]>
Commit | Line | Data |
---|---|---|
428c995c JN |
1 | #! /usr/bin/perl |
2 | ||
3 | # Copyright (C) 2011 | |
4 | # Jérémie Nikaes <jeremie.nikaes@ensimag.imag.fr> | |
5 | # Arnaud Lacurie <arnaud.lacurie@ensimag.imag.fr> | |
6 | # Claire Fousse <claire.fousse@ensimag.imag.fr> | |
7 | # David Amouyal <david.amouyal@ensimag.imag.fr> | |
8 | # Matthieu Moy <matthieu.moy@grenoble-inp.fr> | |
9 | # License: GPL v2 or later | |
10 | ||
11 | # Gateway between Git and MediaWiki. | |
12 | # https://github.com/Bibzball/Git-Mediawiki/wiki | |
13 | # | |
14 | # Known limitations: | |
15 | # | |
16 | # - Only wiki pages are managed, no support for [[File:...]] | |
17 | # attachments. | |
18 | # | |
19 | # - Poor performance in the best case: it takes forever to check | |
20 | # whether we're up-to-date (on fetch or push) or to fetch a few | |
21 | # revisions from a large wiki, because we use exclusively a | |
22 | # page-based synchronization. We could switch to a wiki-wide | |
23 | # synchronization when the synchronization involves few revisions | |
24 | # but the wiki is large. | |
25 | # | |
26 | # - Git renames could be turned into MediaWiki renames (see TODO | |
27 | # below) | |
28 | # | |
29 | # - login/password support requires the user to write the password | |
30 | # cleartext in a file (see TODO below). | |
31 | # | |
32 | # - No way to import "one page, and all pages included in it" | |
33 | # | |
34 | # - Multiple remote MediaWikis have not been very well tested. | |
35 | ||
36 | use strict; | |
37 | use MediaWiki::API; | |
38 | use DateTime::Format::ISO8601; | |
39 | use encoding 'utf8'; | |
40 | ||
41 | # use encoding 'utf8' doesn't change STDERROR | |
42 | # but we're going to output UTF-8 filenames to STDERR | |
43 | binmode STDERR, ":utf8"; | |
44 | ||
45 | use URI::Escape; | |
46 | use warnings; | |
47 | ||
48 | # Mediawiki filenames can contain forward slashes. This variable decides by which pattern they should be replaced | |
49 | use constant SLASH_REPLACEMENT => "%2F"; | |
50 | ||
51 | # It's not always possible to delete pages (may require some | |
52 | # priviledges). Deleted pages are replaced with this content. | |
53 | use constant DELETED_CONTENT => "[[Category:Deleted]]\n"; | |
54 | ||
55 | # It's not possible to create empty pages. New empty files in Git are | |
56 | # sent with this content instead. | |
57 | use constant EMPTY_CONTENT => "<!-- empty page -->\n"; | |
58 | ||
59 | # used to reflect file creation or deletion in diff. | |
60 | use constant NULL_SHA1 => "0000000000000000000000000000000000000000"; | |
61 | ||
62 | my $remotename = $ARGV[0]; | |
63 | my $url = $ARGV[1]; | |
64 | ||
65 | # Accept both space-separated and multiple keys in config file. | |
66 | # Spaces should be written as _ anyway because we'll use chomp. | |
67 | my @tracked_pages = split(/[ \n]/, run_git("config --get-all remote.". $remotename .".pages")); | |
68 | chomp(@tracked_pages); | |
69 | ||
70 | # Just like @tracked_pages, but for MediaWiki categories. | |
71 | my @tracked_categories = split(/[ \n]/, run_git("config --get-all remote.". $remotename .".categories")); | |
72 | chomp(@tracked_categories); | |
73 | ||
74 | my $wiki_login = run_git("config --get remote.". $remotename .".mwLogin"); | |
75 | # TODO: ideally, this should be able to read from keyboard, but we're | |
76 | # inside a remote helper, so our stdin is connect to git, not to a | |
77 | # terminal. | |
78 | my $wiki_passwd = run_git("config --get remote.". $remotename .".mwPassword"); | |
1d6abac9 | 79 | my $wiki_domain = run_git("config --get remote.". $remotename .".mwDomain"); |
428c995c JN |
80 | chomp($wiki_login); |
81 | chomp($wiki_passwd); | |
1d6abac9 | 82 | chomp($wiki_domain); |
428c995c JN |
83 | |
84 | # Import only last revisions (both for clone and fetch) | |
85 | my $shallow_import = run_git("config --get --bool remote.". $remotename .".shallow"); | |
86 | chomp($shallow_import); | |
87 | $shallow_import = ($shallow_import eq "true"); | |
88 | ||
93f0d338 MM |
89 | # Dumb push: don't update notes and mediawiki ref to reflect the last push. |
90 | # | |
91 | # Configurable with mediawiki.dumbPush, or per-remote with | |
92 | # remote.<remotename>.dumbPush. | |
93 | # | |
94 | # This means the user will have to re-import the just-pushed | |
95 | # revisions. On the other hand, this means that the Git revisions | |
96 | # corresponding to MediaWiki revisions are all imported from the wiki, | |
97 | # regardless of whether they were initially created in Git or from the | |
98 | # web interface, hence all users will get the same history (i.e. if | |
99 | # the push from Git to MediaWiki loses some information, everybody | |
100 | # will get the history with information lost). If the import is | |
101 | # deterministic, this means everybody gets the same sha1 for each | |
102 | # MediaWiki revision. | |
103 | my $dumb_push = run_git("config --get --bool remote.$remotename.dumbPush"); | |
104 | unless ($dumb_push) { | |
105 | $dumb_push = run_git("config --get --bool mediawiki.dumbPush"); | |
106 | } | |
107 | chomp($dumb_push); | |
108 | $dumb_push = ($dumb_push eq "true"); | |
109 | ||
428c995c JN |
110 | my $wiki_name = $url; |
111 | $wiki_name =~ s/[^\/]*:\/\///; | |
9fb79503 MM |
112 | # If URL is like http://user:password@example.com/, we clearly don't |
113 | # want the password in $wiki_name. While we're there, also remove user | |
114 | # and '@' sign, to avoid author like MWUser@HTTPUser@host.com | |
115 | $wiki_name =~ s/^.*@//; | |
428c995c JN |
116 | |
117 | # Commands parser | |
118 | my $entry; | |
119 | my @cmd; | |
120 | while (<STDIN>) { | |
121 | chomp; | |
122 | @cmd = split(/ /); | |
123 | if (defined($cmd[0])) { | |
124 | # Line not blank | |
125 | if ($cmd[0] eq "capabilities") { | |
126 | die("Too many arguments for capabilities") unless (!defined($cmd[1])); | |
127 | mw_capabilities(); | |
128 | } elsif ($cmd[0] eq "list") { | |
129 | die("Too many arguments for list") unless (!defined($cmd[2])); | |
130 | mw_list($cmd[1]); | |
131 | } elsif ($cmd[0] eq "import") { | |
132 | die("Invalid arguments for import") unless ($cmd[1] ne "" && !defined($cmd[2])); | |
133 | mw_import($cmd[1]); | |
134 | } elsif ($cmd[0] eq "option") { | |
135 | die("Too many arguments for option") unless ($cmd[1] ne "" && $cmd[2] ne "" && !defined($cmd[3])); | |
136 | mw_option($cmd[1],$cmd[2]); | |
137 | } elsif ($cmd[0] eq "push") { | |
138 | mw_push($cmd[1]); | |
139 | } else { | |
140 | print STDERR "Unknown command. Aborting...\n"; | |
141 | last; | |
142 | } | |
143 | } else { | |
144 | # blank line: we should terminate | |
145 | last; | |
146 | } | |
147 | ||
148 | BEGIN { $| = 1 } # flush STDOUT, to make sure the previous | |
149 | # command is fully processed. | |
150 | } | |
151 | ||
152 | ########################## Functions ############################## | |
153 | ||
154 | # MediaWiki API instance, created lazily. | |
155 | my $mediawiki; | |
156 | ||
157 | sub mw_connect_maybe { | |
158 | if ($mediawiki) { | |
159 | return; | |
160 | } | |
161 | $mediawiki = MediaWiki::API->new; | |
162 | $mediawiki->{config}->{api_url} = "$url/api.php"; | |
163 | if ($wiki_login) { | |
164 | if (!$mediawiki->login({ | |
165 | lgname => $wiki_login, | |
166 | lgpassword => $wiki_passwd, | |
1d6abac9 | 167 | lgdomain => $wiki_domain, |
428c995c JN |
168 | })) { |
169 | print STDERR "Failed to log in mediawiki user \"$wiki_login\" on $url\n"; | |
170 | print STDERR "(error " . | |
171 | $mediawiki->{error}->{code} . ': ' . | |
172 | $mediawiki->{error}->{details} . ")\n"; | |
173 | exit 1; | |
174 | } else { | |
175 | print STDERR "Logged in with user \"$wiki_login\".\n"; | |
176 | } | |
177 | } | |
178 | } | |
179 | ||
180 | sub get_mw_first_pages { | |
181 | my $some_pages = shift; | |
182 | my @some_pages = @{$some_pages}; | |
183 | ||
184 | my $pages = shift; | |
185 | ||
186 | # pattern 'page1|page2|...' required by the API | |
187 | my $titles = join('|', @some_pages); | |
188 | ||
189 | my $mw_pages = $mediawiki->api({ | |
190 | action => 'query', | |
191 | titles => $titles, | |
192 | }); | |
193 | if (!defined($mw_pages)) { | |
194 | print STDERR "fatal: could not query the list of wiki pages.\n"; | |
195 | print STDERR "fatal: '$url' does not appear to be a mediawiki\n"; | |
196 | print STDERR "fatal: make sure '$url/api.php' is a valid page.\n"; | |
197 | exit 1; | |
198 | } | |
199 | while (my ($id, $page) = each(%{$mw_pages->{query}->{pages}})) { | |
200 | if ($id < 0) { | |
201 | print STDERR "Warning: page $page->{title} not found on wiki\n"; | |
202 | } else { | |
203 | $pages->{$page->{title}} = $page; | |
204 | } | |
205 | } | |
206 | } | |
207 | ||
208 | sub get_mw_pages { | |
209 | mw_connect_maybe(); | |
210 | ||
211 | my %pages; # hash on page titles to avoid duplicates | |
212 | my $user_defined; | |
213 | if (@tracked_pages) { | |
214 | $user_defined = 1; | |
215 | # The user provided a list of pages titles, but we | |
216 | # still need to query the API to get the page IDs. | |
217 | ||
218 | my @some_pages = @tracked_pages; | |
219 | while (@some_pages) { | |
220 | my $last = 50; | |
221 | if ($#some_pages < $last) { | |
222 | $last = $#some_pages; | |
223 | } | |
224 | my @slice = @some_pages[0..$last]; | |
225 | get_mw_first_pages(\@slice, \%pages); | |
226 | @some_pages = @some_pages[51..$#some_pages]; | |
227 | } | |
228 | } | |
229 | if (@tracked_categories) { | |
230 | $user_defined = 1; | |
231 | foreach my $category (@tracked_categories) { | |
232 | if (index($category, ':') < 0) { | |
233 | # Mediawiki requires the Category | |
234 | # prefix, but let's not force the user | |
235 | # to specify it. | |
236 | $category = "Category:" . $category; | |
237 | } | |
238 | my $mw_pages = $mediawiki->list( { | |
239 | action => 'query', | |
240 | list => 'categorymembers', | |
241 | cmtitle => $category, | |
242 | cmlimit => 'max' } ) | |
243 | || die $mediawiki->{error}->{code} . ': ' . $mediawiki->{error}->{details}; | |
244 | foreach my $page (@{$mw_pages}) { | |
245 | $pages{$page->{title}} = $page; | |
246 | } | |
247 | } | |
248 | } | |
249 | if (!$user_defined) { | |
250 | # No user-provided list, get the list of pages from | |
251 | # the API. | |
252 | my $mw_pages = $mediawiki->list({ | |
253 | action => 'query', | |
254 | list => 'allpages', | |
255 | aplimit => 500, | |
256 | }); | |
257 | if (!defined($mw_pages)) { | |
258 | print STDERR "fatal: could not get the list of wiki pages.\n"; | |
259 | print STDERR "fatal: '$url' does not appear to be a mediawiki\n"; | |
260 | print STDERR "fatal: make sure '$url/api.php' is a valid page.\n"; | |
261 | exit 1; | |
262 | } | |
263 | foreach my $page (@{$mw_pages}) { | |
264 | $pages{$page->{title}} = $page; | |
265 | } | |
266 | } | |
267 | return values(%pages); | |
268 | } | |
269 | ||
270 | sub run_git { | |
271 | open(my $git, "-|:encoding(UTF-8)", "git " . $_[0]); | |
272 | my $res = do { local $/; <$git> }; | |
273 | close($git); | |
274 | ||
275 | return $res; | |
276 | } | |
277 | ||
278 | ||
279 | sub get_last_local_revision { | |
280 | # Get note regarding last mediawiki revision | |
281 | my $note = run_git("notes --ref=$remotename/mediawiki show refs/mediawiki/$remotename/master 2>/dev/null"); | |
282 | my @note_info = split(/ /, $note); | |
283 | ||
284 | my $lastrevision_number; | |
285 | if (!(defined($note_info[0]) && $note_info[0] eq "mediawiki_revision:")) { | |
286 | print STDERR "No previous mediawiki revision found"; | |
287 | $lastrevision_number = 0; | |
288 | } else { | |
289 | # Notes are formatted : mediawiki_revision: #number | |
290 | $lastrevision_number = $note_info[1]; | |
291 | chomp($lastrevision_number); | |
292 | print STDERR "Last local mediawiki revision found is $lastrevision_number"; | |
293 | } | |
294 | return $lastrevision_number; | |
295 | } | |
296 | ||
3c1ed90e MM |
297 | # Remember the timestamp corresponding to a revision id. |
298 | my %basetimestamps; | |
299 | ||
428c995c JN |
300 | sub get_last_remote_revision { |
301 | mw_connect_maybe(); | |
302 | ||
303 | my @pages = get_mw_pages(); | |
304 | ||
305 | my $max_rev_num = 0; | |
306 | ||
307 | foreach my $page (@pages) { | |
308 | my $id = $page->{pageid}; | |
309 | ||
310 | my $query = { | |
311 | action => 'query', | |
312 | prop => 'revisions', | |
3c1ed90e | 313 | rvprop => 'ids|timestamp', |
428c995c JN |
314 | pageids => $id, |
315 | }; | |
316 | ||
317 | my $result = $mediawiki->api($query); | |
318 | ||
319 | my $lastrev = pop(@{$result->{query}->{pages}->{$id}->{revisions}}); | |
320 | ||
3c1ed90e MM |
321 | $basetimestamps{$lastrev->{revid}} = $lastrev->{timestamp}; |
322 | ||
428c995c JN |
323 | $max_rev_num = ($lastrev->{revid} > $max_rev_num ? $lastrev->{revid} : $max_rev_num); |
324 | } | |
325 | ||
326 | print STDERR "Last remote revision found is $max_rev_num.\n"; | |
327 | return $max_rev_num; | |
328 | } | |
329 | ||
330 | # Clean content before sending it to MediaWiki | |
331 | sub mediawiki_clean { | |
332 | my $string = shift; | |
333 | my $page_created = shift; | |
334 | # Mediawiki does not allow blank space at the end of a page and ends with a single \n. | |
335 | # This function right trims a string and adds a \n at the end to follow this rule | |
336 | $string =~ s/\s+$//; | |
337 | if ($string eq "" && $page_created) { | |
338 | # Creating empty pages is forbidden. | |
339 | $string = EMPTY_CONTENT; | |
340 | } | |
341 | return $string."\n"; | |
342 | } | |
343 | ||
344 | # Filter applied on MediaWiki data before adding them to Git | |
345 | sub mediawiki_smudge { | |
346 | my $string = shift; | |
347 | if ($string eq EMPTY_CONTENT) { | |
348 | $string = ""; | |
349 | } | |
350 | # This \n is important. This is due to mediawiki's way to handle end of files. | |
351 | return $string."\n"; | |
352 | } | |
353 | ||
354 | sub mediawiki_clean_filename { | |
355 | my $filename = shift; | |
356 | $filename =~ s/@{[SLASH_REPLACEMENT]}/\//g; | |
357 | # [, ], |, {, and } are forbidden by MediaWiki, even URL-encoded. | |
358 | # Do a variant of URL-encoding, i.e. looks like URL-encoding, | |
359 | # but with _ added to prevent MediaWiki from thinking this is | |
360 | # an actual special character. | |
361 | $filename =~ s/[\[\]\{\}\|]/sprintf("_%%_%x", ord($&))/ge; | |
362 | # If we use the uri escape before | |
363 | # we should unescape here, before anything | |
364 | ||
365 | return $filename; | |
366 | } | |
367 | ||
368 | sub mediawiki_smudge_filename { | |
369 | my $filename = shift; | |
370 | $filename =~ s/\//@{[SLASH_REPLACEMENT]}/g; | |
371 | $filename =~ s/ /_/g; | |
372 | # Decode forbidden characters encoded in mediawiki_clean_filename | |
373 | $filename =~ s/_%_([0-9a-fA-F][0-9a-fA-F])/sprintf("%c", hex($1))/ge; | |
374 | return $filename; | |
375 | } | |
376 | ||
377 | sub literal_data { | |
378 | my ($content) = @_; | |
379 | print STDOUT "data ", bytes::length($content), "\n", $content; | |
380 | } | |
381 | ||
382 | sub mw_capabilities { | |
383 | # Revisions are imported to the private namespace | |
384 | # refs/mediawiki/$remotename/ by the helper and fetched into | |
385 | # refs/remotes/$remotename later by fetch. | |
386 | print STDOUT "refspec refs/heads/*:refs/mediawiki/$remotename/*\n"; | |
387 | print STDOUT "import\n"; | |
388 | print STDOUT "list\n"; | |
389 | print STDOUT "push\n"; | |
390 | print STDOUT "\n"; | |
391 | } | |
392 | ||
393 | sub mw_list { | |
394 | # MediaWiki do not have branches, we consider one branch arbitrarily | |
395 | # called master, and HEAD pointing to it. | |
396 | print STDOUT "? refs/heads/master\n"; | |
397 | print STDOUT "\@refs/heads/master HEAD\n"; | |
398 | print STDOUT "\n"; | |
399 | } | |
400 | ||
401 | sub mw_option { | |
402 | print STDERR "remote-helper command 'option $_[0]' not yet implemented\n"; | |
403 | print STDOUT "unsupported\n"; | |
404 | } | |
405 | ||
406 | sub fetch_mw_revisions_for_page { | |
407 | my $page = shift; | |
408 | my $id = shift; | |
409 | my $fetch_from = shift; | |
410 | my @page_revs = (); | |
411 | my $query = { | |
412 | action => 'query', | |
413 | prop => 'revisions', | |
414 | rvprop => 'ids', | |
415 | rvdir => 'newer', | |
416 | rvstartid => $fetch_from, | |
417 | rvlimit => 500, | |
418 | pageids => $id, | |
419 | }; | |
420 | ||
421 | my $revnum = 0; | |
422 | # Get 500 revisions at a time due to the mediawiki api limit | |
423 | while (1) { | |
424 | my $result = $mediawiki->api($query); | |
425 | ||
426 | # Parse each of those 500 revisions | |
427 | foreach my $revision (@{$result->{query}->{pages}->{$id}->{revisions}}) { | |
428 | my $page_rev_ids; | |
429 | $page_rev_ids->{pageid} = $page->{pageid}; | |
430 | $page_rev_ids->{revid} = $revision->{revid}; | |
431 | push(@page_revs, $page_rev_ids); | |
432 | $revnum++; | |
433 | } | |
434 | last unless $result->{'query-continue'}; | |
435 | $query->{rvstartid} = $result->{'query-continue'}->{revisions}->{rvstartid}; | |
436 | } | |
437 | if ($shallow_import && @page_revs) { | |
438 | print STDERR " Found 1 revision (shallow import).\n"; | |
439 | @page_revs = sort {$b->{revid} <=> $a->{revid}} (@page_revs); | |
440 | return $page_revs[0]; | |
441 | } | |
442 | print STDERR " Found ", $revnum, " revision(s).\n"; | |
443 | return @page_revs; | |
444 | } | |
445 | ||
446 | sub fetch_mw_revisions { | |
447 | my $pages = shift; my @pages = @{$pages}; | |
448 | my $fetch_from = shift; | |
449 | ||
450 | my @revisions = (); | |
451 | my $n = 1; | |
452 | foreach my $page (@pages) { | |
453 | my $id = $page->{pageid}; | |
454 | ||
455 | print STDERR "page $n/", scalar(@pages), ": ". $page->{title} ."\n"; | |
456 | $n++; | |
457 | my @page_revs = fetch_mw_revisions_for_page($page, $id, $fetch_from); | |
458 | @revisions = (@page_revs, @revisions); | |
459 | } | |
460 | ||
461 | return ($n, @revisions); | |
462 | } | |
463 | ||
464 | sub import_file_revision { | |
465 | my $commit = shift; | |
466 | my %commit = %{$commit}; | |
467 | my $full_import = shift; | |
468 | my $n = shift; | |
469 | ||
470 | my $title = $commit{title}; | |
471 | my $comment = $commit{comment}; | |
472 | my $content = $commit{content}; | |
473 | my $author = $commit{author}; | |
474 | my $date = $commit{date}; | |
475 | ||
476 | print STDOUT "commit refs/mediawiki/$remotename/master\n"; | |
477 | print STDOUT "mark :$n\n"; | |
478 | print STDOUT "committer $author <$author\@$wiki_name> ", $date->epoch, " +0000\n"; | |
479 | literal_data($comment); | |
480 | ||
481 | # If it's not a clone, we need to know where to start from | |
482 | if (!$full_import && $n == 1) { | |
483 | print STDOUT "from refs/mediawiki/$remotename/master^0\n"; | |
484 | } | |
485 | if ($content ne DELETED_CONTENT) { | |
486 | print STDOUT "M 644 inline $title.mw\n"; | |
487 | literal_data($content); | |
488 | print STDOUT "\n\n"; | |
489 | } else { | |
490 | print STDOUT "D $title.mw\n"; | |
491 | } | |
492 | ||
493 | # mediawiki revision number in the git note | |
494 | if ($full_import && $n == 1) { | |
495 | print STDOUT "reset refs/notes/$remotename/mediawiki\n"; | |
496 | } | |
497 | print STDOUT "commit refs/notes/$remotename/mediawiki\n"; | |
498 | print STDOUT "committer $author <$author\@$wiki_name> ", $date->epoch, " +0000\n"; | |
499 | literal_data("Note added by git-mediawiki during import"); | |
500 | if (!$full_import && $n == 1) { | |
501 | print STDOUT "from refs/notes/$remotename/mediawiki^0\n"; | |
502 | } | |
503 | print STDOUT "N inline :$n\n"; | |
504 | literal_data("mediawiki_revision: " . $commit{mw_revision}); | |
505 | print STDOUT "\n\n"; | |
506 | } | |
507 | ||
508 | # parse a sequence of | |
509 | # <cmd> <arg1> | |
510 | # <cmd> <arg2> | |
511 | # \n | |
512 | # (like batch sequence of import and sequence of push statements) | |
513 | sub get_more_refs { | |
514 | my $cmd = shift; | |
515 | my @refs; | |
516 | while (1) { | |
517 | my $line = <STDIN>; | |
518 | if ($line =~ m/^$cmd (.*)$/) { | |
519 | push(@refs, $1); | |
520 | } elsif ($line eq "\n") { | |
521 | return @refs; | |
522 | } else { | |
523 | die("Invalid command in a '$cmd' batch: ". $_); | |
524 | } | |
525 | } | |
526 | } | |
527 | ||
528 | sub mw_import { | |
529 | # multiple import commands can follow each other. | |
530 | my @refs = (shift, get_more_refs("import")); | |
531 | foreach my $ref (@refs) { | |
532 | mw_import_ref($ref); | |
533 | } | |
534 | print STDOUT "done\n"; | |
535 | } | |
536 | ||
537 | sub mw_import_ref { | |
538 | my $ref = shift; | |
539 | # The remote helper will call "import HEAD" and | |
540 | # "import refs/heads/master". | |
541 | # Since HEAD is a symbolic ref to master (by convention, | |
542 | # followed by the output of the command "list" that we gave), | |
543 | # we don't need to do anything in this case. | |
544 | if ($ref eq "HEAD") { | |
545 | return; | |
546 | } | |
547 | ||
548 | mw_connect_maybe(); | |
549 | ||
550 | my @pages = get_mw_pages(); | |
551 | ||
552 | print STDERR "Searching revisions...\n"; | |
553 | my $last_local = get_last_local_revision(); | |
554 | my $fetch_from = $last_local + 1; | |
555 | if ($fetch_from == 1) { | |
556 | print STDERR ", fetching from beginning.\n"; | |
557 | } else { | |
558 | print STDERR ", fetching from here.\n"; | |
559 | } | |
560 | my ($n, @revisions) = fetch_mw_revisions(\@pages, $fetch_from); | |
561 | ||
562 | # Creation of the fast-import stream | |
563 | print STDERR "Fetching & writing export data...\n"; | |
564 | ||
565 | $n = 0; | |
566 | my $last_timestamp = 0; # Placeholer in case $rev->timestamp is undefined | |
567 | ||
568 | foreach my $pagerevid (sort {$a->{revid} <=> $b->{revid}} @revisions) { | |
569 | # fetch the content of the pages | |
570 | my $query = { | |
571 | action => 'query', | |
572 | prop => 'revisions', | |
573 | rvprop => 'content|timestamp|comment|user|ids', | |
574 | revids => $pagerevid->{revid}, | |
575 | }; | |
576 | ||
577 | my $result = $mediawiki->api($query); | |
578 | ||
579 | my $rev = pop(@{$result->{query}->{pages}->{$pagerevid->{pageid}}->{revisions}}); | |
580 | ||
581 | $n++; | |
582 | ||
583 | my %commit; | |
584 | $commit{author} = $rev->{user} || 'Anonymous'; | |
585 | $commit{comment} = $rev->{comment} || '*Empty MediaWiki Message*'; | |
586 | $commit{title} = mediawiki_smudge_filename( | |
587 | $result->{query}->{pages}->{$pagerevid->{pageid}}->{title} | |
588 | ); | |
589 | $commit{mw_revision} = $pagerevid->{revid}; | |
590 | $commit{content} = mediawiki_smudge($rev->{'*'}); | |
591 | ||
592 | if (!defined($rev->{timestamp})) { | |
593 | $last_timestamp++; | |
594 | } else { | |
595 | $last_timestamp = $rev->{timestamp}; | |
596 | } | |
597 | $commit{date} = DateTime::Format::ISO8601->parse_datetime($last_timestamp); | |
598 | ||
599 | print STDERR "$n/", scalar(@revisions), ": Revision #$pagerevid->{revid} of $commit{title}\n"; | |
600 | ||
601 | import_file_revision(\%commit, ($fetch_from == 1), $n); | |
602 | } | |
603 | ||
604 | if ($fetch_from == 1 && $n == 0) { | |
605 | print STDERR "You appear to have cloned an empty MediaWiki.\n"; | |
606 | # Something has to be done remote-helper side. If nothing is done, an error is | |
607 | # thrown saying that HEAD is refering to unknown object 0000000000000000000 | |
608 | # and the clone fails. | |
609 | } | |
610 | } | |
611 | ||
612 | sub error_non_fast_forward { | |
fd47d7b9 MM |
613 | my $advice = run_git("config --bool advice.pushNonFastForward"); |
614 | chomp($advice); | |
615 | if ($advice ne "false") { | |
616 | # Native git-push would show this after the summary. | |
617 | # We can't ask it to display it cleanly, so print it | |
618 | # ourselves before. | |
619 | print STDERR "To prevent you from losing history, non-fast-forward updates were rejected\n"; | |
620 | print STDERR "Merge the remote changes (e.g. 'git pull') before pushing again. See the\n"; | |
621 | print STDERR "'Note about fast-forwards' section of 'git push --help' for details.\n"; | |
622 | } | |
428c995c JN |
623 | print STDOUT "error $_[0] \"non-fast-forward\"\n"; |
624 | return 0; | |
625 | } | |
626 | ||
627 | sub mw_push_file { | |
628 | my $diff_info = shift; | |
629 | # $diff_info contains a string in this format: | |
630 | # 100644 100644 <sha1_of_blob_before_commit> <sha1_of_blob_now> <status> | |
631 | my @diff_info_split = split(/[ \t]/, $diff_info); | |
632 | ||
633 | # Filename, including .mw extension | |
634 | my $complete_file_name = shift; | |
635 | # Commit message | |
636 | my $summary = shift; | |
93f0d338 MM |
637 | # MediaWiki revision number. Keep the previous one by default, |
638 | # in case there's no edit to perform. | |
639 | my $newrevid = shift; | |
428c995c JN |
640 | |
641 | my $new_sha1 = $diff_info_split[3]; | |
642 | my $old_sha1 = $diff_info_split[2]; | |
643 | my $page_created = ($old_sha1 eq NULL_SHA1); | |
644 | my $page_deleted = ($new_sha1 eq NULL_SHA1); | |
645 | $complete_file_name = mediawiki_clean_filename($complete_file_name); | |
646 | ||
647 | if (substr($complete_file_name,-3) eq ".mw") { | |
648 | my $title = substr($complete_file_name,0,-3); | |
649 | ||
650 | my $file_content; | |
651 | if ($page_deleted) { | |
652 | # Deleting a page usually requires | |
653 | # special priviledges. A common | |
654 | # convention is to replace the page | |
655 | # with this content instead: | |
656 | $file_content = DELETED_CONTENT; | |
657 | } else { | |
658 | $file_content = run_git("cat-file blob $new_sha1"); | |
659 | } | |
660 | ||
661 | mw_connect_maybe(); | |
662 | ||
663 | my $result = $mediawiki->edit( { | |
664 | action => 'edit', | |
665 | summary => $summary, | |
666 | title => $title, | |
3c1ed90e | 667 | basetimestamp => $basetimestamps{$newrevid}, |
428c995c JN |
668 | text => mediawiki_clean($file_content, $page_created), |
669 | }, { | |
670 | skip_encoding => 1 # Helps with names with accentuated characters | |
3c1ed90e MM |
671 | }); |
672 | if (!$result) { | |
673 | if ($mediawiki->{error}->{code} == 3) { | |
674 | # edit conflicts, considered as non-fast-forward | |
675 | print STDERR 'Warning: Error ' . | |
676 | $mediawiki->{error}->{code} . | |
677 | ' from mediwiki: ' . $mediawiki->{error}->{details} . | |
678 | ".\n"; | |
679 | return ($newrevid, "non-fast-forward"); | |
680 | } else { | |
681 | # Other errors. Shouldn't happen => just die() | |
682 | die 'Fatal: Error ' . | |
683 | $mediawiki->{error}->{code} . | |
684 | ' from mediwiki: ' . $mediawiki->{error}->{details}; | |
685 | } | |
686 | } | |
93f0d338 | 687 | $newrevid = $result->{edit}->{newrevid}; |
ac86ec0f | 688 | print STDERR "Pushed file: $new_sha1 - $title\n"; |
428c995c JN |
689 | } else { |
690 | print STDERR "$complete_file_name not a mediawiki file (Not pushable on this version of git-remote-mediawiki).\n" | |
691 | } | |
3c1ed90e | 692 | return ($newrevid, "ok"); |
428c995c JN |
693 | } |
694 | ||
695 | sub mw_push { | |
696 | # multiple push statements can follow each other | |
697 | my @refsspecs = (shift, get_more_refs("push")); | |
428c995c JN |
698 | my $pushed; |
699 | for my $refspec (@refsspecs) { | |
700 | my ($force, $local, $remote) = $refspec =~ /^(\+)?([^:]*):([^:]*)$/ | |
701 | or die("Invalid refspec for push. Expected <src>:<dst> or +<src>:<dst>"); | |
702 | if ($force) { | |
703 | print STDERR "Warning: forced push not allowed on a MediaWiki.\n"; | |
704 | } | |
705 | if ($local eq "") { | |
706 | print STDERR "Cannot delete remote branch on a MediaWiki\n"; | |
707 | print STDOUT "error $remote cannot delete\n"; | |
708 | next; | |
709 | } | |
710 | if ($remote ne "refs/heads/master") { | |
711 | print STDERR "Only push to the branch 'master' is supported on a MediaWiki\n"; | |
712 | print STDOUT "error $remote only master allowed\n"; | |
713 | next; | |
714 | } | |
715 | if (mw_push_revision($local, $remote)) { | |
716 | $pushed = 1; | |
717 | } | |
718 | } | |
719 | ||
720 | # Notify Git that the push is done | |
721 | print STDOUT "\n"; | |
722 | ||
93f0d338 | 723 | if ($pushed && $dumb_push) { |
428c995c JN |
724 | print STDERR "Just pushed some revisions to MediaWiki.\n"; |
725 | print STDERR "The pushed revisions now have to be re-imported, and your current branch\n"; | |
726 | print STDERR "needs to be updated with these re-imported commits. You can do this with\n"; | |
727 | print STDERR "\n"; | |
728 | print STDERR " git pull --rebase\n"; | |
729 | print STDERR "\n"; | |
730 | } | |
731 | } | |
732 | ||
733 | sub mw_push_revision { | |
734 | my $local = shift; | |
735 | my $remote = shift; # actually, this has to be "refs/heads/master" at this point. | |
736 | my $last_local_revid = get_last_local_revision(); | |
737 | print STDERR ".\n"; # Finish sentence started by get_last_local_revision() | |
738 | my $last_remote_revid = get_last_remote_revision(); | |
93f0d338 | 739 | my $mw_revision = $last_remote_revid; |
428c995c JN |
740 | |
741 | # Get sha1 of commit pointed by local HEAD | |
742 | my $HEAD_sha1 = run_git("rev-parse $local 2>/dev/null"); chomp($HEAD_sha1); | |
743 | # Get sha1 of commit pointed by remotes/$remotename/master | |
744 | my $remoteorigin_sha1 = run_git("rev-parse refs/remotes/$remotename/master 2>/dev/null"); | |
745 | chomp($remoteorigin_sha1); | |
746 | ||
747 | if ($last_local_revid > 0 && | |
748 | $last_local_revid < $last_remote_revid) { | |
749 | return error_non_fast_forward($remote); | |
750 | } | |
751 | ||
752 | if ($HEAD_sha1 eq $remoteorigin_sha1) { | |
753 | # nothing to push | |
754 | return 0; | |
755 | } | |
756 | ||
757 | # Get every commit in between HEAD and refs/remotes/origin/master, | |
758 | # including HEAD and refs/remotes/origin/master | |
759 | my @commit_pairs = (); | |
760 | if ($last_local_revid > 0) { | |
761 | my $parsed_sha1 = $remoteorigin_sha1; | |
762 | # Find a path from last MediaWiki commit to pushed commit | |
763 | while ($parsed_sha1 ne $HEAD_sha1) { | |
764 | my @commit_info = grep(/^$parsed_sha1/, split(/\n/, run_git("rev-list --children $local"))); | |
765 | if (!@commit_info) { | |
766 | return error_non_fast_forward($remote); | |
767 | } | |
768 | my @commit_info_split = split(/ |\n/, $commit_info[0]); | |
769 | # $commit_info_split[1] is the sha1 of the commit to export | |
770 | # $commit_info_split[0] is the sha1 of its direct child | |
771 | push(@commit_pairs, \@commit_info_split); | |
772 | $parsed_sha1 = $commit_info_split[1]; | |
773 | } | |
774 | } else { | |
775 | # No remote mediawiki revision. Export the whole | |
776 | # history (linearized with --first-parent) | |
777 | print STDERR "Warning: no common ancestor, pushing complete history\n"; | |
778 | my $history = run_git("rev-list --first-parent --children $local"); | |
779 | my @history = split('\n', $history); | |
780 | @history = @history[1..$#history]; | |
781 | foreach my $line (reverse @history) { | |
782 | my @commit_info_split = split(/ |\n/, $line); | |
783 | push(@commit_pairs, \@commit_info_split); | |
784 | } | |
785 | } | |
786 | ||
787 | foreach my $commit_info_split (@commit_pairs) { | |
788 | my $sha1_child = @{$commit_info_split}[0]; | |
789 | my $sha1_commit = @{$commit_info_split}[1]; | |
790 | my $diff_infos = run_git("diff-tree -r --raw -z $sha1_child $sha1_commit"); | |
791 | # TODO: we could detect rename, and encode them with a #redirect on the wiki. | |
792 | # TODO: for now, it's just a delete+add | |
793 | my @diff_info_list = split(/\0/, $diff_infos); | |
794 | # Keep the first line of the commit message as mediawiki comment for the revision | |
795 | my $commit_msg = (split(/\n/, run_git("show --pretty=format:\"%s\" $sha1_commit")))[0]; | |
796 | chomp($commit_msg); | |
797 | # Push every blob | |
798 | while (@diff_info_list) { | |
3c1ed90e | 799 | my $status; |
428c995c JN |
800 | # git diff-tree -z gives an output like |
801 | # <metadata>\0<filename1>\0 | |
802 | # <metadata>\0<filename2>\0 | |
803 | # and we've split on \0. | |
804 | my $info = shift(@diff_info_list); | |
805 | my $file = shift(@diff_info_list); | |
3c1ed90e MM |
806 | ($mw_revision, $status) = mw_push_file($info, $file, $commit_msg, $mw_revision); |
807 | if ($status eq "non-fast-forward") { | |
808 | # we may already have sent part of the | |
809 | # commit to MediaWiki, but it's too | |
810 | # late to cancel it. Stop the push in | |
811 | # the middle, but still give an | |
812 | # accurate error message. | |
813 | return error_non_fast_forward($remote); | |
814 | } | |
815 | if ($status ne "ok") { | |
816 | die("Unknown error from mw_push_file()"); | |
817 | } | |
93f0d338 MM |
818 | } |
819 | unless ($dumb_push) { | |
820 | run_git("notes --ref=$remotename/mediawiki add -m \"mediawiki_revision: $mw_revision\" $sha1_commit"); | |
821 | run_git("update-ref -m \"Git-MediaWiki push\" refs/mediawiki/$remotename/master $sha1_commit $sha1_child"); | |
428c995c JN |
822 | } |
823 | } | |
824 | ||
825 | print STDOUT "ok $remote\n"; | |
826 | return 1; | |
827 | } |