Created
August 19, 2012 15:14
-
-
Save spiculator/3395437 to your computer and use it in GitHub Desktop.
convert livelib.ru page like http://www.livelib.ru/reader/teak/print into zim-wiki format
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/perl -w | |
use strict; | |
use utf8; | |
use encoding "utf8"; | |
binmode( STDIN, ':utf8' ); | |
binmode( STDOUT, ':utf8' ); | |
use Encode qw(decode_utf8); | |
my @month_names = qw/Январь Февраль Март Апрель Май Июнь Июль Август Сентябрь Октябрь Ноябрь Декабрь/; | |
my %months = (); | |
$months{shift @month_names} = $_ foreach 1 .. 12; | |
sub format_date($) { | |
shift =~ /^(.*) (\d+)/ or die; | |
my ($year, $mname) = ($2, $1); | |
my $mnum = $months{$mname} or die; | |
$mnum = sprintf "%02d", $mnum; | |
return "[[Date:$year:$mnum|$year-$mnum]]"; | |
} | |
my ($date, $title, $author); | |
while(<>) { | |
$_ = decode_utf8( $_, Encode::FB_CROAK ); | |
if( /<tr><td colspan="5"><h2 style="margin-top: 14px; border:0;">(.*)<\/h2><\/td><\/tr>/ ) { | |
$date = format_date($1); | |
print "\n"; | |
} elsif( /<a href="\/book\/\d+" style="font-weight: bold">(.*)<\/a><br>/ ) { | |
die if defined $title; | |
$title = $1; | |
} elsif( /<a href="\/author\/\d+" style="font-size:80%">(.*)<\/a>/ ) { | |
die if defined $author; | |
$author = $1; | |
} elsif( /<span style="font-size:80%">(.*)<\/span>/ ) { | |
die if defined $author; | |
$author = $1; | |
} elsif( /^\s*<\/tr>\s*$/ ) { | |
die "no author" unless defined $author; | |
die "no title" unless defined $title; | |
die "no date" unless defined $date; | |
print "$date **$title** ($author)\n"; | |
undef $title; | |
undef $author; | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The thing I love about Perl is that I spent 6 lines out of 43 to make it work with unicode.