Re: Can Text::CSV_XS return key-value pairs?

You can use Text::CSV_XS for handling both the steps, splitting on "|" and splitting on ";". I wasn't able to achieve this using the csv method, so I had to go more procedural:

#!/usr/bin/perl
use warnings;
use strict;

use Text::CSV_XS qw(csv);

use Data::Dumper;


my $file = shift;

my $csv = 'Text::CSV_XS'->new({
    sep_char         => '|',
    quote_char       => undef,
    empty_is_undef   => 1,
    allow_whitespace => 1,
}) or die 'Text::CSV_XS'->error_diag;

open my $CSV, '<', $file or die $!;

my %structure;

my $inner = 'Text::CSV_XS'->new({
    sep_char => ';',
    quote_char => undef,
    allow_whitespace => 1,
}) or die 'Text::CSV_XS'->error_diag;

while (my $row = $csv->getline($CSV)) {
    $inner->parse($row->[1]);
    $structure{ $row->[0] } = [ $inner->fields ];
}

print Dumper(\%structure);
[download]

Update: Fixed along the suggestions by Tux below.

($q=q:Sq=~/;[c](.)(.)/;chr(-||-|5+lengthSq)`"S|oS2"`map{chr |+ord
}map{substrSq`S_+|`|}3E|-|`7**2-3:)=~y+S|`+$1,++print+eval$q,q,a,
[download]

Comment on Re: Can Text::CSV_XS return key-value pairs? Select or Download Code

Replies are listed 'Best First'.
Re^2: Can Text::CSV_XS return key-value pairs? by Tux (Canon) on Jun 11, 2017 at 12:45 UTC
That is close, but not what the OP requested. The second `Text::CSV_XS->new` also needs an `allow_whitespace => 1`. I'd also put the constructor outside of the second loop. No need to create it on every iteration. And yes, it is safer than using `split`, but I did not want to be pedantic Enjoy, Have FUN! H.Merijn	[reply] [d/l] [select]
Re^2: Can Text::CSV_XS return key-value pairs? by Lady_Aleena (Priest) on Jun 13, 2017 at 14:05 UTC
choroba, could you please include where you got the headings/headers? You used the method `new()` instead of the function `csv()`. Since `new()` doesn't have a header option, I don't know where the headers are invoked if they are not the first line of the file being parsed. As for the secondary `new()` splitting (or whatever does it) at `;`, that would be triggered whenever a header has a `+` at the end of it (and the `+` being removing from the header after parsing). So, for example, my movies.txt has the following headers... `'headers' => ['title','start year','end year',qw(media format+ Wikiped +ia allmovie IMDb TV.com Flixster genre+ source company)],` [download] `format` and `genre` would be parsed at the `;` while all other fields are strings. So for a file where I only want a key-value pair for each line, it looks like I will have to put in 2 headers, and if the second one has a `+` it is to be parsed with the value being split (or whatever) on the `;`. I gravitated directly to `csv()` because it looked easier to use than `new()` since I could not figure out what did what. Like what makes an array of hashes and what makes a hash of hashes (what I use mostly). The whatchamacallits (like getline, parse, etc) are not grouped together in such a way as to make it obvious to me. So, would you please expand the code so I can see everything you are doing? I am a bit lost. *No matter how hysterical I get, my problems are not time sensitive. So, relax, have a cookie, and a very nice day!* Lady Aleena	[reply] [d/l] [select]
Re^3: Can Text::CSV_XS return key-value pairs? by choroba (Cardinal) on Jun 13, 2017 at 23:19 UTC
I used `new` instead of `csv` because it gives you more control over what you can do. Here, no headers are defined at all, which means the row is split into an array reference `$row`. The second part of the row is processed by the very same module, as you recognised, to split the string on semicolons—if I understand your comment correctly, you'd need to do that for each column with the + in the general case. The output structure is being built on line 31, where the first part of the line is used as the key, and the result of the secondary processing is used as the value. ($q=q:Sq=~/;[c](.)(.)/;chr(-\|\|-\|5+lengthSq)`"S\|oS2"`map{chr \|+ord }map{substrSq`S_+\|`\|}3E\|-\|`7**2-3:)=~y+S\|`+$1,++print+eval$q,q,a, [download]	[reply] [d/l] [select]