reg ex expression - finding long character strings

[lawrence]

"Garp" <ga***@no7.blueyonder.co.uk> wrote in message news:<_v*********************@news-text.cableinet.net>...

"lawrence" <lk******@geocities.com> wrote in message
news:da**************************@posting.google.c om...

This reg ex will find me any block of strings 40 or more characters in
length without a white space, yes?

[^ ]{40}
To get it to include tabs and newlines, do I to this?

[^ \n\t]{40}

\s is the whitespace token, if that's easier for you.

Good, but now my question is how to insert the white space that I
want. If I do this:

$string = ereg_replace ([^\s]{40}, " ", $string);

Then the text gets obliterated and replaced by a white space. That is
not what I want. I simply want to break up long strings (mostly urls)
that threaten to destroy the format of a page. This is especially true
of Internet Explorer, which tends to expand DIV tags to fit the
contents (Netscape lets long urls burts outside the boundries of the
DIV.)

Go look at this page using IE 5 or 6:

http://www.publicdomainsoftware.org

You'll see a comment (right now it is the second one down) that looks
like this:

>>>>>>>>>>> Misty, I assume you're the one who came up with these interesting
photos of vegetables? Are they from the ARE garden?
http://www.publicdomainsoftware.org/...egetables2.JPG ...read
more>>>>>>>>>>>

That long url is distorting the whole page. I need to break it up.

I suppose I could hit the whole string with explode() and break them
on the white space and then loop through the array and test each entry
for a length of more than 30 or 40 or so, and then stitch it all back
together with implode, but I was assuming I could do it all more
elegantly with regular expressions. I don't know much about regular
expressions, but if someone does, please let me know.

[Pedro Graca]

lawrence wrote:

I suppose I could hit the whole string with explode() and break them
on the white space and then loop through the array and test each entry
for a length of more than 30 or 40 or so, and then stitch it all back
together with implode, but I was assuming I could do it all more
elegantly with regular expressions. I don't know much about regular
expressions, but if someone does, please let me know.

Try this. Change as you see fit.
<?php
function compress_url($txt, $size=40) {
$rx = '=(http://\S{' . ($size-7) . ',})=e';
$compressed_txt = preg_replace($rx,
"'[<a class=\"compressed\" href=\"$1\" title=\"$1\">'
. substr('$1', 0, $size-10)
. '...'
. substr('$1', -7)
. '</a>]'",
$txt);
return $compressed_txt;
}
$txt = '
Misty, I assume you\'re the one who came up with these interesting
photos of vegetables? Are they from the ARE garden?
http://www.publicdomainsoftware.org/...egetables2.JPG ...read
more';

# # # # # # # # # # # # # # # # # # # # # # # #
#
# Remember to define a "compressed" class in your stylesheet
#
# # # # # # # # # # # # # # # # # # # # # # # #

echo compress_url($txt);
?>
Happy Coding :-)

--
USENET would be a better place if everybody read: : mail address :
http://www.catb.org/~esr/faqs/smart-questions.html : is valid for :
http://www.netmeister.org/news/learn2quote2.html : "text/plain" :
http://www.expita.com/nomime.html : to 10K bytes :

[lawrence]

Pedro Graca <he****@hotpop.com> wrote in message news:<sl*******************@ID-203069.user.uni-berlin.de>...

lawrence wrote:

I suppose I could hit the whole string with explode() and break them
on the white space and then loop through the array and test each entry
for a length of more than 30 or 40 or so, and then stitch it all back
together with implode, but I was assuming I could do it all more
elegantly with regular expressions. I don't know much about regular
expressions, but if someone does, please let me know.

Try this. Change as you see fit.
<?php
function compress_url($txt, $size=40) {
$rx = '=(http://\S{' . ($size-7) . ',})=e';
$compressed_txt = preg_replace($rx,
"'[<a class=\"compressed\" href=\"$1\" title=\"$1\">'
. substr('$1', 0, $size-10)
. '...'
. substr('$1', -7)
. '</a>]'",
$txt);
return $compressed_txt;
}

That looks brilliant, though I have trouble reading it. When you write:

http://\S{' . ($size-7) . ',

are the dots saying "one or more of this white space"?

[Pedro Graca]

lawrence wrote:

Pedro Graca <he****@hotpop.com> wrote in message news:<sl*******************@ID-203069.user.uni-berlin.de>...

Try this. Change as you see fit.
<?php
function compress_url($txt, $size=40) {
$rx = '=(http://\S{' . ($size-7) . ',})=e';
$compressed_txt = preg_replace($rx,
"'[<a class=\"compressed\" href=\"$1\" title=\"$1\">'
. substr('$1', 0, $size-10)
. '...'
. substr('$1', -7)
. '</a>]'",
$txt);
return $compressed_txt;
}

That looks brilliant, though I have trouble reading it. When you write:

http://\S{' . ($size-7) . ',

are the dots saying "one or more of this white space"?

No. They are the string concatenator; they are not part of the regular
expression.

If I want to find 40 or more non-whitespace characters in a regular
expression I do

\S{40,}

In the function, I made the length a parameter, so that should be

\S{$size,} *** DOES NOT WORK LIKE THIS!

but, for that specific function I'm already using "http://" (7 chars),
so, that part of the regexp is

\S{$size-7,} *** DOES NOT WORK LIKE THIS!

So, that $rx line concatenates these three strings:
=(http://\S{
$size - 7 *** the result of the subtraction
,})=e

giving, for $size=40

=(http://\S{33,})=e

so it will match http urls (and not https, ftp, mailto, ...) longer than
40 characters.
HTH

--
USENET would be a better place if everybody read: : mail address :
http://www.catb.org/~esr/faqs/smart-questions.html : is valid for :
http://www.netmeister.org/news/learn2quote2.html : "text/plain" :
http://www.expita.com/nomime.html : to 10K bytes :

[lawrence]

Pedro Graca <he****@hotpop.com> wrote in message news:<sl*******************@ID-203069.user.uni-berlin.de>...

lawrence wrote:

Pedro Graca <he****@hotpop.com> wrote in message news:<sl*******************@ID-203069.user.uni-berlin.de>...

Try this. Change as you see fit.
<?php
function compress_url($txt, $size=40) {
$rx = '=(http://\S{' . ($size-7) . ',})=e';
$compressed_txt = preg_replace($rx,
"'[<a class=\"compressed\" href=\"$1\" title=\"$1\">'
. substr('$1', 0, $size-10)
. '...'
. substr('$1', -7)
. '</a>]'",
$txt);
return $compressed_txt;
}

That looks brilliant, though I have trouble reading it. When you write:

http://\S{' . ($size-7) . ',

are the dots saying "one or more of this white space"?

No. They are the string concatenator; they are not part of the regular
expression.

If I want to find 40 or more non-whitespace characters in a regular
expression I do

\S{40,}

In the function, I made the length a parameter, so that should be

\S{$size,} *** DOES NOT WORK LIKE THIS!

but, for that specific function I'm already using "http://" (7 chars),
so, that part of the regexp is

\S{$size-7,} *** DOES NOT WORK LIKE THIS!

So, that $rx line concatenates these three strings:
=(http://\S{
$size - 7 *** the result of the subtraction
,})=e

giving, for $size=40

=(http://\S{33,})=e

so it will match http urls (and not https, ftp, mailto, ...) longer than
40 characters.

So you can, so to speak, go in and out of "regex mode" by using a
single quote:

'

I assume this is simply the way PHP is built. And when I wanted to use
a real ' I suppose I would do this:

\'

[Pedro Graca]

lawrence wrote:

Pedro Graca <he****@hotpop.com> wrote in message news:<sl*******************@ID-203069.user.uni-berlin.de>...

lawrence wrote:

> Pedro Graca <he****@hotpop.com> wrote in message news:<sl*******************@ID-203069.user.uni-berlin.de>...
>> Try this. Change as you see fit.
>>
>>
>> <?php
>> function compress_url($txt, $size=40) {
>> $rx = '=(http://\S{' . ($size-7) . ',})=e';
>> $compressed_txt = preg_replace($rx,
>> "'[<a class=\"compressed\" href=\"$1\" title=\"$1\">'
>> . substr('$1', 0, $size-10)
>> . '...'
>> . substr('$1', -7)
>> . '</a>]'",
>> $txt);
>> return $compressed_txt;
>> }
>
> That looks brilliant, though I have trouble reading it. When you write:
>
> http://\S{' . ($size-7) . ',
>
> are the dots saying "one or more of this white space"?
No. They are the string concatenator; they are not part of the regular
expression.

If I want to find 40 or more non-whitespace characters in a regular
expression I do

\S{40,}

In the function, I made the length a parameter, so that should be

\S{$size,} *** DOES NOT WORK LIKE THIS!

but, for that specific function I'm already using "http://" (7 chars),
so, that part of the regexp is

\S{$size-7,} *** DOES NOT WORK LIKE THIS!

So, that $rx line concatenates these three strings:
=(http://\S{
$size - 7 *** the result of the subtraction
,})=e

giving, for $size=40

=(http://\S{33,})=e

so it will match http urls (and not https, ftp, mailto, ...) longer than
40 characters.

So you can, so to speak, go in and out of "regex mode" by using a
single quote:

'

No, not quite!
This is all standard string management:
http://www.php.net/manual/en/language.types.string.php

I prefer to use single quotes most of the time.

$x = 'abc'; // $x holds a three-character string
$x = $x . 14; // PHP automagically transforms the number 14 into a
// two-character string; $x now holds a five-character
// string
$x .= 'yz'; // add tow more characters to $x
// making it "abc14xy" (without the quotes)

It's the exact same thing with the regexp above :)
Instead of it being constant, it is a /dynamic/ regexp.

I assume this is simply the way PHP is built. And when I wanted to use
a real ' I suppose I would do this:

\'

If it's inside single quotes, yes.

--
USENET would be a better place if everybody read: | to email me: use |
http://www.catb.org/~esr/faqs/smart-questions.html | my name in "To:" |
http://www.netmeister.org/news/learn2quote2.html | header, textonly |
http://www.expita.com/nomime.html | no attachments. |