ViewVC Help

View File | Revision Log | View Changeset | Root Listing

root/Oni2/ValBot/Python/check_interwiki_links.py

Comparing ValBot/Python/check_interwiki_links.py (file contents):
Revision 1169 by iritscen, Mon Feb 21 23:59:20 2022 UTC vs.
Revision 1198 by iritscen, Sun Sep 21 21:50:56 2025 UTC

#	Line 1 | Line 1
1	<	import os
2	<
3	<	from urllib.parse import urljoin
1	>	# Check Interwiki Links
2	>	# by iritscen@yahoo.com
3	>	# Looks at each link on a page (or all the pages in a category) which uses a registered interwiki prefix and loads the linked page, verifying that it exists and that
4	>	# any section link, if present, is valid as well. The output will use the word "ERROR" when it cannot validate the interwiki link.
5	>	# Recommended viewing width:
6	>	# |---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----|
7
8	+	import bs4
9		import pywikibot
10		import re
11	+	import requests # for listing members with dir() when debugging
12
13	<	from pywikibot.bot import QuitKeyboardInterrupt
13	>	from bs4 import BeautifulSoup
14		from pywikibot import pagegenerators
15	<	from pywikibot.tools.formatter import color_format
15	>	from pywikibot.bot import QuitKeyboardInterrupt
16		from pywikibot.comms.http import fetch
17		from pywikibot.specialbots import UploadRobot
18	<	from bs4 import BeautifulSoup
18	>	from pywikibot.tools.formatter import color_format
19	>	from urllib.parse import urljoin
20	>
21	>	class IWLink:
22	>	def __init__(self, iw_prefix, prefix_url, full_url, page_name, page_name_only, page_slug, hosting_page, curl_response):
23	>	self.iw_prefix = iw_prefix # e.g. "wp"
24	>	self.prefix_url = prefix_url # e.g. "https://en.wikipedia.org/wiki/"
25	>	self.full_url = full_url # e.g. "https://en.wikipedia.org/wiki/Marathon_(series)#Rampancy"
26	>	self.page_name = page_name # "Marathon (series)#Rampancy"
27	>	self.page_name_only = page_name # "Marathon (series)"
28	>	self.page_slug = page_slug # "Marathon_(series)#Rampancy"
29	>	self.hosting_page = hosting_page # "Easter eggs"; page where the link was found
30	>	self.curl_response = curl_response # a class defined in the Requests library
31
32		# Parallel arrays based on https://wiki.oni2.net/Special:Interwiki
33		interwiki_prefixes = ('acronym', 'cache', 'commons', 'dictionary', 'google', 'metawikimedia', 'mw', 'wikibooks', 'wikidata', 'wikimedia', 'wikinews', 'wikipedia', 'wikiquote', 'wikisource', 'wikispecies', 'wikiversity', 'wikivoyage', 'wikt', 'wiktionary', 'wp')
34
35		interwiki_urls = ('http://www.acronymfinder.com/~/search/af.aspx?string=exact&Acronym=', 'http://www.google.com/search?q=cache:', 'https://commons.wikimedia.org/wiki/', 'http://www.dict.org/bin/Dict?Database=*&Form=Dict1&Strategy=*&Query=', 'http://www.google.com/search?q=', 'https://meta.wikimedia.org/wiki/', 'https://www.mediawiki.org/wiki/', 'https://en.wikibooks.org/wiki/', 'https://www.wikidata.org/wiki/', 'https://foundation.wikimedia.org/wiki/', 'https://en.wikinews.org/wiki/', 'https://en.wikipedia.org/wiki/', 'https://en.wikiquote.org/wiki/', 'https://wikisource.org/wiki/', 'https://species.wikimedia.org/wiki/', 'https://en.wikiversity.org/wiki/', 'https://en.wikivoyage.org/wiki/', 'https://en.wiktionary.org/wiki/', 'https://en.wiktionary.org/wiki/', 'https://en.wikipedia.org/wiki/')
36
37	+	# Initialize globals
38	+	debug = 0
39		pages_checked = 0
40		iw_found = 0
41	<	problems_found = 0
41	>	errors_issued = 0
42	>	unintended_redirects_found = 0
43	>	name_printed = 0
44	>
45	>	# Prints the name of a page on which something occurred, if it has not been printed before
46	>	def possibly_print(the_link):
47	>	global debug
48	>	global name_printed
49	>
50	>	if not name_printed and not debug:
51	>	pywikibot.stdout('')
52	>	pywikibot.stdout('From page "{}":'.format(the_link.hosting_page))
53	>	name_printed = 1
54	>
55	>	# Search a page for the section specified in the link
56	>	def find_section(the_link, print_result):
57	>	global errors_issued
58	>
59	>	# Isolate section link
60	>	_, anchor_name = the_link.page_slug.split('#')
61	>
62	>	# Convert dot-notation hex entities to proper characters
63	>	replacements = [(r'\.22', '"'), (r'\.27', "'"), (r'\.28', '('), (r'\.29', ')')]
64	>	for pattern, replacement in replacements:
65	>	anchor_name = re.sub(pattern, replacement, anchor_name)
66	>
67	>	# Read linked page to see if it really has this anchor link
68	>	soup = BeautifulSoup(the_link.curl_response.text, 'html.parser')
69	>	tags_to_search = ['span', 'div', 'h2', 'h3', 'h4']
70	>	found_section = False
71	>	for tag_name in tags_to_search:
72	>	for the_tag in soup.find_all(tag_name):
73	>	if the_tag.get('id') == anchor_name:
74	>	found_section = True
75	>	break
76	>	if found_section:
77	>	break
78	>
79	>	# Tell user what we found
80	>	if found_section == False:
81	>	possibly_print(the_link)
82	>	pywikibot.stdout('   ERROR: Could not find section "{0}" on {1} page "{2}".'.format(anchor_name, the_link.iw_prefix, the_link.page_name))
83	>	errors_issued = errors_issued + 1
84	>	elif print_result == True:
85	>	pywikibot.stdout('   The section "{0}" was found on {1} page "{2}".'.format(anchor_name, the_link.iw_prefix, the_link.page_name))
86	>
87	>	# For a link that redirected us to another page, extract the name of the target page from the target page's source
88	>	def find_canonical_link(the_link):
89	>	# Extract link from this markup which contains name of redirected-to page:
90	>	# <link rel="canonical" href="https://en.wikipedia.org/wiki/Page_name"/>
91	>	canonical_name = the_link.curl_response.text.split('<link rel="canonical" href="')[-1]
92	>	prefix_length = len(the_link.prefix_url)
93	>	canonical_name = canonical_name[prefix_length:]
94	>	tag_end = canonical_name.find('">')
95	>
96	>	if tag_end == -1:
97	>	pywikibot.stdout('   ERROR: The {0} link "{1}" is a redirect page, but this script could not isolate the target page name.'.format(the_link.iw_prefix, the_link.page_slug))
98	>	errors_issued = errors_issued + 1
99	>	else:
100	>	canonical_name = canonical_name[:tag_end]
101	>	if len(canonical_name) > 100:
102	>	# Certain things can cause the trim to fail; report error and avoid slamming the output with massive page source from a failed trim
103	>	pywikibot.stdout('   ERROR: The {0} link "{1}" is a redirect to "{2}…" (string overflow).'.format(the_link.iw_prefix, the_link.page_slug, canonical_name[:100]))
104	>	errors_issued = errors_issued + 1
105	>	else:
106	>	the_link.page_name = canonical_name.replace('_', ' ')
107	>	if '#' in the_link.page_slug:
108	>	the_link.page_name_only, _ = the_link.page_slug.split('#')
109	>	pywikibot.stdout('   The {0} link "{1}" is a redirect to "{2}", which is a valid page. Checking for section on that page….'.format(the_link.iw_prefix, the_link.page_name_only, the_link.page_name))
110	>	find_section(the_link, True)
111	>	else:
112	>	pywikibot.stdout('   The {0} link "{1}" is a redirect to "{2}", which is a valid page.'.format(the_link.iw_prefix, the_link.page_slug, the_link.page_name))
113	>
114	>	# Test an interwiki link and look for a section link if applicable
115	>	def test_interwiki_link(the_link):
116	>	global errors_issued
117	>	global unintended_redirects_found
118	>
119	>	the_link.curl_response = fetch(the_link.full_url)
120	>
121	>	# One way we tell that a redirect occurred is by checking fetch's history, as it automatically follows redirects. This will catch formal redirects which come from
122	>	# pages such as Special:PermanentLink.
123	>	if the_link.curl_response.history != []:
124	>	possibly_print(the_link)
125	>
126	>	# If linked page is in all caps, e.g. WP:BEANS, it's likely a deliberate use of a redirect
127	>	if the_link.page_slug.startswith('WP:') and the_link.page_slug == the_link.page_slug.upper():
128	>	pywikibot.stdout('   Got redirection code "{0}" for {1} link "{2}". This appears to be a deliberate use of a Wikipedia shortcut. Checking the target page….'.format(the_link.curl_response.history[0], the_link.iw_prefix, the_link.page_slug))
129	>	find_canonical_link(the_link)
130	>	else:
131	>	permalink1 = 'Special:PermanentLink/'.lower()
132	>	permalink2 = 'Special:Permalink/'.lower()
133	>	page_slug_lower = the_link.page_slug.lower()
134	>	if page_slug_lower.startswith(permalink1) or page_slug_lower.startswith(permalink2):
135	>	pywikibot.stdout('   Got redirection code "{0}" for {1} permanent revision link "{2}". Checking the target page….'.format(the_link.curl_response.history[0], the_link.iw_prefix, the_link.page_slug))
136	>	find_canonical_link(the_link)
137	>	else:
138	>	pywikibot.stdout('   ERROR: Unrecognized type of redirection (code "{0}") for {1} link "{2}". You should check the link manually.'.format(the_link.curl_response.history[0], the_link.iw_prefix, the_link.page_slug))
139	>	errors_issued = errors_issued + 1
140	>	elif the_link.curl_response.status_code != 200:
141	>	possibly_print(the_link)
142	>	pywikibot.stdout('   ERROR: Got response code {0} for {1} link "{2}". The page may not exist.'.format(the_link.curl_response.status_code, the_link.iw_prefix, the_link.page_slug))
143	>	errors_issued = errors_issued + 1
144	>	# However the usual way that a redirect occurs is that MediaWiki redirects us sneakily using JavaScript, while returning code OK 200 as if the link was correct; this
145	>	# happens when a redirect page is accessed. We must detect these soft redirects by looking at the page source to find the redirect note inserted at the top of the
146	>	# page for the reader.
147	>	elif 'Redirected from <a' in the_link.curl_response.text:
148	>	unintended_redirects_found = unintended_redirects_found + 1
149	>	possibly_print(the_link)
150	>	pywikibot.stdout('   WARNING: Got silently redirected by {0} link "{1}". Checking the target page….'.format(the_link.iw_prefix, the_link.page_slug))
151	>	find_canonical_link(the_link) # calls find_section() at end
152	>	elif '#' in the_link.page_slug:
153	>	find_section(the_link, False)
154
155		# Searches the given page text for interwiki links
156	<	def scan_for_iw_links(page_text):
157	<	global pages_checked
158	<	global iw_found
159	<	global problems_found
160	<	pages_checked = pages_checked + 1
161	<	cur = 0
162	<
163	<	for prefix in interwiki_prefixes:
164	<	# Isolate strings that start with "[[prefix:" and end with "|" or "]"
165	<	iw_link = "\[\[" + prefix + ":[^|\]]*(\||\])"
166	<	for match in re.finditer(iw_link, page_text):
167	<	# Extract just the page title from this regex match
168	<	s = match.start() + 2 + len(prefix) + 1
169	<	e = match.end() - 1
170	<
171	<	# Sometimes we used a space char. instead of a '_', so fix that before querying
172	<	page_title = page_text[s:e].replace(' ', '_')
173	<
174	<	# Construct full URL for the particular wiki
175	<	iw_url = interwiki_urls[cur] + page_title
176	<	pywikibot.output('Found {0} link {1}.'.format(prefix, page_title))
177	<	iw_found = iw_found + 1
178	<
179	<	# Adjust URL if this is a foreign-language WP link
180	<	if re.match("^[a-zA-Z]{2}:", page_title):
181	<	lang_code = page_title[0:2] + "."
182	<	# "wp:" is the Wikipedia: namespace, not a language
183	<	if lang_code != "wp." and lang_code != "WP.":
184	<	iw_url = iw_url.replace('en.', lang_code)
185	<	iw_url = iw_url.replace(page_title[0:3], '')
186	<
187	<	# Test the URL
188	<	#pywikibot.output('Testing URL {}...'.format(iw_url))
189	<	response = fetch(iw_url)
190	<
191	<	# Redirects are followed automatically by fetch() and treated as "200"s, so the
192	<	# way we tell that a redirect occurred is by checking the history
193	<	if response.history != []:
194	<	pywikibot.output('WARNING: Redirected from {}.'.format(response.history))
195	<	problems_found = problems_found + 1
196	<	elif response.status_code != 200:
197	<	#pywikibot.output('WARNING: Got response code {}.'.format(response.status_code)) # commented out because fetch() already prints such a msg
198	<	problems_found = problems_found + 1
199	<	elif '#' in page_title:
200	<	# Isolate section link
201	<	pywikibot.output('Detected section link on page {0}.'.format(page_title))
202	<	page_name, anchor_name = page_title.split('#')
203	<
204	<	# Convert dot-notation hex entities to proper characters
205	<	anchor_name = anchor_name.replace('.22', '"')
206	<	anchor_name = anchor_name.replace('.27', '\'')
207	<	anchor_name = anchor_name.replace('.28', '(')
208	<	anchor_name = anchor_name.replace('.29', ')')
209	<
210	<	# Read linked page to see if it really has this anchor link
211	<	soup = BeautifulSoup(response.text, 'html.parser')
212	<	found_section = False
213	<	for tag in soup.findAll('a'):
214	<	link = tag.get('href', None)
215	<	if not link:
216	<	#pywikibot.output('It is not a link.')
217	<	continue
218	<	#pywikibot.output('Got link {0}.'.format(link))
219	<	if not link.startswith('#'):
220	<	continue
221	<
91	<	if link == '#' + anchor_name:
92	<	pywikibot.output('Found section link!')
93	<	found_section = True
94	<	break
95	<	if found_section == False:
96	<	pywikibot.output('Could not find section {0} on page {1}.'.format(anchor_name, page_name))
97	<	problems_found = problems_found + 1
98	<	cur = cur + 1
156	>	def scan_for_interwiki_links(page_text, page_name):
157	>	global debug
158	>	global pages_checked
159	>	global iw_found
160	>	global name_printed
161	>	pages_checked = pages_checked + 1
162	>	cur_prefix = 0
163	>	name_printed = 0
164	>
165	>	for prefix in interwiki_prefixes:
166	>	# Isolate strings that start with "[[prefix:" and end with "|" or "]"
167	>	iw_link = r"\[\[" + prefix + r":[^|\]]*(\||\])"
168	>	for match in re.finditer(iw_link, page_text):
169	>	the_link = IWLink(prefix, interwiki_urls[cur_prefix], "", "", "", "", page_name, "")
170	>
171	>	# Extract just the page title from this regex match
172	>	s = match.start() + 2 + len(the_link.iw_prefix) + 1
173	>	e = match.end() - 1
174	>
175	>	# Use underscores in slug used to constructed URL, but retain spaces for printable name
176	>	the_link.page_slug = page_text[s:e].replace(' ', '_')
177	>	the_link.page_name = page_text[s:e]
178	>	if debug: pywikibot.stdout('   Validating {0} link "{1}"'.format(the_link.iw_prefix, the_link.page_name))
179	>	iw_found = iw_found + 1
180	>
181	>	# Construct full URL for the particular wiki
182	>	the_link.full_url = the_link.prefix_url + the_link.page_slug
183	>
184	>	# Adjust URL if this is a foreign-language WP link
185	>	if re.match("^[a-zA-Z]{2}:", the_link.page_slug):
186	>	lang_code = the_link.page_slug[0:2] + "."
187	>	# "wp:" is the Wikipedia: namespace, not a language
188	>	if lang_code != "wp." and lang_code != "WP.":
189	>	the_link.full_url = the_link.full_url.replace('en.', lang_code)
190	>	the_link.full_url = the_link.full_url.replace(the_link.page_slug[0:3], '')
191	>
192	>	# Test the URL
193	>	test_interwiki_link(the_link)
194	>	cur_prefix = cur_prefix + 1
195	>
196	>	# Print a wrap-up message
197	>	def print_summary():
198	>	global pages_checked
199	>	global iw_found
200	>	global errors_issued
201	>	global unintended_redirects_found
202	>
203	>	page_str = "pages"
204	>	if pages_checked == 1:
205	>	page_str = "page"
206	>
207	>	link_str = "links"
208	>	if iw_found == 1:
209	>	link_str = "link"
210	>
211	>	pywikibot.stdout('Checked {0} {1} and found {2} interwiki {3}.'.format(pages_checked, page_str, iw_found, link_str))
212	>
213	>	error_str = "errors were"
214	>	if errors_issued == 1:
215	>	error_str = "error was"
216	>
217	>	pywikibot.stdout('{0} {1} encountered in validating these links.'.format(errors_issued, error_str))
218	>
219	>	warning_str = "likely-unintended redirects were"
220	>	if unintended_redirects_found == 1:
221	>	warning_str = "likely-unintended redirect was"
222
223	<	def main(*args):
101	<	cat_name = ''
102	<	page_name = ''
223	>	pywikibot.stdout('{0} {1} encountered in validating these links.'.format(unintended_redirects_found, warning_str))
224
225	<	local_args = pywikibot.handle_args(args)
226	<	genFactory = pagegenerators.GeneratorFactory()
225	>	# Main function
226	>	def main(*args):
227	>	global debug
228	>	search_cat = ''
229	>	search_page = ''
230	>
231	>	# Process arguments
232	>	local_args = pywikibot.handle_args(args)
233	>	for arg in local_args:
234	>	if arg.startswith('-cat:'):
235	>	search_cat = arg[5:]
236	>	elif arg.startswith('-page:'):
237	>	search_page = arg[6:]
238	>	elif arg == '-dbg':
239	>	debug = 1
240	>	else:
241	>	pywikibot.stdout('Unknown argument "{}". Exiting.'.format(arg))
242	>	return
243	>
244	>	#pywikibot.stdout('The members of the requests.models.Response class are:')
245	>	#pywikibot.stdout(format(dir(requests.models.Response)))
246	>	#return
247	>
248	>	# Check specified page or loop through specified category and check all pages
249	>	site = pywikibot.Site()
250	>	if search_cat != '':
251	>	cat_obj = pywikibot.Category(site, search_cat)
252	>	generator = pagegenerators.CategorizedPageGenerator(cat_obj, recurse=True)
253	>	for page in pagegenerators.PreloadingGenerator(generator, 100):
254	>	if debug: pywikibot.stdout('Checking page "{}"'.format(page.title()))
255	>	scan_for_interwiki_links(page.text, page.title())
256	>	elif search_page != '':
257	>	page = pywikibot.Page(site, search_page)
258	>	if debug: pywikibot.stdout('Checking page "{}"'.format(page.title()))
259	>	scan_for_interwiki_links(page.text, page.title())
260
261	<	for arg in local_args:
262	<	if arg.startswith('-cat:'):
109	<	cat_name = arg[5:]
110	<	elif arg.startswith('-page:'):
111	<	page_name = arg[6:]
112	<
113	<	site = pywikibot.Site()
114	<
115	<	# This line of code enumerates the methods in the 'page' class
116	<	#pywikibot.stdout(format(dir(page)))
117	<
118	<	if cat_name != '':
119	<	cat_obj = pywikibot.Category(site, cat_name)
120	<	generator = pagegenerators.CategorizedPageGenerator(cat_obj, recurse=True)
121	<	for page in pagegenerators.PreloadingGenerator(generator, 100):
122	<	pywikibot.stdout('Checking page {0}'.format(page.title()))
123	<	scan_for_iw_links(page.text)
124	<	elif page_name != '':
125	<	page = pywikibot.Page(site, page_name)
126	<	pywikibot.stdout('Checking page {0}'.format(page.title()))
127	<	scan_for_iw_links(page.text)
128	<
129	<	global pages_checked
130	<	global iw_found
131	<	global problems_found
132	<	pywikibot.stdout('Checked {0} page(s) and found {1} interwiki link(s) with {2} problem(s).'.format(pages_checked, iw_found, problems_found))
261	>	# Print the results
262	>	print_summary()
263
264		if __name__ == '__main__':
265	<	main()
265	>	main()

Diff Legend

–	Removed lines
+	Added lines
<	Changed lines (old)
>	Changed lines (new)