feat: scraper cleanup

This commit is contained in:
Barrett Ruth 2025-09-18 21:49:25 -04:00
parent 002b75b0ab
commit ca6f8417c0
5 changed files with 296 additions and 178 deletions

View file

@ -2,9 +2,10 @@
import json
import sys
from typing import Any
import requests
from bs4 import BeautifulSoup
from bs4 import BeautifulSoup, Tag
def parse_problem_url(contest_id: str, problem_letter: str) -> str:
@ -23,7 +24,6 @@ def extract_problem_from_row(row, contest_id: str) -> dict[str, str] | None:
task_name = task_link.get_text(strip=True)
task_href = task_link.get("href", "")
if not task_href:
return None
@ -50,13 +50,11 @@ def scrape_contest_problems(contest_id: str) -> list[dict[str, str]]:
soup = BeautifulSoup(response.text, "html.parser")
task_table = soup.find("table", class_="table")
if not task_table:
if not task_table or not isinstance(task_table, Tag):
return []
rows = task_table.find_all("tr")[1:]
problems = []
rows = task_table.find_all("tr")[1:] # skip header
problems: list[dict[str, str]] = []
for row in rows:
problem = extract_problem_from_row(row, contest_id)
if problem:
@ -92,7 +90,6 @@ def extract_test_case_from_headers(sample_headers, i: int) -> tuple[str, str] |
input_text = input_pre.get_text().strip().replace("\r", "")
output_text = output_pre.get_text().strip().replace("\r", "")
if not input_text or not output_text:
return None
@ -109,19 +106,17 @@ def scrape(url: str) -> list[tuple[str, str]]:
response.raise_for_status()
soup = BeautifulSoup(response.text, "html.parser")
sample_headers = soup.find_all(
"h3", string=lambda x: x and "sample" in x.lower() if x else False
)
tests = []
tests: list[tuple[str, str]] = []
i = 0
while i < len(sample_headers):
test_case = extract_test_case_from_headers(sample_headers, i)
if test_case:
tests.append(test_case)
i += 2
i += 2 # move from "Sample Input n" to after "Sample Output n"
else:
i += 1
@ -134,88 +129,111 @@ def scrape(url: str) -> list[tuple[str, str]]:
def main() -> None:
if len(sys.argv) < 2:
result: dict[str, str | bool] = {
"success": False,
"error": "Usage: atcoder.py metadata <contest_id> OR atcoder.py tests <contest_id> <problem_letter>",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: atcoder.py metadata <contest_id> OR atcoder.py tests <contest_id> <problem_letter>",
}
)
)
sys.exit(1)
mode: str = sys.argv[1]
if mode == "metadata":
if len(sys.argv) != 3:
result = {
"success": False,
"error": "Usage: atcoder.py metadata <contest_id>",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: atcoder.py metadata <contest_id>",
}
)
)
sys.exit(1)
contest_id: str = sys.argv[2]
problems: list[dict[str, str]] = scrape_contest_problems(contest_id)
if not problems:
result = {
"success": False,
"error": f"No problems found for contest {contest_id}",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"No problems found for contest {contest_id}",
}
)
)
sys.exit(1)
result = {
"success": True,
"contest_id": contest_id,
"problems": problems,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": True,
"contest_id": contest_id,
"problems": problems,
}
)
)
elif mode == "tests":
if len(sys.argv) != 4:
result = {
"success": False,
"error": "Usage: atcoder.py tests <contest_id> <problem_letter>",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: atcoder.py tests <contest_id> <problem_letter>",
}
)
)
sys.exit(1)
contest_id: str = sys.argv[2]
test_contest_id: str = sys.argv[2]
problem_letter: str = sys.argv[3]
problem_id: str = contest_id + problem_letter.lower()
problem_id: str = f"{test_contest_id}_{problem_letter.lower()}"
url: str = parse_problem_url(contest_id, problem_letter)
url: str = parse_problem_url(test_contest_id, problem_letter)
print(f"Scraping: {url}", file=sys.stderr)
tests: list[tuple[str, str]] = scrape(url)
if not tests:
result = {
"success": False,
"error": f"No tests found for {contest_id} {problem_letter}",
"problem_id": problem_id,
"url": url,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"No tests found for {test_contest_id} {problem_letter}",
"problem_id": problem_id,
"url": url,
}
)
)
sys.exit(1)
test_list: list[dict[str, str]] = []
for input_data, output_data in tests:
test_list.append({"input": input_data, "expected": output_data})
test_list: list[dict[str, str]] = [
{"input": i, "expected": o} for i, o in tests
]
result = {
"success": True,
"problem_id": problem_id,
"url": url,
"tests": test_list,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": True,
"problem_id": problem_id,
"url": url,
"tests": test_list,
}
)
)
else:
result = {
"success": False,
"error": f"Unknown mode: {mode}. Use 'metadata' or 'tests'",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"Unknown mode: {mode}. Use 'metadata' or 'tests'",
}
)
)
sys.exit(1)

View file

@ -2,9 +2,10 @@
import json
import sys
from typing import Any
import cloudscraper
from bs4 import BeautifulSoup
from bs4 import BeautifulSoup, Tag
def scrape(url: str) -> list[tuple[str, str]]:
@ -17,12 +18,12 @@ def scrape(url: str) -> list[tuple[str, str]]:
input_sections = soup.find_all("div", class_="input")
output_sections = soup.find_all("div", class_="output")
individual_inputs = {}
individual_outputs = {}
individual_inputs: dict[str, list[str]] = {}
individual_outputs: dict[str, list[str]] = {}
for inp_section in input_sections:
inp_pre = inp_section.find("pre")
if not inp_pre:
if not inp_pre or not isinstance(inp_pre, Tag):
continue
test_line_divs = inp_pre.find_all(
@ -51,7 +52,7 @@ def scrape(url: str) -> list[tuple[str, str]]:
for out_section in output_sections:
out_pre = out_section.find("pre")
if not out_pre:
if not out_pre or not isinstance(out_pre, Tag):
continue
test_line_divs = out_pre.find_all(
@ -95,12 +96,12 @@ def scrape(url: str) -> list[tuple[str, str]]:
for inp_section in input_sections:
inp_pre = inp_section.find("pre")
if not inp_pre:
if not inp_pre or not isinstance(inp_pre, Tag):
continue
divs = inp_pre.find_all("div")
if divs:
lines = [div.get_text().strip() for div in divs]
lines = [div.get_text().strip() for div in divs if isinstance(div, Tag)]
text = "\n".join(lines)
else:
text = inp_pre.get_text().replace("\r", "").strip()
@ -108,12 +109,12 @@ def scrape(url: str) -> list[tuple[str, str]]:
for out_section in output_sections:
out_pre = out_section.find("pre")
if not out_pre:
if not out_pre or not isinstance(out_pre, Tag):
continue
divs = out_pre.find_all("div")
if divs:
lines = [div.get_text().strip() for div in divs]
lines = [div.get_text().strip() for div in divs if isinstance(div, Tag)]
text = "\n".join(lines)
else:
text = out_pre.get_text().replace("\r", "").strip()
@ -152,7 +153,9 @@ def scrape_contest_problems(contest_id: str) -> list[dict[str, str]]:
)
for link in problem_links:
href: str = link.get("href", "")
if not isinstance(link, Tag):
continue
href: str = str(link.get("href", ""))
if f"/contest/{contest_id}/problem/" in href:
problem_letter: str = href.split("/")[-1].lower()
problem_name: str = link.get_text(strip=True)
@ -183,86 +186,110 @@ def scrape_sample_tests(url: str) -> list[tuple[str, str]]:
def main() -> None:
if len(sys.argv) < 2:
result: dict[str, str | bool] = {
"success": False,
"error": "Usage: codeforces.py metadata <contest_id> OR codeforces.py tests <contest_id> <problem_letter>",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: codeforces.py metadata <contest_id> OR codeforces.py tests <contest_id> <problem_letter>",
}
)
)
sys.exit(1)
mode: str = sys.argv[1]
if mode == "metadata":
if len(sys.argv) != 3:
result: dict[str, str | bool] = {
"success": False,
"error": "Usage: codeforces.py metadata <contest_id>",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: codeforces.py metadata <contest_id>",
}
)
)
sys.exit(1)
contest_id: str = sys.argv[2]
problems: list[dict[str, str]] = scrape_contest_problems(contest_id)
if not problems:
result: dict[str, str | bool] = {
"success": False,
"error": f"No problems found for contest {contest_id}",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"No problems found for contest {contest_id}",
}
)
)
sys.exit(1)
result: dict[str, str | bool | list] = {
"success": True,
"contest_id": contest_id,
"problems": problems,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": True,
"contest_id": contest_id,
"problems": problems,
}
)
)
elif mode == "tests":
if len(sys.argv) != 4:
result: dict[str, str | bool] = {
"success": False,
"error": "Usage: codeforces.py tests <contest_id> <problem_letter>",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: codeforces.py tests <contest_id> <problem_letter>",
}
)
)
sys.exit(1)
contest_id: str = sys.argv[2]
tests_contest_id: str = sys.argv[2]
problem_letter: str = sys.argv[3]
problem_id: str = contest_id + problem_letter.lower()
problem_id: str = tests_contest_id + problem_letter.lower()
url: str = parse_problem_url(contest_id, problem_letter)
url: str = parse_problem_url(tests_contest_id, problem_letter)
tests: list[tuple[str, str]] = scrape_sample_tests(url)
if not tests:
result: dict[str, str | bool] = {
"success": False,
"error": f"No tests found for {contest_id} {problem_letter}",
"problem_id": problem_id,
"url": url,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"No tests found for {tests_contest_id} {problem_letter}",
"problem_id": problem_id,
"url": url,
}
)
)
sys.exit(1)
test_list: list[dict[str, str]] = []
for input_data, output_data in tests:
test_list.append({"input": input_data, "expected": output_data})
result: dict[str, str | bool | list] = {
"success": True,
"problem_id": problem_id,
"url": url,
"tests": test_list,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": True,
"problem_id": problem_id,
"url": url,
"tests": test_list,
}
)
)
else:
result: dict[str, str | bool] = {
"success": False,
"error": f"Unknown mode: {mode}. Use 'metadata' or 'tests'",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"Unknown mode: {mode}. Use 'metadata' or 'tests'",
}
)
)
sys.exit(1)

View file

@ -2,6 +2,7 @@
import json
import sys
from typing import Any
import requests
from bs4 import BeautifulSoup
@ -16,7 +17,9 @@ def parse_problem_url(problem_input: str) -> str | None:
def process_problem_element(
element, current_category: str, all_categories: dict
element,
current_category: str | None,
all_categories: dict[str, list[dict[str, str]]],
) -> str | None:
if element.name == "h1":
category_name = element.get_text().strip()
@ -52,7 +55,7 @@ def scrape_all_problems() -> dict[str, list[dict[str, str]]]:
response.raise_for_status()
soup = BeautifulSoup(response.text, "html.parser")
all_categories = {}
all_categories: dict[str, list[dict[str, str]]] = {}
problem_links = soup.find_all(
"a", href=lambda x: x and "/problemset/task/" in x
@ -127,59 +130,79 @@ def scrape(url: str) -> list[tuple[str, str]]:
def main() -> None:
if len(sys.argv) < 2:
result: dict[str, str | bool] = {
"success": False,
"error": "Usage: cses.py metadata OR cses.py tests <problem_id_or_url>",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: cses.py metadata OR cses.py tests <problem_id_or_url>",
}
)
)
sys.exit(1)
mode: str = sys.argv[1]
if mode == "metadata":
if len(sys.argv) != 2:
result = {
"success": False,
"error": "Usage: cses.py metadata",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: cses.py metadata",
}
)
)
sys.exit(1)
all_categories: dict[str, list[dict[str, str]]] = scrape_all_problems()
if not all_categories:
result = {
"success": False,
"error": "Failed to scrape CSES problem categories",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Failed to scrape CSES problem categories",
}
)
)
sys.exit(1)
result = {
"success": True,
"categories": all_categories,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": True,
"categories": all_categories,
}
)
)
elif mode == "tests":
if len(sys.argv) != 3:
result = {
"success": False,
"error": "Usage: cses.py tests <problem_id_or_url>",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": "Usage: cses.py tests <problem_id_or_url>",
}
)
)
sys.exit(1)
problem_input: str = sys.argv[2]
url: str | None = parse_problem_url(problem_input)
if not url:
result = {
"success": False,
"error": f"Invalid problem input: {problem_input}. Use either problem ID (e.g., 1068) or full URL",
"problem_id": problem_input if problem_input.isdigit() else None,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"Invalid problem input: {problem_input}. Use either problem ID (e.g., 1068) or full URL",
"problem_id": problem_input
if problem_input.isdigit()
else None,
}
)
)
sys.exit(1)
tests: list[tuple[str, str]] = scrape(url)
@ -189,33 +212,42 @@ def main() -> None:
)
if not tests:
result = {
"success": False,
"error": f"No tests found for {problem_input}",
"problem_id": problem_id,
"url": url,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"No tests found for {problem_input}",
"problem_id": problem_id,
"url": url,
}
)
)
sys.exit(1)
test_list: list[dict[str, str]] = []
for input_data, output_data in tests:
test_list.append({"input": input_data, "expected": output_data})
test_list: list[dict[str, str]] = [
{"input": i, "expected": o} for i, o in tests
]
result = {
"success": True,
"problem_id": problem_id,
"url": url,
"tests": test_list,
}
print(json.dumps(result))
print(
json.dumps(
{
"success": True,
"problem_id": problem_id,
"url": url,
"tests": test_list,
}
)
)
else:
result = {
"success": False,
"error": f"Unknown mode: {mode}. Use 'metadata' or 'tests'",
}
print(json.dumps(result))
print(
json.dumps(
{
"success": False,
"error": f"Unknown mode: {mode}. Use 'metadata' or 'tests'",
}
)
)
sys.exit(1)