refresh header names on json load

[MojoFacets.git] / lib / MojoFacets / Data.pm

diff --git a/lib/MojoFacets/Data.pm b/lib/MojoFacets/Data.pm
index c5a8b8a..eebc114 100644 (file)
--- a/lib/MojoFacets/Data.pm
+++ b/lib/MojoFacets/Data.pm
@@ -10,6 +10,7 @@ use File::Slurp;
 use JSON;
 use Encode;
 use locale;
+use File::Find;
 
 sub index {
        my $self = shift;
@@ -17,11 +18,16 @@ sub index {
        my $path = $self->app->home->rel_dir('data');
        die "no data dir $path" unless -d $path;
 
-       opendir(my $dir, $path) || die $!;
-       my @files = 
-               grep { -f "$path/$_" && $_ =~ m/\.js(on)?$/ }
-               readdir $dir;
-       close($dir);
+       my @files;
+       find( sub {
+               my $file = $File::Find::name;
+               if ( -f $file && $file =~ m/\.(js(on)?|txt)$/ ) {
+                       $file =~ s/$path\/*//;
+                       push @files, $file;
+               } else {
+                       warn "IGNORE: $file\n";
+               }
+       }, $path);
 
        $self->render( files => [ @files ] );
 }
@@ -40,9 +46,49 @@ sub load {
        # we could use Mojo::JSON here, but it's too slow
 #      $data = from_json read_file $path;
        $data = read_file $path;
-       Encode::_utf8_on($data);
-       warn "# json snippet: ", substr($data,0,200);
-       $data = from_json $data;
+       warn "# data snippet: ", substr($data,0,200);
+       if ( $path =~ m/\.js/ ) {
+               Encode::_utf8_on($data);
+               $data = from_json $data;
+               $self->session( 'header' => 0 ); # generate later
+       } elsif ( $path =~ m/\.txt/ ) {
+               my @lines = split(/\r?\n/, $data);
+               $data = { items => [] };
+
+               my $headers = shift @lines;
+               my $multiline = $headers =~ s/\^//g;
+               my @header = split(/\|/, $headers );
+               warn "# header ", dump( @header );
+               $self->session( 'header' => [ @header ] );
+               $self->session( 'columns' => [ @header ] );
+               while ( my $line = shift @lines ) {
+                       chomp $line;
+                       my @v = split(/\|/, $line);
+                       while ( @lines && $#v < $#header ) {
+                               $line = $lines[0];
+                               $line =~ s/\^//g;
+                               chomp $line;
+                               my @more_v = split(/\|/, $line);
+                               if ( $#v + $#more_v > $#header ) {
+                                       warn "short line: ",dump( @v );
+                                       last;
+                               }
+                               shift @lines;
+                               $v[ $#v ] .= shift @more_v;
+                               push @v, @more_v if @more_v;
+
+                               if ( $#v > $#header ) {
+                                       die "# splice $#header ", dump( @v );
+                                       @v = splice @v, 0, $#header;
+                               }
+                       }
+                       my $item;
+                       $item->{ $header[$_] || "f_$_" } = [ $v[$_] ] foreach ( 0 .. $#v );
+                       push @{ $data->{items} }, $item;
+               }
+       } else {
+               warn "file format unknown $path";
+       }
 
        $stats = {};
 
@@ -71,8 +117,15 @@ sub load {
                        if $stats->{$n}->{array} == $stats->{$n}->{count};
        }
 
+       $self->session( 'header' => [
+               sort { $stats->{$b}->{count} <=> $stats->{$a}->{count} }
+               grep { defined $stats->{$_}->{count} } keys %$stats
+       ] ) unless $self->session( 'header' );
+
        warn dump($stats);
 
+       $self->session( 'filters' => {} );
+
        $self->redirect_to( '/data/columns' );
 }
 
@@ -80,16 +133,15 @@ sub load {
 sub columns {
     my $self = shift;
 
+       $self->redirect_to( '/data/index' ) unless $self->session('header');
 
        my @columns;
-       @columns = @{ $self->session('columns') } if $self->session('columns');
+       @columns = grep { defined $stats->{$_}->{count} } @{ $self->session('columns') } if $self->session('columns');
 
-       foreach my $c ( sort { $stats->{$b}->{count} <=> $stats->{$a}->{count} } keys %$stats ) {
+       foreach my $c ( @{ $self->session( 'header' ) } ) {
                push @columns, $c unless grep { /^\Q$c\E$/ } @columns;
        }
 
-       $self->redirect_to( '/data/index' ) unless @columns;
-
     $self->render(
                message => 'Select columns to display',
                stats => $stats,
@@ -157,7 +209,7 @@ sub filter {
 
        $self->session( 'offset' => 0 );
 
-       $self->redirect_to('/data/table');
+       $self->redirect_to('/data/items');
 }
 
 sub _filter_item {
@@ -196,7 +248,7 @@ sub _data_items {
        } @{ $data->{items} };
 }
 
-sub table {
+sub items {
     my $self = shift;
 
        $self->redirect_to('/data/index') unless $data->{items};
@@ -207,6 +259,7 @@ sub table {
        my $sort    = $self->_perm_scalar('sort', 'a');
        my $offset  = $self->_perm_scalar('offset', 0);
        my $limit   = $self->_perm_scalar('limit', 20);
+       $self->_perm_scalar('show', 'table');
 
        # fix offset when changing limit
        $offset = int( $offset / $limit ) * $limit;
@@ -243,7 +296,7 @@ sub order {
        my $self = shift;
        $self->session('order', $self->param('order'));
        $self->session('sort', $self->param('sort'));
-       $self->redirect_to('/data/table');
+       $self->redirect_to('/data/items');
 }
 
 sub _is_numeric {
@@ -261,7 +314,7 @@ sub facet {
                my $f = $self->session('filters');
                delete $f->{$remove};
                $self->session( 'filters' => $f );
-               $self->redirect_to( '/data/table' );
+               $self->redirect_to( '/data/items' );
        }
 
        my $facet;