[SOLVED] preg_replace and regex

[aolong]

I'm having trouble with this one...

 

I need to replace everything from beginning of string up to and including the occurrence of "= Gauge32: " OR "=STRING: " with "" (maybe the ":[space]" would be better). Problem is I'm not too adept at regex and I get only something akin to "/[^.*:\s]/" which I know is horribly malformed.

 

$output = shell_exec("./script.sh $propertyIP");

$pattern = 'the expression I can't seem to write';

$replace = '';

 

echo "<pre>" . preg_replace($pattern,$replace,$output) . "</pre>";

 

clues?

[Ken2k7]

preg_replace('#=( Gauge32|STRING): #', '', $str);

?

[aolong]

preg_replace('#=( Gauge32|STRING): #', '', $str);

?

 

This only replaces "Gauge32: " or "STRING: " with ""; I need to include all characters from beginning of line - "a-zA-z",", "1-9", ".", "-", and "::" - example string to replace: "COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoLoadAverage5Min.0 Gauge32: "

[aolong]

Correction. Here are more accurate examples of the strings to be replaced:

 

"SNMPv2-MIB::sysDescr.0 = STRING: "

"COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoUpTime.0 = Timeticks: "

"COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoLoadAverage1Min.0 = Gauge32: "

 

perhaps matching everything up to ":[space]"?

 

[MadTechie]

So this

$result = preg_replace('/\s=\s.*?:\s$/m', '', $str);

 

SNMPv2-MIB::sysDescr.0 = STRING:

COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoUpTime.0 = Timeticks:

COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoLoadAverage1Min.0 = Gauge32:

 

to

SNMPv2-MIB::sysDescr.0

COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoUpTime.0

COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoLoadAverage1Min.0

[aolong]

I got it... this works:

 

preg_replace('/(.*):\s/', '', $output);

 

now if I could just find the "solved" button...

[MadTechie]

But that will wipe out the whole line, i thought you just wanted the

= STRING: 

removed

in anycase solved ?

 

(bottom right topic solved)

[aolong]

Probably my fault, perhaps I was not clear that there are some characters in $output after ":[space]". It's those characters I want preserved, without the preceeding... Thank you for your help.

 

[nrg_alpha]

.* is usually not a good idea. You can read this thread to understand why (post#11 and 14 - it deals with .+ but the principal is the same). Granted, if each string containing ": " is on their own lines, it might not be an issue.. so it is circumstantial..

 

Just note that the capture (the set of parenthesis) surrounding .* isn't necessary.. the pattern could simply be:

/.*:\s/

 

Granted, you mentioned from the beginning of the string, so you should include ^ at the beginning, and if there are multiple lines, you could add the m modifier after the closing delimiter:

/^.*:\s/m

 

In the event you find out you don't want to replace just anything under the sun that contains :[space], you could perhaps use something like:

 

Example:

$str = 'SNMPv2-MIB::sysDescr.0 = STRING: 
COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoUpTime.0 = Timeticks: 
COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoLoadAverage1Min.0 = Gauge32: ';

$str = preg_replace('#^[^=]+=\s(?:Gauge32|STRING):\s#m', '', $str);
echo $str;

 

Output:

COLUBRIS-USAGE-INFORMATION-MIB::coUsInfoUpTime.0 = Timeticks:

 

You can simply append any additional to the conditional (?: ... | .... ) to extend this list of targets to wipe out.