issue #505 option to disable error logs in batch jobs

Author: ElienVandermaesenVITO

openeo/rest/datacube.py

@@ -2421,6 +2421,7 @@ def execute_batch(
         job_options: Optional[dict] = None,
         validate: Optional[bool] = None,
         auto_add_save_result: bool = True,
+        enable_print=True,
         # TODO: deprecate `format_options` as keyword arguments
         **format_options,
     ) -> BatchJob:
@@ -2438,6 +2439,7 @@ def execute_batch(
         :param validate: Optional toggle to enable/prevent validation of the process graphs before execution
             (overruling the connection's ``auto_validate`` setting).
         :param auto_add_save_result: Automatically add a ``save_result`` node to the process graph if there is none yet.
+        :param enable_print: whether to print error logs
 
         .. versionchanged:: 0.32.0
             Added ``auto_add_save_result`` option
@@ -2473,7 +2475,8 @@ def execute_batch(
         )
         return job.run_synchronous(
             outputfile=outputfile,
-            print=print, max_poll_interval=max_poll_interval, connection_retry_interval=connection_retry_interval
+            print=print, max_poll_interval=max_poll_interval, connection_retry_interval=connection_retry_interval,
+            enable_print=enable_print
         )
 
     def create_job(

openeo/rest/job.py

@@ -236,19 +236,30 @@ def logs(
 
     def run_synchronous(
             self, outputfile: Union[str, Path, None] = None,
-            print=print, max_poll_interval=60, connection_retry_interval=30
+            print=print, max_poll_interval=60, connection_retry_interval=30, enable_print=True
     ) -> BatchJob:
-        """Start the job, wait for it to finish and download result"""
+        """
+        Start the job, wait for it to finish and download result
+
+        :param outputfile: The path of a file to which a result can be written
+        :param print: print/logging function to show progress/status
+        :param max_poll_interval: maximum number of seconds to sleep between status polls
+        :param connection_retry_interval: how long to wait when status poll failed due to connection issue
+        :param enable_print: whether to print error logs
+        :return:
+        """
         self.start_and_wait(
-            print=print, max_poll_interval=max_poll_interval, connection_retry_interval=connection_retry_interval
+            print=print, max_poll_interval=max_poll_interval, connection_retry_interval=connection_retry_interval,
+            enable_print=enable_print
         )
         # TODO #135 support multi file result sets too?
         if outputfile is not None:
             self.download_result(outputfile)
         return self
 
     def start_and_wait(
-            self, print=print, max_poll_interval: int = 60, connection_retry_interval: int = 30, soft_error_max=10
+            self, print=print, max_poll_interval: int = 60, connection_retry_interval: int = 30, soft_error_max=10,
+            enable_print=True
     ) -> BatchJob:
         """
         Start the batch job, poll its status and wait till it finishes (or fails)
@@ -257,6 +268,7 @@ def start_and_wait(
         :param max_poll_interval: maximum number of seconds to sleep between status polls
         :param connection_retry_interval: how long to wait when status poll failed due to connection issue
         :param soft_error_max: maximum number of soft errors (e.g. temporary connection glitches) to allow
+        :param enable_print: whether to print error logs
         :return:
         """
         # TODO rename `connection_retry_interval` to something more generic?
@@ -314,13 +326,13 @@ def soft_error(message: str):
             poll_interval = min(1.25 * poll_interval, max_poll_interval)
 
         if status != "finished":
-            # TODO: allow to disable this printing logs (e.g. in non-interactive contexts)?
             # TODO: render logs jupyter-aware in a notebook context?
-            print(f"Your batch job {self.job_id!r} failed. Error logs:")
-            print(self.logs(level=logging.ERROR))
-            print(
-                f"Full logs can be inspected in an openEO (web) editor or with `connection.job({self.job_id!r}).logs()`."
-            )
+            if enable_print:
+                print(f"Your batch job {self.job_id!r} failed. Error logs:")
+                print(self.logs(level=logging.ERROR))
+                print(
+                    f"Full logs can be inspected in an openEO (web) editor or with `connection.job({self.job_id!r}).logs()`."
+                )
             raise JobFailedException(
                 f"Batch job {self.job_id!r} didn't finish successfully. Status: {status} (after {elapsed()}).",
                 job=self,

tests/rest/test_job.py

@@ -223,6 +223,57 @@ def test_execute_batch_with_soft_errors(con100, requests_mock, tmpdir, error_res
     assert path.read() == "tiffdata"
     assert job.logs() == []
 
+def test_execute_batch_with_error_with_error_logs_disabled(con100, requests_mock, tmpdir):
+    requests_mock.get(API_URL + "/file_formats", json={"output": {"GTiff": {"gis_data_types": ["raster"]}}})
+    requests_mock.get(API_URL + "/collections/SENTINEL2", json={"foo": "bar"})
+    requests_mock.post(API_URL + "/jobs", status_code=201, headers={"OpenEO-Identifier": "f00ba5"})
+    requests_mock.post(API_URL + "/jobs/f00ba5/results", status_code=202)
+    requests_mock.get(
+        API_URL + "/jobs/f00ba5",
+        [
+            {"json": {"status": "submitted"}},
+            {"json": {"status": "queued"}},
+            {"json": {"status": "running", "progress": 15}},
+            {"json": {"status": "running", "progress": 80}},
+            {"json": {"status": "error", "progress": 100}},
+        ],
+    )
+    requests_mock.get(
+        API_URL + "/jobs/f00ba5/logs",
+        json={
+            "logs": [
+                {"id": "12", "level": "info", "message": "starting"},
+                {"id": "34", "level": "error", "message": "nope"},
+            ]
+        },
+    )
+
+    path = tmpdir.join("tmp.tiff")
+    log = []
+
+    try:
+        with fake_time():
+            con100.load_collection("SENTINEL2").execute_batch(
+                outputfile=path, out_format="GTIFF",
+                max_poll_interval=.1, print=log.append, enable_print=False
+            )
+        pytest.fail("execute_batch should fail")
+    except JobFailedException as e:
+        assert e.job.status() == "error"
+        assert [(l.level, l.message) for l in e.job.logs()] == [
+            ("info", "starting"),
+            ("error", "nope"),
+        ]
+
+    assert log == [
+        "0:00:01 Job 'f00ba5': send 'start'",
+        "0:00:02 Job 'f00ba5': submitted (progress N/A)",
+        "0:00:04 Job 'f00ba5': queued (progress N/A)",
+        "0:00:07 Job 'f00ba5': running (progress 15%)",
+        "0:00:12 Job 'f00ba5': running (progress 80%)",
+        "0:00:20 Job 'f00ba5': error (progress 100%)",
+    ]
+
 
 @pytest.mark.parametrize(["error_response", "expected"], [
     (