eregi_replace

[Anon]

Hi,

I've been trying to understand this eregi_replace thing, but I'm stuck with the following:

I have this eregi_replace:
$text= eregi_replace("flash size=1[flash]","<object classid=\"clsid😃27CDB6E-AE6D-11cf-96B8-444553540000\" codebase=\"http://download.macromedia.com/pub/shockwave/cabs/flash/swflash.cab#version=5,0,0,0\" width=50 height=50><param name=movie value="\\1\"><param name=quality value=high><param name=wmode value=transparent><embed src=\"\1\" width=\"50\" height=\"50\" quality=high wmode=transparent type=\"application/x-shockwave-flash\" pluginspage=\"http://www.macromedia.com/shockwave/download/index.cgi?P1_Prod_Version=ShockwaveFlash\"></embed></object>",$text);

I want the text between [flash size=1] and [/flash] to become the \1 (since it's between brackets).
(.*) is everything right?

But it gives the following error:
Warning: Unexpected character in input: '\' (ASCII=92) state=1 in /home/sites/site95/web/ubb.php on line 52
Parse error: parse error, unexpected T_LNUMBER in /home/sites/site95/web/ubb.php on line 52

I've already tried to remove all of the \", but then I get this error:
Warning: REG_BADRPT in /home/sites/site95/web/ubb.php on line 54

I did get something to work, but it just copied the message like 10 times, and it was completey f*cked up.

Can somebody help me out?
Thanks,

Menno

[Anon]

Deep breath....

First, we're not doing any variable interpolation so we can use single quotes to delimit our strings instead of double quotes. Then we can drop the \ escapes from the double quotes that are there.

Next, I think I'll throw a few newlines in to the replacement text so's I can read it better.

I think I see the problem: an extra \ in the <param> value= attribute's value, where the replacement text should go. Take that out, but let's keep going in case there's something else.

Nope, looks fine. I end up with:

$text= eregi_replace('flash size=1[flash]',
'<object classid="clsid😃27CDB6E-AE6D-11cf-96B8-444553540000" codebase="http://download.macromedia.com/pub/shockwave/cabs/flash/swflash.cab#version=5,0,0,0" width="50" height="50">
<param name="movie" value="\1">
<param name="quality" value="high">
<param name="wmode" value="transparent">
<embed src="\1" width="50" height="50" quality="high" wmode="transparent" type="application/x-shockwave-flash" pluginspage="http://www.macromedia.com/shockwave/download/index.cgi?P1_Prod_Version=ShockwaveFlash">
</embed></object>',
$text);

Have a go with that.

[Anon]

ah, it needed some tweaking, but it works now...
I had to escape the ?, since it gave the REG_BADRPT error again...

Thanks for the help!

$text= eregi_replace('flash size=1[/flash]','<object classid="clsid😃27CDB6E-AE6D-11cf-96B8-444553540000" codebase="http://download.macromedia.com/pub/shockwave/cabs/flash/swflash.cab#version=5,0,0,0" width="50" height="50">
<param name="movie" value="\1">
<param name="quality" value="high">
<param name="wmode" value="transparent">
<embed src="\1" width="50" height="50" quality="high" wmode="transparent" type="application/x-shockwave-flash" pluginspage="http://www.macromedia.com/shockwave/download/index.cgi?P1_Prod_Version=ShockwaveFlash">
</embed></object>',$text);

[Anon]

hmm, i had to use this code, since that one didn't work...

$text= preg_replace("#[flash size=1](.*?)[/flash]#si","<object classid=\"clsid😃27CDB6E-AE6D-11cf-96B8-444553540000\" codebase=\"http://download.macromedia.com/pub/shockwave/cabs/flash/swflash.cab#version=5,0,0,0\" width=\"50\" height=\"50\"><param name=\"movie\" value=\"\1\"><param name=\"quality\" value=\"high\"><param name=\"wmode\" value=\"transparent\"><embed src=\"\1\" width=\"50\" height=\"50\" quality=\"high\" wmode=\"transparent\" type=\"application/x-shockwave-flash\" pluginspage=\"http://www.macromedia.com/shockwave/download/index.cgi?P1_Prod_Version=ShockwaveFlash\"></embed></object>",$text);

I just looked at the xmb forum codes...
but what does the # and #si do?

[Anon]

You can use ani delimiter. The classic one is "/expression/", but "#expression#" does the same. The 'si' tells your regexp to ignore case (i) and let . match all, including newlines. Just read the doc....