remove hash from first column name

[MojoFacets.git] / lib / MojoFacets / Import / CSV.pm

diff --git a/lib/MojoFacets/Import/CSV.pm b/lib/MojoFacets/Import/CSV.pm
index 260c002..e8fa196 100644 (file)
--- a/lib/MojoFacets/Import/CSV.pm
+++ b/lib/MojoFacets/Import/CSV.pm
@@ -10,26 +10,57 @@ use Data::Dump qw(dump);
 
 __PACKAGE__->attr('full_path');
 
+sub ext { '\.[ct]sv$' };
+
 sub data {
        my $self = shift;
 
        my $path = $self->full_path;
 
        my $encoding = 'utf-8';
-       if ( $path =~ m/\.(\w+).csv/i ) {
+       if ( $path =~ m/\.([\w\-]+).[ct]sv/i ) {
                $encoding = $1;
        }
 
        my $data = { items => [] };
        my @header;
 
-       my $csv = Text::CSV->new ( { binary => 1, eol => $/ } )
+       open my $fh, "<:encoding($encoding)", $path or die "$path: $!";
+       my $first = <$fh>;
+       my $possible_delimiters;
+       while ( $first =~ s/(\W)// ) {
+               $possible_delimiters->{$1}++;
+       }
+       warn "# possible_delimiters = ",dump($possible_delimiters);
+       seek $fh,0,0; # rewind for Text::CSV
+
+       my @sep_by_usage = sort { $possible_delimiters->{$b} <=> $possible_delimiters->{$a} } keys %$possible_delimiters;
+       my $sep_char = shift @sep_by_usage;
+       while ( $sep_char =~ m/^\s$/ ) {
+               last if $sep_char eq "\t" && $path =~ m/\.tsv$/i;
+               warn "## skip whitespace separator ",dump($sep_char);
+               $sep_char = shift @sep_by_usage;
+       }
+
+       while ( $sep_char =~ m/^\"$/ ) {
+               warn "## skip quote separator ",dump($sep_char);
+               $sep_char = shift @sep_by_usage;
+       }
+
+       if ( $sep_char !~ m/,/ && $possible_delimiters->{','} && $path =~ m/\.csv/i ) {
+               $sep_char = ',';
+               warn "## csv file detected so prefer , as separator";
+       }
+
+       warn "sep_char = [$sep_char] for $path\n";
+
+       my $csv = Text::CSV->new ( { binary => 1, eol => $/, sep_char => $sep_char } )
                or die "Cannot use CSV: ".Text::CSV->error_diag ();
 
-       open my $fh, "<:encoding($encoding)", $path or die "$path: $!";
        while ( my $row = $csv->getline( $fh ) ) {
                if ( ! @header ) {
                        @header = @$row;
+                       $header[0] =~ s/^#// if $path =~ m/\.tsv/i; # remove hash from 1st column
                        next;
                }
                my $item;