From 6ee3c62f9737ecf2d502e412d9235fce70b995c2 Mon Sep 17 00:00:00 2001
From: Trenton Holmes <797416+stumpylog@users.noreply.github.com>
Date: Thu, 7 May 2026 06:20:49 -0700
Subject: [PATCH] Handle user queries with Tantivy operators like - or + in
 them

---
 src/documents/search/_query.py           | 14 ++++-
 src/documents/tests/search/test_query.py | 74 ++++++++++++++++++++++++
 2 files changed, 87 insertions(+), 1 deletion(-)

diff --git a/src/documents/search/_query.py b/src/documents/search/_query.py
index 65df260c3..04d79d1ef 100644
--- a/src/documents/search/_query.py
+++ b/src/documents/search/_query.py
@@ -76,6 +76,11 @@ _YEAR_RANGE_RE = regex.compile(
     regex.IGNORECASE,
 )
 _SIMPLE_QUERY_TOKEN_RE = regex.compile(r"\S+")
+# Tantivy syntax error: " - " and " + " with spaces on both sides are invalid because
+# the NOT/MUST operators require no space between the operator and the term.
+# In natural-language queries (e.g., "H52.1 - Kurzsichtigkeit"), the dash is a separator.
+_SPACED_OPERATOR_RE = regex.compile(r"\s+[-+]\s+")
+_TRAILING_OPERATOR_RE = regex.compile(r"\s+[-+]+\s*$")
 
 
 def _fmt(dt: datetime) -> str:
@@ -430,7 +435,14 @@ def normalize_query(query: str) -> str:
             query,
             timeout=_REGEX_TIMEOUT,
         )
-        return regex.sub(r" {2,}", " ", query, timeout=_REGEX_TIMEOUT).strip()
+        query = regex.sub(r" {2,}", " ", query, timeout=_REGEX_TIMEOUT).strip()
+        # Strip trailing dangling operators before Tantivy sees them.
+        query = _TRAILING_OPERATOR_RE.sub("", query, timeout=_REGEX_TIMEOUT).strip()
+        # Replace " - " / " + " with a space: Tantivy requires no space between
+        # the operator and its operand (-term / +term), so spaces on both sides
+        # means this is a natural-language separator, not a query operator.
+        query = _SPACED_OPERATOR_RE.sub(" ", query, timeout=_REGEX_TIMEOUT).strip()
+        return query
     except TimeoutError:  # pragma: no cover
         raise ValueError("Query too complex to process (normalization timed out)")
 
diff --git a/src/documents/tests/search/test_query.py b/src/documents/tests/search/test_query.py
index 11297eb92..0db224c5d 100644
--- a/src/documents/tests/search/test_query.py
+++ b/src/documents/tests/search/test_query.py
@@ -443,6 +443,25 @@ class TestParseUserQuery:
             q = parse_user_query(query_index, "created:today", UTC)
         assert isinstance(q, tantivy.Query)
 
+    @pytest.mark.parametrize(
+        "raw_query",
+        [
+            pytest.param("h52.1 - kurzsichtigkeit", id="icd_code_dash_description"),
+            pytest.param("H52.1 - asd", id="icd_code_uppercase"),
+            pytest.param("h52.1 -", id="trailing_minus"),
+            pytest.param(". -", id="dot_trailing_minus"),
+            pytest.param("h52. -", id="partial_code_trailing_minus"),
+            pytest.param(".12 -", id="dot_number_trailing_minus"),
+            pytest.param("h52.1 - ku", id="partial_word_after_dash"),
+        ],
+    )
+    def test_spaced_dash_queries_do_not_raise(
+        self,
+        query_index: tantivy.Index,
+        raw_query: str,
+    ) -> None:
+        assert isinstance(parse_user_query(query_index, raw_query, UTC), tantivy.Query)
+
 
 class TestYearRangeRewriting:
     """Whoosh-style year-only date ranges must be rewritten to ISO 8601."""
@@ -548,6 +567,61 @@ class TestNormalizeQuery:
     def test_normalize_no_commas_unchanged(self) -> None:
         assert normalize_query("bank statement") == "bank statement"
 
+    @pytest.mark.parametrize(
+        ("raw", "expected"),
+        [
+            pytest.param(
+                "h52.1 - kurzsichtigkeit",
+                "h52.1 kurzsichtigkeit",
+                id="icd_code_dash_description",
+            ),
+            pytest.param(
+                "H52.1 - asd",
+                "H52.1 asd",
+                id="icd_code_uppercase_dash",
+            ),
+            pytest.param(
+                "h52.1 -",
+                "h52.1",
+                id="trailing_minus",
+            ),
+            pytest.param(
+                ". -",
+                ".",
+                id="dot_trailing_minus",
+            ),
+            pytest.param(
+                "h52. -",
+                "h52.",
+                id="partial_code_trailing_minus",
+            ),
+            pytest.param(
+                "foo - bar - baz",
+                "foo bar baz",
+                id="multiple_dashes",
+            ),
+            pytest.param(
+                "foo + bar",
+                "foo bar",
+                id="spaced_plus_operator",
+            ),
+        ],
+    )
+    def test_normalize_strips_dangling_operators(self, raw: str, expected: str) -> None:
+        assert normalize_query(raw) == expected
+
+    @pytest.mark.parametrize(
+        "query",
+        [
+            pytest.param("term -other", id="adjacent_not_operator"),
+            pytest.param("-term", id="leading_not_operator"),
+            pytest.param("+term", id="leading_must_operator"),
+            pytest.param("foo -bar +baz", id="mixed_adjacent_operators"),
+        ],
+    )
+    def test_normalize_preserves_valid_operators(self, query: str) -> None:
+        assert normalize_query(query) == query
+
 
 class TestPermissionFilter:
     """