Context Navigation

abc2_website @ 486

Revision 486, 23.5 kB (checked in by lincoln, 6 years ago)
fix microgap support in abc2_website - was expecting XMLTV date/time format not epoch format
Property svn:executable set to ``*

Line
1	#!/usr/bin/perl -w
2
3	# ABC/ABC2 au_tv guide grabber - runs from "Shepherd" master grabber
4	# * written by ltd
5	# * uses ABC website for ABC or ABC2 data
6	# * when used in conjunction with Shepherd, shepherd can collect other channels
7	# using other grabbers
8	# * this does NOT use any config file - all settings are passed in from shepherd
9
10	# changelog:
11	# 1.50 22sep06 added support for "shepherd" master grabber script
12	# 1.51 02oct06 --ready option
13	# 1.52 03oct06 split out abc grabber into its own grabber
14	# 1.55 09oct06 formalize --cheap option
15	# 1.56 20oct06 misc cleanups
16	# 1.60 11nov06 fix midday time calculation
17	# 1.70 16nov06 also use "printable" TV guide to determine 'station close'
18	# 2.00 23nov06 simplified
19
20	use strict;
21
22	my $progname = "abc2_website";
23	my $chan_id = "ABC2";
24	my $version = "2.06";
25
26	use LWP::UserAgent;
27	use XMLTV;
28	use POSIX qw(strftime mktime);
29	use Getopt::Long;
30	use HTML::TreeBuilder;
31	use Data::Dumper;
32	use Storable;
33
34	#
35	# constants
36	#
37	my $urls;
38	$urls->{station_close}->{ABC} = "http://www.abc.net.au/tv/guide/abctvweekguide.htm";
39	$urls->{station_close}->{ABC2} = "http://www.abc.net.au/tv/guide/abc2weekguide.htm";
40	$urls->{guide}->{ABC} = "http://www.abc.net.au/tv/guide/netw";
41	$urls->{guide}->{ABC2} = "http://www.abc.net.au/tv/guide/abc2";
42
43	#
44	# some initial cruft
45	#
46
47	my $script_start_time = time;
48	my $gmt_offset;
49	my %stats;
50	my $channels, my $opt_channels, my $gaps;
51	my $tv_guide;
52	my $data_cache;
53	my @station_close_data;
54	my $writer;
55	my %amp = ( nbsp => ' ', qw{ amp & lt < gt > apos ' quot " } );
56
57	my $ua;
58	$ua = LWP::UserAgent->new('timeout' => 30, 'keep_alive' => 30, 'agent' => "Shepherd / $progname $version");
59	$ua->env_proxy;
60	# $ua->cookie_jar({});
61	$ua->conn_cache(LWP::ConnCache->new());
62	$\| = 1;
63
64	#
65	# parse command line
66	#
67
68	my $opt_days = 7; # default
69	my $opt_offset = 0; # default
70	my $opt_timezone = "1000"; # default
71	my $opt_outputfile = $progname.".xmltv"; # default
72	my $opt_configfile = $progname.".conf"; # ignored
73	my $opt_cache_file = $progname.".storable.cache";
74	my $opt_old_cache_file = $progname.".cache";
75	my $opt_channels_file= "";
76	my $opt_gaps_file= "";
77	my $opt_no_cache = 0;
78	my $opt_cheap = 0;
79	my $opt_fast = 0;
80	my $opt_warper = 0;
81	my $opt_obfuscate = 0;
82	my $opt_do_extra_days = 0;
83	my $opt_help = 0;
84	my $opt_version = 0;
85	my $opt_desc = 0;
86	my $opt_dont_retry = 0;
87	my $debug = 0;
88	my $lang = "en";
89	my $region = 94;
90	my $time_offset = 0;
91
92	GetOptions(
93	'region=i' => \$region,
94	'days=i' => \$opt_days,
95	'offset=i' => \$opt_offset,
96	'timezone=s' => \$opt_timezone,
97	'channels_file=s' => \$opt_channels_file,
98	'gaps_file=s' => \$opt_gaps_file,
99	'output=s' => \$opt_outputfile,
100	'config-file=s' => \$opt_configfile,
101	'cache-file=s' => \$opt_cache_file,
102	'do-extra-days' => \$opt_do_extra_days,
103	'fast' => \$opt_fast,
104	'no-cache' => \$opt_no_cache,
105	'cheap' => \$opt_cheap,
106	'debug+' => \$debug,
107	'warper' => \$opt_warper,
108	'lang=s' => \$lang,
109	'obfuscate' => \$opt_obfuscate,
110	'no-retry' => \$opt_dont_retry,
111	'help' => \$opt_help,
112	'verbose' => \$opt_help,
113	'version' => \$opt_version,
114	'ready' => \$opt_version,
115	'desc' => \$opt_desc,
116	'v' => \$opt_help);
117
118	&help if ($opt_help);
119
120	if ($opt_version \|\| $opt_desc) {
121	printf "%s %s\n",$progname,$version;
122	printf "%s is a details-aware grabber that collects decent quality data using the ABC website for %s only.",$progname,$chan_id if $opt_desc;
123	exit(0);
124	}
125
126	die "no channel file specified, see --help for instructions\n", if ($opt_channels_file eq "");
127
128	#
129	# go go go!
130	#
131
132	my $starttime = time;
133
134	&log(sprintf "going to %s%s %s%d%s days%s of data into %s (%s%s)",
135	($opt_gaps_file ne "" ? "micro-gap " : ""),
136	($opt_cheap ? "verify (cache-validate)" : "grab"),
137	($opt_do_extra_days ? "somewhere between " : ""),
138	$opt_days,
139	($opt_do_extra_days ? " to 14" : ""),
140	($opt_offset ? " (skipping first $opt_offset days)" : ""),
141	$opt_outputfile,
142	($opt_no_cache ? "without caching" : "with caching"),
143	($opt_warper ? ", anonymously" : ""));
144
145	# read channels file
146	if (-r $opt_channels_file) {
147	local (@ARGV, $/) = ($opt_channels_file);
148	no warnings 'all'; eval <>; die "$@" if $@;
149	} else {
150	die "WARNING: channels file $opt_channels_file could not be read: $!\n";
151	}
152	die "nothing to do; $chan_id not in channels lineup!\n" if (!defined $channels->{$chan_id});
153
154	# if just filling in microgaps, parse gaps
155	if ($opt_gaps_file ne "") {
156	if (-r $opt_gaps_file) {
157	local (@ARGV, $/) = ($opt_gaps_file);
158	no warnings 'all'; eval <>; die "$@" if $@;
159	} else {
160	die "WARNING: gaps_file $opt_gaps_file could not be read: $!\n";
161	}
162	}
163
164	&read_cache if ($opt_no_cache == 0);
165
166	my %writer_args = ( encoding => 'ISO-8859-1' );
167	my $fh = new IO::File(">$opt_outputfile") \|\| die "can't open $opt_outputfile: $!";
168	$writer_args{OUTPUT} = $fh;
169
170	$writer = new XMLTV::Writer(%writer_args);
171	$writer->start( { 'source-info-name' => "$progname $version", 'generator-info-name' => "$progname $version"} );
172	$writer->write_channel( { 'display-name' => [[ $chan_id, $lang ]], 'id' => $channels->{$chan_id} } );
173
174	&get_station_close($channels->{$chan_id}, $urls->{station_close}->{$chan_id});
175	&get_abc_data($channels->{$chan_id}, $urls->{guide}->{$chan_id});
176	&write_cache if ($opt_no_cache == 0);
177
178	$writer->end;
179
180	&print_stats;
181	exit(0);
182
183	######################################################################################################
184	# help
185
186	sub help
187	{
188	print<<EOF
189	$progname $version
190
191	options are as follows:
192	--help show these help options
193	--days=N fetch 'n' days of data (default: $opt_days)
194	--output=file send xml output to file (default: "$opt_outputfile")
195	--config-file=file (ignored - historically used by grabbers not not this one)
196	--no-cache don't use a cache to optimize (reduce) number of web queries
197	--cheap validate contents of cache - fetch summary only, not details
198	--cache-file=file where to store cache (default "$opt_cache_file")
199	--fast don't run slow - get data as quick as you can - not recommended
200	--debug increase debug level
201	--warper fetch data using WebWarper web anonymizer service
202	--obfuscate pretend to be a proxy servicing multiple clients
203	--do-extra-days fetch extra (14 days) from ABC website
204	--no-retry if webserver is rejecting our request, don't retry (default: do retry)
205	--lang=[s] set language of xmltv output data (default $lang)
206
207	--region=N set region for where to collect data from (default: $region)
208	--channels_file=file where to get channel data from (if not set manually)
209	--timezone=HHMM timezone for channel data (default: $opt_timezone)
210	EOF
211	;
212
213	exit(0);
214	}
215
216	######################################################################################################
217	# populate cache
218
219	sub read_cache
220	{
221	# convert old-style Data::Dumper cache to new format
222	if (-r $opt_old_cache_file) {
223	local (@ARGV, $/) = ($opt_old_cache_file);
224	no warnings 'all'; eval <>; die "$@" if $@;
225
226	&write_cache;
227	unlink $opt_old_cache_file;
228	}
229
230	if (-r $opt_cache_file) {
231	my $store = Storable::retrieve($opt_cache_file);
232	$data_cache = $store->{data_cache};
233	} else {
234	printf "WARNING: no programme cache $opt_cache_file - have to fetch all details\n";
235
236	# try to write to it - if directory doesn't exist this will then cause an error
237	&write_cache;
238	}
239	}
240
241	######################################################################################################
242	# write out updated cache
243
244	sub write_cache
245	{
246	# cleanup old entries from cache
247	for my $cache_key (keys %{$data_cache}) {
248	my ($starttime, @rest) = split(/,/,$cache_key);
249	if ($starttime < (time-86400)) {
250	delete $data_cache->{$cache_key};
251	$stats{removed_items_from_cache}++;
252	}
253	}
254
255	my $store;
256	$store->{data_cache} = $data_cache;
257	Storable::store($store, $opt_cache_file);
258	}
259
260	######################################################################################################
261
262	sub get_abc_data
263	{
264	my ($xmlid,$urlbase) = @_;
265	my $try_to_add_abc_detail;
266	my $unprocessed_programmes = 0;
267	my $stop_fetching = 0;
268	my @unprocessed_progname, my @unprocessed_starttime, my @unprocessed_url;
269
270	my $days_left = $opt_days;
271	my $to_skip = $opt_offset;
272	my $daynum = 0;
273	my @gap_s, my @gap_e;
274
275	while ($days_left > 0) {
276	my $currtime = $starttime + ($daynum * 86400);
277	$days_left--;
278	$daynum++;
279
280	if ($to_skip > 0) {
281	$to_skip--;
282	next;
283	}
284
285	if ($opt_gaps_file ne "") { # micro-gap mode!
286	my $found_gap_match = 0;
287
288	if ((defined $gaps) && (defined $gaps->{$chan_id})) {
289	foreach my $g (@{($gaps->{$chan_id})}) {
290	my ($s, $e) = split(/-/,$g);
291	if (($s >= $currtime) && ($e <= ($currtime+86400))) {
292	$found_gap_match++;
293	push(@gap_s,$s);
294	push(@gap_e,$e);
295	printf "including day %d channel '%s' gap start %d, gap end %d\n",
296	$daynum, $chan_id, $s, $e if $debug;
297	}
298	}
299	}
300	next if (!$found_gap_match); # no gaps for this day - skip!
301	}
302
303	my @timeattr = localtime($currtime); # 0=sec,1=min,2=hour,3=day,4=month,5=year,6=wday,7=yday,8=isdst
304	$timeattr[0] = 0; # zero seconds
305
306	my $url = sprintf "%s/%s.htm",$urlbase, POSIX::strftime("%Y%m/%Y%m%d",localtime($currtime));
307
308	my $data;
309	my $tree;
310	my $tries = 0;
311	while (($tries < 5) && (!defined $tree)) {
312	$tries++;
313	&log((sprintf "fetching %s summary data: day %d of %d%s",
314	$xmlid, $daynum, $opt_days, ($tries > 1 ? " (try $tries)" : "")));
315	$data = &get_url($url);
316	$tree = HTML::TreeBuilder->new_from_content($data) if ((defined $data) && ($data ne ""));
317
318	if ((!$tree) && ($tries < 5)) {
319	# slow down ..
320	my $waittime = ($tries*20)+int(rand(20));
321	&log("failed to fetch $url, retrying in $waittime secs");
322	sleep($waittime);
323	$stats{slept_for} += $waittime;
324	}
325	}
326	if (!defined $tree) {
327	&log("failed to fetch $url after $tries attempts; skipping");
328	next;
329	}
330
331	my $seen_programmes = 0;
332	my $seen_pm = 0;
333
334	for ($tree->look_down('_tag' => 'div', 'class' => 'scheduleDiv')) {
335	foreach my $tree_tr ($_->look_down('_tag' => 'tr')) {
336	if (my $tree_row = $tree_tr->look_down('_tag' => 'th', 'scope' => 'row')) {
337	if ($tree_row->as_text() =~ /^(\d+):(\d+)(.)m/) {
338	$timeattr[2] = $1; # hour
339	$timeattr[1] = $2; # min
340
341	if ($3 eq "p") {
342	# pm
343	$timeattr[2] += 12 if ($timeattr[2] != 12);
344	$seen_pm = 1;
345	}
346	my $found_time = mktime(@timeattr);
347
348	# handle programmes that are after midnight
349	if (($seen_pm) && ($3 eq "a")) {
350	if ($timeattr[2] == 12) {
351	$found_time += (126060); # 12:xx am
352	} else {
353	$found_time += (246060);
354	}
355	}
356
357	if ($tree_tr->look_down('_tag' => 'td')) {
358	foreach my $prog ($tree_tr->look_down('_tag' => 'a')) {
359	my $programme = $prog->as_text();
360	my $progurl = $prog->attr('href');
361
362	if ($progurl =~ /^\/tv\/guide\//) {
363	printf "day %d time '%s' (%s) prog: %s url: %s\n",
364	$daynum,$tree_row->as_text(),POSIX::strftime("%Y%m%d%H%M", localtime($found_time)),
365	$programme,$progurl if ($debug && $debug > 1);
366
367	$unprocessed_progname[$unprocessed_programmes] = $programme;
368	$unprocessed_starttime[$unprocessed_programmes] = $found_time;
369	$unprocessed_url[$unprocessed_programmes] = "http://www.abc.net.au".$progurl;
370	$unprocessed_programmes++;
371	$seen_programmes++;
372	} else {
373	printf "ignoring prog %s because url %s is not a detail page\n",
374	$programme,$progurl if $debug;
375	}
376	}
377	}
378	}
379	}
380	}
381	}
382
383	if ($seen_programmes > 0) {
384	$stats{abc_daily_pages}++;
385
386	if (defined $station_close_data[$daynum]) {
387	# get station-close time from the previously-fetched "weekly programme guide"
388
389	$unprocessed_progname[$unprocessed_programmes] = "Station Close";
390	$unprocessed_starttime[$unprocessed_programmes] = $station_close_data[$daynum];
391	$unprocessed_url[$unprocessed_programmes] = "";
392	$unprocessed_programmes++;
393	} else {
394	# throw away last programme from each day - we can't use it as
395	# we don't have a 'stop' time for it
396
397	printf "throwing away '%s' (%s) because we won't have a valid stop time\n",
398	$unprocessed_progname[$unprocessed_programmes-1],
399	POSIX::strftime("%Y%m%d%H%M", localtime($unprocessed_starttime[$unprocessed_programmes-1]))
400	if $debug;
401	$unprocessed_progname[$unprocessed_programmes-1] = "";
402	}
403	} else {
404	# if we were trying to fetch more than 7 days, stop on first day with no programmes
405	goto STOP_FETCHING if ($daynum > 7);
406	}
407	}
408
409	STOP_FETCHING:
410	# have 'n' days of this channel unprocessed - process it!
411	&log((sprintf "have summary data, now fetching detail pages for up to %d programmes..",$unprocessed_programmes-2));
412
413	for (my $i = 0; $i < ($unprocessed_programmes-1); $i++) {
414	next if ($unprocessed_progname[$i] eq "");
415
416	# if we are micro-gap fetching, only include programmes which match our micro gaps
417	if ($opt_gaps_file ne "") {
418	my $found_gap_match = 0;
419	for (my $g_num = 0; $g_num < $#gap_s; $g_num++) {
420	$found_gap_match++
421	if ((($gap_s[$g_num] <= $unprocessed_starttime[$i]) &&
422	($gap_s[$g_num] <= $unprocessed_starttime[$i+1])) \|\|
423	(($gap_e[$g_num] <= $unprocessed_starttime[$i]) &&
424	($gap_e[$g_num] <= $unprocessed_starttime[$i+1])));
425	}
426	next if (!$found_gap_match);
427
428	$stats{programme_gaps_used}++;
429	printf "gap-fetching: including prog '%s', start %d, end %d\n", $unprocessed_progname[$i],
430	$unprocessed_starttime[$i], $unprocessed_starttime[$i+1] if $debug;
431	}
432
433	$stats{programmes}++;
434	my $prog;
435
436	my $cache_key = sprintf "%d,%d,%s,%s", $unprocessed_starttime[$i], $unprocessed_starttime[$i+1], $xmlid, $unprocessed_progname[$i];
437
438	$prog->{'channel'} = $xmlid;
439	$prog->{'start'} = POSIX::strftime("%Y%m%d%H%M", localtime($unprocessed_starttime[$i]));
440	$prog->{'stop'} = POSIX::strftime("%Y%m%d%H%M", localtime($unprocessed_starttime[$i+1]));
441	$prog->{'title'} = [[ $unprocessed_progname[$i], $lang ]];
442
443	if (defined $data_cache->{$cache_key}) {
444	$stats{used_cached_data}++;
445	} else {
446	if ((!$opt_cheap) && ($unprocessed_url[$i] ne "")) {
447	$stats{portal_detail_pages}++;
448	&get_one_abc_event($cache_key, $unprocessed_url[$i]);
449
450	if (($stats{portal_detail_pages} % 25) == 1) {
451	&log((sprintf " .. at %s detail page %d of %d (used %d cached entries)",
452	$xmlid, ($i+1), $unprocessed_programmes-2,
453	(defined $stats{used_cached_data} ? $stats{used_cached_data} : 0)));
454
455	if (!$opt_fast) {
456	# slow down ..
457	my $waittime = 3 + int(rand(10));
458	sleep($waittime);
459	$stats{slept_for} += $waittime;
460	}
461	}
462	}
463	}
464
465	if (defined $data_cache->{$cache_key}) {
466	$prog->{'sub-title'} = [[ $data_cache->{$cache_key}->{subtitle}, $lang ]]
467	if $data_cache->{$cache_key}->{subtitle};
468	$prog->{'desc'} = [[ $data_cache->{$cache_key}->{desc}, $lang ]]
469	if $data_cache->{$cache_key}->{desc};
470	$prog->{'category'} = [[ $data_cache->{$cache_key}->{genre}, $lang ]]
471	if $data_cache->{$cache_key}->{genre};
472	$prog->{'previously-shown'} = { } if (defined $data_cache->{$cache_key}->{repeat});
473	$prog->{'subtitles'} = [ { 'type' => 'teletext' } ] if (defined $data_cache->{$cache_key}->{cc});
474	$prog->{'rating'} = [ [ $data_cache->{$cache_key}->{rating}, 'ABA', undef] ]
475	if (defined $data_cache->{$cache_key}->{rating});
476	}
477
478	&cleanup($prog);
479	$writer->write_programme($prog);
480	}
481	}
482
483	######################################################################################################
484
485	sub get_one_abc_event
486	{
487	my ($cache_key, $url) = @_;
488	my $seen_programme = 0;
489
490	my $tries = 0;
491	my $tree;
492	while (($tries < 5) && (!$tree)) {
493	$tries++;
494	my $data = &get_url($url);
495	$tree = HTML::TreeBuilder->new_from_content($data) if ((defined $data) && ($data ne ""));
496
497	if ((!$tree) && ($tries < 5)) {
498	# slow down ..
499	my $waittime = ($tries*20)+int(rand(20));
500	&log("failed to fetch $url on try $tries, retrying in $waittime secs");
501	sleep($waittime);
502	$stats{slept_for} += $waittime;
503	}
504	}
505	if (!defined $tree) {
506	&log("failed to fetch $url after $tries attempts; skipping");
507	return;
508	}
509
510	if (my $inner_tree = $tree->look_down('_tag' => 'div', 'class' => 'column2')) {
511	my $event_title = undef, my $event_subtitle = undef, my $event_description = undef, my $event_genre = undef;
512
513	if (my $prog_h2 = $inner_tree->look_down('_tag' => 'h2')) {
514	my $full_title = $prog_h2->as_HTML();
515	($event_title,$event_subtitle) = split(/<br>/,$full_title);
516
517	$event_title =~ s/(<[a-zA-Z0-9]+\>)//g; # remove html tags
518	$event_title =~ s/(^\n\|\n$)//g; # strip trailing/leading blank lines
519
520	if ($event_subtitle) {
521	$event_subtitle =~ s/(<[\/a-zA-Z0-9]+\>)//g; # remove html tags
522	$event_subtitle =~ s/(^\n\|\n$)//g; # strip trailing/leading blank lines
523	$data_cache->{$cache_key}->{subtitle} = $event_subtitle;
524	}
525	}
526
527	my $paranum = 0;
528	my $seen_genre = 0;
529	foreach my $para ($inner_tree->look_down('_tag' => 'p')) {
530	$paranum++;
531
532	if (($paranum > 1) && (!($para->as_text() =~ /^Go to website/)) && (!($para->as_text() =~ /^Send to a Friend/))) {
533	if (my $try_genre = $para->look_down('_tag' => 'a')) {
534	$data_cache->{$cache_key}->{genre} = $try_genre->as_text();
535	$seen_genre = 1;
536	}
537
538	if (!$seen_genre) {
539	$data_cache->{$cache_key}->{desc} .= $para->as_text() . "\n";
540	} else {
541	$data_cache->{$cache_key}->{repeat} = 1 if ($para->as_text() =~ /Repeat/);
542	$data_cache->{$cache_key}->{cc} = 1 if ($para->as_text() =~ /CC/);
543	$data_cache->{$cache_key}->{rating} = $1 if ($para->as_text() =~ /(M\|PG\|G)/);
544	}
545	}
546	}
547
548	if (defined $data_cache->{$cache_key}->{desc}) {
549	$data_cache->{$cache_key}->{desc} =~ s/(^\n\|\n$)//g; # strip trailing/leading blank lines
550	$data_cache->{$cache_key}->{desc} =~ s/(^\s+\|\s+$)//g; # strip trailing/leading spaces
551	delete $data_cache->{$cache_key}->{desc} if ($data_cache->{$cache_key}->{desc} eq "");
552	}
553
554	$seen_programme++;
555	$stats{added_cached_data}++;
556
557	&write_cache if (($opt_no_cache == 0) &&
558	(($stats{added_cached_data} % 30) == 0)); # incrementally write
559	}
560
561	if ($seen_programme == 0) {
562	printf "WARNING: failed to parse any programme data from '%s' - blocked/rate-limited/format-changed?\n",$url;
563	$stats{failed_to_parse_portal_detail_page}++;
564	}
565	}
566
567	######################################################################################################
568	# logic to fetch a page via http
569
570	sub get_url
571	{
572	my $url = shift;
573	my $response;
574	my $attempts = 0;
575	my ($raw, $page, $base);
576
577	$url =~ s#^http://#http://webwarper.net/ww/# if $opt_warper;
578	my $request = HTTP::Request->new(GET => $url);
579	$request->header('Accept-Encoding' => 'gzip');
580
581	if ($opt_obfuscate) {
582	$ua->agent('Mozilla/5.0 (Windows; U; Windows NT 5.1; en-us');
583	my $randomaddr = sprintf "203.%d.%d.%d",rand(255),rand(255),(rand(254)+1);
584	$request->header('Via' => '1.0 proxy:81 (Squid/2.3.STABLE3)');
585	$request->header('X-Forwarded-For' => $randomaddr);
586	}
587	$response = $ua->request($request);
588	if (!($response->is_success)) {
589	$stats{http_failed_requests}++;
590	return undef;
591	}
592
593	$stats{bytes_fetched} += do {use bytes; length($response->content)};
594	$stats{http_successful_requests}++;
595
596	if ($response->header('Content-Encoding') &&
597	$response->header('Content-Encoding') eq 'gzip') {
598	$stats{compressed_pages} += do {use bytes; length($response->content)};
599	$response->content(Compress::Zlib::memGunzip($response->content));
600	}
601	return $response->content;
602	}
603
604	######################################################################################################
605
606	sub log
607	{
608	my ($entry) = @_;
609	printf "[%d] %s\n", time, $entry;
610	}
611
612	######################################################################################################
613
614	sub print_stats
615	{
616	printf "STATS: %s v%s completed in %d seconds", $progname, $version, (time-$script_start_time);
617	foreach my $key (sort keys %stats) {
618	printf ", %d %s",$stats{$key},$key;
619	}
620	printf "\n";
621	}
622
623	######################################################################################################
624	# descend a structure and clean up various things, including stripping
625	# leading/trailing spaces in strings, translations of html stuff etc
626	# -- taken & modified from Michael 'Immir' Smith's excellent tv_grab_au
627
628	sub cleanup {
629	my $x = shift;
630	if (ref $x eq "REF") { cleanup($_) }
631	elsif (ref $x eq "HASH") { cleanup(\$_) for values %$x }
632	elsif (ref $x eq "ARRAY") { cleanup(\$_) for @$x }
633	elsif (defined $$x) {
634	$$x =~ s/&(#(\d+)\|(.*?));/ $2 ? chr($2) : $amp{$3}\|\|' ' /eg;
635	$$x =~ s/[^\x20-\x7f]/ /g;
636	$$x =~ s/(^\s+\|\s+$)//g;
637	}
638	}
639
640	######################################################################################################
641
642	# strptime type date parsing - BUT - if no timezone is present, treat time as being in localtime
643	# rather than the various other perl implementation which treat it as being in UTC/GMT
644	sub parse_xmltv_date
645	{
646	my $datestring = shift;
647	my @t; # 0=sec,1=min,2=hour,3=day,4=month,5=year,6=wday,7=yday,8=isdst
648	my $tz_offset = 0;
649
650	# work out GMT offset - we only do this once
651	if (!$gmt_offset) {
652	my $tzstring = strftime("%z", localtime(time));
653
654	$gmt_offset = (6060) int(substr($tzstring,1,2)); # hr
655	$gmt_offset += (60 * int(substr($tzstring,3,2))); # min
656	$gmt_offset *= -1 if (substr($tzstring,0,1) eq "-"); # +/-
657	}
658
659	if ($datestring =~ /^(\d{4})(\d{2})(\d{2})(\d{2})(\d{2})/) {
660	($t[5],$t[4],$t[3],$t[2],$t[1],$t[0]) = (int($1)-1900,int($2)-1,int($3),int($4),int($5),0);
661	($t[6],$t[7],$t[8]) = (-1,-1,-1);
662
663	# if input data has a timezone offset, then offset by that
664	if ($datestring =~ /\+(\d{2})(\d{2})/) {
665	$tz_offset = $gmt_offset - (($1(6060)) + ($2*60));
666	} elsif ($datestring =~ /\-(\d{2})(\d{2})/) {
667	$tz_offset = $gmt_offset + (($1(6060)) + ($2*60));
668	}
669
670	my $e = mktime(@t);
671	return ($e+$tz_offset) if ($e > 1);
672	}
673	return undef;
674	}
675
676	######################################################################################################
677
678	sub get_station_close
679	{
680	my ($xmlid,$url) = @_;
681	my $tries = 0;
682	my $tree;
683	while (($tries < 5) && (!$tree)) {
684	$tries++;
685	&log("fetching (weekly) station close data for $xmlid");
686	my $data = &get_url($url);
687	$tree = HTML::TreeBuilder->new_from_content($data) if ((defined $data) && ($data ne ""));
688
689	if ((!$tree) && ($tries < 5)) {
690	# slow down ..
691	my $waittime = ($tries*20)+int(rand(20));
692	&log("failed to fetch $url, retrying in $waittime secs");
693	sleep($waittime);
694	$stats{slept_for} += $waittime;
695	}
696	}
697	if (!defined $tree) {
698	&log("failed to fetch $url after $tries attempts; skipping");
699	return;
700	}
701
702	my $to_skip = $opt_offset;
703	my $daynum = 0;
704	my $last_td_text;
705
706	foreach my $tree_td ($tree->look_down('_tag' => 'td')) {
707	if ($tree_td->as_text() =~ /^\.\.\.programs start at /) {
708	if (defined $last_td_text) {
709	if ($to_skip > 0) {
710	$to_skip--;
711	} else {
712	# 0=sec,1=min,2=hour,3=day,4=month,5=year,6=wday,7=yday,8=isdst
713	my @timeattr = localtime($starttime + ($daynum*86400));
714	$timeattr[0] = 0; # zero seconds
715
716	if ($last_td_text =~ /^(\d+):(\d+)(.)m/) {
717	$timeattr[2] = $1; # hour
718	$timeattr[1] = $2; # min
719
720	if ($3 eq "p") {
721	# pm
722	$timeattr[2] += 12 if ($timeattr[2] != 12);
723	}
724	my $found_time = mktime(@timeattr);
725
726	if ($3 eq "a") {
727	# am - must be tomorrow
728	if ($timeattr[2] == 12) {
729	$found_time += (126060); # 12:xx am
730	} else {
731	$found_time += (246060);
732	}
733	}
734
735	$daynum++;
736	$station_close_data[$daynum] = $found_time;
737
738	printf "station close time for day %d is %s\n",
739	$daynum, POSIX::strftime("%Y%m%d%H%M", localtime($found_time))
740	if $debug;
741	}
742	}
743	}
744	}
745	$last_td_text = $tree_td->as_text();
746	}
747	}
748
749	######################################################################################################

Note: See TracBrowser for help on using the browser.

Context Navigation

root/grabbers/abc2_website @ 486

Download in other formats: