perl regular expression problem

true_atlantis · 01-08-2004, 06:17 PM

i have a perl script that i created, and what it needs to do is process a bunch of files and put the info from those files into a different format into a summary file. when processing, i need to take out all of the and i come into a problem. i use this

foreach $line (@fields_add){
$line =~ s/ / /i;
}

to replace the with a space, but it does not get all of them.
basically if you look at the perl script, what it does is just seperate the lines of the files at the occurance of ;; and then puts them in a format for another script. so, if you look at where it shows

1/2 This run is too short to be a segment;;

that will be removed, but if you look at

 Lat/lons are for start and stop of imaging;;

it will skip that ... anyone know why this is not working correctly???

file to be processed 'f000411t01.log' :

f000411t01;;p01_r01;;1;;1;;Preflight;;;;Jack Holbrook;;;;;;;;;;1738;;1740;;4 - Pre/Postflight;;;;
f000411t01;;p02_r01;;2;;1;;Mobile Check;;;;Dee Porter;;;;;;;;;;19:07:09;;19:09:01;;6 - Tarmac Check;;;;
f000411t01;;p03_r01;;3;;1;;Maui 3-A, HI;;02072;;Gene Carl Feldman;;20-49.6;;157-8.6;;20-44.1;;156-22.7;;20:26:51;;20:34:29;;1 - Science;;28/27;;
f000411t01;;p03_r02;;3;;2;;Molokai 2-A, HI;;02072;;Gene Carl Feldman;;21-5.6;;156-42.2;;21-10.3;;157-20.8;;20:41:41;;20:47:53;;1 - Science;;97/98;;
f000411t01;;p03_r03;;3;;3;;MODIS/MOBY;;02026;;Green;;20-53.55;;157-5.1;;20-36.9;;157-5.4;;21:06:30;;21:19:53;;1 - Science;;Control point = 20-49.6, 157-08.25 Looking at MOBY @ approx. 21:14:33 Lat/lons are for start and stop of imaging;;
f000411t01;;p03_r04;;3;;4;;Hawaii 4-A, HI;;02072;;Gene Carl Feldman;;20-0.9;;156-0.3;;19-59.0;;155-42.1;;21:31:15;;21:35:01;;1 - Science;;58/57;;
f000411t01;;p03_r05;;3;;5;;Hawaii 5-B, HI;;02072;;Gene Carl Feldman;;19-54.6;;155-47.6;;19-56.5;;156-5.3;;21:40:52;;21:44:21;;1 - Science;;60/59 5-A was flown on 000410;;
f000411t01;;p03_r06;;3;;6;;Oahu 4-A, HI;;02072;;Gene Carl Feldman;;21-18.0;;157-45.8;;21-19.8;;158-18.1;;22:03:02;;22:06:03;;1 - Science;;7/8;;
f000411t01;;p03_r07;;3;;7;;Oahu 3-A, HI;;02072;;Gene Carl Feldman;;21-25.8;;158-9.3;;21-23.9;;157-53.4;;22:11:37;;22:15:00;;1 - Science;;5/6;;
f000411t01;;p03_r08;;3;;8;;Oahu 1-A, HI;;02072;;Gene Carl Feldman;;21-27.8;;157-45.7;;21-29.9;;158-3.0;;22:20:21;;22:23:47;;1 - Science;;1/2 This run is too short to be a segment;;
f000411t01;;p03_r09;;3;;9;;Niihau 2-A, HI;;02072;;Gene Carl Feldman;;21-49.7;;159-59.4;;21-52.1;;160-16.9;;22:40:03;;22:43:29;;1 - Science;;47/48;;
f000411t01;;p03_r10;;3;;10;;Kauai 5-A, HI;;02072;;Gene Carl Feldman;;21-59.5;;160-3.5;;21-54.7;;159-29.0;;22:51:15;;22:57:22;;1 - Science;;44/43;;
f000411t01;;p04_r01;;4;;1;;Postflight;;;;Chuck Sarture;;;;;;;;;;00:40:00;;00:42:00;;4 - Pre/Postflight;;;;

perl script 'convert.pl' :

Code:

#!/usr/bin/perl
$LID    = 100;
$FLT    = 0;
$F_PATH = 1;
$LAT1   = 7;
$LON1   = 8;
$LAT2   = 9;
$LON2   = 10;

@filelist  = <f*.log>;
#@filelist = <f000411t01.log>;

$arc_file  = "Arc_file";
$info_file = "info_file";
%out_dat = ();
$hide = 0;

if($ARGV[0] eq "-h" || $ARGV[0] eq "-H"){
	$hide = 1;
}

open (OUTPUT, ">$arc_file") || die "Cant create $arc_file";


print($a);

foreach $F (@filelist){
	if($hide == 0){
		print("Processing file $F...\n");
	}
	open (FILE, $F) || die "Cant open $arg_file";
	while ($line = <FILE>) {
		@fields = split(/;;/, $line);
		if($fields[$LAT1]){
			$LID++;
			#first latitude and longitude
			($lat1_d, $lat1_m) = split(/-/, $fields[$LAT1]);
			$lat1_d += $lat1_m/60;
			($lon1_d, $lon1_m) = split(/-/, $fields[$LON1]);
			$lon1_d = ($lon1_d * -1) + (($lon1_m/60) * -1);
			
			#second lat and lon
			($lat2_d, $lat2_m) = split(/-/, $fields[$LAT2]);
			$lat2_d += $lat2_m/60;
			($lon2_d, $lon2_m) = split(/-/, $fields[$LAT2]);
			$lon2_d = ($lon2_d * -1) + (($lon2_m/60) * -1);

			#output to lat/lon file
			print (OUTPUT "$LID\n");
			print (OUTPUT "$lon1_d\t$lat1_d\n");
			print (OUTPUT "$lon2_d\t$lat2_d\n");
			print (OUTPUT "end\n");

			#store the rest of the info into a hash
			$UID = "$fields[0]_$fields[1]";

			@fields_add = @fields[1 .. 6, 11 .. $#fields];

			#take out the <br>
			foreach $line (@fields_add){
				$line =~ s/<br>/ /i;
			}
						
			$out_dat{$LID} = (join("\t", $UID, join("\t", @fields_add)));
		}
	}
close(FILE);
}
print (OUTPUT "end\n");
close(OUTPUT);
open(OUTPUT, ">$info_file");
	
for $key (sort {$a<=>$b} keys %out_dat){
	print(OUTPUT "$key\t$out_dat{$key}\n");
}

eric.r.turner · 01-08-2004, 11:59 PM

Quote:

Originally posted by true_atlantis

foreach $line (@fields_add){
$line =~ s/ / /i;
}

anyone know why this is not working correctly???

Yeah, it should be:

Code:

foreach $line (@fields_add){
   $line =~ s/<br>/ /ig;
}

The g option means do a global replace. The line of data that you're having problems with has more than one in it. Without the g option the regex only matches the first instance of .

Regards,
Eric

thangappan · 05-26-2009, 11:41 PM

Perl is a very effective and simple programming language.

I solution is using 'g' in the substitution it can be solved.

while (<$FH>){
if ($_ =~ s/ //g){
print "$_\n";
}
}

Input:
testing testing 
 testing testing

Output:
testing testing
testing testing

syg00 · 05-27-2009, 12:00 AM

That's a *five* year old thread ....

eric.r.turner · 05-27-2009, 06:35 AM

Quote:

Originally Posted by syg00

That's a *five* year old thread ....

And he didn't provide any new information! Weird.