vgi-python 0.11.0__tar.gz → 0.12.0__tar.gz

{vgi_python-0.11.0 → vgi_python-0.12.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vgi-python
-Version: 0.11.0
+Version: 0.12.0
 Summary: Vector Gateway Interface - Connect DuckDB to external programs via Apache Arrow
 Project-URL: Homepage, https://query.farm
 Project-URL: Repository, https://github.com/Query-farm/vgi-python

{vgi_python-0.11.0 → vgi_python-0.12.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "vgi-python"
-version = "0.11.0"
+version = "0.12.0"
 description = "Vector Gateway Interface - Connect DuckDB to external programs via Apache Arrow"
 readme = "README.md"
 keywords = [

{vgi_python-0.11.0 → vgi_python-0.12.0}/tests/landing/_checker.py

@@ -63,7 +63,7 @@ def check(
     """Return a list of failure strings (empty == conformant)."""
     base = base_url.rstrip("/")
     fails: list[str] = []
-    schema = json.loads(schema_path.read_text())
+    schema = json.loads(schema_path.read_text(encoding="utf-8"))
     columns_schema = {**schema, **schema["$defs"]["columns"]}
 
     # 1) describe.json — schema-valid.
@@ -105,11 +105,17 @@ def check(
     # 4) Golden diff (cross-language equality).
     if golden_path is not None:
         got = _canonical(normalize(doc))
-        want = golden_path.read_text() if golden_path.exists() else None
+        want = golden_path.read_text(encoding="utf-8") if golden_path.exists() else None
         if want is None:
             fails.append(f"golden {golden_path} does not exist (run with --update)")
         elif got.strip() != want.strip():
-            fails.append(f"describe.json does not match golden {golden_path.name} (normalized diff)")
+            import difflib
+
+            diff = list(
+                difflib.unified_diff(want.strip().splitlines(), got.strip().splitlines(), "golden", "live", lineterm="")
+            )
+            snippet = "\n".join(diff[:40])
+            fails.append(f"describe.json does not match golden {golden_path.name}:\n{snippet}")
 
     return fails
 
@@ -127,7 +133,7 @@ def main() -> int:
             ap.error("--update requires --golden")
         _, body, _ = _get(f"{args.url.rstrip('/')}/describe.json")
         args.golden.parent.mkdir(parents=True, exist_ok=True)
-        args.golden.write_text(_canonical(normalize(json.loads(body))) + "\n")
+        args.golden.write_text(_canonical(normalize(json.loads(body))) + "\n", encoding="utf-8")
         print(f"wrote golden {args.golden}")
         return 0
 

{vgi_python-0.11.0 → vgi_python-0.12.0}/tests/landing/describe.expected.json

@@ -3,7 +3,7 @@
     {
       "attach_options": [],
       "counts": {
-        "functions": 162,
+        "functions": 165,
         "schemas": 2,
         "tables": 44,
         "views": 3
@@ -13,6 +13,240 @@
       "implementation_version": null,
       "name": "example",
       "schemas": [
+        {
+          "functions": [],
+          "name": "data",
+          "tables": [
+            {
+              "cols": 1,
+              "comment": "Function-backed table with inlined cardinality (10000 rows)",
+              "name": "cardinality_inlined_table"
+            },
+            {
+              "cols": 3,
+              "comment": "Colors table with ENUM-derived statistics",
+              "name": "colors"
+            },
+            {
+              "cols": 3,
+              "comment": "Department reference table",
+              "name": "departments"
+            },
+            {
+              "cols": 4,
+              "comment": "Employee table with FK to departments",
+              "name": "employees"
+            },
+            {
+              "cols": 3,
+              "comment": "Catalog table echoing pushed-down filters (filter-pushdown-through-view tests).",
+              "name": "filter_echo_table"
+            },
+            {
+              "cols": 1,
+              "comment": "123456 integers; stats served by the sequence function, not the table",
+              "name": "funny_numbers"
+            },
+            {
+              "cols": 3,
+              "comment": "Table with generated columns backed by sequence(10)",
+              "name": "generated_sequence"
+            },
+            {
+              "cols": 2,
+              "comment": "5x5 grid of points with spatial statistics",
+              "name": "geo_points"
+            },
+            {
+              "cols": 1,
+              "comment": "A large sequence of integers from 0 to 1,000,000",
+              "name": "large_sequence"
+            },
+            {
+              "cols": 4,
+              "comment": "Late-materialization table (1000 rows, unique rowid)",
+              "name": "late_mat"
+            },
+            {
+              "cols": 4,
+              "comment": "Late-materialization table with deliberately non-unique rowid (contract violation)",
+              "name": "late_mat_dup"
+            },
+            {
+              "cols": 4,
+              "comment": "Late-materialization table with NULLs in the ord column",
+              "name": "late_mat_nulls"
+            },
+            {
+              "cols": 1,
+              "comment": "Multi-branch: empty branches list — used by multi_branch_empty_branches.test",
+              "name": "multi_branch_empty"
+            },
+            {
+              "cols": 1,
+              "comment": "Multi-branch with complementary branch_filters — exercises pruning",
+              "name": "multi_branch_filtered_numbers"
+            },
+            {
+              "cols": 1,
+              "comment": "Multi-branch: sequence(50) + read_parquet — used by multi_branch_heterogeneous.test",
+              "name": "multi_branch_hetero"
+            },
+            {
+              "cols": 1,
+              "comment": "Multi-branch: sequence(50) + iceberg_scan — used by multi_branch_iceberg.test",
+              "name": "multi_branch_iceberg"
+            },
+            {
+              "cols": 1,
+              "comment": "Multi-branch: VGI + read_csv — used by multi_branch_pushdown_incapable.test",
+              "name": "multi_branch_nopushdown"
+            },
+            {
+              "cols": 1,
+              "comment": "Multi-branch: UNION of sequence(50) + sequence(50) — used by multi_branch_scan.test",
+              "name": "multi_branch_numbers"
+            },
+            {
+              "cols": 2,
+              "comment": "Multi-branch: column reconciliation — used by multi_branch_reconciliation.test",
+              "name": "multi_branch_recon"
+            },
+            {
+              "cols": 1,
+              "comment": "Multi-branch with two writable=True arms — used by multi_branch_two_writable.test",
+              "name": "multi_branch_two_writable"
+            },
+            {
+              "cols": 1,
+              "comment": "First 100 integers (demonstrates explicit columns)",
+              "name": "numbers"
+            },
+            {
+              "cols": 4,
+              "comment": "Product table with column defaults",
+              "name": "products"
+            },
+            {
+              "cols": 3,
+              "comment": "Projects with composite PK and FK to departments",
+              "name": "projects"
+            },
+            {
+              "cols": 6,
+              "comment": "rff_hive — native read_parquet over Hive glob with bbox.* required filters.",
+              "name": "rff_hive"
+            },
+            {
+              "cols": 6,
+              "comment": "rff_hive_mixed — native read_parquet, top-level 'id' + bbox.* required filters.",
+              "name": "rff_hive_mixed"
+            },
+            {
+              "cols": 2,
+              "comment": "rff_multi — mixed top-level + struct subfield requirements.",
+              "name": "rff_multi"
+            },
+            {
+              "cols": 1,
+              "comment": "rff_nested — requires a filter on the 3-deep nested path wrapper.mid.leaf.",
+              "name": "rff_nested"
+            },
+            {
+              "cols": 2,
+              "comment": "rff_none — control table with no required_field_filter_paths (opt-out fast path).",
+              "name": "rff_none"
+            },
+            {
+              "cols": 2,
+              "comment": "rff_parquet — native read_parquet delegation with bbox.* required filters.",
+              "name": "rff_parquet"
+            },
+            {
+              "cols": 3,
+              "comment": "rff_rowid — row_id virtual column + required bbox.* filters.",
+              "name": "rff_rowid"
+            },
+            {
+              "cols": 2,
+              "comment": "rff_simple — requires a filter referencing column 'a'.",
+              "name": "rff_simple"
+            },
+            {
+              "cols": 2,
+              "comment": "rff_struct — requires filters on both struct subfields s.a and s.b.",
+              "name": "rff_struct"
+            },
+            {
+              "cols": 3,
+              "comment": "Table with row_id at column index 0",
+              "name": "rowid_first"
+            },
+            {
+              "cols": 3,
+              "comment": "Table with row_id at column index 2",
+              "name": "rowid_last"
+            },
+            {
+              "cols": 3,
+              "comment": "Table with row_id at column index 1",
+              "name": "rowid_middle"
+            },
+            {
+              "cols": 2,
+              "comment": "Table with string row_id",
+              "name": "rowid_string"
+            },
+            {
+              "cols": 2,
+              "comment": "Table with struct row_id",
+              "name": "rowid_struct"
+            },
+            {
+              "cols": 3,
+              "comment": "Function-backed table over the secret-using secret_demo function",
+              "name": "secret_demo_table"
+            },
+            {
+              "cols": 1,
+              "comment": "Function-backed table over the no-arg ten_thousand function",
+              "name": "ten_thousand_table"
+            },
+            {
+              "cols": 4,
+              "comment": "Columns-based: prunes by filter AND time-travels (AT → version arg).",
+              "name": "tt_pushdown_cols"
+            },
+            {
+              "cols": 4,
+              "comment": "Function-backed: prunes by filter AND time-travels (AT read at init).",
+              "name": "tt_pushdown_fn"
+            },
+            {
+              "cols": 4,
+              "comment": "Table with constraints that evolve across versions",
+              "name": "versioned_constraints"
+            },
+            {
+              "cols": 2,
+              "comment": "Versioned data table demonstrating time travel with schema evolution",
+              "name": "versioned_data"
+            },
+            {
+              "cols": 1,
+              "comment": "Numbers with volatile stats (TTL=0, always re-fetched)",
+              "name": "volatile_numbers"
+            }
+          ],
+          "views": [
+            {
+              "cols": 1,
+              "comment": "Numbers less than 10",
+              "def": "SELECT * FROM numbers WHERE value < 10",
+              "name": "small_numbers"
+            }
+          ]
+        },
         {
           "functions": [
             {
@@ -883,26 +1117,69 @@
             {
               "args": [
                 {
-                  "desc": "dummy input",
-                  "name": "x",
+                  "desc": "Value to clamp",
+                  "name": "val",
+                  "type": "ANY"
+                },
+                {
+                  "default": "0",
+                  "desc": "Lower bound (inclusive)",
+                  "name": "lo",
+                  "named": true,
+                  "type": "int64"
+                },
+                {
+                  "default": "100",
+                  "desc": "Upper bound (inclusive)",
+                  "name": "hi",
+                  "named": true,
                   "type": "int64"
                 }
               ],
-              "doc": "Return the authenticated principal name.",
-              "name": "whoami",
-              "returns": "string",
+              "doc": "Clamp a value between lo and hi (defaults: 0..100)",
+              "name": "vgi_clamp",
               "type": "scalar"
             },
             {
               "args": [
                 {
-                  "desc": "Total rows to attempt to generate",
-                  "name": "count",
-                  "type": "int64"
+                  "desc": "First factor",
+                  "name": "x",
+                  "type": "ANY"
+                },
+                {
+                  "desc": "Second factor",
+                  "name": "y",
+                  "type": "ANY"
                 }
               ],
-              "doc": "DELIBERATELY BROKEN: emits a batch tagged with a partition_id well above DuckDB's BATCH_INCREMENT=10^13 per-pipeline cap. C++ extension rejects at parse time.",
-              "name": "broken_batch_index_overflow",
+              "doc": "Multiply two values",
+              "name": "vgi_multiply",
+              "type": "scalar"
+            },
+            {
+              "args": [
+                {
+                  "desc": "dummy input",
+                  "name": "x",
+                  "type": "int64"
+                }
+              ],
+              "doc": "Return the authenticated principal name.",
+              "name": "whoami",
+              "returns": "string",
+              "type": "scalar"
+            },
+            {
+              "args": [
+                {
+                  "desc": "Total rows to attempt to generate",
+                  "name": "count",
+                  "type": "int64"
+                }
+              ],
+              "doc": "DELIBERATELY BROKEN: emits a batch tagged with a partition_id well above DuckDB's BATCH_INCREMENT=10^13 per-pipeline cap. C++ extension rejects at parse time.",
+              "name": "broken_batch_index_overflow",
               "type": "table"
             },
             {
@@ -1972,12 +2249,6 @@
               "name": "struct_settings",
               "type": "table"
             },
-            {
-              "args": [],
-              "doc": "Distributed sum using simple callback API",
-              "name": "sum_all_columns_simple_distributed",
-              "type": "table_in_out"
-            },
             {
               "args": [],
               "doc": "Generates 10000 integers from 0 to 9999",
@@ -2127,6 +2398,18 @@
               "name": "versioned_data_scan",
               "type": "table"
             },
+            {
+              "args": [
+                {
+                  "desc": "Number of rows to generate",
+                  "name": "n",
+                  "type": "ANY"
+                }
+              ],
+              "doc": "Table macro returning range of values",
+              "name": "vgi_range_table",
+              "type": "table"
+            },
             {
               "args": [],
               "doc": "buffer_input variant using batch_index to reconstruct order",
@@ -2376,256 +2659,28 @@
               "doc": "Computes column-wise sums across all batches",
               "name": "sum_all_columns",
               "type": "table_in_out"
+            },
+            {
+              "args": [],
+              "doc": "Distributed sum using simple callback API",
+              "name": "sum_all_columns_simple_distributed",
+              "type": "table_in_out"
             }
           ],
           "name": "main",
           "tables": [],
           "views": [
-            {
-              "cols": 1,
-              "comment": "First 10 integers",
-              "def": "SELECT * FROM sequence(10)",
-              "name": "first_ten"
-            },
             {
               "cols": 0,
               "comment": "Even numbers from 0 to 98",
               "def": "SELECT * FROM sequence(100) WHERE n % 2 = 0",
               "name": "even_numbers"
-            }
-          ]
-        },
-        {
-          "functions": [],
-          "name": "data",
-          "tables": [
-            {
-              "cols": 1,
-              "comment": "A large sequence of integers from 0 to 1,000,000",
-              "name": "large_sequence"
-            },
-            {
-              "cols": 1,
-              "comment": "Function-backed table over the no-arg ten_thousand function",
-              "name": "ten_thousand_table"
-            },
-            {
-              "cols": 3,
-              "comment": "Function-backed table over the secret-using secret_demo function",
-              "name": "secret_demo_table"
-            },
-            {
-              "cols": 1,
-              "comment": "Function-backed table with inlined cardinality (10000 rows)",
-              "name": "cardinality_inlined_table"
-            },
-            {
-              "cols": 2,
-              "comment": "Versioned data table demonstrating time travel with schema evolution",
-              "name": "versioned_data"
-            },
-            {
-              "cols": 4,
-              "comment": "Function-backed: prunes by filter AND time-travels (AT read at init).",
-              "name": "tt_pushdown_fn"
-            },
-            {
-              "cols": 4,
-              "comment": "Columns-based: prunes by filter AND time-travels (AT → version arg).",
-              "name": "tt_pushdown_cols"
-            },
-            {
-              "cols": 1,
-              "comment": "First 100 integers (demonstrates explicit columns)",
-              "name": "numbers"
-            },
-            {
-              "cols": 2,
-              "comment": "5x5 grid of points with spatial statistics",
-              "name": "geo_points"
-            },
-            {
-              "cols": 1,
-              "comment": "Numbers with volatile stats (TTL=0, always re-fetched)",
-              "name": "volatile_numbers"
-            },
-            {
-              "cols": 1,
-              "comment": "123456 integers; stats served by the sequence function, not the table",
-              "name": "funny_numbers"
-            },
-            {
-              "cols": 1,
-              "comment": "Multi-branch: UNION of sequence(50) + sequence(50) — used by multi_branch_scan.test",
-              "name": "multi_branch_numbers"
-            },
-            {
-              "cols": 1,
-              "comment": "Multi-branch with complementary branch_filters — exercises pruning",
-              "name": "multi_branch_filtered_numbers"
-            },
-            {
-              "cols": 1,
-              "comment": "Multi-branch: sequence(50) + read_parquet — used by multi_branch_heterogeneous.test",
-              "name": "multi_branch_hetero"
-            },
-            {
-              "cols": 1,
-              "comment": "Multi-branch: sequence(50) + iceberg_scan — used by multi_branch_iceberg.test",
-              "name": "multi_branch_iceberg"
-            },
-            {
-              "cols": 2,
-              "comment": "Multi-branch: column reconciliation — used by multi_branch_reconciliation.test",
-              "name": "multi_branch_recon"
-            },
-            {
-              "cols": 1,
-              "comment": "Multi-branch: VGI + read_csv — used by multi_branch_pushdown_incapable.test",
-              "name": "multi_branch_nopushdown"
             },
             {
               "cols": 1,
-              "comment": "Multi-branch: empty branches list — used by multi_branch_empty_branches.test",
-              "name": "multi_branch_empty"
-            },
-            {
-              "cols": 1,
-              "comment": "Multi-branch with two writable=True arms — used by multi_branch_two_writable.test",
-              "name": "multi_branch_two_writable"
-            },
-            {
-              "cols": 3,
-              "comment": "Colors table with ENUM-derived statistics",
-              "name": "colors"
-            },
-            {
-              "cols": 3,
-              "comment": "Table with row_id at column index 0",
-              "name": "rowid_first"
-            },
-            {
-              "cols": 3,
-              "comment": "Table with row_id at column index 1",
-              "name": "rowid_middle"
-            },
-            {
-              "cols": 3,
-              "comment": "Table with row_id at column index 2",
-              "name": "rowid_last"
-            },
-            {
-              "cols": 2,
-              "comment": "Table with string row_id",
-              "name": "rowid_string"
-            },
-            {
-              "cols": 2,
-              "comment": "Table with struct row_id",
-              "name": "rowid_struct"
-            },
-            {
-              "cols": 4,
-              "comment": "Late-materialization table (1000 rows, unique rowid)",
-              "name": "late_mat"
-            },
-            {
-              "cols": 4,
-              "comment": "Late-materialization table with deliberately non-unique rowid (contract violation)",
-              "name": "late_mat_dup"
-            },
-            {
-              "cols": 4,
-              "comment": "Late-materialization table with NULLs in the ord column",
-              "name": "late_mat_nulls"
-            },
-            {
-              "cols": 3,
-              "comment": "Table with generated columns backed by sequence(10)",
-              "name": "generated_sequence"
-            },
-            {
-              "cols": 3,
-              "comment": "Department reference table",
-              "name": "departments"
-            },
-            {
-              "cols": 4,
-              "comment": "Product table with column defaults",
-              "name": "products"
-            },
-            {
-              "cols": 4,
-              "comment": "Employee table with FK to departments",
-              "name": "employees"
-            },
-            {
-              "cols": 3,
-              "comment": "Projects with composite PK and FK to departments",
-              "name": "projects"
-            },
-            {
-              "cols": 3,
-              "comment": "Catalog table echoing pushed-down filters (filter-pushdown-through-view tests).",
-              "name": "filter_echo_table"
-            },
-            {
-              "cols": 2,
-              "comment": "rff_simple — requires a filter referencing column 'a'.",
-              "name": "rff_simple"
-            },
-            {
-              "cols": 2,
-              "comment": "rff_struct — requires filters on both struct subfields s.a and s.b.",
-              "name": "rff_struct"
-            },
-            {
-              "cols": 1,
-              "comment": "rff_nested — requires a filter on the 3-deep nested path wrapper.mid.leaf.",
-              "name": "rff_nested"
-            },
-            {
-              "cols": 2,
-              "comment": "rff_multi — mixed top-level + struct subfield requirements.",
-              "name": "rff_multi"
-            },
-            {
-              "cols": 2,
-              "comment": "rff_none — control table with no required_field_filter_paths (opt-out fast path).",
-              "name": "rff_none"
-            },
-            {
-              "cols": 3,
-              "comment": "rff_rowid — row_id virtual column + required bbox.* filters.",
-              "name": "rff_rowid"
-            },
-            {
-              "cols": 2,
-              "comment": "rff_parquet — native read_parquet delegation with bbox.* required filters.",
-              "name": "rff_parquet"
-            },
-            {
-              "cols": 6,
-              "comment": "rff_hive — native read_parquet over Hive glob with bbox.* required filters.",
-              "name": "rff_hive"
-            },
-            {
-              "cols": 6,
-              "comment": "rff_hive_mixed — native read_parquet, top-level 'id' + bbox.* required filters.",
-              "name": "rff_hive_mixed"
-            },
-            {
-              "cols": 4,
-              "comment": "Table with constraints that evolve across versions",
-              "name": "versioned_constraints"
-            }
-          ],
-          "views": [
-            {
-              "cols": 1,
-              "comment": "Numbers less than 10",
-              "def": "SELECT * FROM numbers WHERE value < 10",
-              "name": "small_numbers"
+              "comment": "First 10 integers",
+              "def": "SELECT * FROM sequence(10)",
+              "name": "first_ten"
             }
           ]
         }