Re: Parsing COD text help

I would suggest you use the html source and use HTML::TokeParser to parse it out. The following might be a good place to start. (I saved the html source into the file "APMA.txt")

#!/usr/bin/perl

use HTML::TokeParser;
use strict vars;

my $file = 'APMA.txt';
my $stream = HTML::TokeParser->new($file);
my $tok;

while( $tok = $stream->get_token) {
    if( $tok->[1] eq 'a' && $tok->[2]{'href'} =~ m/course_nbr/ ) {
        print "Class: " . $stream->get_text('/a') . " ";
    }
    if( $tok->[1] eq 'span' && exists($tok->[2]{'class'}) && $tok->[2]
+{'class'} eq "title") {
        print $stream->get_text('/span') . "\n";
    }
    if( $tok->[1] eq 'span' && exists($tok->[2]{'title'}) && !exists($
+tok->[2]{'class'}) ) {
        print "Title = $tok->[2]{'title'}: ";
        print $stream->get_text('/span'), "\n";
    }
}

exit;
[download]

partial output:

Class: APMA 109 
Calculus I
Title = Schedule Number: 92861
Title = Section Number: 0001
Title = Credit Hours: 04.0
Title = Time: 0900-0950
Title = Day:MTWRFS: MWF
Title = Olsson Hall: OLS  011
Title = Instructor: Oberhauser, James P.
Title = Enrollment:Authorized/Actual: 55/2
Title = Grading Method: O
Title = Time: 0830-0920
Title = Day:MTWRFS: T
Title = Olsson Hall: OLS  005
Title = Instructor: 
Title = Schedule Number: 90063
Title = Section Number: 0002
Title = Credit Hours: 04.0
Title = Time: 1000-1050
Title = Day:MTWRFS: MWF
Title = Olsson Hall: OLS  120
Title = Instructor: Beck, Mary
Title = Enrollment:Authorized/Actual: 55/4
Title = Grading Method: O
Title = Time: 0830-0920
Title = Day:MTWRFS: R
Title = Olsson Hall: OLS  120
Title = Instructor: 

...

Class: APMA 111 
Single Variable Calculus
Title = Schedule Number: 93433
Title = Section Number: 0001
Title = Credit Hours: 04.0
Title = Time: 1100-1150
Title = Day:MTWRFS: MWF
Title = Olsson Hall: OLS  011
Title = Instructor: 
Title = Enrollment:Authorized/Actual: 55/55
Title = Grading Method: O
Title = Time: 0830-0920
Title = Day:MTWRFS: T
Title = Instructor: Castiglione, Jason

...
[download]

Further parsing is easily done to get exactly what you want.
Take a look at Perl & LWP for more information.

Hope this helps,
davidj

Comment on Re: Parsing COD text help Select or Download Code

Replies are listed 'Best First'.
Re^2: Parsing COD text help by Aristotle (Chancellor) on Jul 26, 2004 at 02:57 UTC
Of course, you should never say HTML::TokeParser without the ::Simple. `:)` #!/usr/bin/perl use strict; use warnings; use HTML::TokeParser::Simple; my $file = 'APMA.txt'; my $stream = HTML::TokeParser::Simple->new( $file ); while( my $t = $stream->get_token ) { if( $t->is_start_tag( 'a' ) and $t->return_attr( 'href' ) =~ m/course_nbr/ ) { print "Class: ", $stream->get_text( '/a' ), " "; } elsif( $t->is_start_tag( 'span' ) ) { my $class = $t->return_attr( 'class' ); my $title = $t->return_attr( 'title' ); if( defined $title and not defined $class ) { print "Title = $title: ", $stream->get_text( '/span' ), "\ +n", } elsif( defined $class and $class eq 'title' ) { print $stream->get_text( '/span' ), "\n"; } } } [download] Makeshifts last the longest.	[reply] [d/l]

Replies are listed 'Best First'.

Re^2: Parsing COD text help
by Aristotle (Chancellor) on Jul 26, 2004 at 02:57 UTC

HTML::TokeParser

::Simple

:)

#!/usr/bin/perl
use strict;
use warnings;

use HTML::TokeParser::Simple;

my $file = 'APMA.txt';
my $stream = HTML::TokeParser::Simple->new( $file );

while( my $t = $stream->get_token ) {
    if(
        $t->is_start_tag( 'a' )
        and $t->return_attr( 'href' ) =~ m/course_nbr/
    ) {
        print "Class: ", $stream->get_text( '/a' ), " ";
    }
    elsif( $t->is_start_tag( 'span' ) ) {
        my $class = $t->return_attr( 'class' );
        my $title = $t->return_attr( 'title' );

        if( defined $title and not defined $class ) {
            print "Title = $title: ", $stream->get_text( '/span' ), "\
+n",
        }
        elsif( defined $class and $class eq 'title' ) {
            print $stream->get_text( '/span' ), "\n";
        }
    }
}
[download]

Makeshifts last the longest.

[reply]
[d/l]