Regexp question

Textblock (indented for clarity):

Hello, my nickname is Sandman, here is
a list of things I like:

<ol>
<li> Perl
<li> PHP
<li> MySQL
</ol>

I want to replace newlines with "<br />\n" in this textblock, for displaying it
to a browser - but I don't want to replace it inside some specified html-tags -
in this case "ol", but also "ul", "pre" and "xmp".

So, the resulting text block should be:

Hello, my nickname is Sandman, here is<br />
a list of things I like:<br />
<br />
<ol>
<li> Perl
<li> PHP
<li> MySQL
</ol>

Is there a fancy regexp for this? I know how to exclude certain characters with
the [^...] 'switch', but what about whole blocks of text? In pseudo-code:
[^<(ol|ul|pre|xmp)>].

--
Sandman[.net]

In article <mr-E6670E.12155716082004@individual.net>, Sandman <mr@sandman.net>
wrote:

> Textblock (indented for clarity):
>
> Hello, my nickname is Sandman, here is
> a list of things I like:
>
> <ol>
> <li> Perl
> <li> PHP
> <li> MySQL
> </ol>
>
> I want to replace newlines with "<br />\n" in this textblock, for displaying
> it
> to a browser - but I don't want to replace it inside some specified html-tags
> -
> in this case "ol", but also "ul", "pre" and "xmp".
>
> So, the resulting text block should be:
>
> Hello, my nickname is Sandman, here is<br />
> a list of things I like:<br />
> <br />
> <ol>
> <li> Perl
> <li> PHP
> <li> MySQL
> </ol>
>
> Is there a fancy regexp for this? I know how to exclude certain characters
> with
> the [^...] 'switch', but what about whole blocks of text? In pseudo-code:
> [^<(ol|ul|pre|xmp)>].

I have now solved this, and this is the solution (posted for people searching
for this topic on google):

-----------------------------------------
#!/usr/bin/php
<?
$text="Hello, my nickname is Sandman, here is
a list of things I like:

<ol>
<li> Perl
<li> PHP
<li> MySQL
</ol>

Do you like it?

I do";

print preg_replace_callback(
"!(<(ol|ul|pre|xmp)[^>]*>.*?</\\2>\n?)|\n!is",
create_function('$m', 'return $m[1] ? $m[1] : "<br />\n";'),
$text
);
?>
-----------------------------------------
Output:
-----------------------------------------
Hello, my nickname is Sandman, here is<br />
a list of things I like:<br />
<br />
<ol>
<li> Perl
<li> PHP
<li> MySQL
</ol>
<br />
Do you like it?<br />
<br />
I do
-----------------------------------------

--
Sandman[.net]

In article <pan.2004.08.17.12.17.43.610000@bubbleboy.digiserv .net>,
"Ian.H" <ian@WINDOZEdigiserv.net> wrote:

> On Tue, 17 Aug 2004 13:01:05 +0200, Sandman wrote:
>
>
> [ snip ]
>
> > <ol>
> > <li> Perl
> ^^^
> > <li> PHP ^^^
> > <li> MySQL ^^^
> > </ol>
> > <br />
> > Do you like it?<br />
> > <br />
> > I do
>
>
> Would be better to produce valid code would it not? Some missing tags ;)

That's the input text, I don't touch that in my script. It could be anything
coming from anyone. I just make sure I don't add any <br /> to the <ol>
container space.

--
Sandman[.net]

On Tue, 17 Aug 2004 12:16:09 GMT, "Ian.H" <ian@WINDOZEdigiserv.net> wrote:

>On Tue, 17 Aug 2004 13:01:05 +0200, Sandman wrote:
>
>[ snip ]
>
>> <ol>
>> <li> Perl
> ^^^
>> <li> PHP ^^^
>> <li> MySQL ^^^
>> </ol>
>> <br />
>> Do you like it?<br />
>> <br />
>> I do
>
>
>Would be better to produce valid code would it not? Some missing tags ;)

Well, that's perfectly valid HTML. In fact it's almost identical to the Lists
example in sec. 10.1 of the standard.

But adding <br /> does imply the OP is going for XHTML in which case yes, it's
wrong.

--
Andy Hassall / <andy@andyh.co.uk> / <http://www.andyh.co.uk>
<http://www.andyhsoftware.co.uk/space> Space: disk usage analysis tool

In article <ans4i0tk7g4uopk6oe96e5f67evlqdk5e3@4ax.com>,
Andy Hassall <andy@andyh.co.uk> wrote:

> On Tue, 17 Aug 2004 12:16:09 GMT, "Ian.H" <ian@WINDOZEdigiserv.net> wrote:
>
> >On Tue, 17 Aug 2004 13:01:05 +0200, Sandman wrote:
> >
> >[ snip ]
> >
> >> <ol>
> >> <li> Perl
> > ^^^
> >> <li> PHP ^^^
> >> <li> MySQL ^^^
> >> </ol>
> >> <br />
> >> Do you like it?<br />
> >> <br />
> >> I do
> >
> >
> >Would be better to produce valid code would it not? Some missing tags ;)
>
> Well, that's perfectly valid HTML. In fact it's almost identical to the
> Lists
> example in sec. 10.1 of the standard.
>
> But adding <br /> does imply the OP is going for XHTML in which case yes,
> it's
> wrong.

The OP, being me, strives for correct transitional XHTML. But you have to
understand, as I noted earlier, that I am not the one controlling the input to
the script. I am merely preserving code within it, not altering it.

--
Sandman[.net]