> Is this also something that can be done?

• Yes
• No
• Depends

How exactly do you want me to do that? Not sure if I can upload something here?

Please see Markup in the Monastery, Writeup Formatting Tips and How do I post a question effectively? Also maybe take a look at Perl Monks Approved HTML tags and What shortcuts can I use for linking to other information? Definitely see How do I compose an effective node title?

As you say your data is plain text, it may be enough to enclose each example in  <p> (paragraph) tags. If the text has some sort of internal structure, use  <code> ... </code> or  <c> ... </c> tags to preserve that structure. (Blocks of text enclosed in code tags will be easily downloadable by the monks and this encourages the monks to work on your problem!)

Please, please don't post 50K of text! Hopefully, each text example will be just a few score or few hundred words; I would say 1000 words max per example. Please try to post examples showing the full range of complexity of your data. Please use  <readmore> tags to reduce data "clutter."

---

Give a man a fish:  <%-{-{-{-<

cd {top_level_directory_where_all_files_are_located}

find * -type f | search_script.pl > hit_list.txt
[download]

#!/usr/bin/perl

use strict;
use warnings;

while (<STDIN>) {
    chomp;
    next unless -f();
    my $text = do {
        if ( open( my $fh, $_ )) {
            local $/;
            <$fh>;
        }
        else {
            warn "Unable to read $_\n";
        }
    };
    if ( $text =~ /\s(recommend\S*\s+to\s+vote\s+\S+)/ ) {
        ( my $hit = $1 ) =~ s/\s+/ /g;
        print "$_: $hit\n";
    }
    else {
        print "$_: NO_MATCH\n";
    }
}
[download]

find * -type f | search_script.pl > hit_list.txt 2> search.errlog
[download]

The regex proposed above will match all the inflections on "recommend" (-ed, -ing, -s, -ation), and will capture the matched phrase only up to the word that follows "vote". (You can extend the capture to include more words before and/or after, if you like, by adding more \s+\S+\s+ elements inside the parens.)

When there's a match, all kinds of white-space between words is allowed, and it's all normalized to a single space before output, to ensure one line of output per file.

This actually really looks like something I can use! Thank you very much!

I will have to add something in the code that would differentiate the different proposals from each other and match the recommendations and proposals, since there could be more proposals per filing, and thus more recommendations. This, however, is a very very nice start for me:D

I suggested the code above several hours before you provided a sample of data ("Filing Example" below), so I was unaware that you were dealing with HTML data. That changes things. For example, in HTML, some "whitespace" looks like this:

... Board recommends a vote FOR Proposal No. 2. ...
[download]

and given the variety of distinct sources (which presumably use distinct HTML/CSS formats and styles), I'd expect a variety of structural differences in the tags that appear in and around the patterns of interest.

BTW, on the matter of "html" vs. "txt", it doesn't matter what a given file name looks like - what matters is what the content looks like. If the content has HTML tags, it's HTML data, and needs to be treated as such, regardless of what the file name might be.

If it's typical for texts of this sort to always include a single table near the top of the document that lists the proposals with number, name, and result, it may be that your best bet is Corion's idea about HTML::TableExtractor. It's just a matter of knowing which table in the overall file is the one you want.

Aside from that, any other practical approach will involve parsing the HTML first to get its plain-text content before you do anything that involves string comparisons or regex matches.