lav-outils/podcasts/scripts/make-all-podcasts.pl

#!/usr/bin/perl

# Copyright (C) 2019 Quentin GIBEAUX <qgibeaux@april.org>
# Copyright (C) 2019 Frédéric COUCHET <fcouchet@april.org>
# Copyright (C) 2019 Christian Pierre MOMON <cmomon@april.org>
#
# This file is part of lav-outils from "April/Libre à vous !"
#
# This script is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as
# published by the Free Software Foundation, either version 3 of the
# License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.

use strict;
use Getopt::Long;
use JSON;
use Data::Dumper;
use utf8;

my ($help,$config,$verbose,$dryrun);
my $meta_data_script = "lav-outils/podcasts/scripts/make-metadata-image-podcast.sh";

my $verbose;
my $mp3;
my $textwebpage="<ul>\n";

GetOptions ("help" => \$help,
	"config=s"   => \$config,
	"mp3"  => \$mp3,
	"verbose"  => \$verbose,
	"dryrun"  => \$dryrun);

if($help) {
	usage();
} elsif( not $config ) {
	print " /!\\ Missing config arg\n\n";
	usage();
}

sub usage {
	print <<EOS
	Exec commands to cut and process LAV! podcast with json file containing the timestamps and sections to be cut.
	Needs JSON perl library (apt install libjson-perl)
$0 --config conf_file.json
  --config conf_file.json	the configuration file containing stuff
  --mp3 			generate mp3 files
  --help 			show this message
  --verbose			increase verbosity
  --dryrun			print commands without executing them
EOS
	  ;
}

sub read_config {
	my ($filename) = @_;

	my $json_text = do { #read all the file in one shot
		open(my $json_fh, '<:encoding(UTF-8)', $filename)
		  or die("Can't open \$filename\": $!\n");
		local $/;
		<$json_fh>
	  };

	my $json = JSON->new;
	my $data = $json->decode($json_text);

	return $data;
}

sub process {
	my ($config,$mp3,$verbose,$dryrun)=@_;
	my $data = read_config($config);

	my $short_date = $data->{short_date};
	my $year = substr $short_date,0,4;
	my $long_date = $data->{long_date};
	my $source_name = "libre-a-vous-$short_date";
	my $title = "Libre à vous ! du $long_date sur Cause Commune";
    my $short_description = $data->{short_description};
	my $ffmpeg_bin = $data->{ffmpeg_bin};

	my $option_mp3_meta_data_script;
	my @formats_files;

	if($mp3) {
		@formats_files = qw/ogg mp3/;
		$option_mp3_meta_data_script = "yes";
	}
	else {
		@formats_files = qw/ogg/;
		$option_mp3_meta_data_script = "no";
	}

	for my $chapter (values @{$data->{chapters}}) {
		my $start = $chapter->{start_timestamp};
		my $end = $chapter->{end_timestamp};
		my $short_chapter_name = $chapter->{short_chapter_name};
		my $chapter_title = $chapter->{chapter_title};

		# cutting chapter
		my $command = "$ffmpeg_bin -y -i $source_name.ogg -vn -acodec copy -ss \"$start\" -to \"$end\" $source_name-$short_chapter_name.ogg";
		if($dryrun) {
			print "$command\n";
		} else {
			my @ret = `$command`;
			if($?) {
				print "Error while cutting $short_chapter_name\n";
				if($verbose) {
					print Dumper @ret;
					print Dumper $data;
				}
				return 0;
			}
		}

		# putting metadata
		my $url = "https://media.april.org/audio/radio-cause-commune/libre-a-vous/emissions/$short_date/$source_name-$short_chapter_name.ogg";
		my $command = "$meta_data_script -s \"$source_name-$short_chapter_name.ogg\" -d \"output\" -u \"$url\" -t \"$title - Partie $chapter_title\" -p \"$ffmpeg_bin\" -y \"$year\" -m \"$option_mp3_meta_data_script\"";

		if($dryrun) {
			print "$command\n";
		} else {

			my @ret = `$command`;
			if($?) {
				print "Error while setting metadata in $short_chapter_name\n";
				if($verbose) {
					print Dumper @ret;
					print Dumper $data;
				}
				return 0;
			}
		}

		# Update string for web site

		my $command = "mediainfo --Inform=\"Audio;%Duration/String%\" output.ogg";

		if($dryrun) {
			print "$command\n";
		} else {
			my @ret = `$command`;
			if($?) {
				print "Error while updating string for web site\n";
				if($verbose) {
					print Dumper @ret;
					print Dumper $data;
				}
				return 0;
			} else {
				chomp($ret[0]);
				$ret[0] =~ s/s/secondes/ig;
				$ret[0] =~ s/min/minutes/ig;
				my $url2 = $url =~ s/\.ogg/\.mp3/r;
				$textwebpage = $textwebpage . "<li><a href=\"$url\">$chapter_title</a> (format OGG) (et <a href=\"$url2\">format MP3</a>) ($ret[0])</li>\n";
			}
		}

		# renaming to target OGG and MP3
		foreach my $format (@formats_files) {
			my $target_name = "$source_name-$short_chapter_name.$format";
			my $command = "mv output.$format $target_name";
			if($dryrun) {
				print "$command\n";
			} else {
				my @ret = `$command`;
				if($?) {
					print "Error while renaming $short_chapter_name.$format\n";
					if($verbose) {
						print Dumper @ret;
						print Dumper $data;
					}
					return 0;
				}
			}

			# hashing
			my $command = "sha1sum $target_name > $target_name.sha1";
			if($dryrun) {
				print "$command\n";
			} else {
				my @ret = `$command`;
				if($?) {
					print "Error while hashing $target_name\n";
					if($verbose) {
						print Dumper @ret;
						print Dumper $data;
					}
					return 0;
				}
			}

			# generate .size
			my $command = "du -b $target_name |awk '{print \$1}' > $target_name.size";
			if($dryrun) {
				print "$command\n";
			} else {
				my @ret = `$command`;
				if($?) {
					print "Error while generation size of $target_name\n";
					if($verbose) {
						print Dumper @ret;
						print Dumper $data;
					}
					return 0;
				}
			}

         # generate duration
#			my $command = "ffprobe $target_name -show_entries format=duration -sexagesimal 2> /dev/null |grep duration |sed '1,\$s/duration=//' > $target_name.duration";
			my $command = "mediainfo --Inform='General;%Duration/String3%' $target_name > $target_name.duration";
			if($dryrun) {
				print "$command\n";
			} else {
				my @ret = `$command`;
				if($?) {
					print "Error while generation duration of $target_name\n";
					if($verbose) {
						print Dumper @ret;
						print Dumper $data;
					}
					return 0;
				}
			}

		}

	}

	# putting metadata in main podcast
	my $url = "https://media.april.org/audio/radio-cause-commune/libre-a-vous/emissions/$short_date/$source_name.ogg";
	my $command = "$meta_data_script -s \"$source_name.ogg\" -d \"output\" -u \"$url\" -t \"$title - $short_description\" -p \"$ffmpeg_bin\" -y \"$year\" -m \"$option_mp3_meta_data_script\"";
	if($dryrun) {
		print "$command\n";
	} else {
		my @ret = `$command`;
		if($?) {
			print "Error while setting metadata in $source_name\n";
			if($verbose) {
				print Dumper @ret;
				print Dumper $data;
			}
			return 0;
		}
	}

	# renaming to target OGG and MP3
	foreach my $format (@formats_files) {

		my $command = "mv output.$format $source_name.$format";
		if($dryrun) {
			print "$command\n";
		} else {
			my @ret = `$command`;
			if($?) {
				print "Error while renaming $source_name.$format\n";
				if($verbose) {
					print Dumper @ret;
					print Dumper $data;
				}
				return 0;
			}
		}

		# hashing
		my $command = "sha1sum $source_name.$format > $source_name.$format.sha1";
		if($dryrun) {
			print "$command\n";
		} else {
			my @ret = `$command`;
			if($?) {
				print "Error while hashing $source_name.$format\n";
				if($verbose) {
					print Dumper @ret;
					print Dumper $data;
				}
				return 0;
			}
		}
      # duration 
      my $command = "mediainfo --Inform=\"Audio;%Duration/String3%\" $source_name.$format > $source_name.$format.duration";
      if($dryrun) {
          print "$command\n";
      } else {
          my @ret = `$command`;
          if($?) {
              print "Error while generation duration of $source_name.$format\n";
              if($verbose) {
						print Dumper @ret;
						print Dumper $data;
              }
              return 0;
          }
      }

      # size
      my $command = "du -b $source_name.$format |awk '{print \$1}' > $source_name.$format.size";
      if($dryrun) {
          print "$command\n";
      } else {
          my @ret = `$command`;
          if($?) {
              print "Error while generation size of $source_name.$format\n";
              if($verbose) {
						print Dumper @ret;
						print Dumper $data;
              }
              return 0;
          }
      }

      
	}

	$textwebpage = $textwebpage . "</ul>\n\n";

	binmode(STDOUT, ":utf8");

	my $url = "https://www.april.org/libre-a-vous-diffusee-mardi-$long_date-sur-radio-cause-commune";
	$url=~s/ /-/g;

	print "\nText for the web page of the radio program : $url\n\n";
	print "$textwebpage\n\n";
	print "Add the html code on $url\n\n";

	print "And remove the line : <p>Les podcasts seront disponibles après la diffusion de l'émission (quelques jours après en général).</p>\n\n";

	if(! $mp3) {
		print "MP3 files not generated, please use --mp3 option to generate them\n";
	}

}

process($config,$mp3,$verbose,$dryrun);