00dani/mstdn-ebooks
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity

Merge pull request #12 from SneakyBastardSword/master

Browse source 
precompile regexes
This commit is contained in:
Lynne 2019-02-23 10:23:23 +10:00 committed by GitHub
commit 0e01d5d8ec
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 21 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
.gitignore vendored
View file

@ -9,3 +9,4 @@ toots.db-wal
__pycache__/*
.vscode/
.editorconfig
.*.swp

15
main.py
View file

@ -92,6 +92,13 @@ def handleCtrlC(signal, frame):
signal.signal(signal.SIGINT, handleCtrlC)
patterns = {
"handle": re.compile(r"^.*@(.+)"),
"url": re.compile(r"https?:\/\/(.*)"),
"uri": re.compile(r'template="([^"]+)"'),
"pid": re.compile(r"[^\/]+$"),
}
for f in following:
last_toot = c.execute("SELECT id FROM `toots` WHERE userid LIKE ? ORDER BY id DESC LIMIT 1", (f.id,)).fetchone()
if last_toot != None:
@ -102,9 +109,9 @@ for f in following:
#find the user's activitypub outbox
print("WebFingering... (do not laugh at this. WebFinger is a federated protocol. https://wikipedia.org/wiki/WebFinger)")
instance = re.search(r"^.*@(.+)", f.acct)
instance = patterns["handle"].search(f.acct)
if instance == None:
instance = re.search(r"https?:\/\/(.*)", cfg['site']).group(1)
instance = patterns["url"].search(cfg['site']).group(1)
else:
instance = instance.group(1)
@ -114,7 +121,7 @@ for f in following:
try:
r = requests.get("https://{}/.well-known/host-meta".format(instance), timeout=10)
uri = re.search(r'template="([^"]+)"', r.text).group(1)
uri = patterns["uri"].search(r.text).group(1)
uri = uri.format(uri = "{}@{}".format(f.username, instance))
r = requests.get(uri, headers={"Accept": "application/json"}, timeout=10)
j = r.json()
@ -162,7 +169,7 @@ for f in following:
#you might be wondering, "lynne, what if the instance ratelimits you after 40 posts, and they've made 60 since main.py was last run? wouldn't the bot miss 20 posts and never be able to see them?" to which i reply, "it's called mstdn-ebooks not fediverse-ebooks. pleroma support is an afterthought"
done = True
break
pid = re.search(r"[^\/]+$", oi['object']['id']).group(0)
pid = patterns["pid"].search(oi['object']['id']).group(0)
c.execute("REPLACE INTO toots (id, userid, uri, content) VALUES (?, ?, ?, ?)", (
pid,
f.id,