Re: Split lines in file to columns

Here is some code that makes use of a couple of very nice perl features.

1) You can put your data after the _END_ statement and then have while read it in, rather than putting it in an array at the source. Makes for easier setting of test cases.

2) It uses REGEX power of Perl to split the line. This is much more flexible than split, and is the usual way that perl programmers parse stuff.

I have added a couple of lines to your original input to show that a) if it is not REGION but say HQ the value is not replaced. Ditto if it is REGION but the original value is not XYZ. Not sure if that is exactly what you meant but the concept should be useful.

#!/usr/bin/perl -w

while (<DATA>) { #read a line into $_
    chomp;
    $_ =~ m/^\s*(\S+)\s+(\S+)\s+(\S+)\s+(.*)$/;
    #             col1   col_b       xyz    Descr
    my ($col1, $data_b, $xyz, $description ) = ($1, $2, $3, $4);

    if ( ( $description =~ m/Region/ ) && ( $xyz eq "XYZ" ) ) { $xyz =
+ "N/A"; }
    print "Line $. = $_\n"; 
    print "\tXYZ result = $xyz \n";
} #end while DATA
exit 1;
__END__
ValuesInColumn1    DataColumnB    XYZ    RowDescription|RowCode|Suppli
+er ID::Region
ValuesInColumn1    DataColumnB    XYZ    RowDescription|RowCode|Suppli
+er ID::Region
ValuesInColumn1    DataColumnB    XYZ    RowDescription|RowCode|Suppli
+er ID::Region
ValuesInColumn1    DataColumnB    XYZ    RowDescription at RowCode
ValuesInColumn1    DataColumnB    ABC    RowDescription at RowCode
ValuesInColumn1    DataColumnB    XYZ    RowDescription|RowCode|Suppli
+er ID::HQ
ValuesInColumn1    DataColumnB    BCD    RowDescription|RowCode|Suppli
+er ID::Region
[download]

The result of running the above is:

Line 1 = ValuesInColumn1    DataColumnB    XYZ    RowDescription|RowCo
+de|Supplier ID::Region
    XYZ result = N/A 
Line 2 = ValuesInColumn1    DataColumnB    XYZ    RowDescription|RowCo
+de|Supplier ID::Region
    XYZ result = N/A 
Line 3 = ValuesInColumn1    DataColumnB    XYZ    RowDescription|RowCo
+de|Supplier ID::Region
    XYZ result = N/A 
Line 4 = ValuesInColumn1    DataColumnB    XYZ    RowDescription at Ro
+wCode
    XYZ result = XYZ 
Line 5 = ValuesInColumn1    DataColumnB    ABC    RowDescription at Ro
+wCode
    XYZ result = ABC 
Line 6 = ValuesInColumn1    DataColumnB    XYZ    RowDescription|RowCo
+de|Supplier ID::HQ
    XYZ result = XYZ 
Line 7 = ValuesInColumn1    DataColumnB    BCD    RowDescription|RowCo
+de|Supplier ID::Region
    XYZ result = BCD
[download]

Comment on Re: Split lines in file to columns Select or Download Code

Replies are listed 'Best First'.

Re^2: Split lines in file to columns
by haukex (Archbishop) on Mar 24, 2019 at 22:26 UTC

You can put your data after the _END_ statement

Normally, one would use the __DATA__ token for this purpose - see Special Literals in perldata.

Update: Another nitpick: The special variables $1 etc. should only be used if the match succeeds. And the two lines could be shortened to: my ($col1, $data_b, $xyz, $description) = /^\s*(\S+)\s+(\S+)\s+(\S+)\s+(.*)$/ or die "Failed to parse: $_";

[reply]
[d/l]
[select]