in reply to Perl script help to convert .txt file to .csv

use strict; 
use warnings; 

print "\n Running script for Jiggs \n";
my $infile = "foot.txt";                   # <foot.txt> is not legal q
+uotation in perl5

open my $in,  "<", $infile     or die $!;  # Don't quote a variable ag
+ain
open my $out, ">", "foot1.txt" or die $!;

my $line = <$in>;                          # Is $line the header? If s
+o, don't call it "line"
$line =~ s/\s+\Z/\n/;                      # When you chomp, you don't
+ get a "new" newline
$line =~ s/ +/,/g;
print $out $line;                          # Don't quote variable. It 
+now still has a newline
print $out scalar <$in>;                   # What is this second line?
+ Something special?

while (<$in>) {
    s/\s+\Z/\n/;
    s/ +/,/g;
    s/,length=/,/g;
    print $out $_;
    }

close $in; 
close $out; 
 
print "\n Done!\n";

[download]

Note that this might up being invalid CSV. Use Text::CSV or Text::CSV_XS to ensure valid CSV output.


Enjoy, Have FUN! H.Merijn

Replies are listed 'Best First'.
Re^2: Perl script help to convert .txt file to .csv
by Seabass (Novice) on Dec 20, 2011 at 17:11 UTC
    Cool, thanks for the reply!

    Using your example, I found out there is no need for the header part at all. I removed that part of the code, and then used the s//g; to get rid of the rest of the field names.

    Here is the code: 
    
use strict; 
use warnings; 

print "\n Running script for Jiggs \n";
my $infile = "foot.txt";                   

open my $in,  "<", $infile     or die $!;  
open my $out, ">", "foot1.txt" or die $!;

while (<$in>) {
    
    s/\s+\Z/\n/;
    s/ +/,/g;
    s/,length=/,/g;
    s/,xy=/,/g;
    s/,region=/,/g;
    s/,run=/,/g;
    
    print $out $_;
    }

close $in; 
close $out; 
 
print "\n Done!\n";

    [download]
    Here is the output: 
    
>G9JVYGV01AJE8V,135,0104_0349,1,R_2011_09_20_15_00_06_
GGTGGTAGTGAAGAAGAGGAGATGAAAGTGGAAGAGGTTGAGGATGAGAAGGTTGAATTG
GAAGAAGAAGATGAGAAGGTTGAAGTGGAAGATGAGAAGGTTGAAGTGGAAGAAGATGAA
GTGGAAGAGAGGAGC
>G9JVYGV01A4910,90,0353_0150,1,R_2011_09_20_15_00_06_
GGTGCATGGCATTGTAGATGGTTGCTTGATAGTTGCCCATACGTGTACTACACTTGCAGA
GTGAAGCAACCATCTACAATGCCATGCACC
>G9JVYGV01A0SVP,70,0302_0163,1,R_2011_09_20_15_00_06_
GCACCATTCAGCACAGATATAGTAGCCACATCAACACAAGTTACCTAACTATATCTGTGC
TGAATGGTGC
>G9JVYGV01A221U,89,0328_0160,1,R_2011_09_20_15_00_06_
CTGGACATTTACATCCATAAGTAGGAGTTAGGACTCTGCACCAGCCTCTTGAGCTTGTGA
CGTCTCTTCTCCTCCTCCGGACTGGGACA
>G9JVYGV01BVCPK,46,0650_0134,1,R_2011_09_20_15_00_06_
GCAAGATCGCAAGCCAAGCAACGTTTCACGAACTGGCCAGAATGAG
>G9JVYGV01AOU3I,81,0166_0220,1,R_2011_09_20_15_00_06_
TCATTGACATCTGTGCAGCTGCAGGAGCGGATATGAGGAGATGGTTCTATCTGCACAGAT
GTCAATGAGTGTGACAGTGAT
>G9JVYGV01A0JEL,61,0299_0171,1,R_2011_09_20_15_00_06_
CGAGTGAAGGCATTGGTGATGCTGGTGTGAAGAGTGAGGGCATCGCCAATGCCTTCACTC
G
>G9JVYGV01AUKIG,119,0231_0198,1,R_2011_09_20_15_00_06_
GGCCACCAGGGCTTAACTTCCTGTGCCTCACCATCACGCAGTTGTCAGAGGATCCACATT
GAACAAAGTAGCAATTCTTTCCACTCTGTGACACACCAACATTCTTATACAGCACCAGG
>G9JVYGV01AJ8F7,29,0113_1333,1,R_2011_09_20_15_00_06_
CTGCTTCCAAGCCTCCAACCTCTAACCAG
>G9JVYGV01AMQ87,79,0142_0233,1,R_2011_09_20_15_00_06_
AGAGTCTCCTCATTGTTCTTTCCAAGTCCTCTATTGCTGAGCCTGGTTTCGTACCTTCTC
AGCTAGGCCCTCTTTCTCT
>G9JVYGV01A4W45,85,0348_3895,1,R_2011_09_20_15_00_06_
GCTTCACATCTCAGAAATATAACCGCTAATGATCTGAAACAAGTTACAATCTGACATTCT
GAAACCAAATGAAAGCAGCATAAAC
>G9JVYGV01A7TPA,66,0382_0140,1,R_2011_09_20_15_00_06_
ATGGCTTACCTCACTGTCGATGGAGATCGAATGCAAGCGATGTCCATCGACAGTGAGGTA
AGCCAT

    [download]

    Almost there, but I need to put a comma after the last field before the sequence. Then remove the returns within the sequence.

    If each entry starts with a carot, then I need 6 fields seperated by commas and then the new line: the opening entry, the length=, xy=, region=, run=, AAGGTTGGCC /n).

    Apologies for not being clearer, but thanks for the help so far.

[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom