[EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> Please help!!! I'm trying to complete an assignment to determine the number
> of HTML tags within a string at each level of nesting. ie: for the string
>
> "<p><strong>a</strong><a>b</a></p><strong>c<italic>d<a>e</a></italic></strong>"
>
> <p> is at level 0
> <strong> and <a> are at level 1
> etc...
>
> My code is below. The problem is that apparently my RegEx is wrong (in the
> while loop), and so the loop is infinite, and no values are printed--just "".
> I'm just learning perl, and RegEx is really confusing to me. Any feedback
> would be appreciated!
Okay -- first of all, be patient. You mailed this three times.
> #!usr/bin/perl
> #Exercise 8.9
> #To determine how many HTML tags are included in a string at each level of nesting.
>
> use strict;
> use warnings;
>
> my $string =
>"<p><strong>a</strong><a>b</a></p><strong>c<italic>d<a>e</a></italic></strong>";
>
> print("The HTML string is:\n$string\n\n");
>
> our @counter;
> count($string, 0);
>
> for (0 .. $#counter){
> print("There are $counter[$_] valid level",$_," html tags.\n");
> }
>
> sub count
> {
> my $string = shift();
> my $level = shift();
>
> while ($string =~ m!.*?<([a-z]*?)>(.*)</\1>!g){
The regex is not bad, given that sample input.
There are two bugs and some unnecesary clutter, though:
m!
.*? # This doesn't do anything useful.
# Since there's no anchor at the front, there's
# no need to match the text in between <tags>.
# Just let perl skip them.
<([a-z]*?)> # Here you don't need the "?".
# Since '>' is outside the range [a-z], the star
# has to consume all the letters, and greediness
# can't enter into it.
# But you ought to consider whether you really
# want to accept *zero* or more letters here.
(.*) # And _this_ might be a good place to use the
# nongreedy "*?" quantifier.
</\1>
!xg;
> #should remember html tag as 1 and contents as 2
> $counter[$level]++;
> print("Tag is \"\1\", contains \"\2\" ");
Backreferences (\1, \2, etc) are only valid inside the regex, where
you've used \1 correctly.
Outside the regex, the captured substrings are in the digit variables
$1, $2 ... $n.
> print("at level $level\n");
> count($string, $level+1);
>
> #recursively calling count with the string and the level increased by one
Here you recursively process the original $string, which matches at the
same place and recurses again on the original $string, which matches at
the same place and...
You want to recurse on "the bit in the middle", which was captured in the
second set of parens.
> }
> }
>
Since it's homework, I'll leave you to it.
But I really hope your instructor has explained that the only time
you should use regexes on HTML is in the context of an exercise.
In real-life, go directly to HTML::Parser.
In fact, I'm probably assuming too much and you should go get
HTML::Parser *right now* :)
--
Steve
perldoc -qa.j | perl -lpe '($_)=m("(.*)")'
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
