mirror of
https://gitlab.com/chicken-riders/RcGcDb.git
synced 2025-02-23 00:54:09 +00:00
Additional work done on the ratelimiting
This commit is contained in:
parent
c1831b992b
commit
a4462369bb
43
src/bot.py
43
src/bot.py
|
@ -12,7 +12,7 @@ from src.argparser import command_line_args
|
|||
from src.config import settings
|
||||
from src.database import db_cursor
|
||||
from src.exceptions import *
|
||||
from src.misc import get_paths
|
||||
from src.misc import get_paths, get_domain
|
||||
from src.msgqueue import messagequeue
|
||||
from src.queue_handler import DBHandler
|
||||
from src.wiki import Wiki, process_cats, process_mwmsgs, essential_info, essential_feeds
|
||||
|
@ -41,11 +41,11 @@ for wiki in db_cursor.execute('SELECT DISTINCT wiki FROM rcgcdw'):
|
|||
# Start queueing logic
|
||||
|
||||
|
||||
def calculate_delay() -> float:
|
||||
def calculate_delay_for_group(group_length: int) -> float:
|
||||
"""Calculate the delay between fetching each wiki to avoid rate limits"""
|
||||
min_delay = 60 / settings["max_requests_per_minute"]
|
||||
if (len(all_wikis) * min_delay) < settings["minimal_cooldown_per_wiki_in_sec"]:
|
||||
return settings["minimal_cooldown_per_wiki_in_sec"] / len(all_wikis)
|
||||
if (group_length * min_delay) < settings["minimal_cooldown_per_wiki_in_sec"]:
|
||||
return settings["minimal_cooldown_per_wiki_in_sec"] / group_length
|
||||
else:
|
||||
return min_delay
|
||||
|
||||
|
@ -62,15 +62,18 @@ def generate_targets(wiki_url: str) -> defaultdict:
|
|||
return combinations
|
||||
|
||||
|
||||
async def wiki_scanner():
|
||||
"""Wiki scanner is spawned as a task which purpose is to continuously run over wikis in the DB, fetching recent changes
|
||||
to add messages based on the changes to message queue later handled by message_sender coroutine."""
|
||||
try:
|
||||
while True:
|
||||
calc_delay = calculate_delay()
|
||||
fetch_all = db_cursor.execute(
|
||||
'SELECT webhook, wiki, lang, display, wikiid, rcid, postid FROM rcgcdw GROUP BY wiki')
|
||||
async def generate_domain_groups(): # oh boy, I cannot wait to learn about async generators
|
||||
combinations = defaultdict(list)
|
||||
fetch_all = db_cursor.execute('SELECT webhook, wiki, lang, display, wikiid, rcid FROM rcgcdw GROUP BY wiki')
|
||||
for db_wiki in fetch_all.fetchall():
|
||||
combinations[get_domain(db_wiki["wiki"])].append(db_wiki)
|
||||
for item in combinations.values():
|
||||
yield item
|
||||
|
||||
|
||||
async def scan_group(group: list):
|
||||
calc_delay = calculate_delay_for_group(len(group))
|
||||
for db_wiki in group:
|
||||
logger.debug("Wiki {}".format(db_wiki["wiki"]))
|
||||
if db_wiki["wiki"] not in all_wikis:
|
||||
logger.info("Registering new wiki locally: {}".format(db_wiki["wiki"]))
|
||||
|
@ -133,7 +136,21 @@ async def wiki_scanner():
|
|||
await formatter_exception_logger(db_wiki["wiki"], change, traceback.format_exc())
|
||||
if recent_changes:
|
||||
DBHandler.add(db_wiki["wiki"], change["rcid"])
|
||||
await asyncio.sleep(delay=2.0) # temporary measure until rate limiting is not implemented
|
||||
await asyncio.sleep(delay=calc_delay)
|
||||
|
||||
|
||||
async def wiki_scanner():
|
||||
"""Wiki scanner is spawned as a task which purpose is to continuously run over wikis in the DB, fetching recent changes
|
||||
to add messages based on the changes to message queue later handled by message_sender coroutine."""
|
||||
try:
|
||||
while True:
|
||||
async for group in generate_domain_groups():
|
||||
asyncio.create_task(scan_group(group))
|
||||
|
||||
fetch_all = db_cursor.execute(
|
||||
'SELECT webhook, wiki, lang, display, wikiid, rcid, postid FROM rcgcdw GROUP BY wiki')
|
||||
for db_wiki in fetch_all.fetchall():
|
||||
|
||||
if db_wiki["wikiid"] is not None:
|
||||
header = settings["header"]
|
||||
header["Accept"] = "application/hal+json"
|
||||
|
|
|
@ -17,6 +17,12 @@ def get_paths(wiki: str, request) -> tuple:
|
|||
return WIKI_API_PATH, WIKI_SCRIPT_PATH, WIKI_ARTICLE_PATH, WIKI_JUST_DOMAIN
|
||||
|
||||
|
||||
def get_domain(url: str) -> str:
|
||||
"""Get domain of given URL"""
|
||||
parsed_url = urlparse(url)
|
||||
return ".".join(urlunparse((*parsed_url[0:2], "", "", "", "")).split(".")[-2:]) # something like gamepedia.com, fandom.com
|
||||
|
||||
|
||||
class LinkParser(HTMLParser):
|
||||
|
||||
new_string = ""
|
||||
|
|
Loading…
Reference in a new issue