added pytest test suite (174 tests)

174 tests covering URL normalization, FTS5 query sanitization, SSRF/CSRF
guards, sharing-mode logic, DB schema and upsert paths, handler
end-to-end flows, and gateway body-size / mesh-whitelist guards. Each
recent bug-fix commit (6ffd38d, 1bc695f, 8dffd8c) has an explicit
regression test in test_regressions.py. One xfail documents a minor
latent bug in clean_url where port 80 is not stripped from upgraded
https URLs.

tests/test_handlers_pages.py

@@ -0,0 +1,174 @@
+"""Tests for `handle_bulk_action`, edit flow, and the bulk-delete confirm step.
+
+The bulk-delete confirmation flow is a data-loss guard added in commit
+8dffd8c — a stray POST without `confirmed=1` must render the confirmation
+page instead of actually deleting.
+"""
+from db import get_db, return_db
+from handlers import (
+    handle_bulk_action,
+    handle_edit_form,
+    handle_edit_submit,
+    handle_pages,
+)
+
+
+def _all_urls(seeded_db):
+    db = get_db()
+    try:
+        return {r["url"] for r in db.execute("SELECT url FROM pages").fetchall()}
+    finally:
+        return_db(db)
+
+
+def _page_id(seeded_db, url):
+    db = get_db()
+    try:
+        return db.execute("SELECT id FROM pages WHERE url = ?", (url,)).fetchone()["id"]
+    finally:
+        return_db(db)
+
+
+def test_bulk_delete_without_confirmed_renders_confirm_page(seeded_db, csrf_session):
+    """Regression for 8dffd8c: bulk delete must NOT delete until confirmed=1 is set."""
+    pid = _page_id(seeded_db, "https://example.com/rust-intro")
+    urls_before = _all_urls(seeded_db)
+
+    resp = handle_bulk_action({
+        "ids": [str(pid)],
+        "action": ["delete"],
+    })
+    assert resp["status"] == 200
+    assert "confirm delete" in resp["body"].lower()
+    assert "Rust Intro" in resp["body"]
+    # Must still show a hidden confirmed=1 field in the follow-up form.
+    assert 'name="confirmed" value="1"' in resp["body"]
+
+    # Crucially: nothing should have been deleted.
+    assert _all_urls(seeded_db) == urls_before
+
+
+def test_bulk_delete_with_confirmed_actually_deletes(seeded_db, csrf_session):
+    pid = _page_id(seeded_db, "https://example.com/rust-intro")
+
+    resp = handle_bulk_action({
+        "ids": [str(pid)],
+        "action": ["delete"],
+        "confirmed": ["1"],
+    })
+    # Confirmed delete redirects back to /pages.
+    assert resp["status"] in (302, 303)
+
+    urls = _all_urls(seeded_db)
+    assert "https://example.com/rust-intro" not in urls
+    # Other pages untouched.
+    assert "https://example.com/python-tips" in urls
+
+
+def test_bulk_delete_with_no_ids_redirects(seeded_db, csrf_session):
+    resp = handle_bulk_action({
+        "ids": [],
+        "action": ["delete"],
+        "confirmed": ["1"],
+    })
+    assert resp["status"] in (302, 303)
+    assert _all_urls(seeded_db) == {
+        "https://example.com/rust-intro",
+        "https://example.com/python-tips",
+        "https://example.com/ocaml-why",
+        "https://news.example.org/mesh",
+    }
+
+
+def test_bulk_delete_rejects_non_integer_ids(seeded_db, csrf_session):
+    resp = handle_bulk_action({
+        "ids": ["not-a-number"],
+        "action": ["delete"],
+        "confirmed": ["1"],
+    })
+    assert resp["status"] == 400
+
+
+def test_bulk_retag_add_mode_merges_tags(seeded_db, csrf_session):
+    pid = _page_id(seeded_db, "https://example.com/python-tips")
+
+    handle_bulk_action({
+        "ids": [str(pid)],
+        "action": ["retag"],
+        "bulk_tags": ["scripting, tutorials"],
+        "tag_mode": ["add"],
+    })
+    db = get_db()
+    try:
+        rows = db.execute(
+            "SELECT t.name FROM tags t JOIN page_tags pt ON pt.tag_id = t.id "
+            "WHERE pt.page_id = ? ORDER BY t.name",
+            (pid,),
+        ).fetchall()
+    finally:
+        return_db(db)
+    tags = [r["name"] for r in rows]
+    assert "python" in tags  # existing kept
+    assert "scripting" in tags  # new added
+    assert "tutorials" in tags
+
+
+def test_bulk_retag_replace_mode_overwrites_tags(seeded_db, csrf_session):
+    pid = _page_id(seeded_db, "https://example.com/python-tips")
+
+    handle_bulk_action({
+        "ids": [str(pid)],
+        "action": ["retag"],
+        "bulk_tags": ["one, two"],
+        "tag_mode": ["replace"],
+    })
+    db = get_db()
+    try:
+        rows = db.execute(
+            "SELECT t.name FROM tags t JOIN page_tags pt ON pt.tag_id = t.id "
+            "WHERE pt.page_id = ?",
+            (pid,),
+        ).fetchall()
+    finally:
+        return_db(db)
+    tags = {r["name"] for r in rows}
+    assert tags == {"one", "two"}
+    assert "python" not in tags
+
+
+def test_edit_form_renders_current_values(seeded_db, csrf_session):
+    pid = _page_id(seeded_db, "https://example.com/rust-intro")
+    resp = handle_edit_form(pid)
+    assert resp["status"] == 200
+    assert "Rust Intro" in resp["body"]
+    # Existing tags should appear in the tag field.
+    assert "rust" in resp["body"]
+
+
+def test_edit_form_404_for_unknown_page(temp_db, csrf_session):
+    resp = handle_edit_form(99999)
+    assert resp["status"] == 404
+
+
+def test_edit_submit_updates_title_and_note(seeded_db, csrf_session):
+    pid = _page_id(seeded_db, "https://example.com/rust-intro")
+    handle_edit_submit(pid, {
+        "title": ["New Rust Title"],
+        "note": ["new annotation"],
+        "tags": ["rust, updated"],
+    })
+    db = get_db()
+    try:
+        row = db.execute("SELECT title, note FROM pages WHERE id = ?", (pid,)).fetchone()
+    finally:
+        return_db(db)
+    assert row["title"] == "New Rust Title"
+    assert row["note"] == "new annotation"
+
+
+def test_handle_pages_lists_indexed_pages(seeded_db, csrf_session):
+    resp = handle_pages({})
+    assert resp["status"] == 200
+    # Every seeded page title appears on the list page.
+    for title in ("Rust Intro", "Python Tips", "Why OCaml", "Mesh Networking"):
+        assert title in resp["body"]