PyReverseETL 3.1.6__tar.gz → 3.2.0__tar.gz

{pyreverseetl-3.1.6 → pyreverseetl-3.2.0}/Cargo.lock

@@ -3199,7 +3199,7 @@ dependencies = [
 
 [[package]]
 name = "pyreverseetl"
-version = "3.1.6"
+version = "3.2.0"
 dependencies = [
  "chrono",
  "pyo3",
@@ -3210,7 +3210,7 @@ dependencies = [
 
 [[package]]
 name = "pyreverseetl-core"
-version = "3.1.6"
+version = "3.2.0"
 dependencies = [
  "async-trait",
  "aws-config",

{pyreverseetl-3.1.6 → pyreverseetl-3.2.0}/Cargo.toml

@@ -3,7 +3,7 @@ members = ["core", "python"]
 resolver = "2"
 
 [workspace.package]
-version = "3.1.6"
+version = "3.2.0"
 authors = ["Georgi Mammen Mullassery <mullassery@gmail.com>"]
 license = "Apache-2.0"
 repository = "https://github.com/Mullassery/PyReverseETL"

{pyreverseetl-3.1.6 → pyreverseetl-3.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PyReverseETL
-Version: 3.1.6
+Version: 3.2.0
 Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Developers
 Classifier: Programming Language :: Python :: 3
@@ -249,6 +249,16 @@ Python CLI / API  →  pyreverseetl._core (PyO3 bindings)  →  pyreverseetl_cor
 - **`python/pyreverseetl/`** — the installed Python package: `cli.py` (the `pyreverseetl`
   command) and `server.py` (an optional Flask REST wrapper around the same engine).
 
+## Validation gate
+
+```bash
+pyreverseetl execute-activation my-activation --validation-report statguardian-report.json
+```
+
+The report must be a JSON object with `"passed": true`; anything else blocks
+the sync before reading or writing data. This is a file-based gate until a live
+StatGuardian service endpoint is available.
+
 ## Documentation
 
 - [ROADMAP_HONEST.md](ROADMAP_HONEST.md) — what's tested vs. untested vs. broken vs. not

{pyreverseetl-3.1.6 → pyreverseetl-3.2.0}/README.md

@@ -226,6 +226,16 @@ Python CLI / API  →  pyreverseetl._core (PyO3 bindings)  →  pyreverseetl_cor
 - **`python/pyreverseetl/`** — the installed Python package: `cli.py` (the `pyreverseetl`
   command) and `server.py` (an optional Flask REST wrapper around the same engine).
 
+## Validation gate
+
+```bash
+pyreverseetl execute-activation my-activation --validation-report statguardian-report.json
+```
+
+The report must be a JSON object with `"passed": true`; anything else blocks
+the sync before reading or writing data. This is a file-based gate until a live
+StatGuardian service endpoint is available.
+
 ## Documentation
 
 - [ROADMAP_HONEST.md](ROADMAP_HONEST.md) — what's tested vs. untested vs. broken vs. not

{pyreverseetl-3.1.6 → pyreverseetl-3.2.0}/core/src/connectors/database_advanced.rs

@@ -222,27 +222,29 @@ impl super::DestinationConnector for AdvancedDatabaseDestination {
     }
 
     async fn test_connection(&self) -> crate::Result<ConnectionTest> {
-        Ok(ConnectionTest::success(&format!(
-            "Connected to {} database",
-            self.config.database_type
-        )))
+        // Honest: this connector has no real driver, so it cannot connect.
+        Ok(ConnectionTest::failure(
+            &format!(
+                "{} destination is not implemented: no real connection was attempted",
+                self.config.database_type
+            ),
+            Some("advanced database connector is a stub (RECHECK-054)".to_string()),
+        ))
     }
 
     async fn write_record(&self, _record: &Record) -> crate::Result<()> {
-        // In production: execute INSERT/UPDATE based on strategy
-        Ok(())
+        Err(crate::Error::ConnectorError(format!(
+            "{} destination write is not implemented; refusing to report a write that did not happen",
+            self.config.database_type
+        )))
     }
 
-    async fn write_batch(&self, records: &[Record]) -> crate::Result<usize> {
-        // In production:
-        // 1. If schema defined: CREATE TABLE IF NOT EXISTS
-        // 2. Choose strategy:
-        //    - INSERT: simple insert all
-        //    - UPSERT: use ON CONFLICT or MERGE
-        //    - BULK_LOAD: write to S3/GCS, COPY into database
-        // 3. Handle partitioning for warehouses
-        // 4. Create clustering indexes for BigQuery
-        Ok(records.len())
+    async fn write_batch(&self, _records: &[Record]) -> crate::Result<usize> {
+        Err(crate::Error::ConnectorError(format!(
+            "{} destination batch write is not implemented; refusing to report {} rows written when none were",
+            self.config.database_type,
+            _records.len()
+        )))
     }
 
     async fn validate_records(&self, _records: &[Record]) -> crate::Result<()> {
@@ -283,38 +285,42 @@ impl super::SourceConnector for AdvancedDatabaseSource {
     }
 
     async fn test_connection(&self) -> crate::Result<ConnectionTest> {
-        Ok(ConnectionTest::success(&format!(
-            "Connected to {} database: {}",
-            self.config.database_type, self.config.table
-        )))
+        // Honest: this connector has no real driver, so it cannot connect.
+        Ok(ConnectionTest::failure(
+            &format!(
+                "{} source is not implemented: no real connection was attempted",
+                self.config.database_type
+            ),
+            Some("advanced database connector is a stub (RECHECK-054)".to_string()),
+        ))
     }
 
     async fn detect_schema(&self) -> crate::Result<super::Schema> {
-        // In production: query database metadata
-        // SELECT column_name, data_type FROM information_schema.columns WHERE table_name = ?
-        Ok(super::Schema {
-            fields: vec![],
-            sample_records: vec![],
-        })
+        Err(crate::Error::ConnectorError(format!(
+            "{} schema detection is not implemented",
+            self.config.database_type
+        )))
     }
 
     async fn read_all(&self) -> crate::Result<Vec<Record>> {
-        // In production: SELECT * FROM table
-        Ok(vec![])
+        Err(crate::Error::ConnectorError(format!(
+            "{} read is not implemented; refusing to report an empty result as a successful read",
+            self.config.database_type
+        )))
     }
 
     async fn read_batch(&self, _offset: u64, _limit: u64) -> crate::Result<Vec<Record>> {
-        // In production: SELECT * FROM table OFFSET ? LIMIT ?
-        Ok(vec![])
+        Err(crate::Error::ConnectorError(format!(
+            "{} read_batch is not implemented",
+            self.config.database_type
+        )))
     }
 
     async fn read_incremental(&self, _last_value: &str) -> crate::Result<Vec<Record>> {
-        // In production:
-        // If incremental_column is timestamp:
-        //   SELECT * FROM table WHERE incremental_column > ? ORDER BY incremental_column
-        // If incremental_column is id:
-        //   SELECT * FROM table WHERE id > ? ORDER BY id LIMIT ?
-        Ok(vec![])
+        Err(crate::Error::ConnectorError(format!(
+            "{} read_incremental is not implemented",
+            self.config.database_type
+        )))
     }
 
     fn capabilities(&self) -> Vec<Capability> {
@@ -370,8 +376,13 @@ mod tests {
         let dest = AdvancedDatabaseDestination { config };
         assert_eq!(dest.name(), "Snowflake");
 
+        // The connector is a stub: it must report failure, not fake success.
         let test = dest.test_connection().await.unwrap();
-        assert!(test.success);
+        assert!(!test.success);
+        assert!(test.message.contains("not implemented"));
+
+        // And a write must error rather than silently claim rows were written.
+        assert!(dest.write_batch(&[]).await.is_err());
     }
 
     #[tokio::test]

{pyreverseetl-3.1.6 → pyreverseetl-3.2.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "PyReverseETL"
-version = "3.1.6"
+version = "3.2.0"
 description = "The Open Operational Data Activation Runtime"
 readme = "README.md"
 license = "Apache-2.0"

{pyreverseetl-3.1.6 → pyreverseetl-3.2.0}/python/pyreverseetl/cli.py

@@ -31,6 +31,28 @@ import sys
 DEFAULT_STATE_PATH = os.path.join(os.path.expanduser("~"), ".pyreverseetl", "state.json")
 
 
+def validation_report_allows(path: str) -> tuple[bool, dict]:
+    """Read a StatGuardian-style JSON report and decide whether sync may run.
+
+    Expected shape: ``{"passed": bool, "violations": [...], ...}``. A missing
+    or malformed file fails closed. This is a file-based gate until
+    StatGuardian exposes a service endpoint; it does not call StatGuardian.
+    """
+    try:
+        with open(path, encoding="utf-8") as handle:
+            report = json.load(handle)
+    except (OSError, ValueError) as exc:
+        return False, {"reason": f"cannot read validation report: {exc}"}
+    if not isinstance(report, dict) or "passed" not in report:
+        return False, {"reason": "validation report must be a JSON object with a 'passed' field"}
+    if report.get("passed") is not True:
+        return False, {
+            "reason": "validation report did not pass",
+            "violations": report.get("violations", []),
+        }
+    return True, {}
+
+
 class CLIInterface:
     """Command-line interface for PyReverseETL workflow integration."""
 
@@ -162,6 +184,7 @@ class CLIInterface:
         schema_store_path: str | None = None,
         idempotency_store_path: str | None = None,
         dlq_store_path: str | None = None,
+        validation_report_path: str | None = None,
     ) -> dict:
         """Execute an activation: run a real sync through the Rust engine.
 
@@ -200,6 +223,11 @@ class CLIInterface:
                 changed is still sent. Only applies to the adapter-based
                 destinations (webhook/salesforce/hubspot/marketo). Omit to
                 disable idempotency checking entirely (default).
+            validation_report_path: Optional path to a JSON validation report
+                (``{"passed": bool, ...}``). If supplied and not passing, the
+                sync is blocked before any source read or destination write.
+                This is a file-based StatGuardian gate; it does not call the
+                StatGuardian service.
             dlq_store_path: Optional path to a SQLite file used as a
                 dead-letter queue (quarantine) for records the destination
                 rejects. Each failed record is stored with its run id, the
@@ -220,6 +248,15 @@ class CLIInterface:
             }
 
         activation = self.activations[activation_id]
+        if validation_report_path:
+            allowed, detail = validation_report_allows(validation_report_path)
+            if not allowed:
+                return {
+                    "status": "error",
+                    "activation_id": activation_id,
+                    "message": "Blocked by validation gate",
+                    "validation": detail,
+                }
         workflow = self.workflows.get(activation["workflow_id"])
         if workflow is None:
             return {
@@ -501,6 +538,14 @@ def main():
                     sys.exit(1)
                 dlq_store_path = sys.argv[idx + 1]
 
+            validation_report_path = None
+            if "--validation-report" in sys.argv:
+                idx = sys.argv.index("--validation-report")
+                if idx + 1 >= len(sys.argv):
+                    print(json.dumps({"error": "--validation-report requires a file path"}))
+                    sys.exit(1)
+                validation_report_path = sys.argv[idx + 1]
+
             result = cli.execute_activation(
                 activation_id,
                 limit,
@@ -509,6 +554,7 @@ def main():
                 schema_store_path,
                 idempotency_store_path,
                 dlq_store_path,
+                validation_report_path,
             )
             print(json.dumps(result))