[pm-h] Question about using HTML::TableExtract

[Robert Boone]

On 5/14/07, Richard Price <raprice at gmail.com> wrote:
> I am an intermediate perl user.  I taught myself Perl by reading "Learning
> Perl," with some online tutorials and I have some other reference texts.  I
> can generally do what I need to with with Perl, but my code is far from
> elegant.  I understand the very basics of object-oriented programming in
> Perl, but I generally need sample code to get started with modules from
> cpan.  I am a professor at Rice University and have found Perl to be
> invaluable for extracting data for my research, especially the regular
> expression capabilities of Perl.  I have been unable to attend any of the
> monthly meetings, but hope to in the future.
>
>  For my current project, I am trying to extract historical financial
> statement data from www.marketwatch.com.  The url is
> http://www.marketwatch.com/tools/quotes/financials.asp?symb=ABSD&sid=0&report=2&freq=0.
> I use WWW::Mechanize to download the webpage and then I use
> HTML::TableExtract to extract the text that I need.  I want to transpose the
> table at depth=1, count=1 after extracting it so that each year is a row and
> each variable is a column.  I have not been able to find any documentation
> on how to extract a column from a table using HTML::TableExtract.
>
>  The following simple program downloads the data using WWW::Mechanize and
> extracts the table with HTML::TableExtract and prints the output of each
> row.
>
>  #!/usr/bin/perl
>
>  use HTML::TableExtract;
>  use WWW::Mechanize;
>  use strict;
>
>  my $marketwatch = WWW::Mechanize->new( autocheck => 1 );
> $marketwatch->get("http://www.marketwatch.com/tools/quotes/financials.asp?symb=ABSD&sid=0&report=2&freq=0
> ");
>
>  chomp(my $html = $marketwatch->content);
>
>  my $table = HTML::TableExtract->new(keep_html=>0, depth =>
> 1, count => 1, br_translate => 0 );
>  $table->parse($html);
>
>  foreach my $row ($table->rows) {
>      print join("\t", @$row), "\n";
>  }
>
>  I am not able to figure out how to use the columns method.  My intuition
> makes me think it should be something like the following (but my intuition
> is wrong):
>
>  foreach my $column ($table->columns) {
>      print join("\t", @$column), "\n";
>  }
>
>  The error message I get says:  Can't locate object method "columns" via
> package "HTML::TableExtract".  The documentation doesn't shed much light
> (for me anyway).  I can see in the code of the module that the columns
> method belongs to HTML::TableExtract::Table, but I can't figure out how to
> use it.
>
>  I appreciate any help.  For an experienced programmer, I am sure this is
> trivial, but I am the closest thing to a programmer in my department, and I
> don't really have anyone around me that I can get help from.
>
> _______________________________________________
> Houston mailing list
> Houston at pm.org
> http://mail.pm.org/mailman/listinfo/houston
> Website: http://houston.pm.org/
>

It looks like you need to call method columns from a
HTML::TableExtract::Table object and not a HTML::TableExtract object.

>From the docs and your email maybe something like this could get you started:

my $table = HTML::TableExtract->new(keep_html=>0, depth => 1, count =>
1, br_translate => 0 );
$table->parse($html);

my $t = $table->table(1,1);

foreach my $row ($t->columns) {
    print join("\t", @$row), "\n";
}