From dfd01567cba515e8ebe9e9adae7b983a88a9b804 Mon Sep 17 00:00:00 2001
From: JustAnotherArchivist <JustAnotherArchivist@users.noreply.github.com>
Date: Sun, 7 Apr 2024 03:29:42 +0000
Subject: [PATCH] Fix org listings not including archived repos

While the 'All' listing currently only excludes archived repos, this is a more general solution that, as long as they don't redesign the repository list yet again, should work even with further category display changes.
---
 github-list-repos | 39 ++++++++++++++++++++++++++-------------
 1 file changed, 26 insertions(+), 13 deletions(-)

diff --git a/github-list-repos b/github-list-repos
index 4ac522a..ab45060 100755
--- a/github-list-repos
+++ b/github-list-repos
@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+import collections
 import html
 import logging
 import re
@@ -47,7 +48,7 @@ def p(repoName):
 for user in users:
 	r = get(f'https://github.com/{user}')
 	if '<div id="org-repositories"' in r.text:
-		# Organisation, complete list under /orgs/ with ?page=2 pagination
+		# Organisation, archived repositories don't appear on /orgs/ + pagination, so need to also iterate over all the 'type' parameters
 		if mode == NAME_OPTION:
 			musername = re.search(r'<meta property="profile:username" content="([^"]*)" />', r.text)
 			if not musername:
@@ -60,18 +61,30 @@ for user in users:
 			print(html.unescape(musername.group(1).strip().replace('\n', ' ').replace('\r', ' ')))
 			print(html.unescape(mfullname.group(1).strip().replace('\n', ' ').replace('\r', ' ')))
 			sys.exit(0)
-		r = get(f'https://github.com/orgs/{user}/repositories')
-		page = 1
-		while True:
-			for m in re.finditer(r'<a itemprop="name codeRepository"\s(?:[^>]*\s)?data-hovercard-url="/([^/>"]+/[^/>"]+)/hovercard"', r.text):
-				p(m.group(1))
-			for m in re.finditer(r'<a data-testid="listitem-title-link"\s(?:[^>]*\s)?href="/([^/>"]+/[^/>"]+)"', r.text):
-				p(m.group(1))
-			if '<a class="next_page"' not in r.text and '<a rel="next"' not in r.text:
-				# End of pagination
-				break
-			page += 1
-			r = get(f'https://github.com/orgs/{user}/repositories?page={page}')
+		types = collections.deque()
+		types.append('')
+		seen = set()
+		def maybe_p(repoName):
+			if repoName not in seen:
+				p(repoName)
+				seen.add(repoName)
+		while types:
+			type_ = types.popleft()
+			j = '&' if type_ else '?'
+			r = get(f'https://github.com/orgs/{user}/repositories{type_}')
+			if not type_:
+				types.extend(x.split('"')[1] for x in re.findall(r'href="\?type=[^"]*', r.text))
+			page = 1
+			while True:
+				for m in re.finditer(r'<a itemprop="name codeRepository"\s(?:[^>]*\s)?data-hovercard-url="/([^/>"]+/[^/>"]+)/hovercard"', r.text):
+					maybe_p(m.group(1))
+				for m in re.finditer(r'<a data-testid="listitem-title-link"\s(?:[^>]*\s)?href="/([^/>"]+/[^/>"]+)"', r.text):
+					maybe_p(m.group(1))
+				if '<a class="next_page"' not in r.text and '<a rel="next"' not in r.text:
+					# End of pagination
+					break
+				page += 1
+				r = get(f'https://github.com/orgs/{user}/repositories{type_}{j}page={page}')
 	else:
 		# User, ?tab=repositories + cursor pagination
 		if mode == NAME_OPTION: