fastprocesses 0.22.4__tar.gz → 0.22.6__tar.gz

{fastprocesses-0.22.4 → fastprocesses-0.22.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: fastprocesses
-Version: 0.22.4
+Version: 0.22.6
 Summary: A library to create a FastAPI-based  OGC API Processes wrapper around existing projects.
 License-File: AUTHORS.md
 Author: Stefan Schuhart
@@ -39,7 +39,7 @@ A library to create a FastAPI-based OGC API Processes wrapper around existing pr
 
 AI helped to create this code.
 
-## Version: 0.22.4
+## Version: 0.22.6
 
 ### Description
 

{fastprocesses-0.22.4 → fastprocesses-0.22.6}/README.md

@@ -4,7 +4,7 @@ A library to create a FastAPI-based OGC API Processes wrapper around existing pr
 
 AI helped to create this code.
 
-## Version: 0.22.4
+## Version: 0.22.6
 
 ### Description
 

{fastprocesses-0.22.4 → fastprocesses-0.22.6}/pyproject.toml

@@ -79,7 +79,7 @@ exclude = [
 
 [tool.poetry]
 name = "fastprocesses"
-version = "0.22.4"
+version = "0.22.6"
 description = "A library to create a FastAPI-based  OGC API Processes wrapper around existing projects."
 authors = ["Stefan Schuhart <stefan.schuhart@gv.hamburg.de>"]
 readme = "README.md"

{fastprocesses-0.22.4 → fastprocesses-0.22.6}/src/fastprocesses/api/manager.py

@@ -116,6 +116,7 @@ class AsyncExecutionStrategy(ExecutionStrategy):
                 "processID": process_id,
                 "created": datetime.now(timezone.utc),
                 "progress": 0,
+                "message": "Waiting for a worker to pick up the task.",
                 "links": [
                     Link.model_validate(
                         {
@@ -183,6 +184,7 @@ class AsyncExecutionStrategy(ExecutionStrategy):
                 "processID": process_id,
                 "created": datetime.now(timezone.utc),
                 "progress": 0,
+                "message": "Waiting for a worker to pick up the task.",
                 "links": [
                     Link.model_validate(
                         {

{fastprocesses-0.22.4 → fastprocesses-0.22.6}/src/fastprocesses/core/config.py

@@ -128,6 +128,13 @@ class OGCProcessesSettings(BaseSettings):
         default=10,
         description="Timeout in seconds for synchronous execution waiting for result."
     )
+    FP_SKIP_INPUT_VALIDATION: bool = Field(
+        default=False,
+        description=(
+            "Skip JSON-Schema input validation on the worker. "
+            "Intended for debugging only — never enable in production."
+        ),
+    )
     FP_LOG_LEVEL: str = Field(
         default="INFO",
         description=(

{fastprocesses-0.22.4 → fastprocesses-0.22.6}/src/fastprocesses/worker/celery_app.py

@@ -89,6 +89,12 @@ def execute_process(self, process_id: str, serialized_data: str | bytes):
         process_id,
         job_id,
     )
+    update_job_status(
+        job_id,
+        0,
+        "Worker started, task picked up.",
+        process_id=process_id,
+    )
 
     # Check cache before running the pipeline — avoids re-execution for
     # identical inputs (especially important for parallel/scatter processes

{fastprocesses-0.22.4 → fastprocesses-0.22.6}/src/fastprocesses/worker/chord_tasks.py

@@ -72,10 +72,13 @@ def execute_parallel_item(
     """
     item = temp_result_cache.get(item_key)
     if item is None:
-        raise RuntimeError(
-            f"Claim-check data not found in cache for key '{item_key}'. "
-            "The temp_result_cache entry may have expired or been evicted."
+        logger.error(
+            "Claim-check data not found for key '{}' (job {}). "
+            "Cache entry may have expired or been evicted.",
+            item_key,
+            job_id,
         )
+        return {"__error__": f"Claim-check data not found in cache for key '{item_key}'"}
     try:
         process = cast(
             BaseParallelProcess,
@@ -99,14 +102,14 @@ def execute_parallel_item(
         _increment_and_report_progress(job_id, total)
         return {"__claim_check__": rkey}
     except Exception as exc:
-        update_job_status(
+        logger.error(
+            "Parallel subtask failed for process {} (job {}): {}",
+            process_id,
             job_id,
-            0,
-            f"Parallel subtask failed: {exc}",
-            JobStatusCode.FAILED,
-            process_id=process_id,
+            exc,
+            exc_info=True,
         )
-        raise
+        return {"__error__": str(exc)}
 
 
 def _run_parallel(
@@ -168,6 +171,29 @@ def finalize_parallel(
     actual results, caches the final output, and updates the job status.
     """
     process = cast(BaseParallelProcess, get_process_registry().get_process(process_id))
+
+    errors = [r for r in sub_results if isinstance(r, dict) and "__error__" in r]
+    if errors:
+        error_msg = errors[0]["__error__"]
+        update_job_status(
+            job_id,
+            0,
+            f"A parallel subtask failed: {error_msg}",
+            JobStatusCode.FAILED,
+        )
+        _cleanup_progress_counter(job_id)
+        temp_result_cache.delete(meta_key)
+        logger.error(
+            "Parallel chord failed for process {} (job {}): {}/{} subtask(s) failed. "
+            "First error: {}",
+            process_id,
+            job_id,
+            len(errors),
+            len(sub_results),
+            error_msg,
+        )
+        return None
+
     try:
         actual_results: list[dict] = []
         for result_ref in sub_results:
@@ -242,10 +268,13 @@ def execute_scatter_step(
     """
     payload = temp_result_cache.get(payload_key)
     if payload is None:
-        raise RuntimeError(
-            f"Claim-check payload not found in cache for key '{payload_key}'. "
-            "The temp_result_cache entry may have expired or been evicted."
+        logger.error(
+            "Claim-check payload not found for key '{}' (job {}). "
+            "Cache entry may have expired or been evicted.",
+            payload_key,
+            job_id,
         )
+        return {"__error__": f"Claim-check payload not found in cache for key '{payload_key}'"}
     data: dict = payload["step_input"]
     try:
         process = cast(
@@ -277,14 +306,15 @@ def execute_scatter_step(
         _record_scatter_step_done(job_id, step_name, total)
         return {"__claim_check__": rkey}
     except Exception as exc:
-        update_job_status(
+        logger.error(
+            "Scatter step '{}' failed for process {} (job {}): {}",
+            step_name,
+            process_id,
             job_id,
-            0,
-            f"Scatter step '{step_name}' failed: {exc}",
-            JobStatusCode.FAILED,
-            process_id=process_id,
+            exc,
+            exc_info=True,
         )
-        raise
+        return {"__error__": str(exc), "__step__": step_name}
 
 
 def _run_scatter(
@@ -341,7 +371,7 @@ def finalize_scatter(
     process_id: str,
     job_id: str,
     payload_key: str,
-) -> dict:
+) -> dict | None:
     """
     Chord callback for ``BaseScatterProcess``.
 
@@ -350,6 +380,29 @@ def finalize_scatter(
     and updates the job status.
     """
     process = cast(BaseScatterProcess, get_process_registry().get_process(process_id))
+
+    errors = [r for r in step_results if isinstance(r, dict) and "__error__" in r]
+    if errors:
+        first = errors[0]
+        error_msg = (
+            f"Step '{first['__step__']}' failed: {first['__error__']}"
+            if "__step__" in first
+            else first["__error__"]
+        )
+        update_job_status(job_id, 0, error_msg, JobStatusCode.FAILED)
+        _cleanup_scatter_roster(job_id)
+        temp_result_cache.delete(payload_key)
+        logger.error(
+            "Scatter chord failed for process {} (job {}): {}/{} step(s) failed. "
+            "First error: {}",
+            process_id,
+            job_id,
+            len(errors),
+            len(step_results),
+            error_msg,
+        )
+        return None
+
     try:
         named_results: dict[str, dict] = {}
         for step_name, result_ref in zip(step_names, step_results):

{fastprocesses-0.22.4 → fastprocesses-0.22.6}/src/fastprocesses/worker/pipeline.py

@@ -1,6 +1,7 @@
 import json
 
 from fastprocesses.core.base_process import BaseProcess
+from fastprocesses.core.config import OGCProcessesSettings
 from fastprocesses.core.exceptions import (
     InputValidationError,
     ProcessClassNotFoundError,
@@ -49,17 +50,26 @@ def _run_pipeline(
     Returns the (possibly modified) data dict after remote input resolution.
     """
     # Step 1: validate wire-format inputs against the process description schema
-    update_job_status(
-        job_id, 0,
-        "Pre-process step 1: Validating inputs against process description.",
-        JobStatusCode.RUNNING
-    )
-    try:
-        process.validate_inputs(data["inputs"])
-    except ValueError as e:
-        logger.error(f"Input validation failed for process {process_id}: {e}")
-        update_job_status(job_id, 0, str(e), JobStatusCode.FAILED)
-        raise InputValidationError(process_id, repr(e))
+    settings = OGCProcessesSettings()
+    if settings.FP_SKIP_INPUT_VALIDATION:
+        logger.warning(
+            "FP_SKIP_INPUT_VALIDATION is enabled — skipping input validation "
+            "for process {} (job {}). Do not use this in production.",
+            process_id,
+            job_id,
+        )
+    else:
+        update_job_status(
+            job_id, 0,
+            "Pre-process step 1: Validating inputs against process description.",
+            JobStatusCode.RUNNING
+        )
+        try:
+            process.validate_inputs(data["inputs"])
+        except ValueError as e:
+            logger.error(f"Input validation failed for process {process_id}: {e}")
+            update_job_status(job_id, 0, str(e), JobStatusCode.FAILED)
+            raise InputValidationError(process_id, repr(e))
 
     # Step 2: resolve remote inputs (URI strings → downloaded data)
     update_job_status(