Changeset 1205 for ValBot/Python/check_intrawiki_section_links.py

Timestamp:

Feb 9, 2026, 4:24:16 AM (7 hours ago)

Author:

iritscen

Message:

ValBot: Revised logic in check_intrawiki_section_links.py as MediaWiki now apparently returns response 301 when the user is redirected by a redirect page.

File:

• ValBot/Python/check_intrawiki_section_links.py (modified) (6 diffs)

ValBot/Python/check_intrawiki_section_links.py

@@ -16 +16 @@
 from pywikibot.bot import QuitKeyboardInterrupt
 from pywikibot import pagegenerators
-from pywikibot.tools.formatter import color_format
 from pywikibot.comms.http import fetch
 from pywikibot.specialbots import UploadRobot
@@ -79 +78 @@
       if search_terms[-1].startswith('-'):
          search_terms.pop()
-      # Remake text directive with the terms separated by spaces as they should be in the page text
+      # Remake text directive with the terms separated by spaces as they should be in the page
+      # text
       newSep = ' ' 
       search_string = newSep.join(search_terms)
@@ -114 +114 @@
       pywikibot.stdout('   The section "{0}" was found on page "{1}".'.format(anchor_name, target_page_name_human))
 
-# For a link that redirected us to another page, extract the name of the target page from
-# the target page's source
+# For a link that redirected us to another page, extract the name of the target page from the
+# target page's source
 def find_canonical_link(page_text, page_name, page_slug):
    # Extract link from this markup which contains name of redirected-to page:
@@ -129 +129 @@
       canonical_name = canonical_name[:tag_end]
       if len(canonical_name) > 100:
-         # Certain things can cause the trim to fail; report error and avoid slamming the
-         # output with massive page source from a failed trim
+         # Certain things can cause the trim to fail; report error and avoid slamming the output
+         # with massive page source from a failed trim
          pywikibot.stdout('   ERROR: The link "{}" is a redirect to "{2}…" (string overflow).'.format(page_slug, canonical_name[:100]))
          errors_issued = errors_issued + 1
@@ -152 +152 @@
    # automatically follows redirects. This will catch formal redirects which come from pages
    # such as Special:PermanentLink.
-   if response.history != []:
-      permalink1 = 'Special:PermanentLink/'.lower()
-      permalink2 = 'Special:Permalink/'.lower()
-      page_slug_lower = page_slug.lower()
-      if page_slug_lower.startswith(permalink1) or page_slug_lower.startswith(permalink2):
-         if debug:
-            possibly_print(page_name)
-            pywikibot.stdout('   Got redirection code "{0}" for permanent revision link "{1}". Checking the target page….'.format(response.history[0], page_slug))
-         find_canonical_link(response.text, page_name, page_slug)
-      else:
+   permalink1 = 'Special:PermanentLink/'.lower()
+   permalink2 = 'Special:Permalink/'.lower()
+   page_slug_lower = page_slug.lower()
+   if response.history != [] and (page_slug_lower.startswith(permalink1) or page_slug_lower.startswith(permalink2)):
+      if debug:
          possibly_print(page_name)
-         pywikibot.stdout('   ERROR: Unrecognized type of redirection (code "{0}") for link "{1}". You should check the link manually.'.format(response.history[0], page_slug))
-         advice_issued += 1
-   elif response.status_code != 200:
-      possibly_print(page_name)
-      pywikibot.stdout('   ERROR: Got response code {0} on URL {1}. The target page may not exist.'.format(response.status_code, iw_url))
-      errors_issued += 1
-   # 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 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 page for the reader.
+         pywikibot.stdout('   Got redirection code "{0}" for permanent revision link "{1}". Checking the target page….'.format(response.history[0], page_slug))
+      find_canonical_link(response.text, page_name, page_slug)
+   # However the usual way that a redirect occurs is that a redirect page is visited and
+   # MediaWiki sends us to the new page using JavaScript while returning code 301. Formerly it
+   # used to return 200 as if the link was correct, so rather than looking for code 301 we
+   # detect these soft redirects by looking at the page source to find the redirect note that
+   # gets inserted at the top of the page for the reader.
    elif 'Redirected from <a' in response.text:
       if debug:
@@ -178 +170 @@
          pywikibot.stdout('   Got silently redirected by link "{}". Checking the target page….'.format(page_slug))
       find_canonical_link(response.text, page_name, page_slug)
+   # This handles response codes other than 200 and 301 (301 is returned in the above case of a
+   # silent redirect)
+   elif response.status_code != 200:
+      possibly_print(page_name)
+      pywikibot.stdout('   ERROR: Got response code {0} on URL {1}. The target page may not exist.'.format(response.status_code, iw_url))
+      errors_issued += 1
    else: # URL is OK, so proceed
       find_section(response.text, page_name, page_slug, False)