test: Consolidate tests (#126)

Author: isaacvando

.github/workflows/ci.yml

@@ -114,7 +114,7 @@ jobs:
 
     services:
       paradedb:
-        image: paradedb/paradedb:0.23.0-pg${{ matrix.postgres-version }}
+        image: paradedb/paradedb:0.23.4-pg${{ matrix.postgres-version }}
         env:
           POSTGRES_USER: postgres
           POSTGRES_PASSWORD: postgres

paradedb/functions.py

@@ -38,15 +38,6 @@ def _validate_non_negative_int(name: str, value: int | None) -> None:
         raise ValueError(f"{name} must be zero or positive.")
 
 
-def _validate_sample_rate(sample_rate: float | None) -> None:
-    if sample_rate is None:
-        return
-    if isinstance(sample_rate, bool) or not isinstance(sample_rate, int | float):
-        raise TypeError("sample_rate must be a float between 0.0 and 1.0.")
-    if sample_rate < 0 or sample_rate > 1:
-        raise ValueError("sample_rate must be between 0.0 and 1.0.")
-
-
 class Score(Func):
     """BM25 score annotation."""
 
@@ -100,19 +91,9 @@ def as_sql(  # type: ignore[override]
 
 
 class Snippets(Func):
-    """Multiple-snippets annotation.
-
-    Wraps ``pdb.snippets(column, ...)`` which returns a text array of all
-    matching snippet fragments, with named SQL parameters.
-
-    See: https://docs.paradedb.com/documentation/full-text/highlight#multiple-snippets
-    """
-
     function = FN_SNIPPETS
     output_field = ArrayField(base_field=CharField())
 
-    _VALID_SORT_BY = ("score", "position")
-
     def __init__(
         self,
         field: str,
@@ -124,10 +105,6 @@ def __init__(
         offset: int | None = None,
         sort_by: Literal["score", "position"] | None = None,
     ) -> None:
-        if sort_by is not None and sort_by not in self._VALID_SORT_BY:
-            raise ValueError(
-                f"sort_by must be one of {self._VALID_SORT_BY!r}, got {sort_by!r}"
-            )
         _validate_non_negative_int("max_num_chars", max_num_chars)
         _validate_non_negative_int("limit", limit)
         _validate_non_negative_int("offset", offset)
@@ -320,7 +297,6 @@ def paradedb_verify_index(
     using: str = DEFAULT_DB_ALIAS,
 ) -> list[dict[str, Any]]:
     """Run ``pdb.verify_index()`` for one BM25 index."""
-    _validate_sample_rate(sample_rate)
     sql = [f"SELECT * FROM {FN_VERIFY_INDEX}(%s::regclass"]
     params: list[Any] = [index]
     if heapallindexed:
@@ -356,7 +332,6 @@ def paradedb_verify_all_indexes(
     using: str = DEFAULT_DB_ALIAS,
 ) -> list[dict[str, Any]]:
     """Run ``pdb.verify_all_indexes()`` across BM25 indexes."""
-    _validate_sample_rate(sample_rate)
     sql = [f"SELECT * FROM {FN_VERIFY_ALL_INDEXES}("]
     params: list[Any] = []
     named_params: list[tuple[str, str, Any]] = []

paradedb/queryset.py

@@ -43,15 +43,6 @@ def facets(
         include_rows: bool = True,
         exact: bool | None = None,
     ) -> dict[str, object] | tuple[list[Any], dict[str, object]]:
-        # Faceted queries require pdb.agg() OVER () with ORDER BY + LIMIT and a ParadeDB
-        # search condition in the WHERE clause to trigger the custom scan.
-        #
-        # Example:
-        # SELECT id, description, pdb.agg('{"value_count":{"field":"id"}}') OVER ()
-        # FROM mock_items
-        # WHERE description ||| 'running shoes'
-        # ORDER BY id
-        # LIMIT 5;
         if not fields and agg is None:
             raise ValueError("facets() requires fields or agg.")
 

pyproject.toml

@@ -138,6 +138,7 @@ markers = [
 [tool.coverage.run]
 source = ["paradedb"]
 branch = true
+omit = ["api.py", "__init__.py"]
 
 [tool.coverage.report]
 exclude_lines = [