If this is OT, someone holler, it does apply to spamassassin directly but is actually more perl questions.
Theo posted an answer[1] to this query already, but it largely flew over my thick skull. (Sorry Theo). See this message for the above reference[1]: From: Theo Van Dinter <[EMAIL PROTECTED]> Subject: Re: [SAtalk] Re: [Q perl regex] X-Spam-Status: To: [EMAIL PROTECTED] Date: Wed, 12 Jun 2002 16:15:29 -0400 Message-ID: <[EMAIL PROTECTED]> Tony also posted a nifty way to do this inside procmail[2]. It works and is being employed in that way. (Thanks Tony) See this message for the reference[2] From: "Tony L. Svanstrom" <[EMAIL PROTECTED]> Subject: Re: [SAtalk] [Q perl regex] X-Spam-Status: To: [EMAIL PROTECTED] cc: [EMAIL PROTECTED] Date: Wed, 12 Jun 2002 20:59:05 +0200 (CEST) Message-ID: <[EMAIL PROTECTED]> To cut to the chase here: I'm still looking for a way to do something similar, although my original query is now satisfied. An example will explain: I want to capture this style of header: X-Spam-Status: No, hits=-106.4 required=5.0 tests=Spama,BUGZILLA_BUG,NO_REAL_NAME,UNIFIED_PATCH version=2.21 (the leading white space is \t) what I want to do with it isn't really important to the query at hand. But in general.. reassemble it into a single line for further processing. I have worked out a painstaking way that will work (at least in my tests so far), but wanted to see a better way with some of that obfuscated but powerfull code perl is famous for. Only problem is it needs to be heavily commented else it will probably be like ancient hebrew to me. So, although its asking quite a lot of list readers patience and any respondants good will, I'm pretty sure there are at least a few other dopes like me who would benefit. The following code is a start at a planned script that will check score totals against some other homeboy scripts. That is, this is a debugging script for other work. Its aim is limited here to just grabbing those lines of folded data. That is not the final aim but I need to get that in hand first. The final bit will be to compare the line to a similar line aquired with another script. The idea being to see if its working properly. I'm reading a directory of numbered files, finding the target line and reassembling it in a single line. But I suspect this can be done in much less coding. #!/usr/local/bin/perl -w ## Make sure we have a cmdline arg if(!$ARGV[0]){ print "No cmdline argument given.. we need a directory name\n"; exit; } ## Establish a target $dir = shift; ## Open our target dir and scoop all numberd filenames into an array opendir(DIR,"$dir")|| die "Cannot open directory $dir: $!"; @files = grep {/^[0-9]+$/} readdir(DIR); ## Step thru the files for(@files){ $message = $_; open(FILE,"<$dir/$message"); while(<FILE>){ chomp; ## If we see a blank line, join our captured lines into a single line ## And stuff into a hash indexed by filename if(/^$/){ $my_str = join(' ',@lines_we_want); $output{"$dir/$message"} = $my_str; ## Zero out our array and break out of the while loop @lines_we_want = ''; last; } ## If we see our target line, turn on grab mode if(/^X-Spam-Status: /){ $grab = "true"; } ## If we see a line beginning with a Cap, and it isn't ## or target line, turn grab mode off. if(!/^X-Spam-Status:/ && /^[A-Z]/){ $grab = ''; } ## While in grab mode, massage the grabbed line by removing ## leading white space if($grab){ ($str = $_) =~ s/^\s+//g; push @lines_we_want,$str; } } } ## Print out our findings in an orderly fashion foreach $key (keys %output){ print "$key\n"; print "$output{$key}\n\n"; } _______________________________________________________________ Don't miss the 2002 Sprint PCS Application Developer's Conference August 25-28 in Las Vegas - http://devcon.sprintpcs.com/adp/index.cfm?source=osdntextlink _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
