Skip invalid and protected candidates

web_view.py

@@ -1,7 +1,6 @@
 #!/usr/bin/python3
 
 import html
-import itertools
 import json
 import re
 import sys
@@ -37,6 +36,14 @@ class Hit(typing.TypedDict):
     timestamp: str
 
 
+class PageInfo(typing.TypedDict, total=False):
+    """Wikipedia page metadata."""
+
+    pageid: int
+    title: str
+    protection: list[dict[str, str]]
+
+
 def render_error(message: str) -> str:
     """Render shared error page."""
     return flask.render_template("error.html", message=message)
@@ -115,13 +122,60 @@ def search_count_with_link(q: str, redirect_to: str | None = None) -> int:
     return count
 
 
+def chunks(items: list[int], size: int) -> typing.Iterator[list[int]]:
+    """Yield fixed-size chunks from a list."""
+    for offset in range(0, len(items), size):
+        yield items[offset : offset + size]
+
+
+def get_page_info_by_id(pageids: list[int]) -> dict[int, PageInfo]:
+    """Fetch metadata for pages by page ID."""
+    if not pageids:
+        return {}
+
+    pages: dict[int, PageInfo] = {}
+    for pageid_chunk in chunks(pageids, 50):
+        ret = api.api_get(
+            {
+                "prop": "info",
+                "inprop": "protection",
+                "pageids": "|".join(str(pageid) for pageid in pageid_chunk),
+            }
+        )
+        for page in ret["query"]["pages"]:
+            pageid = typing.cast(int, page["pageid"])
+            pages[pageid] = typing.cast(PageInfo, page)
+    return pages
+
+
+def is_edit_protected(page: PageInfo | None) -> bool:
+    """Check if a page has edit protection."""
+    if not page:
+        return False
+    return any(
+        protection.get("type") == "edit" or protection.get("action") == "edit"
+        for protection in page.get("protection", [])
+    )
+
+
+def filter_edit_protected_hits(hits: list[Hit]) -> list[Hit]:
+    """Remove candidates that cannot be edited because of page protection."""
+    page_info = get_page_info_by_id([hit["pageid"] for hit in hits])
+    return [
+        hit for hit in hits if not is_edit_protected(page_info.get(hit["pageid"]))
+    ]
+
+
 def search_no_link(q: str, redirect_to: str | None = None) -> tuple[int, list[Hit]]:
     """Search for mentions of article title with no link included."""
     exclude = f' -linksto:"{q}"'
     if redirect_to:
         exclude += f' -linksto:"{redirect_to}"'
     query = run_search(article_title_to_search_query(q) + exclude, "max")
-    return (query["searchinfo"]["totalhits"], query["search"])
+    return (
+        query["searchinfo"]["totalhits"],
+        filter_edit_protected_hits(query["search"]),
+    )
 
 
 @app.before_request
@@ -317,6 +371,9 @@ def match_type(q: str, snippet: str) -> str | None:
 
 def _record_skip(from_title: str, hit_title: str) -> None:
     """Record that a candidate was skipped or saved for this article."""
+    hit_title = hit_title.strip()
+    if not hit_title:
+        return
     skipped: dict[str, list[str]] = flask.session.get("skipped", {})
     article_skipped = skipped.get(from_title, [])
     if hit_title not in article_skipped:
@@ -371,10 +428,8 @@ def article_page(url_title: str) -> str | Response:
             render_mediawiki_error(e), 502
         )
 
-    # Filter out candidates already processed this session
-    session_skipped: set[str] = set(
-        flask.session.get("skipped", {}).get(from_title, [])
-    )
+    article_skipped = flask.session.get("skipped", {}).get(from_title, [])
+    session_skipped: set[str] = set(article_skipped)
 
     # If a specific candidate was requested, move it to the front
     title_param = flask.request.args.get("title")
@@ -382,12 +437,6 @@ def article_page(url_title: str) -> str | Response:
         hits = [h for h in hits if h["title"] == title_param] + \
                [h for h in hits if h["title"] != title_param]
 
-    # Record and apply explicit skip-past
-    after = flask.request.args.get("after")
-    if after:
-        _record_skip(from_title, after)
-        session_skipped.add(after)
-
     hits = [h for h in hits if h["title"] not in session_skipped
             and h["title"] != from_title and h["title"] != case_flip_first(from_title)]
 
@@ -404,6 +453,7 @@ def article_page(url_title: str) -> str | Response:
         with_link=with_link,
         hits=hits,
         url_title=url_title,
+        skipped_titles=article_skipped,
         saves_this_session=saves_this_session,
     )