Regex on a array

adobepro has asked for the wisdom of the Perl Monks concerning the following question:

Hi! I opened a file and placed it into an array - now before I print to the browser the contents of the array, I want to find two specific lines and remove them.

How I want it to be setup is like so:
1.) Print to the browser the xml version, dtd + xsl location
2.) Read in the xml file and store it in an array.
3.) Remove the xml file version, dtd + xsl location in the actual xml file read in from step 2.
4.) Print out to the browser the xml file data ONLY (minusing the xml version/xsl/dtd location that was removed on step 3)

That's it. The reason being is that becuase the xml data files I'm recieving have the dtd/xsl location different than were it's actually going to be - so I need to remove that from the xml data file.

The other problem, is that we have 4 reports type, but right now I'm just testing for just one type. Is the a way to use the regex to start from particular characters and end at a particular characters with a wildcard in the middle? I want to find the string and replace with no value. e.g.:

@lines =~ s/<!DOCTYPE XMLDATA SYSTEM \"artist_royalty.dtd\">//; #????
[download]

It could be a artist_royalty.dtd or license_royalty.dtd or mechanical_royalty.dtd - so I'd like to do the wild card before the report type and end off with --> .dtd\"> for the replace search.

Here's my code, everything works great EXCEPT for the Regex on the array which has the xml file contents:

print "<?xml version=\"1.0\"?>\n";
[download]

# The conditional statement below checks to see what report type the user is requesting
# Then prints out the dtd + xsl for that report type.

             
if ($doctype eq "m") {
print "<!DOCTYPE XMLDATA SYSTEM \"http://www.mydotcom.com/DRM/clients/
+XML_FORM/XML_DTD/mechanical_royalty_rightrax.dtd\">\n";
print "<?xml-stylesheet type=\"text/xsl\" href=\"http://www.mydotcom.c
+om/DRM/clients/XML_FORM/XSL_templates/M_template_rightrax.xsl\"?>\n";
}

elsif ($doctype eq "a") {
print "<!DOCTYPE XMLDATA SYSTEM \"http://www.mydotcom.com/DRM/clients/
+XML_FORM/XML_DTD/artist_royalty_rightrax.dtd\">\n";
print "<?xml-stylesheet type=\"text/xsl\" href=\"http://www.mydotcom.c
+om/DRM/clients/XML_FORM/XSL_templates/A_template_rightrax.xsl\"?>\n";
 
else {
print "<!DOCTYPE XMLDATA SYSTEM \"http://www.mydotcom.com/DRM/clients/
+XML_FORM/XML_DTD/license_royalty_rightrax.dtd\">\n";
print "<?xml-stylesheet type=\"text/xsl\" href=\"http://www.mydotcom.c
+om/DRM/clients/XML_FORM/XSL_templates/L_template_rightrax.xsl\"?>\n";

# The snippet below opens then reads the XML data file from the server
+<br>
# and prints it to the browser - so that they can view it - and not kn
+ow<br>
# where the doc originates from.  
    
open (XMLFILE, "/opt/www/docs/$document") || &open_error("http://$serv
+er$document");
@lines = <XMLFILE>;
close (XMLFILE);
@lines =~ s/<?xml version=\"1.0\"?>//; #????????????
@lines =~ s/<?xml-stylesheet type=\"text/xsl\" href=\"A_template.xsl\"
+?>//; #????????????
@lines =~ s/<!DOCTYPE XMLDATA SYSTEM \"artist_royalty.dtd\">; #???????
+?????
foreach $line (@lines)
{
    #chop $line;
    print "$line";
}
[download]

Comment on Regex on a array Select or Download Code

Replies are listed 'Best First'.
Re: Regex on a array by Coyote (Deacon) on Jan 29, 2001 at 10:18 UTC
Seriously consider using one of the XML modules (XML::Twig is my personal fav) to do this rather than using a regex. Use the strict pragma and the -w flag, move your regexen to the foreach loop, and be sure to escape any metacharacters in your regular expressions. ---- Coyote	[reply]
Re: Regex on a array by jeroenes (Priest) on Jan 29, 2001 at 11:36 UTC
In general, a regex on an array is best done using grep. Makes your code look better. If you keep your newlines intact, you can do the following. Escape the dot (perlre)! `print grep{! /<!DOCTYPE XMLDATA SYSTEM \"artist_royalty\.dtd\">/} @lin +es;` [download] This will do most of your work. Hope this helps, Jeroen "We are not alone"(FZ) Update $code or die pointed me to stupid omission. Added that damn exclamation mark.	[reply] [d/l]
Re: Re: Regex on a array by $code or die (Deacon) on Jan 29, 2001 at 13:06 UTC
I think that does the opposite. Could be wrong, but don't you need: `print grep{!/<!DOCTYPE XMLDATA SYSTEM \"artist_royalty\.dtd\">/} @line +s;` [download] $code or die Using perl at The Spiders Web	[reply] [d/l]
Re: Regex on a array by adobepro (Initiate) on Jan 29, 2001 at 10:18 UTC
I fixed it!!!Thanks to Chromtic!! I looked at his reply to the removal of html tags reply and did this and it works like a charm! THANKS Chrom!! `foreach $line (@lines) { #chop $line; $line =~ s/<\?xml\ version=\"1.0\"\?>//igs; next regex - etc... next regex - etc... print "$line"; }` [download]	[reply] [d/l]