Regular expression to replace text NOT within HTML tags: php, regular expressions, regex

by: TerryAtOpusPosted on 2009-11-19 at 12:47:05ID: 25864743

Something like this perhaps:

$string = 'This is a <a href="hyperlink.html">hyperlink</a>.'; 
$pattern = '@(<a href[^>]*>)[^<]*(</a>)@'; 
$replacement = '$1link$2'; 
$new_string = preg_replace($pattern, $replacement, $string); 
print $new_string; 
############################################################
#Result is: 
This is a <a href="hyperlink.html">link</a>.

1:
2:
3:
4:
5:
6:
7:
8:
9:
10:
11:
12:
13:
14:

Select all Open in new window

by: bounsyPosted on 2009-11-19 at 14:42:04ID: 25865963

In a general sense, keep in mind that text inside of a tag is always between "<" and ">" (with no "<" or ">" in between) while text outside of a tag is always between ">" and "<" (with no "<" or ">" in between).

Note that the above solution does not look at the content of the link text, it relaces any link text with the work link.

$string = 'This is a <a href="hyperlink.html">hyperlink</a>. And so is <a href="hyperlink.html">this one</a>.'; 
$pattern = '@(<a [^>]*>)hyperlink(<)@'; 
$replacement = '$1link$2'; 
$new_string = preg_replace($pattern, $replacement, $string); 
print $new_string; 
############################################################
#Result is: 
This is a <a href="hyperlink.html">link</a>. And so is <a href="hyperlink.html">this one</a>.

1:
2:
3:
4:
5:
6:
7:
8:

Select all Open in new window

by: ftsaoPosted on 2009-11-20 at 05:59:40ID: 25870240

@bounsy

The example I posted was a little too simple (sorry) as it's not working with the actual HTML code I am trying to search/replace.

The code I have is like the following where we have to account for the word "hyperlink" not occuring within a "<" and ">" boundary and also that the acutal link text may not be ">hyperlink</a>" but ">this is hyperlink text</a>."

The following string does not currently get replaced properly:

$string = 'This is a hyperlink demo to <a href="hyperlink.html">test hyperlink text</a>. And so is <a href="hyperlink.html">this one</a>. We also need to account for hyperlink text that is not within an html tag boundary though. Like this fake hyperlink.';

by: ftsaoPosted on 2009-11-20 at 06:00:43ID: 25870250

Correction:

within a "<" and ">" boundary

should be

within a ">" and "<" boundary

by: TerryAtOpusPosted on 2009-11-22 at 12:20:34ID: 25883493

I think I get it. Is this what you want?

$string = 'This is a hyperlink demo to <a href="hyperlink.html">test hyperlink text</a>. And so is <a href="hyperlink.html">this one</a>. We also need to account for hyperlink text that is not within an html tag boundary though. Like this fake hyperlink.'; 
$pattern = '@(?![^<>]*>)hyperlink@'; 
$replacement = 'link'; 
$new_string = preg_replace($pattern, $replacement, $string); 
print $new_string; 
############################################################
#Result is: 
This is a link demo to <a href="hyperlink.html">test link text</a>. And so is <a href="hyperlink.html">this one</a>. We also need to account for link text that is not within an html tag boundary though. Like this fake link.

1:
2:
3:
4:
5:
6:
7:
8:
9:
10:
11:
12:
13:
14:

Select all Open in new window

by: TerryAtOpusPosted on 2009-11-22 at 12:23:35ID: 25883505

And if you don't want to replace "hyperlink" when it's part of a larger word (eg phyperlink or hyperlinks), then use this:

$pattern = '@(?![^<>]*>)\bhyperlink\b@';

by: ftsaoPosted on 2009-11-23 at 06:00:55ID: 31657566

Thanks so much TerryAtOpus for sticking through with me :). Your solution worked great.

I just recently started switching from the ereg_replace() function in PHP to preg_replace() since the former is deprecated now. I'm not sure if ereg_replace would have also been able to accomplish the same thing since it's a POSIX regexp vs PCRE regexp.

In any case, I'm being exposed to more advanced regexps.

Is my following explanation of your solution correct in deciphering what the REGEXP is doing?

$pattern = '@(?![^<>]*>)hyperlink@';

We "look ahead" for a pattern (?!<pattern>) matching zero or more occurences of a starting "<" and ending ">" which would match a starting or ending HTML tag. If a match is made, then there will be no hit.

Next we check for a the closing part of a tag ">" followed by our word. This would account for text following after a final HTML tag.

Actually, I'm wondering if we can edit your REGEXP...

FROM:
$pattern = '@(?![^<>]*>)hyperlink@';

TOP:
$pattern = '@(?![^<]*>)hyperlink@';

Is it necessary to include the right angle bracket inside of [^<>]?

by: TerryAtOpusPosted on 2009-11-23 at 13:51:17ID: 25891948

Have copied and pasted your private comment for others' benefit:

> Is my following explanation of your solution correct in deciphering what the REGEXP is doing?
>
> $pattern = '@(?![^<>]*>)hyperlink@';
>
> We "look ahead" for a pattern (?!<pattern>) matching zero or more occurences of a starting "<" and ending ">"
> which would match a starting or ending HTML tag. If a match is made, then there will be no hit.
>
> Next we check for a the closing part of a tag ">" followed by our word. This would account for text following after a final HTML tag.
>
>
> Actually, I'm wondering if we can edit your REGEXP...
>
> FROM:
> $pattern = '@(?![^<>]*>)hyperlink@';
>
> TOP:
> $pattern = '@(?![^<]*>)hyperlink@';
>
> Is it necessary to include the right angle bracket inside of [^<>]?

Well spotted - the pattern @(?![^<]*>)hyperlink@ works just fine.

I think you've got something wrong in your interpretation though - where the ^ character is the first character within square brackets, it acts as a NOT for the characters listed in brackets. ie:
[^<>]
means match one character that is NOT a < or > character.

ie with the pattern
(?![^<]*>)
we are saying lookahead and don't match if we find a > character before a < character. ie if we're within a tag.

I am still learning the potential of lookaheads myself - they can be tough to get your head around! One key limitation you might be interested to know is that you can't use * or + as wildcards in a lookbehind - they have to be a fixed length.

Question

Regular expression to replace text NOT within HTML tags

Subscribe now for full access to Experts Exchange and get

Instant Access to this Solution

php

regular expressions

regex

Regular Expressions

Miscellaneous Web Development

Trusted by hundreds of thousands everyday for fast, accurate and reliable tech support.

See what Experts Exchange can do for you.

Got a question?

We've got the answer.

Need individual assistance?

Our experts are ready to help.

Want to learn from the best?

Read articles from industry experts.

Working on a long term project?

Store your work and research.

What Makes Experts Exchange Unique?

Trusted by the world's most respected brands.

Faithfully serving IT professionals since 1996.

Related Solutions

Free Tech Articles

Cloud Class Webinars

Join the Community

Give a Little. Get a Lot.

Answers

3 Ways to Join

The Experts

The Experts

Testimonials

Testimonials

Testimonials

Business Clients

Business Clients

In the Press

In the Press

In the Press