broken regex

[elwood]

I'm trying to delete words in $line that are in the array words, but it appears the regex isn't working. The regex is susposed to check each individual word, and if it exists, deleted it from $line.

  foreach (@words) {
    $line =~ s/([%\d])$_([%\d])//g;
    print "$line\n";
  }

Does anyone have any ideas? Thanks

Edit davorg: Added code tags

[davorg]

(Please read Writeup Formatting Tips and format your posts so they are easier to read)

It's hard (maybe impossible) to know what the correct regex is without knowing the format of the data that you are dealing with.

[elwood]

The data in $line is a simple sentance.
The @words array is a list of individual words, one per entry, to remove. i.e. $line="the cat sat on the mat"
and the @words might contain
"on"
"the"

[hgolden]

Then you probably want your regex to be something more like s/\b$_\b//g which will match word boundaries. Just to make sure that there's nothing more that you need to do than is coming across in your message, what were you aiming to do with the %\d?

Hays

[davorg]

Ok. Now we're getting somewhere. In <code> tags, your code looks like this:

foreach (@words) {
  $line =~ s/([%\d])$_([%\d])//g;
  print "$line\n";
}

So the regex is saying this:

Look for either a percent sign or a digit (which will be captured in $1) followed by the string in $_ followed by another percent sign or a digit (which will be captured in $2)

And if all that is found then it's replaced by an empty string.

So as your data doesn't appear to contain any digits or percent signs, then none of that is ever going to match. I think that what you actually wanted was far simpler.

foreach (@words) {
  $line =~ s/\b$_\b//g;
  print "$line\n";
}

And I think you may want to move the print statement out of the loop - tho' it might be there for debugging purposes.

See perlretut for a good introduction to regexes.

[robartes]

As the man said, we need your data to be able to tell something about this. From looking at your regex (mangled by the fact that it's not in code tags), it seems like it is not looking for a word, but for a word preceded and followed by a % or a digit. Is that what you intend to do?

[elwood]

sorry about the bad formatting, I'll try again


  foreach (@words) {
    $line =~ s/(%\d)$_(%\d)//g;
    print "$line\n";
    }


It was my understanding that (%\d) detected a word boundary (space, begining of line, end of line)

[hgolden]

That's why we're here. \b is the word boundary metacharacter. And thanks for the code tags! :)

Hays

[ikegami]

sorry about the bad formatting, I'll try again

On PerlMonks, put <c>..</c> tags around your code. It's like <pre>..</pre>, but it handles escaping special characters for your (such as &, <, >, [ and ]). It also wraps long lines while offering a link to download the unwrapped version.