Hey guys help me in solving this match problem

[thendal]

Hey guys!

I have a problem,Iam storing a set of data in a flat file
where iam searching a particular field which carries special characters in it.
Flat file looks like

FIELD1|SEARCHABLE Field|filed3
name|Name(area)|notes
name1|name-city(code)|notes1
name2|Name2-city2|notes 2

When user selects the drop down city and searchs the flat file.

It will match the data's with '-' not '('

what may be the cause.Explain please any function to escape the special characters

For matching

Iam using the code
if($userselect=~/$searchable_field_from_Flatfile/)
{
}

advise me guys...






[sig][/sig]

 

[MikeLacey]

Hi thendal,

When you're searching for a ( or ) or any special character you have to precede it with a backslash in your pattern, like this example which will match lines starting with a left bracket:

    /^\(/

[sig]<p>Mike<br><a href=mailto:michael.j.lacey@ntlworld.com>michael.j.lacey@ntlworld.com</a><br><a href=

http://www.cargill.com/>

Cargill's Corporate Web Site</a><br> [/sig]

 

[thendal]

Mike:-

Iam not searching the '('.

for example I need to match two same variable .

$x='test(test)';
$y='test(test)';
if($x=~/$y/)
{
print &quot;It matched&quot;;
}

but when i use 'eq' it works apart from that .Is there any way to make the below code to work.

[sig][/sig]

 

[MikeLacey]

could you try:

$y='test\(test\)';

[sig]<p>Mike<br><a href=mailto:michael.j.lacey@ntlworld.com>michael.j.lacey@ntlworld.com</a><br><a href=

http://www.cargill.com/>

Cargill's Corporate Web Site</a><br> [/sig]

 

[thendal]

Thanks mike it worked .

just curisoity why we need to escape the special characters
in =~// and not for eq .
when values coming dynamically we can't specify escape character explictily .so i have used quotemeta.
for escaping

ie) $y=quotemeta($y);

Thank u very much mike for ur valuable timely help.
[sig][/sig]

 

[MikeLacey]

eq doesn't do pattern matching at all -- it just compares strings

now the ( and ) characters are a bit special in patterns (regular expressions they're called really I suppose)

let's say you want to do somthing like this:

on every line ending with 'ing', put the 'ing' at the beginning of the line (this is a really silly example, but bear with me)

[tt]
while(<>){
[tab]if(/ing$/){
[tab][tab]s/(.*)ing$/ing$1/;
[tab]}
[tab]print;
}
[/tt]

if you run that on a file with the following lines:

something
anything
nothing
another line

you get the following output

ingsometh
inganyth
ingnoth
another line

the (.*) in the s/// thing *remembers* what it matches (any number of any characters in this case '.*') and puts it into $1 --- and you can take the $1 thing and do something with it... [sig]<p>Mike<br><a href=mailto:michael.j.lacey@ntlworld.com>michael.j.lacey@ntlworld.com</a><br><a href=

http://www.cargill.com/>

Cargill's Corporate Web Site</a><br> [/sig]

 

[thendal]

Mike:-

Thanks for making me to understand.

Still mike
what makes me to confuss is
for example take a string &quot;test-test&quot;
$x='test-test';
$y='test-test';

if($x=~/$y/)
{
print &quot;I will be matched&quot;;
}

and another String
$x='test(test)';
$y='test(test)';
if($x=~/$y/)
{
print &quot;I won't be matched&quot;;
}

why is it so.


Mike once again iam thanking for giving me wonderfull tips and making me to understand PERL better(I love this language)
[sig][/sig]

 

[MikeLacey]

Hi thendal,

written like this it works:

$x='this test(test) and that';
$y='test\(test\)';
if($x=~/$y/)
{
print &quot;I will be matched&quot;;
}

Each time you use a special character, like a left or right backet, in a search pattern the pattern matching routine needs to know -- Is the character special this time or not? If it isn't (if you really *are* searching for a left bracket) it needs to be &quot;flagged&quot;, if you like, with a '\' character.

Here's an example of what brackets mean to the pattern matching routine:

$x = &quot;Thus said Larry&quot;; # give $x a value
$x =~ s/(.*Larry)/$1 Wall/; # search and replace in $x
print &quot;$x\n&quot;; # print out $x

This fragment of code prints out:

[red]Thus said Larry Wall[/red]

The pattern '.*Larry' matches the whole thing.

The brackets around that pattern *remember* what was between the brackets, and puts it into $1.

So when I put '$1 Wall' in the second half of the search and replace command it knows that $1 is what was in the brackets in the first half of the command.

A bit tricky this, but ok after a bit, honest...

Is that any clearer? Regular expressions (regexes) come from Unix and are the wierdest bit of Perl, I think. Keep at it. Regexes are *very* powerful and well worth taking the trouble to learn.

Have a look at the documentation with the command &quot;perldoc perlre&quot;. You might have to change directory to where your perl program is unless it's on the path.
[sig]<p>Mike<br><a href=mailto:michael.j.lacey@ntlworld.com>michael.j.lacey@ntlworld.com</a><br><a href=

http://www.cargill.com/>

Cargill's Corporate Web Site</a><br> [/sig]

 

[thendal]

Mike:-

Thank u very much .I need to spend much more time in understanding regular expressions.Ya its very powerfull
even '.' has a meaning in it.

Thanks mike.
[sig][/sig]

 

[MikeLacey]

a pleasure <s> [sig]<p>Mike<br><a href=mailto:michael.j.lacey@ntlworld.com>michael.j.lacey@ntlworld.com</a><br><a href=

http://www.cargill.com/>

Cargill's Corporate Web Site</a><br> [/sig]