Fix User-Agent header, timeouts, and JSON error handling

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>

add_links/api.py

@@ -272,6 +272,10 @@ def call_get_diff(title: str, section_num: int, section_text: str) -> str:
     }
 
     s = get_session()
-    ret = s.post(get_query_url(), data=data).json()
+    r = s.post(get_query_url(), data=data)
+    try:
+        ret = r.json()
+    except JSONDecodeError:
+        raise MediawikiError(f"HTTP {r.status_code}: {r.text[:200]!r}")
     check_for_error(ret)
     return typing.cast(str, ret["query"]["pages"][0]["revisions"][0]["diff"]["body"])

add_links/mediawiki_api.py

@@ -4,8 +4,14 @@ 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"
@@ -27,10 +33,13 @@ def parse_page(enwiki: str) -> dict[str, Any]:
     return parse
 
 
-def call(params: dict[str, str | int]) -> dict[str, typing.Any]:
+def call(params: dict[str, str | int], timeout: int = 4) -> dict[str, typing.Any]:
     """Make GET request to mediawiki API."""
-    data = mediawiki_oauth.api_post_request(params)
-    return cast(dict[str, Any], data.json())
+    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:
@@ -92,7 +101,7 @@ def edit_page(
         "summary": summary,
         "section": section,
     }
-    ret = call(params)
+    ret = call(params, timeout=30)
     if "edit" not in ret:
         print("params")
         pprint(params)

add_links/mediawiki_oauth.py

@@ -8,6 +8,8 @@ import requests
 from flask import current_app, session
 from requests_oauthlib import OAuth1Session
 
+from .api import ua
+
 wiki_hostname = "en.wikipedia.org"
 api_url = f"https://{wiki_hostname}/w/api.php"
 
@@ -27,7 +29,7 @@ def get_edit_proxy() -> dict[str, str]:
         return {}
 
 
-def api_post_request(params: dict[str, str | int]) -> requests.Response:
+def api_post_request(params: dict[str, str | int], timeout: int = 4) -> requests.Response:
     """HTTP Post using Oauth."""
     app = current_app
     # url = "https://www.wikidata.org/w/api.php"
@@ -39,8 +41,9 @@ def api_post_request(params: dict[str, str | int]) -> requests.Response:
         resource_owner_key=session["owner_key"],
         resource_owner_secret=session["owner_secret"],
     )
+    oauth.headers.update({"User-Agent": ua})
     proxies = get_edit_proxy()
-    return oauth.post(api_url, data=params, timeout=4, proxies=proxies)
+    return oauth.post(api_url, data=params, timeout=timeout, proxies=proxies)
 
 
 def raw_request(params: typing.Mapping[str, str | int]) -> requests.Response:
@@ -57,6 +60,7 @@ def raw_request(params: typing.Mapping[str, str | int]) -> requests.Response:
         resource_owner_key=session["owner_key"],
         resource_owner_secret=session["owner_secret"],
     )
+    oauth.headers.update({"User-Agent": ua})
     proxies = get_edit_proxy()
     return oauth.get(
         api_url + "?" + urllib.parse.urlencode(params), timeout=4, proxies=proxies