feat: Add --dry-run option to validate. (#778)

* Add --dry-run option to validate.

* Document --dry-run in README.

* Blacken

README.md

@@ -376,6 +376,31 @@ data-validation (--verbose or -v) (--log-level or -ll) validate custom-query row
 The [Examples](https://github.com/GoogleCloudPlatform/professional-services-data-validator/blob/develop/docs/examples.md)
 page provides few examples of how this tool can be used to run custom query row validations.
 
+#### Dry Run Validation
+
+The `validate` command takes a `--dry-run` command line flag that prints source
+and target SQL to stdout as JSON in lieu of performing a validation:
+
+```
+data-validation (--verbose or -v) (--log-level or -ll) validate
+  [--dry-run or -dr]    Prints source and target SQL to stdout in lieu of performing a validation.
+```
+
+For example, this flag can be used as follows:
+
+```shell
+> data-validation validate --dry-run row \
+  -sc my_bq_conn \
+  -tc my_bq_conn \
+  -tbls bigquery-public-data.new_york_citibike.citibike_stations \
+  --primary-keys station_id \
+  --hash '*'
+{
+    "source_query": "SELECT `hash__all`, `station_id`\nFROM ...",
+    "target_query": "SELECT `hash__all`, `station_id`\nFROM ..."
+}
+```
+
 ### YAML Configuration Files
 
 You can customize the configuration for any given validation by providing use

data_validation/__main__.py

@@ -419,11 +419,12 @@ def convert_config_to_yaml(args, config_managers):
     return yaml_config
 
 
-def run_validation(config_manager, verbose=False):
+def run_validation(config_manager, dry_run=False, verbose=False):
     """Run a single validation.
 
     Args:
         config_manager (ConfigManager): Validation config manager instance.
+        dry_run (bool): Print source and target SQL to stdout in lieu of validation.
         verbose (bool): Validation setting to log queries run.
     """
     validator = DataValidation(
@@ -432,7 +433,18 @@ def run_validation(config_manager, verbose=False):
         result_handler=None,
         verbose=verbose,
     )
-    validator.execute()
+    if dry_run:
+        print(
+            json.dumps(
+                {
+                    "source_query": validator.validation_builder.get_source_query().compile(),
+                    "target_query": validator.validation_builder.get_target_query().compile(),
+                },
+                indent=4,
+            )
+        )
+    else:
+        validator.execute()
 
 
 def run_validations(args, config_managers):
@@ -449,15 +461,17 @@ def run_validations(args, config_managers):
                 config_manager.config[consts.CONFIG_FILE],
             )
             try:
-                run_validation(config_manager, verbose=args.verbose)
+                run_validation(
+                    config_manager, dry_run=args.dry_run, verbose=args.verbose
+                )
             except Exception as e:
                 logging.error(
                     "Error %s occured while running config file %s. Skipping it for now.",
                     str(e),
                     config_manager.config[consts.CONFIG_FILE],
                 )
         else:
-            run_validation(config_manager, verbose=args.verbose)
+            run_validation(config_manager, dry_run=args.dry_run, verbose=args.verbose)
 
 
 def store_yaml_config_file(args, config_managers):

data_validation/cli_tools.py

@@ -399,6 +399,13 @@ def _configure_validate_parser(subparsers):
         "validate", help="Run a validation and optionally store to config"
     )
 
+    validate_parser.add_argument(
+        "--dry-run",
+        "-dr",
+        action="store_true",
+        help="Prints source and target SQL to stdout in lieu of performing a validation.",
+    )
+
     validate_subparsers = validate_parser.add_subparsers(dest="validate_cmd")
 
     column_parser = validate_subparsers.add_parser(