spaces removed in backreference

Allasso has asked for the wisdom of the Perl Monks concerning the following question:

I have this routine that takes all the spaces in between a string of adjacent html tags. Only problem is it also takes the spaces out of the tags themselves. I cannot understand why this is happening.

eg: <span class="c"> becomes <spanclass="c">

        $newtoken = "";
        while ("$token" ne "$newtoken") {
            $newtoken = "$token";
            $token =~ s@((?:<[^>]*>)*) +((?:<[^>]*>)*)@\1\2@g;
        }
[download]

If I put a "dot" metacharacter as below, it fixes it, but I don't understand why. If I am back-referencing the whole tag, why would anything inside the back-reference be affected?

        $newtoken = "";
        while ("$token" ne "$newtoken") {
            $newtoken = "$token";
            $token =~ s@((?:<[^>]*>).*) +((?:<[^>]*>).*)@\1\2@g;
        }
[download]

Comment on spaces removed in backreference Select or Download Code

Replies are listed 'Best First'.
Re: spaces removed in backreference by graff (Chancellor) on Apr 25, 2010 at 05:10 UTC
There is probably an easy way to do this with a real parser (and it's not easy to imagine a situation where it makes sense to worry about spaces between html tags). But putting all that aside, I wonder why you wouldn't just do something like this, using look-behind and look-ahead assertions: `s/(?<=>) +(?=<)//g;` [download] Or like this, using captures (possibly a little bit "less efficient", but not enough to worry about it): `s/(>) +(<)/$1$2/g;` [download] (No need to "loop until done" -- the "g" modifier takes care of the whole string.) I also wonder why you use backslashes instead of dollar signs for your captures in the replacement part. If you had `use warnings;` in your code, you would have been told: "\1 better written as $1..." (and I suspect there's a good reason why, but it escapes me at the moment).	[reply] [d/l] [select]
Re^2: spaces removed in backreference by Allasso (Monk) on Apr 25, 2010 at 09:49 UTC
`s/(?<=>) +(?=<)//g; s/(>) +(<)/$1$2/g;` [download] Don't know why I never thought of these... Thanks for the tip on use warnings	[reply] [d/l]
Re: spaces removed in backreference by ig (Vicar) on Apr 25, 2010 at 03:10 UTC
`((?:<[^>]>))` can match nothing (because of the * quantifier), so your original expression matches every sting of one or more space characters. When you add the . between the `(?:<[^>]>)` and the ``, then there must be a tag, which can be followed by anything. A better solution might be `((?:<[^>]*>)+)`	[reply] [d/l] [select]
Re^2: spaces removed in backreference by AnomalousMonk (Archbishop) on Apr 25, 2010 at 04:43 UTC
A better solution might be `((?:<[^>]>)+)`* `(?:<[^>]>)+` or even just plain `(?:<[^>]>)` allows a pair of tags with no intervening blanks to match or to be excluded from matching, respectively, thereby potentially destroying open-tag/close-tag synchronization: in either case, the "pair" of tags between which blanks are eliminated is incorrect. This can be remedied by changing the quantifier on `$blank`, but, as another reply has suggested, a proper HTML parser is really the best approach. `>perl -wMstrict -le "my $s = '<foo a=b></foo> <bar c=d> </bar> '; my $t; ($t = $s) =~ s@((?:<[^>]>)+) +((?:<[^>]>)+)@$1$2@g; print qq{'$t'}; my $tag = qr{ < [^>]* > }xms; my $blank = qr{ [ ] }xms; ($t = $s) =~ s{ ($tag) $blank+ ($tag) }{$1$2}xmsg; print qq{'$t'}; ($t = $s) =~ s{ ($tag) $blank* ($tag) }{$1$2}xmsg; print qq{'$t'}; print qq{'$s'}; " '<foo a=b></foo><bar c=d> </bar> ' '<foo a=b></foo><bar c=d> </bar> ' '<foo a=b></foo> <bar c=d></bar> ' '<foo a=b></foo> <bar c=d> </bar> '` [download] Update: Added another example using `*` quantifier.	[reply] [d/l] [select]
Re: spaces removed in backreference by Anonymous Monk on Apr 25, 2010 at 03:16 UTC
You have a zero-length match (in $1), followed by space, followed byzero-length match ($2) `use re 'debug';` to see exactly what your regex matches use a real HTML parser, like HTML::TreeBuilder or YAPE::HTML	[reply] [d/l]
Re: spaces removed in backreference by Allasso (Monk) on Apr 25, 2010 at 14:47 UTC
FWIW, I am actually not parsing HTML, I am generating it from RTF docs. I wanted a simple rtf2html converter that didn't mark up the page with every possible style anyone can think of (which every converter I have tried does), and just give the basic p, blockquote, i, b, u, and colored text spans. It does 90% of the tedious work, and is much easier than trying to fix the scary html that the other converters spit out. (ps, I like my html pages to look good and readable, and I don't like spaces where they aren't needed.)	[reply]
Re: spaces removed in backreference by Allasso (Monk) on Apr 25, 2010 at 09:46 UTC
thanks for all the input, really helpful information. Very much appreciated. This mainly a learning experience for me, so is why I haven't thought about a "real" parser. Good suggestion though, it would probably be helpful.	[reply]