Edward Betts
95ca5f755d
mediawiki_oauth: set User-Agent on all OAuth1Session instances so Wikimedia doesn't reject token and API requests with 403; add timeout parameter to api_post_request (default 4s). mediawiki_api: add APIError exception; wrap .json() in call() to raise APIError with status code and response body on decode failure; raise timeout to 30s for edit POSTs. api: wrap call_get_diff .json() with the same JSONDecodeError guard, raising MediawikiError with HTTP status and body on failure. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
110 lines
2.9 KiB
Python
110 lines
2.9 KiB
Python
"""Interface with the mediawiki API."""
|
|
|
|
import typing
|
|
from pprint import pprint
|
|
from typing import Any, cast
|
|
|
|
import requests
|
|
|
|
from . import mediawiki_oauth
|
|
|
|
|
|
class APIError(Exception):
|
|
"""Unexpected response from the MediaWiki API."""
|
|
|
|
wiki_hostname = "en.wikipedia.org"
|
|
wiki_api_php = f"https://{wiki_hostname}/w/api.php"
|
|
user_agent = "add-links/0.1"
|
|
|
|
|
|
def parse_page(enwiki: str) -> dict[str, Any]:
|
|
"""Call mediawiki parse API for given article."""
|
|
params: dict[str, str | int] = {
|
|
"action": "parse",
|
|
"format": "json",
|
|
"formatversion": 2,
|
|
"disableeditsection": 1,
|
|
"page": enwiki,
|
|
"prop": "text|links|headhtml",
|
|
"disabletoc": 1,
|
|
}
|
|
|
|
parse: dict[str, Any] = call(params)["parse"]
|
|
return parse
|
|
|
|
|
|
def call(params: dict[str, str | int], timeout: int = 4) -> dict[str, typing.Any]:
|
|
"""Make GET request to mediawiki API."""
|
|
r = mediawiki_oauth.api_post_request(params, timeout=timeout)
|
|
try:
|
|
return cast(dict[str, Any], r.json())
|
|
except requests.exceptions.JSONDecodeError:
|
|
raise APIError(f"HTTP {r.status_code}: {r.text[:200]!r}")
|
|
|
|
|
|
def article_exists(title: str) -> bool:
|
|
"""Get article text."""
|
|
params: dict[str, str | int] = {
|
|
"action": "query",
|
|
"format": "json",
|
|
"formatversion": 2,
|
|
"titles": title,
|
|
}
|
|
return not call(params)["query"]["pages"][0].get("missing")
|
|
|
|
|
|
def get_content(title: str) -> tuple[str, int]:
|
|
"""Get article text."""
|
|
params: dict[str, str | int] = {
|
|
"action": "query",
|
|
"format": "json",
|
|
"formatversion": 2,
|
|
"prop": "revisions|info",
|
|
"rvprop": "content|timestamp|ids",
|
|
"titles": title,
|
|
}
|
|
data = call(params)
|
|
rev = data["query"]["pages"][0]["revisions"][0]
|
|
content: str = rev["content"]
|
|
revid: int = int(rev["revid"])
|
|
return content, revid
|
|
|
|
|
|
def compare(title: str, new_text: str) -> str:
|
|
"""Generate a diff for the new article text."""
|
|
params: dict[str, str | int] = {
|
|
"format": "json",
|
|
"formatversion": 2,
|
|
"action": "compare",
|
|
"fromtitle": title,
|
|
"toslots": "main",
|
|
"totext-main": new_text,
|
|
"prop": "diff",
|
|
}
|
|
diff: str = call(params)["compare"]["body"]
|
|
return diff
|
|
|
|
|
|
def edit_page(
|
|
pageid: int, section: str | int, text: str, summary: str, baserevid: str, token: str
|
|
) -> str:
|
|
"""Edit a page on Wikipedia."""
|
|
params: dict[str, str | int] = {
|
|
"format": "json",
|
|
"formatversion": 2,
|
|
"action": "edit",
|
|
"pageid": pageid,
|
|
"text": text,
|
|
"baserevid": baserevid,
|
|
"token": token,
|
|
"nocreate": 1,
|
|
"summary": summary,
|
|
"section": section,
|
|
}
|
|
ret = call(params, timeout=30)
|
|
if "edit" not in ret:
|
|
print("params")
|
|
pprint(params)
|
|
print()
|
|
pprint(ret)
|
|
return typing.cast(str, ret["edit"])
|