clickzetta-semantic-model-generator 1.0.10__tar.gz → 1.0.12__tar.gz

{clickzetta_semantic_model_generator-1.0.10 → clickzetta_semantic_model_generator-1.0.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: clickzetta-semantic-model-generator
-Version: 1.0.10
+Version: 1.0.12
 Summary: Curate a Semantic Model for ClickZetta Lakehouse
 License: Apache Software License; BSD License
 Author: qililiang

{clickzetta_semantic_model_generator-1.0.10 → clickzetta_semantic_model_generator-1.0.12}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "clickzetta-semantic-model-generator"
-version = "1.0.10"
+version = "1.0.12"
 description = "Curate a Semantic Model for ClickZetta Lakehouse"
 authors = ["qililiang <qililiang@clickzetta.com>"]
 license = "Apache Software License; BSD License"

{clickzetta_semantic_model_generator-1.0.10 → clickzetta_semantic_model_generator-1.0.12}/semantic_model_generator/clickzetta_utils/clickzetta_connector.py

@@ -491,19 +491,64 @@ def _fetch_columns_via_show(
         table_token = str(table_name).strip()
         if not table_token:
             continue
-        qualified_table = join_quoted_identifiers(
+
+        identifier_candidates: List[str] = []
+        fully_qualified = join_quoted_identifiers(
             *(part for part in (catalog, schema_token, table_token) if part)
         )
-        query = f"SHOW COLUMNS IN {qualified_table}"
-        try:
-            df = session.sql(query).to_pandas()
-        except Exception as exc:
-            logger.debug(
-                "SHOW COLUMNS fallback failed for {}: {}", qualified_table, exc
-            )
-            continue
+        if fully_qualified:
+            identifier_candidates.append(fully_qualified)
+        schema_qualified = (
+            join_quoted_identifiers(schema_token, table_token)
+            if schema_token
+            else ""
+        )
+        if schema_qualified:
+            identifier_candidates.append(schema_qualified)
+        bare_identifier = join_quoted_identifiers(table_token)
+        if bare_identifier:
+            identifier_candidates.append(bare_identifier)
+
+        df = pd.DataFrame()
+        df_source = ""
+        for identifier in identifier_candidates:
+            query = f"SHOW COLUMNS IN {identifier}"
+            try:
+                df = session.sql(query).to_pandas()
+                df_source = "SHOW COLUMNS"
+            except Exception as exc:
+                logger.debug(
+                    "SHOW COLUMNS fallback failed for {}: {}", identifier, exc
+                )
+                df = pd.DataFrame()
+            if df.empty:
+                describe_query = f"DESCRIBE TABLE {identifier}"
+                try:
+                    describe_df = session.sql(describe_query).to_pandas()
+                except Exception as exc:
+                    logger.debug(
+                        "DESCRIBE TABLE fallback failed for {}: {}", identifier, exc
+                    )
+                    describe_df = pd.DataFrame()
+                if not describe_df.empty:
+                    df = describe_df
+                    df_source = "DESCRIBE TABLE"
+            if not df.empty:
+                break
         if df.empty:
             continue
+        if df_source == "DESCRIBE TABLE":
+            if "KIND" in df.columns:
+                df = df[df["KIND"].astype(str).str.upper() == "COLUMN"]
+            rename_map = {}
+            if "NAME" in df.columns:
+                rename_map["NAME"] = "COLUMN_NAME"
+            if "TYPE" in df.columns:
+                rename_map["TYPE"] = "DATA_TYPE"
+            if rename_map:
+                df = df.rename(columns=rename_map)
+            if df.empty:
+                continue
         df.columns = [str(col).upper() for col in df.columns]
         schema_col = next(
             (col for col in ("TABLE_SCHEMA", "SCHEMA_NAME") if col in df.columns), None
@@ -527,8 +572,14 @@ def _fetch_columns_via_show(
         )
 
         normalized = pd.DataFrame()
-        normalized[_TABLE_SCHEMA_COL] = df[schema_col] if schema_col else table_schema
-        normalized[_TABLE_NAME_COL] = df[table_col] if table_col else table_name
+        normalized[_TABLE_SCHEMA_COL] = (
+            df[schema_col]
+            if schema_col
+            else (schema_token or table_schema or "")
+        )
+        normalized[_TABLE_NAME_COL] = (
+            df[table_col] if table_col else table_token
+        )
         normalized[_COLUMN_NAME_COL] = (
             df[column_col] if column_col else df.index.astype(str)
         )