feat(scraper): use backoff

This commit is contained in:
Barrett Ruth 2025-09-21 11:26:54 -04:00
parent 58f9be5f9a
commit 46c615416f
6 changed files with 186 additions and 193 deletions

View file

@ -3,9 +3,9 @@
import json
import re
import sys
import time
from dataclasses import asdict
import backoff
import requests
from bs4 import BeautifulSoup, Tag
@ -169,7 +169,6 @@ def scrape(url: str) -> list[TestCase]:
def scrape_contests() -> list[ContestSummary]:
import concurrent.futures
import random
def get_max_pages() -> int:
try:
@ -197,168 +196,161 @@ def scrape_contests() -> list[ContestSummary]:
except Exception:
return 15
def scrape_page_with_retry(page: int, max_retries: int = 3) -> list[ContestSummary]:
for attempt in range(max_retries):
def scrape_page(page: int) -> list[ContestSummary]:
@backoff.on_exception(
backoff.expo,
(requests.exceptions.RequestException, requests.exceptions.HTTPError),
max_tries=4,
jitter=backoff.random_jitter,
on_backoff=lambda details: print(
f"Request failed on page {page} (attempt {details['tries']}), retrying in {details['wait']:.1f}s: {details['exception']}",
file=sys.stderr,
),
)
@backoff.on_predicate(
backoff.expo,
lambda response: response.status_code == 429,
max_tries=4,
jitter=backoff.random_jitter,
on_backoff=lambda details: print(
f"Rate limited on page {page}, retrying in {details['wait']:.1f}s",
file=sys.stderr,
),
)
def make_request() -> requests.Response:
headers = {
"User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
}
url = f"https://atcoder.jp/contests/archive?page={page}"
response = requests.get(url, headers=headers, timeout=10)
response.raise_for_status()
return response
try:
response = make_request()
except Exception:
return []
soup = BeautifulSoup(response.text, "html.parser")
table = soup.find("table", class_="table")
if not table:
return []
tbody = table.find("tbody")
if not tbody or not isinstance(tbody, Tag):
return []
rows = tbody.find_all("tr")
if not rows:
return []
contests = []
for row in rows:
cells = row.find_all("td")
if len(cells) < 2:
continue
contest_cell = cells[1]
link = contest_cell.find("a")
if not link or not link.get("href"):
continue
href = link.get("href")
contest_id = href.split("/")[-1]
name = link.get_text().strip()
try:
headers = {
"User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
}
url = f"https://atcoder.jp/contests/archive?page={page}"
response = requests.get(url, headers=headers, timeout=10)
name = name.encode().decode("unicode_escape")
except (UnicodeDecodeError, UnicodeEncodeError):
pass
if response.status_code == 429:
backoff_time = (2**attempt) + random.uniform(0, 1)
print(
f"Rate limited on page {page}, retrying in {backoff_time:.1f}s",
file=sys.stderr,
)
time.sleep(backoff_time)
continue
name = (
name.replace("\uff08", "(")
.replace("\uff09", ")")
.replace("\u3000", " ")
)
name = re.sub(
r"[\uff01-\uff5e]", lambda m: chr(ord(m.group()) - 0xFEE0), name
)
response.raise_for_status()
def generate_display_name_from_id(contest_id: str) -> str:
parts = contest_id.replace("-", " ").replace("_", " ")
soup = BeautifulSoup(response.text, "html.parser")
table = soup.find("table", class_="table")
if not table:
return []
tbody = table.find("tbody")
if not tbody or not isinstance(tbody, Tag):
return []
rows = tbody.find_all("tr")
if not rows:
return []
contests = []
for row in rows:
cells = row.find_all("td")
if len(cells) < 2:
continue
contest_cell = cells[1]
link = contest_cell.find("a")
if not link or not link.get("href"):
continue
href = link.get("href")
contest_id = href.split("/")[-1]
name = link.get_text().strip()
try:
name = name.encode().decode("unicode_escape")
except (UnicodeDecodeError, UnicodeEncodeError):
pass
name = (
name.replace("\uff08", "(")
.replace("\uff09", ")")
.replace("\u3000", " ")
)
name = re.sub(
r"[\uff01-\uff5e]", lambda m: chr(ord(m.group()) - 0xFEE0), name
)
def generate_display_name_from_id(contest_id: str) -> str:
parts = contest_id.replace("-", " ").replace("_", " ")
parts = re.sub(
r"\b(jsc|JSC)\b",
"Japanese Student Championship",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(wtf|WTF)\b",
"World Tour Finals",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(ahc)(\d+)\b",
r"Heuristic Contest \2 (AHC)",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(arc)(\d+)\b",
r"Regular Contest \2 (ARC)",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(abc)(\d+)\b",
r"Beginner Contest \2 (ABC)",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(agc)(\d+)\b",
r"Grand Contest \2 (AGC)",
parts,
flags=re.IGNORECASE,
)
return parts.title()
english_chars = sum(1 for c in name if c.isascii() and c.isalpha())
total_chars = len(re.sub(r"\s+", "", name))
if total_chars > 0 and english_chars / total_chars < 0.3:
display_name = generate_display_name_from_id(contest_id)
else:
display_name = name
if "AtCoder Beginner Contest" in name:
match = re.search(r"AtCoder Beginner Contest (\d+)", name)
if match:
display_name = (
f"Beginner Contest {match.group(1)} (ABC)"
)
elif "AtCoder Regular Contest" in name:
match = re.search(r"AtCoder Regular Contest (\d+)", name)
if match:
display_name = f"Regular Contest {match.group(1)} (ARC)"
elif "AtCoder Grand Contest" in name:
match = re.search(r"AtCoder Grand Contest (\d+)", name)
if match:
display_name = f"Grand Contest {match.group(1)} (AGC)"
elif "AtCoder Heuristic Contest" in name:
match = re.search(r"AtCoder Heuristic Contest (\d+)", name)
if match:
display_name = (
f"Heuristic Contest {match.group(1)} (AHC)"
)
contests.append(
ContestSummary(
id=contest_id, name=name, display_name=display_name
)
)
return contests
except requests.exceptions.RequestException as e:
if response.status_code == 429:
continue
print(
f"Failed to scrape page {page} (attempt {attempt + 1}): {e}",
file=sys.stderr,
parts = re.sub(
r"\b(jsc|JSC)\b",
"Japanese Student Championship",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(wtf|WTF)\b",
"World Tour Finals",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(ahc)(\d+)\b",
r"Heuristic Contest \2 (AHC)",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(arc)(\d+)\b",
r"Regular Contest \2 (ARC)",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(abc)(\d+)\b",
r"Beginner Contest \2 (ABC)",
parts,
flags=re.IGNORECASE,
)
parts = re.sub(
r"\b(agc)(\d+)\b",
r"Grand Contest \2 (AGC)",
parts,
flags=re.IGNORECASE,
)
if attempt == max_retries - 1:
return []
except Exception as e:
print(f"Unexpected error on page {page}: {e}", file=sys.stderr)
return []
return []
return parts.title()
english_chars = sum(1 for c in name if c.isascii() and c.isalpha())
total_chars = len(re.sub(r"\s+", "", name))
if total_chars > 0 and english_chars / total_chars < 0.3:
display_name = generate_display_name_from_id(contest_id)
else:
display_name = name
if "AtCoder Beginner Contest" in name:
match = re.search(r"AtCoder Beginner Contest (\d+)", name)
if match:
display_name = f"Beginner Contest {match.group(1)} (ABC)"
elif "AtCoder Regular Contest" in name:
match = re.search(r"AtCoder Regular Contest (\d+)", name)
if match:
display_name = f"Regular Contest {match.group(1)} (ARC)"
elif "AtCoder Grand Contest" in name:
match = re.search(r"AtCoder Grand Contest (\d+)", name)
if match:
display_name = f"Grand Contest {match.group(1)} (AGC)"
elif "AtCoder Heuristic Contest" in name:
match = re.search(r"AtCoder Heuristic Contest (\d+)", name)
if match:
display_name = f"Heuristic Contest {match.group(1)} (AHC)"
contests.append(
ContestSummary(id=contest_id, name=name, display_name=display_name)
)
return contests
max_pages = get_max_pages()
page_results = {}
with concurrent.futures.ThreadPoolExecutor(max_workers=5) as executor:
future_to_page = {
executor.submit(scrape_page_with_retry, page): page
for page in range(1, max_pages + 1)
executor.submit(scrape_page, page): page for page in range(1, max_pages + 1)
}
for future in concurrent.futures.as_completed(future_to_page):