chore(ibis): implement mdl validation scripts (#864)

ibis-server/pyproject.toml

@@ -62,6 +62,7 @@ markers = [
 [tool.ruff]
 line-length = 88
 target-version = "py311"
+exclude = ["tools/"]
 
 [tool.ruff.lint]
 select = [

ibis-server/tools/README.md

@@ -0,0 +1,10 @@
+# Description
+
+This folder contains useful tools and scripts for debugging and validation.
+
+# Tools
+- `mdl_validation.py`: Used to validate a Wren MDL. This script attempts to select all columns in all models.
+  - Requires the `wren_core` library. Run `just install-core` and `just install` before using it.
+```
+poetry run python tools/mdl_validation.py mdl.json function_list/bigquery.csv
+```

ibis-server/tools/mdl_validation.py

@@ -0,0 +1,47 @@
+#
+# This script is used to validate the MDL file by transforming the SQL queries
+# in the MDL file to the SQL queries. This script only validate if it can be transformed
+# by Wren core. It does not validate the SQL can be executed by the data source.
+#
+# Argements:
+# - manifest_json_path: Path to the JSON file
+# - function_list_path: Path to the function list CSV file
+#
+
+import argparse
+import base64
+import json
+
+from wren_core import SessionContext
+
+# Set up argument parsing
+parser = argparse.ArgumentParser(description="Validate the MDL file")
+parser.add_argument("manifest_json_path", help="Path to the JSON file")
+parser.add_argument("function_list_path", help="Path to the function list CSV file")
+
+args = parser.parse_args()
+
+# Read and encode the JSON data
+with open(args.manifest_json_path) as file:
+    mdl = json.load(file)
+    # Convert to JSON string
+    json_str = json.dumps(mdl)
+    # Encode to base64
+    encoded_str = base64.b64encode(json_str.encode("utf-8")).decode("utf-8")
+
+session_context = SessionContext(encoded_str, args.function_list_path)
+error_cases = []
+for model in mdl["models"]:
+    for column in model["columns"]:
+        # ignore hidden columns
+        if column.get("is_hidden"):
+            continue
+        sql = f"select \"{column['name']}\" from \"{model['name']}\""
+        try:
+            planned_sql = session_context.transform_sql(sql)
+        except Exception:
+            error_cases.append((model, column))
+            print(f"Error transforming {model['name']} {column['name']}")
+
+if len(error_cases) > 0:
+    raise Exception(f"Error transforming {len(error_cases)} columns")