Merge pull request #56 from Teqed/feat/lemmy

feat: lemmy
This commit is contained in:
Michael 2023-07-01 20:43:58 +01:00 committed by GitHub
commit f7d015004e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23

View file

@ -74,7 +74,7 @@ def add_user_posts(server, access_token, followings, know_followings, all_known_
count = 0 count = 0
failed = 0 failed = 0
for post in posts: for post in posts:
if post['reblog'] == None and post['url'] != None and post['url'] not in seen_urls: if post.get('reblog') is None and post.get('url') is not None and post.get('url') not in seen_urls:
added = add_post_with_context(post, server, access_token, seen_urls) added = add_post_with_context(post, server, access_token, seen_urls)
if added is True: if added is True:
seen_urls.add(post['url']) seen_urls.add(post['url'])
@ -90,7 +90,7 @@ def add_post_with_context(post, server, access_token, seen_urls):
added = add_context_url(post['url'], server, access_token) added = add_context_url(post['url'], server, access_token)
if added is True: if added is True:
seen_urls.add(post['url']) seen_urls.add(post['url'])
if (post['replies_count'] or post['in_reply_to_id']) and arguments.backfill_with_context > 0: if ('replies_count' in post or 'in_reply_to_id' in post) and getattr(arguments, 'backfill_with_context', 0) > 0:
parsed_urls = {} parsed_urls = {}
parsed = parse_url(post['url'], parsed_urls) parsed = parse_url(post['url'], parsed_urls)
if parsed == None: if parsed == None:
@ -113,6 +113,37 @@ def get_user_posts(user, know_followings, server):
log(f"{user['acct']} is a local user. Skip") log(f"{user['acct']} is a local user. Skip")
know_followings.add(user['acct']) know_followings.add(user['acct'])
return None return None
if re.match(r"^https:\/\/[^\/]+\/c\/", user['url']):
try:
url = f"https://{parsed_url[0]}/api/v3/post/list?community_name={parsed_url[1]}&sort=New&limit=50"
response = get(url)
if(response.status_code == 200):
posts = [post['post'] for post in response.json()['posts']]
for post in posts:
post['url'] = post['ap_id']
return posts
except Exception as ex:
log(f"Error getting community posts for community {parsed_url[1]}: {ex}")
return None
if re.match(r"^https:\/\/[^\/]+\/u\/", user['url']):
try:
url = f"https://{parsed_url[0]}/api/v3/user?username={parsed_url[1]}&sort=New&limit=50"
response = get(url)
if(response.status_code == 200):
comments = [post['post'] for post in response.json()['comments']]
posts = [post['post'] for post in response.json()['posts']]
all_posts = comments + posts
for post in all_posts:
post['url'] = post['ap_id']
return all_posts
except Exception as ex:
log(f"Error getting user posts for user {parsed_url[1]}: {ex}")
return None
try: try:
user_id = get_user_id(parsed_url[0], parsed_url[1]) user_id = get_user_id(parsed_url[0], parsed_url[1])
@ -356,21 +387,24 @@ def get_reply_toots(user_id, server, access_token, seen_urls, reply_since):
) )
def get_all_known_context_urls(server, reply_toots,parsed_urls): def get_all_known_context_urls(server, reply_toots, parsed_urls):
"""get the context toots of the given toots from their original server""" """get the context toots of the given toots from their original server"""
known_context_urls = set( known_context_urls = set()
filter(
lambda url: not url.startswith(f"https://{server}/"), for toot in reply_toots:
itertools.chain.from_iterable( if toot_has_parseable_url(toot, parsed_urls):
get_toot_context(*parse_url(toot["url"] if toot["reblog"] is None else toot["reblog"]["url"],parsed_urls), toot["url"]) url = toot["url"] if toot["reblog"] is None else toot["reblog"]["url"]
for toot in filter( parsed_url = parse_url(url, parsed_urls)
lambda toot: toot_has_parseable_url(toot,parsed_urls), context = get_toot_context(parsed_url[0], parsed_url[1], url)
reply_toots if context is not None:
) for item in context:
), known_context_urls.add(item)
) else:
) log(f"Error getting context for toot {url}")
known_context_urls = set(filter(lambda url: not url.startswith(f"https://{server}/"), known_context_urls))
log(f"Found {len(known_context_urls)} known context toots") log(f"Found {len(known_context_urls)} known context toots")
return known_context_urls return known_context_urls
@ -435,6 +469,11 @@ def parse_user_url(url):
if match is not None: if match is not None:
return match return match
match = parse_lemmy_profile_url(url)
if match is not None:
return match
# Pixelfed profile paths do not use a subdirectory, so we need to match for them last.
match = parse_pixelfed_profile_url(url) match = parse_pixelfed_profile_url(url)
if match is not None: if match is not None:
return match return match
@ -454,6 +493,11 @@ def parse_url(url, parsed_urls):
if match is not None: if match is not None:
parsed_urls[url] = match parsed_urls[url] = match
if url not in parsed_urls:
match = parse_lemmy_url(url)
if match is not None:
parsed_urls[url] = match
if url not in parsed_urls: if url not in parsed_urls:
match = parse_pixelfed_url(url) match = parse_pixelfed_url(url)
if match is not None: if match is not None:
@ -468,7 +512,7 @@ def parse_url(url, parsed_urls):
def parse_mastodon_profile_url(url): def parse_mastodon_profile_url(url):
"""parse a Mastodon Profile URL and return the server and username""" """parse a Mastodon Profile URL and return the server and username"""
match = re.match( match = re.match(
r"https://(?P<server>.*)/@(?P<username>.*)", url r"https://(?P<server>[^/]+)/@(?P<username>[^/]+)", url
) )
if match is not None: if match is not None:
return (match.group("server"), match.group("username")) return (match.group("server"), match.group("username"))
@ -477,7 +521,7 @@ def parse_mastodon_profile_url(url):
def parse_mastodon_url(url): def parse_mastodon_url(url):
"""parse a Mastodon URL and return the server and ID""" """parse a Mastodon URL and return the server and ID"""
match = re.match( match = re.match(
r"https://(?P<server>.*)/@(?P<username>.*)/(?P<toot_id>.*)", url r"https://(?P<server>[^/]+)/@(?P<username>[^/]+)/(?P<toot_id>[^/]+)", url
) )
if match is not None: if match is not None:
return (match.group("server"), match.group("toot_id")) return (match.group("server"), match.group("toot_id"))
@ -486,14 +530,14 @@ def parse_mastodon_url(url):
def parse_pleroma_url(url): def parse_pleroma_url(url):
"""parse a Pleroma URL and return the server and ID""" """parse a Pleroma URL and return the server and ID"""
match = re.match(r"https://(?P<server>.*)/objects/(?P<toot_id>.*)", url) match = re.match(r"https://(?P<server>[^/]+)/objects/(?P<toot_id>[^/]+)", url)
if match is not None: if match is not None:
server = match.group("server") server = match.group("server")
url = get_redirect_url(url) url = get_redirect_url(url)
if url is None: if url is None:
return None return None
match = re.match(r"/notice/(?P<toot_id>.*)", url) match = re.match(r"/notice/(?P<toot_id>[^/]+)", url)
if match is not None: if match is not None:
return (server, match.group("toot_id")) return (server, match.group("toot_id"))
return None return None
@ -501,7 +545,7 @@ def parse_pleroma_url(url):
def parse_pleroma_profile_url(url): def parse_pleroma_profile_url(url):
"""parse a Pleroma Profile URL and return the server and username""" """parse a Pleroma Profile URL and return the server and username"""
match = re.match(r"https://(?P<server>.*)/users/(?P<username>.*)", url) match = re.match(r"https://(?P<server>[^/]+)/users/(?P<username>[^/]+)", url)
if match is not None: if match is not None:
return (match.group("server"), match.group("username")) return (match.group("server"), match.group("username"))
return None return None
@ -509,7 +553,7 @@ def parse_pleroma_profile_url(url):
def parse_pixelfed_url(url): def parse_pixelfed_url(url):
"""parse a Pixelfed URL and return the server and ID""" """parse a Pixelfed URL and return the server and ID"""
match = re.match( match = re.match(
r"https://(?P<server>.*)/p/(?P<username>.*)/(?P<toot_id>.*)", url r"https://(?P<server>[^/]+)/p/(?P<username>[^/]+)/(?P<toot_id>[^/]+)", url
) )
if match is not None: if match is not None:
return (match.group("server"), match.group("toot_id")) return (match.group("server"), match.group("toot_id"))
@ -517,11 +561,26 @@ def parse_pixelfed_url(url):
def parse_pixelfed_profile_url(url): def parse_pixelfed_profile_url(url):
"""parse a Pixelfed Profile URL and return the server and username""" """parse a Pixelfed Profile URL and return the server and username"""
match = re.match(r"https://(?P<server>.*)/(?P<username>.*)", url) match = re.match(r"https://(?P<server>[^/]+)/(?P<username>[^/]+)", url)
if match is not None: if match is not None:
return (match.group("server"), match.group("username")) return (match.group("server"), match.group("username"))
return None return None
def parse_lemmy_url(url):
"""parse a Lemmy URL and return the server, and ID"""
match = re.match(
r"https://(?P<server>[^/]+)/(?:comment|post)/(?P<toot_id>[^/]+)", url
)
if match is not None:
return (match.group("server"), match.group("toot_id"))
return None
def parse_lemmy_profile_url(url):
"""parse a Lemmy Profile URL and return the server and username"""
match = re.match(r"https://(?P<server>[^/]+)/(?:u|c)/(?P<username>[^/]+)", url)
if match is not None:
return (match.group("server"), match.group("username"))
return None
def get_redirect_url(url): def get_redirect_url(url):
"""get the URL given URL redirects to""" """get the URL given URL redirects to"""
@ -559,6 +618,10 @@ def get_all_context_urls(server, replied_toot_ids):
def get_toot_context(server, toot_id, toot_url): def get_toot_context(server, toot_id, toot_url):
"""get the URLs of the context toots of the given toot""" """get the URLs of the context toots of the given toot"""
if toot_url.find("/comment/") != -1:
return get_comment_context(server, toot_id, toot_url)
if toot_url.find("/post/") != -1:
return get_comments_urls(server, toot_id, toot_url)
url = f"https://{server}/api/v1/statuses/{toot_id}/context" url = f"https://{server}/api/v1/statuses/{toot_id}/context"
try: try:
resp = get(url) resp = get(url)
@ -585,6 +648,72 @@ def get_toot_context(server, toot_id, toot_url):
) )
return [] return []
def get_comment_context(server, toot_id, toot_url):
"""get the URLs of the context toots of the given toot"""
comment = f"https://{server}/api/v3/comment?id={toot_id}"
try:
resp = get(comment)
except Exception as ex:
log(f"Error getting comment {toot_id} from {toot_url}. Exception: {ex}")
return []
if resp.status_code == 200:
try:
res = resp.json()
post_id = res['comment_view']['comment']['post_id']
return get_comments_urls(server, post_id, toot_url)
except Exception as ex:
log(f"Error parsing context for comment {toot_url}. Exception: {ex}")
return []
elif resp.status_code == 429:
reset = datetime.strptime(resp.headers['x-ratelimit-reset'], '%Y-%m-%dT%H:%M:%S.%fZ')
log(f"Rate Limit hit when getting context for {toot_url}. Waiting to retry at {resp.headers['x-ratelimit-reset']}")
time.sleep((reset - datetime.now()).total_seconds() + 1)
return get_comment_context(server, toot_id, toot_url)
def get_comments_urls(server, post_id, toot_url):
"""get the URLs of the comments of the given post"""
urls = []
url = f"https://{server}/api/v3/post?id={post_id}"
try:
resp = get(url)
except Exception as ex:
log(f"Error getting post {post_id} from {toot_url}. Exception: {ex}")
return []
if resp.status_code == 200:
try:
res = resp.json()
if res['post_view']['counts']['comments'] == 0:
return []
urls.append(res['post_view']['post']['ap_id'])
except Exception as ex:
log(f"Error parsing post {post_id} from {toot_url}. Exception: {ex}")
url = f"https://{server}/api/v3/comment/list?post_id={post_id}&sort=New&limit=50"
try:
resp = get(url)
except Exception as ex:
log(f"Error getting comments for post {post_id} from {toot_url}. Exception: {ex}")
return []
if resp.status_code == 200:
try:
res = resp.json()
list_of_urls = [comment_info['comment']['ap_id'] for comment_info in res['comments']]
log(f"Got {len(list_of_urls)} comments for post {toot_url}")
urls.extend(list_of_urls)
return urls
except Exception as ex:
log(f"Error parsing comments for post {toot_url}. Exception: {ex}")
elif resp.status_code == 429:
reset = datetime.strptime(resp.headers['x-ratelimit-reset'], '%Y-%m-%dT%H:%M:%S.%fZ')
log(f"Rate Limit hit when getting comments for {toot_url}. Waiting to retry at {resp.headers['x-ratelimit-reset']}")
time.sleep((reset - datetime.now()).total_seconds() + 1)
return get_comments_urls(server, post_id, toot_url)
log(f"Error getting comments for post {toot_url}. Status code: {resp.status_code}")
return []
def add_context_urls(server, access_token, context_urls, seen_urls): def add_context_urls(server, access_token, context_urls, seen_urls):
"""add the given toot URLs to the server""" """add the given toot URLs to the server"""
@ -966,4 +1095,4 @@ if __name__ == "__main__":
get(f"{arguments.on_fail}?rid={runId}") get(f"{arguments.on_fail}?rid={runId}")
except Exception as ex: except Exception as ex:
log(f"Error getting callback url: {ex}") log(f"Error getting callback url: {ex}")
raise raise