exeptions for nl2br --> preg_replace??

[Marco Snoek]

Hi,

Hope you can help me out with this one..

nl2br($string) converts all the \n to <br \>

... but ..

Suppose I have a table or other HTML elements in $string..
All the TD> and TR> and FORM> are trailed by <br />

How can I adjust the function:
preg_replace("/\r\n|\r|\n/", "<br>\n", $r[body])

to ignore lines ending with >
(or: > n spaces )


Thanx in advance,
Marco Snoek

[Andy Hassall]

Re: exeptions for nl2br --&gt; preg_replace??

On Thu, 11 Mar 2004 20:21:33 +0100, "Marco Snoek"
<_[mps]@[webmind.nl]_Dont_you_dare_ send.spam> wrote:
[color=blue]
>Hope you can help me out with this one..
>
>nl2br($strin g) converts all the \n to <br \>
>
>.. but ..
>
>Suppose I have a table or other HTML elements in $string..
>All the TD> and TR> and FORM> are trailed by <br />
>
>How can I adjust the function:
>preg_replace ("/\r\n|\r|\n/", "<br>\n", $r[body])
>
>to ignore lines ending with >
>(or: > n spaces )[/color]

I think you're heading into trouble with this approach; what about the
following perfectly valid HTML:

<form method="get"
action="somethi ng.php"
name="whatever" >

You've now got a whole load of other exceptions to deal with.

The idea of nl2br is to convert _non-HTML_ text data containing newlines to an
HTML equivalent that has <br> elements so that it has newlines in the same
place when inserted into an HTML document.

In general, you can't parse HTML with a single regular expression, as regular
expressions aren't capable of maintaining states as necessary to implement a
HTML parser.

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

[Marco Snoek]

Re: exeptions for nl2br --&gt; preg_replace??

You're absolutely right..
But....
I'm quite sure the trick should work.. I enter the input myself :-)

Can you think of a reg_exp ???

Regards...

Marco


"Andy Hassall" <andy@andyh.co. uk> schreef in bericht
news:8jo1509m0t nchr6pv6qe325sq cb4goi80k@4ax.c om...[color=blue]
> On Thu, 11 Mar 2004 20:21:33 +0100, "Marco Snoek"
> <_[mps]@[webmind.nl]_Dont_you_dare_ send.spam> wrote:
>[color=green]
> >Hope you can help me out with this one..
> >
> >nl2br($strin g) converts all the \n to <br \>
> >
> >.. but ..
> >
> >Suppose I have a table or other HTML elements in $string..
> >All the TD> and TR> and FORM> are trailed by <br />
> >
> >How can I adjust the function:
> >preg_replace ("/\r\n|\r|\n/", "<br>\n", $r[body])
> >
> >to ignore lines ending with >
> >(or: > n spaces )[/color]
>
> I think you're heading into trouble with this approach; what about the
> following perfectly valid HTML:
>
> <form method="get"
> action="somethi ng.php"
> name="whatever" >
>
> You've now got a whole load of other exceptions to deal with.
>
> The idea of nl2br is to convert _non-HTML_ text data containing newlines[/color]
to an[color=blue]
> HTML equivalent that has <br> elements so that it has newlines in the same
> place when inserted into an HTML document.
>
> In general, you can't parse HTML with a single regular expression, as[/color]
regular[color=blue]
> expressions aren't capable of maintaining states as necessary to implement[/color]
a[color=blue]
> HTML parser.
>
> --
> Andy Hassall <andy@andyh.co. uk> / Space: disk usage analysis tool
> <http://www.andyh.co.uk > / <http://www.andyhsoftwa re.co.uk/space>[/color]

[Pedro Graca]

Re: exeptions for nl2br --&gt; preg_replace??

Marco Snoek wrote:[color=blue]
> Can you think of a reg_exp ???[/color]

You need the "lookbehind " stuff.

Open this URLi: http://www.php.net/manual/en/pcre.pattern.syntax.php
and search "lookbehind " in that page
--
--= my mail box only accepts =--
--= Content-Type: text/plain =--
--= Size below 10001 bytes =--