source: ValBot/Python/check_intrawiki_section_links.py@ 1186

Last change on this file since 1186 was 1186, checked in by iritscen, 12 months ago

ValBot: Forgot to update one variable name in check_intrawiki_section_links.py.
File size: 18.2 KB
Line 
1# Check Intrawiki Section Links
2# by iritscen@yahoo.com
3# Looks at each wikilink on a page (or in all the pages in a category) for a section link ('#'),
4# and loads the linked page and verifies that the named section actually exists. It also
5# understands section links generated through a call to Template:SectionLink.
6# Recommended viewing width:
7# |---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- --|
8
9import os
10
11from urllib.parse import urljoin
12
13import pywikibot
14import re
15
16from pywikibot.bot import QuitKeyboardInterrupt
17from pywikibot import pagegenerators
18from pywikibot.tools.formatter import color_format
19from pywikibot.comms.http import fetch
20from pywikibot.specialbots import UploadRobot
21from bs4 import BeautifulSoup
22
23# Tuple of OniGalore's namespaces
24intrawiki_prefixes = ('Image', 'Special', 'Talk', 'User', 'User_talk', 'OniGalore', 'OniGalore_talk', 'File', 'File_talk', 'MediaWiki', 'MediaWiki_talk', 'Template', 'Template_talk', 'Help', 'Help_talk', 'Category', 'Category_talk', 'BSL', 'BSL_talk', 'OBD', 'OBD_talk', 'AE', 'AE_talk', 'Oni2', 'Oni2_talk', 'XML', 'XML_talk')
25
26# URL for main namespace of our wiki
27onigalore_url = 'https://wiki.oni2.net/'
28
29# Tuple of interwiki prefixes, for recognizing and passing over such links
30interwiki_prefixes = ('acronym', 'cache', 'commons', 'dictionary', 'google', 'metawikimedia', 'mw', 'wikibooks', 'wikidata', 'wikimedia', 'wikinews', 'wikipedia', 'wikiquote', 'wikisource', 'wikispecies', 'wikiversity', 'wikivoyage', 'wikt', 'wiktionary', 'wp')
31
32# List of chapter names, for substitution into links that use "{{Cn}}" transclusion
33chapter_names = ['CHAPTER_00_._COMBAT_TRAINING', 'CHAPTER_01_._TRIAL_RUN', 'CHAPTER_02_._ENGINES_OF_EVIL', 'CHAPTER_03_._PUZZLE_PIECES', 'CHAPTER_04_._TIGER_BY_THE_TAIL', 'CHAPTER_05_._HOT_PURSUIT', 'CHAPTER_06_._COUNTERATTACK', 'CHAPTER_07_._A_FRIEND_IN_NEED', 'CHAPTER_08_._AN_INNOCENT_LIFE', 'CHAPTER_09_._TRUTH_AND_CONSEQUENCES', 'CHAPTER_10_._CAT_AND_MOUSE', 'CHAPTER_11_._DREAM_DIVER', 'CHAPTER_12_._SINS_OF_THE_FATHER', 'CHAPTER_13_._PHOENIX_RISING', 'CHAPTER_14_._DAWN_OF_THE_CHRYSALIS']
34
35# Tuple of patterns for recognizing wikilinks
36# Pattern 1: Detect "[[anything]]", "[[any:thing]]", "[[any|thing]]", "[[any:thi|ng]]"
37# Pattern 2: Detect "{{SectionLink|Page|Section name}}", "{{SectionLink||Section name}}"
38link_patterns = ("\[\[[^|\]]*(\||\])", "\{\{SectionLink\|[^|\}]*\|[^|\}]*\}\}")
39
40# Initialize globals
41debug = 0
42pages_checked = 0
43iw_found = 0
44advice_issued = 0
45errors_issued = 0
46name_printed = 0
47
48# Prints the name of a page on which something occurred, if it has not been printed before
49def possibly_print(page_name):
50 global debug
51 global name_printed
52
53 if not name_printed and not debug:
54 pywikibot.stdout('')
55 pywikibot.stdout('From page "{}":'.format(page_name))
56 name_printed = 1
57
58# Search a page for the section specified in the link
59def find_section(page_text, page_name, page_slug, print_result):
60 global errors_issued
61
62 # Isolate section link
63 target_page_name, anchor_name = page_slug.split('#', 1)
64 target_page_name_human = target_page_name.replace('_', ' ')
65 if debug: pywikibot.stdout(' Searching for section link {} on page.'.format(anchor_name))
66
67 # Convert slash character to the dot-notation hex encoding that MediaWiki uses
68 anchor_name = anchor_name.replace('/', '.2F')
69
70 # Read linked page to see if it really has this anchor link
71 soup = BeautifulSoup(page_text, 'html.parser')
72 found_section = False
73 for span_tag in soup.findAll('span'):
74 span_name = span_tag.get('id', None)
75 if span_name == anchor_name:
76 if debug and not print_result: pywikibot.stdout(' Found section in a span!')
77 found_section = True
78 break
79 if found_section == False:
80 # Search for a div with this ID
81 for span_tag in soup.findAll('div'):
82 span_name = span_tag.get('id', None)
83 if span_name == anchor_name:
84 if debug and not print_result: pywikibot.stdout(' Found section in a div!')
85 found_section = True
86 break
87 if found_section == False:
88 possibly_print(page_name)
89 pywikibot.stdout(' ERROR: Could not find section "{0}" on page {1}!'.format(anchor_name, target_page_name_human))
90 errors_issued += 1
91 elif debug and print_result:
92 pywikibot.stdout(' The section "{0}" was found on page "{1}".'.format(anchor_name, target_page_name_human))
93
94# For a link that redirected us to another page, extract the name of the target page from
95# the target page's source
96def find_canonical_link(page_text, page_name, page_slug):
97 # Extract link from this markup which contains name of redirected-to page:
98 # <link rel="canonical" href="https://en.wikipedia.org/wiki/Page_name"/>
99 # "wgPageName":"Namespace:Page_name",
100 canonical_name = page_text.split('"wgPageName":"')[-1]
101 tag_end = canonical_name.find('",')
102
103 if tag_end == -1:
104 pywikibot.stdout(' ERROR: The link "{}" is a redirect page, but this script could not isolate the target page name.'.format(page_slug))
105 errors_issued = errors_issued + 1
106 else:
107 canonical_name = canonical_name[:tag_end]
108 if len(canonical_name) > 100:
109 # Certain things can cause the trim to fail; report error and avoid slamming the
110 # output with massive page source from a failed trim
111 pywikibot.stdout(' ERROR: The link "{}" is a redirect to "{2}…" (string overflow).'.format(page_slug, canonical_name[:100]))
112 errors_issued = errors_issued + 1
113 else:
114 canonical_name = canonical_name.replace('_', ' ')
115 if '#' in page_slug:
116 _, anchor_name = page_slug.split('#')
117 if debug: pywikibot.stdout(' The link "{0}" is a redirect to "{1}#{2}", which is a valid page. Checking section link….'.format(page_slug, canonical_name, anchor_name))
118 find_section(page_text, page_name, page_slug, True)
119 else:
120 pywikibot.stdout(' The link "{0}" is a redirect to "{1}", which is a valid page.'.format(page_slug, canonical_name))
121
122# Test an intrawiki link and look for a section link if applicable
123def test_intrawiki_link(iw_url, page_name, page_slug):
124 global advice_issued
125 global errors_issued
126
127 response = fetch(iw_url)
128
129 # One way we tell that a redirect occurred is by checking fetch's history, as it
130 # automatically follows redirects. This will catch formal redirects which come from pages
131 # such as Special:PermanentLink.
132 if response.history != []:
133
134 permalink1 = 'Special:PermanentLink/'.lower()
135 permalink2 = 'Special:Permalink/'.lower()
136 page_slug_lower = page_slug.lower()
137 if page_slug_lower.startswith(permalink1) or page_slug_lower.startswith(permalink2):
138 if debug:
139 possibly_print(page_name)
140 pywikibot.stdout(' Got redirection code "{0}" for permanent revision link "{1}". Checking the target page….'.format(response.history[0], page_slug))
141 find_canonical_link(response.text, page_name, page_slug)
142 else:
143 possibly_print(page_name)
144 pywikibot.stdout(' ERROR: Unrecognized type of redirection (code "{0}") for link "{1}". You should check the link manually.'.format(response.history[0], page_slug))
145 advice_issued += 1
146 elif response.status_code != 200:
147 possibly_print(page_name)
148 pywikibot.stdout(' ERROR: Got response code {0} on URL {1}. The target page may not exist.'.format(response.status_code, iw_url))
149 errors_issued += 1
150 # However the usual way that a redirect occurs is that MediaWiki redirects us sneakily
151 # using JavaScript, while returning code OK 200 as if the link was correct; this happens
152 # when a redirect page is accessed. We must detect these soft redirects by looking at the
153 # page source to find the redirect note inserted at the top of the page for the reader.
154 elif 'Redirected from <a' in response.text:
155 if debug:
156 possibly_print(page_name)
157 pywikibot.stdout(' Got silently redirected by link "{}". Checking the target page….'.format(page_slug))
158 find_canonical_link(response.text, page_name, page_slug)
159 else: # URL is OK, so proceed
160 find_section(response.text, page_name, page_slug, False)
161
162# Searches the given page text for intrawiki links with section links in them
163def scan_for_intrawiki_links(page_text, page_name):
164 global debug
165 global pages_checked
166 global iw_found
167 global advice_issued
168 global errors_issued
169 global name_printed
170 pages_checked += 1
171 name_printed = 0
172
173 for i, the_pattern in enumerate(link_patterns):
174 if debug:
175 if i == 0:
176 pywikibot.stdout(' Checking page for wikilinks with section names.')
177 elif i == 1:
178 pywikibot.stdout(' Checking page for {{SectionLink}} calls.')
179
180 for match in re.finditer(the_pattern, page_text):
181 found_iw_match = False
182 iw_url = ""
183 page_name2 = page_name
184
185 # Cut out the matched text from the page, isolating just the page+section name
186 target_start = 2 # "[["
187 target_end = 1 # "|" or "]" (we only match the first ending bracket)
188 if i == 1:
189 target_start = 14 # "{{SectionLink|"
190 target_end = 2 # "}}"
191 s = match.start() + target_start # remove the link-opening markup
192 e = match.end() - target_end # remove the link-ending markup
193 page_slug = page_text[s:e]
194
195 # The second link type will look like "Page|Section" or "|Section", so fix that pipe
196 if i == 1:
197 page_slug = page_slug.replace('|', '#')
198
199 # Sometimes we use a space char. instead of a '_', so fix that before querying
200 page_slug = page_slug.replace(' ', '_')
201 if debug: pywikibot.stdout(' Found link {0}.'.format(page_slug))
202
203 # If this link doesn't have a section link in it, then we don't care about it, as
204 # MediaWiki takes care of checking basic intrawiki links
205 if not '#' in page_slug:
206 if debug: pywikibot.stdout(' Link doesn\'t have a section anchor in it. Skipping.')
207 continue
208
209 # If this link has an interwiki prefix, it can be ignored; see check_interwiki_links.py
210 # for the task of checking interwiki page+section links
211 is_interwiki = False
212 if found_iw_match == False:
213 for prefix in interwiki_prefixes:
214 if prefix + ":" in page_slug:
215 if debug: pywikibot.stdout(' Skipping link {} because it is an interwiki link.'.format(page_slug))
216 is_interwiki = True
217 break
218 if is_interwiki:
219 continue
220
221 # If there is a '{' in the link, then probably it's a link built on transcluded text.
222 # If it's a chapter template transclusion like "Quotes/Diary#{{C3}}", expand it using
223 # our "chapter_names" array. If it's another type of transclusion, punt it to the user.
224 if '{' in page_slug:
225 ch_link_pattern = re.compile(r"{{C[0-9]*}}")
226 ch_link = ch_link_pattern.search(page_slug)
227 if debug: pywikibot.stdout(' Found transclusion in link: "{}".'.format(ch_link.group(0)))
228 if ch_link:
229 ch_link_match = ch_link.group(0)
230 ch_num_pattern = re.compile("[0-9]+")
231 ch_num = ch_num_pattern.search(ch_link_match)
232 if ch_num:
233 ch_num_match = int(ch_num.group(0))
234 if ch_num_match >= 0 and ch_num_match <= 14:
235 ch_name = chapter_names[ch_num_match]
236 replace_pattern = re.compile(r"{{C" + ch_num.group(0) + r"}}")
237 page_slug = replace_pattern.sub(ch_name, page_slug)
238 if debug: pywikibot.stdout(' After performing transclusion, link is now "{}".'.format(page_slug))
239 else:
240 possibly_print(page_name)
241 pywikibot.stdout(' ERROR: Link {0} transcludes a chapter name using an out-of-range number, {1}.'.format(page_slug, ch_num_match))
242 errors_issued += 1
243 continue
244 else:
245 possibly_print(page_name)
246 pywikibot.stdout(' ADVICE: Link {} seems to be transcluding a chapter name, but this script couldn\'t read it.'.format(page_slug))
247 advice_issued += 1
248 continue
249 else:
250 possibly_print(page_name)
251 pywikibot.stdout(' ADVICE: Link {0} seems to use transclusion. This script can understand chapter name transclusions such as "{1}" but it doesn\'t recognize this one so it can\'t be verified. You should check the link manually.'.format(page_slug, "{{C7}}"))
252 advice_issued += 1
253 continue
254
255 # If this is a relative "/" link, use the current page as the basis for the URL. Note
256 # that only a leading slash is looked for, so if there's multiple steps down ("/x/y"),
257 # we're out of luck.
258 if page_slug.startswith('/'):
259 page_slug = page_name + page_slug
260 if debug: pywikibot.stdout(' Changed page_slug to {} on account of "/".'.format(page_slug))
261
262 # If this is a relative "../" link, find the parent page, set ourselves to that page,
263 # then remove the relative portion of the link. Note that this is only performed once,
264 # so if there's multiple steps back ("../../"), we're out of luck.
265 if page_slug.startswith('../'):
266 last_slash = page_name.rfind('/')
267 page_name2 = page_name[0:last_slash]
268 if debug: pywikibot.stdout(' Changed page_name to {} on account of "../".'.format(page_name2))
269 page_slug = page_slug[3:len(page_slug)]
270 if debug: pywikibot.stdout(' Changed page_slug to {} on account of "../".'.format(page_slug))
271 # If this is now going to be a bare section link for the parent page, don't add a
272 # slash, otherwise do because we are drilling down to another subpage
273 if page_slug.startswith('#'):
274 page_slug = page_name2 + page_slug
275 else:
276 page_slug = page_name2 + '/' + page_slug
277
278 # If this is a bare section link, build URL based on this page
279 if page_slug.startswith('#'):
280 iw_url = onigalore_url + page_name2
281 iw_found += 1
282 if debug: pywikibot.stdout(' Found link to this very page, {}.'.format(page_slug))
283 found_iw_match = True
284 page_slug = page_name2 + page_slug
285
286 # If there's no ":" in the link (before the section link, where a colon would just be
287 # part of the text) then it's a Main namespace article; proceed with building URL
288 if found_iw_match == False:
289 if not re.search(":.*#", page_slug):
290 iw_url = onigalore_url + page_slug
291 iw_found += 1
292 if debug: pywikibot.stdout(' Link is to a Main namespace page.')
293 found_iw_match = True
294
295 # If there is a ":", match the prefix against the intrawiki prefixes on OniGalore
296 # before building URL
297 if found_iw_match == False:
298 for prefix in intrawiki_prefixes:
299 if prefix + ":" in page_slug:
300 iw_url = onigalore_url + page_slug
301 if debug: pywikibot.stdout(' Identified namespace {}.'.format(prefix))
302 iw_found += 1
303 found_iw_match = True
304 break
305
306 # If we still haven't turned this match into a URL, something's gone wrong
307 if (found_iw_match == False) or (iw_url == ""):
308 possibly_print(page_name)
309 pywikibot.stdout(' ERROR: Couldn\'t figure out link {}.'.format(page_slug))
310 continue
311
312 # Test the URL
313 iw_url = iw_url.replace(' ', '_')
314 if debug: pywikibot.stdout(' Reading page at {}….'.format(iw_url))
315 test_intrawiki_link(iw_url, page_name, page_slug)
316
317# Print a wrap-up message
318def print_summary():
319 global pages_checked
320 global iw_found
321 global advice_issued
322 global errors_issued
323
324 page_str = "pages"
325 if pages_checked == 1:
326 page_str = "page"
327
328 link_str = "links"
329 if iw_found == 1:
330 link_str = "link"
331
332 pywikibot.stdout('Checked {0} {1} and found {2} intrawiki {3}.'.format(pages_checked, page_str, iw_found, link_str))
333 pywikibot.stdout('While attempting to follow section links….')
334
335 if advice_issued == 0:
336 pywikibot.stdout(' No advice on potential problems was issued.')
337 elif advice_issued == 1:
338 pywikibot.stdout(' 1 piece of advice on a potential problem was issued.')
339 else:
340 pywikibot.stdout(' {} pieces of advice on potential problems were issued.'.format(advice_issued))
341
342 error_str = "errors were"
343 if errors_issued == 1:
344 error_str = "error was"
345 pywikibot.stdout(' {0} {1} encountered.'.format(errors_issued, error_str))
346
347# Main function
348def main(*args):
349 global debug
350 search_cat = ''
351 search_page = ''
352
353 # Process arguments
354 local_args = pywikibot.handle_args(args)
355 for arg in local_args:
356 if arg.startswith('-cat:'):
357 search_cat = arg[5:]
358 elif arg.startswith('-page:'):
359 search_page = arg[6:]
360 elif arg == '-dbg':
361 debug = 1
362 else:
363 pywikibot.stdout('Unknown argument "{}".'.format(arg))
364 return
365
366 site = pywikibot.Site()
367
368 # This line of code enumerates the methods in the 'page' class
369 #pywikibot.stdout(format(dir(page)))
370
371 # Check specified page or loop through specified category and check all pages
372 if search_cat != '':
373 cat_obj = pywikibot.Category(site, search_cat)
374 generator = pagegenerators.CategorizedPageGenerator(cat_obj, recurse=True)
375 for page in pagegenerators.PreloadingGenerator(generator, 100):
376 if debug: pywikibot.stdout('Checking page {0}'.format(page.title()))
377 scan_for_intrawiki_links(page.text, page.title())
378 elif search_page != '':
379 page = pywikibot.Page(site, search_page)
380 if debug: pywikibot.stdout('Checking page {0}'.format(page.title()))
381 scan_for_intrawiki_links(page.text, page.title())
382
383 # Print the results
384 print_summary()
385
386if __name__ == '__main__':
387 main()
Note: See TracBrowser for help on using the repository browser.