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 1185 by iritscen, Tue Aug 15 02:03:16 2023 UTC

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

Diff Legend

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