scan: names without HTML entities, and the template says locations match as written

src/aihawk/config.py

@@ -33,7 +33,8 @@ goals: Describe the role, seniority, location and pay you are looking for.
 threshold: 4.0
 
 # Keep a job only if its title holds one of `include` (required) and none of `exclude`, and its
-# location holds one of `locations`, in any language the posting uses. Older postings are left out.
+# location holds one of `locations`. Words are matched as written, so list every spelling you
+# want: italy, italia, milan, milano. Older postings are left out.
 filters:
   include: [Replace with words of the job titles you want]
   exclude: []

src/aihawk/database.py

@@ -135,4 +135,8 @@ def search(location: str, filters: Filters, max_age_days: int) -> tuple[dict[str
         raise DatabaseError(f"the job database cannot be read: {error}") from error
     finally:
         connection.close()
-    return manifest, [Job(key, company, title, place, url, url, text) for key, company, title, place, url, text in rows]
+    # Some sources publish their names with HTML entities still in them.
+    return manifest, [
+        Job(key, html.unescape(company), html.unescape(title), html.unescape(place), url, url, text)
+        for key, company, title, place, url, text in rows
+    ]

tests/test_jobs.py

@@ -19,7 +19,7 @@ def database(tmp_path: Path) -> str:
     rows = [
         ("k1", "https://jobs.lever.co/a/1", "Senior Python Engineer", "Acme", "Rome, Italy", "Build", "2099-01-01"),
         ("k2", "https://jobs.lever.co/a/2", "Python Intern", "Acme", "Berlin", "Learn", None),
-        ("k3", "https://example.com/3", "Accountant", "Beta", "Rome, Italy", "Count", "2099-01-01"),
+        ("k3", "https://example.com/3", "Accountant & Clerk", "B&A", "Rome, Italy", "Count", "2099-01-01"),
         ("k4", "https://example.com/4", "Python Developer", "Old", "Rome, Italy", "Past", "2001-01-01"),
     ]
     connection = duckdb.connect()
@@ -45,6 +45,11 @@ def test_search_keeps_what_the_filters_say_and_leaves_out_what_is_old(tmp_path:
     assert [job.title for job in jobs] == titles
 
 
+def test_names_come_without_html_entities(tmp_path: Path) -> None:
+    _, jobs = search(database(tmp_path), Filters(include=("accountant",)), max_age_days=60)
+    assert [(job.title, job.company) for job in jobs] == [("Accountant & Clerk", "B&A")]
+
+
 def test_search_needs_titles_and_says_so(tmp_path: Path) -> None:
     with pytest.raises(DatabaseError, match="filters.include"):
         search(database(tmp_path), Filters(), max_age_days=60)