers belong to a group and that group belongs to another
etc). I only used a small example to post. I will have to slurp the
whole file rather than passing in every line.

Regards
John

Thanks for the tip! I will try it out.

Regards
John

>>>>> "fc" == freesoft12@gmail com <> writes:

fc> I can't slurp as my orig grammar has a hierarchy to it (meaning that
fc> the identifiers belong to a group and that group belongs to another
fc> etc). I only used a small example to post. I will have to slurp the
fc> whole file rather than passing in every line.

that makes no sense as that is what slurping is - reading in the whole
file at one time into an array or scalar. sometimes parsing is much
easier and faster when the entire file is in a scalar. since P::RD can't
parse incrementally it does better when you slurp. and yes, slurping is
the technical term! check out File::Slurp for more.

uri

--
Uri Guttman ------ -------- http://www.sysarch.com --
----- Perl Code Review , Architecture, Development, Training, Support ------
--------- Gourmet Hot Cocoa Mix ---- http://bestfriendscocoa.com ---------