aboutsummaryrefslogtreecommitdiff
path: root/IkiWiki/Plugin/aggregate.pm
blob: cb165acd2a0e7bb2c647d990d7723bf18a33b291 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
#!/usr/bin/perl
# Blog aggregation plugin.
package IkiWiki::Plugin::aggregate;

use warnings;
use strict;
use IkiWiki 2.00;
use HTML::Parser;
use HTML::Tagset;
use HTML::Entities;
use URI;
use open qw{:utf8 :std};

my %feeds;
my %guids;

sub import { #{{{
	hook(type => "getopt", id => "aggregate", call => \&getopt);
	hook(type => "checkconfig", id => "aggregate", call => \&checkconfig);
	hook(type => "needsbuild", id => "aggregate", call => \&needsbuild);
	hook(type => "preprocess", id => "aggregate", call => \&preprocess);
        hook(type => "delete", id => "aggregate", call => \&delete);
	hook(type => "savestate", id => "aggregate", call => \&savestate);
} # }}}

sub getopt () { #{{{
        eval q{use Getopt::Long};
	error($@) if $@;
        Getopt::Long::Configure('pass_through');
        GetOptions("aggregate" => \$config{aggregate});
} #}}}

sub checkconfig () { #{{{
	if ($config{aggregate} && ! ($config{post_commit} && 
	                             IkiWiki::commit_hook_enabled())) {
		# See if any feeds need aggregation.
		loadstate();
		my @feeds=needsaggregate();
		return unless @feeds;
		if (! lockaggregate()) {
			debug("an aggregation process is already running");
			return;
		}
		# force a later rebuild of source pages
		$IkiWiki::forcerebuild{$_->{sourcepage}}=1
			foreach @feeds;

		# Fork a child process to handle the aggregation.
		# The parent process will then handle building the
		# result. This avoids messy code to clear state
		# accumulated while aggregating.
		defined(my $pid = fork) or error("Can't fork: $!");
		if (! $pid) {
			IkiWiki::loadindex();

			# Aggregation happens without the main wiki lock
			# being held. This allows editing pages etc while
			# aggregation is running.
			aggregate(@feeds);

			IkiWiki::lockwiki;
			# Merge changes, since aggregation state may have
			# changed on disk while the aggregation was happening.
			mergestate();
			expire();
			savestate();
			IkiWiki::unlockwiki;
			exit 0;
		}
		waitpid($pid,0);
		if ($?) {
			error "aggregation failed with code $?";
		}

		clearstate();
		unlockaggregate();
	}
} #}}}

sub needsbuild (@) { #{{{
	my $needsbuild=shift;
	
	loadstate();

	foreach my $feed (values %feeds) {
		if (exists $pagesources{$feed->{sourcepage}} && 
		    grep { $_ eq $pagesources{$feed->{sourcepage}} } @$needsbuild) {
			# Mark all feeds originating on this page as 
			# not yet seen; preprocess will unmark those that
			# still exist.
			markunseen($feed->{sourcepage});
		}
	}
} # }}}

sub preprocess (@) { #{{{
	my %params=@_;

	foreach my $required (qw{name url}) {
		if (! exists $params{$required}) {
			return "[[aggregate ".sprintf(gettext("missing %s parameter"), $required)."]]";
		}
	}

	my $feed={};
	my $name=$params{name};
	if (exists $feeds{$name}) {
		$feed=$feeds{$name};
	}
	else {
		$feeds{$name}=$feed;
	}
	$feed->{name}=$name;
	$feed->{sourcepage}=$params{page};
	$feed->{url}=$params{url};
	my $dir=exists $params{dir} ? $params{dir} : $params{page}."/".IkiWiki::titlepage($params{name});
	$dir=~s/^\/+//;
	($dir)=$dir=~/$config{wiki_file_regexp}/;
	$feed->{dir}=$dir;
	$feed->{feedurl}=defined $params{feedurl} ? $params{feedurl} : "";
	$feed->{updateinterval}=defined $params{updateinterval} ? $params{updateinterval} * 60 : 15 * 60;
	$feed->{expireage}=defined $params{expireage} ? $params{expireage} : 0;
	$feed->{expirecount}=defined $params{expirecount} ? $params{expirecount} : 0;
	delete $feed->{unseen};
	$feed->{lastupdate}=0 unless defined $feed->{lastupdate};
	$feed->{numposts}=0 unless defined $feed->{numposts};
	$feed->{newposts}=0 unless defined $feed->{newposts};
	$feed->{message}=gettext("new feed") unless defined $feed->{message};
	$feed->{error}=0 unless defined $feed->{error};
	$feed->{tags}=[];
	while (@_) {
		my $key=shift;
		my $value=shift;
		if ($key eq 'tag') {
			push @{$feed->{tags}}, $value;
		}
	}

	return "<a href=\"".$feed->{url}."\">".$feed->{name}."</a>: ".
	       ($feed->{error} ? "<em>" : "").$feed->{message}.
	       ($feed->{error} ? "</em>" : "").
	       " (".$feed->{numposts}." ".gettext("posts").
	       ($feed->{newposts} ? "; ".$feed->{newposts}.
	                            " ".gettext("new") : "").
	       ")";
} # }}}

sub delete (@) { #{{{
	my @files=@_;

	# Remove feed data for removed pages.
	foreach my $file (@files) {
		my $page=pagename($file);
		markunseen($page);
	}
} #}}}

sub markunseen ($) { #{{{
	my $page=shift;

	foreach my $id (keys %feeds) {
		if ($feeds{$id}->{sourcepage} eq $page) {
			$feeds{$id}->{unseen}=1;
		}
	}
} #}}}

my $state_loaded=0;

sub loadstate () { #{{{
	return if $state_loaded;
	$state_loaded=1;
	if (-e "$config{wikistatedir}/aggregate") {
		open(IN, "$config{wikistatedir}/aggregate") ||
			die "$config{wikistatedir}/aggregate: $!";
		while (<IN>) {
			$_=IkiWiki::possibly_foolish_untaint($_);
			chomp;
			my $data={};
			foreach my $i (split(/ /, $_)) {
				my ($field, $val)=split(/=/, $i, 2);
				if ($field eq "name" || $field eq "feed" ||
				    $field eq "guid" || $field eq "message") {
					$data->{$field}=decode_entities($val, " \t\n");
				}
				elsif ($field eq "tag") {
					push @{$data->{tags}}, $val;
				}
				else {
					$data->{$field}=$val;
				}
			}
			
			if (exists $data->{name}) {
				$feeds{$data->{name}}=$data;
			}
			elsif (exists $data->{guid}) {
				$guids{$data->{guid}}=$data;
			}
		}

		close IN;
	}
} #}}}

sub savestate () { #{{{
	return unless $state_loaded;
	garbage_collect();
	my $newfile="$config{wikistatedir}/aggregate.new";
	my $cleanup = sub { unlink($newfile) };
	open (OUT, ">$newfile") || error("open $newfile: $!", $cleanup);
	foreach my $data (values %feeds, values %guids) {
		my @line;
		foreach my $field (keys %$data) {
			if ($field eq "name" || $field eq "feed" ||
			    $field eq "guid" || $field eq "message") {
				push @line, "$field=".encode_entities($data->{$field}, " \t\n");
			}
			elsif ($field eq "tags") {
				push @line, "tag=$_" foreach @{$data->{tags}};
			}
			else {
				push @line, "$field=".$data->{$field};
			}
		}
		print OUT join(" ", @line)."\n" || error("write $newfile: $!", $cleanup);
	}
	close OUT || error("save $newfile: $!", $cleanup);
	rename($newfile, "$config{wikistatedir}/aggregate") ||
		error("rename $newfile: $!", $cleanup);
} #}}}

sub garbage_collect () { #{{{
	foreach my $name (keys %feeds) {
		# remove any feeds that were not seen while building the pages
		# that used to contain them
		if ($feeds{$name}->{unseen}) {
			delete $feeds{$name};
		}
	}

	foreach my $guid (values %guids) {
		# any guid whose feed is gone should be removed
		if (! exists $feeds{$guid->{feed}}) {
			unlink pagefile($guid->{page})
				if exists $guid->{page};
			delete $guids{$guid->{guid}};
		}
		# handle expired guids
		elsif ($guid->{expired} && exists $guid->{page}) {
			unlink pagefile($guid->{page});
			delete $guid->{page};
			delete $guid->{md5};
		}
	}
} #}}}

sub mergestate () { #{{{
	# Load the current state in from disk, and merge into it
	# values from the state in memory that might have changed
	# during aggregation.
	my %myfeeds=%feeds;
	my %myguids=%guids;
	clearstate();
	loadstate();

	# All that can change in feed state during aggregation is a few
	# fields.
	foreach my $name (keys %myfeeds) {
		if (exists $feeds{$name}) {
			foreach my $field (qw{message lastupdate numposts
			                      newposts error}) {
				$feeds{$name}->{$field}=$myfeeds{$name}->{$field};
			}
		}
	}

	# New guids can be created during aggregation.
	# It's also possible that guids were removed from the on-disk state
	# while the aggregation was in process. That would only happen if
	# their feed was also removed, so any removed guids added back here
	# will be garbage collected later.
	foreach my $guid (keys %myguids) {
		if (! exists $guids{$guid}) {
			$guids{$guid}=$myguids{$guid};
		}
	}
} #}}}

sub clearstate () { #{{{
	%feeds=();
	%guids=();
	$state_loaded=0;
} #}}}

sub expire () { #{{{
	foreach my $feed (values %feeds) {
		next unless $feed->{expireage} || $feed->{expirecount};
		my $count=0;
		my %seen;
		foreach my $item (sort { $IkiWiki::pagectime{$b->{page}} <=> $IkiWiki::pagectime{$a->{page}} }
		                  grep { exists $_->{page} && $_->{feed} eq $feed->{name} && $IkiWiki::pagectime{$_->{page}} }
		                  values %guids) {
			if ($feed->{expireage}) {
				my $days_old = (time - $IkiWiki::pagectime{$item->{page}}) / 60 / 60 / 24;
				if ($days_old > $feed->{expireage}) {
					debug(sprintf(gettext("expiring %s (%s days old)"),
						$item->{page}, int($days_old)));
					$item->{expired}=1;
				}
			}
			elsif ($feed->{expirecount} &&
			       $count >= $feed->{expirecount}) {
				debug(sprintf(gettext("expiring %s"), $item->{page}));
				$item->{expired}=1;
			}
			else {
				if (! $seen{$item->{page}}) {
					$seen{$item->{page}}=1;
					$count++;
				}
			}
		}
	}
} #}}}

sub needsaggregate () { #{{{
	return values %feeds if $config{rebuild};
	return grep { time - $_->{lastupdate} >= $_->{updateinterval} } values %feeds;
} #}}}

sub aggregate (@) { #{{{
	eval q{use XML::Feed};
	error($@) if $@;
	eval q{use URI::Fetch};
	error($@) if $@;

	foreach my $feed (@_) {
		$feed->{lastupdate}=time;
		$feed->{newposts}=0;
		$feed->{message}=sprintf(gettext("processed ok at %s"),
			displaytime($feed->{lastupdate}));
		$feed->{error}=0;

		debug(sprintf(gettext("checking feed %s ..."), $feed->{name}));

		if (! length $feed->{feedurl}) {
			my @urls=XML::Feed->find_feeds($feed->{url});
			if (! @urls) {
				$feed->{message}=sprintf(gettext("could not find feed at %s"), $feed->{url});
				$feed->{error}=1;
				debug($feed->{message});
				next;
			}
			$feed->{feedurl}=pop @urls;
		}
		my $res=URI::Fetch->fetch($feed->{feedurl});
		if (! $res) {
			$feed->{message}=URI::Fetch->errstr;
			$feed->{error}=1;
			debug($feed->{message});
			next;
		}
		if ($res->status == URI::Fetch::URI_GONE()) {
			$feed->{message}=gettext("feed not found");
			$feed->{error}=1;
			debug($feed->{message});
			next;
		}
		my $content=$res->content;
		my $f=eval{XML::Feed->parse(\$content)};
		if ($@) {
			# One common cause of XML::Feed crashing is a feed
			# that contains invalid UTF-8 sequences. Convert
			# feed to ascii to try to work around.
			$feed->{message}.=" ".sprintf(gettext("(invalid UTF-8 stripped from feed)"));
			$content=Encode::decode_utf8($content);
			$f=eval{XML::Feed->parse(\$content)};
		}
		if ($@) {
			# Another possibility is badly escaped entities.
			$feed->{message}.=" ".sprintf(gettext("(feed entities escaped)"));
			$content=~s/\&(?!amp)(\w+);/&amp;$1;/g;
			$content=Encode::decode_utf8($content);
			$f=eval{XML::Feed->parse(\$content)};
		}
		if ($@) {
			$feed->{message}=gettext("feed crashed XML::Feed!")." ($@)";
			$feed->{error}=1;
			debug($feed->{message});
			next;
		}
		if (! $f) {
			$feed->{message}=XML::Feed->errstr;
			$feed->{error}=1;
			debug($feed->{message});
			next;
		}

		foreach my $entry ($f->entries) {
			add_page(
				feed => $feed,
				copyright => $f->copyright,
				title => defined $entry->title ? decode_entities($entry->title) : "untitled",
				link => $entry->link,
				content => defined $entry->content->body ? $entry->content->body : "",
				guid => defined $entry->id ? $entry->id : time."_".$feed->{name},
				ctime => $entry->issued ? ($entry->issued->epoch || time) : time,
			);
		}
	}
} #}}}

sub add_page (@) { #{{{
	my %params=@_;
	
	my $feed=$params{feed};
	my $guid={};
	my $mtime;
	if (exists $guids{$params{guid}}) {
		# updating an existing post
		$guid=$guids{$params{guid}};
		return if $guid->{expired};
	}
	else {
		# new post
		$guid->{guid}=$params{guid};
		$guids{$params{guid}}=$guid;
		$mtime=$params{ctime};
		$feed->{numposts}++;
		$feed->{newposts}++;

		# assign it an unused page
		my $page=IkiWiki::titlepage($params{title});
		# escape slashes and periods in title so it doesn't specify
		# directory name or trigger ".." disallowing code.
		$page=~s!([/.])!"__".ord($1)."__"!eg;
		$page=$feed->{dir}."/".$page;
		($page)=$page=~/$config{wiki_file_regexp}/;
		if (! defined $page || ! length $page) {
			$page=$feed->{dir}."/item";
		}
		my $c="";
		while (exists $IkiWiki::pagecase{lc $page.$c} ||
		       -e pagefile($page.$c)) {
			$c++
		}

		# Make sure that the file name isn't too long. 
		# NB: This doesn't check for path length limits.
		my $max=POSIX::pathconf($config{srcdir}, &POSIX::_PC_NAME_MAX);
		if (defined $max && length(htmlfn($page)) >= $max) {
			$c="";
			$page=$feed->{dir}."/item";
			while (exists $IkiWiki::pagecase{lc $page.$c} ||
			       -e pagefile($page.$c)) {
				$c++
			}
		}

		$guid->{page}=$page;
		debug(sprintf(gettext("creating new page %s"), $page));
	}
	$guid->{feed}=$feed->{name};
	
	# To write or not to write? Need to avoid writing unchanged pages
	# to avoid unneccessary rebuilding. The mtime from rss cannot be
	# trusted; let's use a digest.
	eval q{use Digest::MD5 'md5_hex'};
	error($@) if $@;
	require Encode;
	my $digest=md5_hex(Encode::encode_utf8($params{content}));
	return unless ! exists $guid->{md5} || $guid->{md5} ne $digest || $config{rebuild};
	$guid->{md5}=$digest;

	# Create the page.
	my $template=template("aggregatepost.tmpl", blind_cache => 1);
	$template->param(title => $params{title})
		if defined $params{title} && length($params{title});
	$template->param(content => htmlescape(htmlabs($params{content}, $feed->{feedurl})));
	$template->param(name => $feed->{name});
	$template->param(url => $feed->{url});
	$template->param(copyright => $params{copyright})
		if defined $params{copyright} && length $params{copyright};
	$template->param(permalink => urlabs($params{link}, $feed->{feedurl}))
		if defined $params{link};
	if (ref $feed->{tags}) {
		$template->param(tags => [map { tag => $_ }, @{$feed->{tags}}]);
	}
	writefile(htmlfn($guid->{page}), $config{srcdir},
		$template->output);

	# Set the mtime, this lets the build process get the right creation
	# time on record for the new page.
	utime $mtime, $mtime, pagefile($guid->{page})
		if defined $mtime && $mtime <= time;
} #}}}

sub htmlescape ($) { #{{{
	# escape accidental wikilinks and preprocessor stuff
	my $html=shift;
	$html=~s/(?<!\\)\[\[/\\\[\[/g;
	return $html;
} #}}}

sub urlabs ($$) { #{{{
	my $url=shift;
	my $urlbase=shift;

	URI->new_abs($url, $urlbase)->as_string;
} #}}}

sub htmlabs ($$) { #{{{
	# Convert links in html from relative to absolute.
	# Note that this is a heuristic, which is not specified by the rss
	# spec and may not be right for all feeds. Also, see Debian
	# bug #381359.
	my $html=shift;
	my $urlbase=shift;

	my $ret="";
	my $p = HTML::Parser->new(api_version => 3);
	$p->handler(default => sub { $ret.=join("", @_) }, "text");
	$p->handler(start => sub {
		my ($tagname, $pos, $text) = @_;
		if (ref $HTML::Tagset::linkElements{$tagname}) {
			while (4 <= @$pos) {
				# use attribute sets from right to left
				# to avoid invalidating the offsets
				# when replacing the values
				my($k_offset, $k_len, $v_offset, $v_len) =
					splice(@$pos, -4);
				my $attrname = lc(substr($text, $k_offset, $k_len));
				next unless grep { $_ eq $attrname } @{$HTML::Tagset::linkElements{$tagname}};
				next unless $v_offset; # 0 v_offset means no value
				my $v = substr($text, $v_offset, $v_len);
				$v =~ s/^([\'\"])(.*)\1$/$2/;
				my $new_v=urlabs($v, $urlbase);
				$new_v =~ s/\"/&quot;/g; # since we quote with ""
				substr($text, $v_offset, $v_len) = qq("$new_v");
			}
		}
		$ret.=$text;
	}, "tagname, tokenpos, text");
	$p->parse($html);
	$p->eof;

	return $ret;
} #}}}

sub pagefile ($) { #{{{
	my $page=shift;

	return "$config{srcdir}/".htmlfn($page);
} #}}}

sub htmlfn ($) { #{{{
	return shift().".".$config{htmlext};
} #}}}

my $aggregatelock;

sub lockaggregate () { #{{{
	# Take an exclusive lock to prevent multiple concurrent aggregators.
	# Returns true if the lock was aquired.
	if (! -d $config{wikistatedir}) {
		mkdir($config{wikistatedir});
	}
	open($aggregatelock, '>', "$config{wikistatedir}/aggregatelock") ||
		error ("cannot open to $config{wikistatedir}/aggregatelock: $!");
	if (! flock($aggregatelock, 2 | 4)) { # LOCK_EX | LOCK_NB
		close($aggregatelock) || error("failed closing aggregatelock: $!");
		return 0;
	}
	return 1;
} #}}}

sub unlockaggregate () { #{{{
	return close($aggregatelock) if $aggregatelock;
	return;
} #}}}

1