Path String Sanitation (aws#158)

Add path str sanitization to create_trial (do not check interspersing) so that when customer loads trial from local path that has trailing/leading spaces by mistake - they got trimmed.

Author: NihalHarish

smdebug/trials/trial.py

@@ -145,9 +145,10 @@ def _fetch():
             num_times_before_warning -= 1
             if num_times_before_warning < 0:
                 self.logger.warning(
-                    "Waiting to read collections files generated by the training job."
-                    "If this has been a while, you might want to check that the "
-                    "trial is pointed at the right path."
+                    f"Waiting to read collections files generated by the training job,"
+                    f"from {self.path}. "
+                    f"If this has been a while, you might want to check that the "
+                    f"trial is pointed at the right path."
                 )
             else:
                 self.logger.debug(

smdebug/trials/utils.py

@@ -10,6 +10,7 @@
 
 
 def create_trial(path, name=None, **kwargs):
+    path = path.strip()  # Remove any accidental leading/trailing whitespace input by the user
     if name is None:
         name = os.path.basename(path)
     s3, bucket_name, prefix_name = is_s3(path)

tests/analysis/trials/test_has_passed_step_scenarios.py

@@ -1,51 +1,20 @@
 # Standard Library
-import json
 import os
 import shutil
 import uuid
-from pathlib import Path
 
 # Third Party
 import pytest
 
 # First Party
-from smdebug.core.collection_manager import CollectionManager
 from smdebug.core.config_constants import INCOMPLETE_STEP_WAIT_WINDOW_KEY
-from smdebug.core.locations import IndexFileLocationUtils
 from smdebug.core.modes import ModeKeys
 from smdebug.core.tensor import StepState
 from smdebug.exceptions import NoMoreData, StepUnavailable
 from smdebug.trials import create_trial
 
-
-def dummy_trial_creator(trial_dir, num_workers, job_ended):
-    Path(trial_dir).mkdir(parents=True, exist_ok=True)
-    cm = CollectionManager()
-    for i in range(num_workers):
-        collection_file_name = f"worker_{i}_collections.json"
-        cm.export(trial_dir, collection_file_name)
-    if job_ended:
-        Path(os.path.join(trial_dir, "training_job_end.ts")).touch()
-
-
-def dummy_step_creator(trial_dir, global_step, mode, mode_step, worker_name):
-    static_step_data = (
-        '{"meta": {"mode": "TRAIN", "mode_step": 0, "event_file_name": ""}, '
-        '"tensor_payload": ['
-        '{"tensorname": "gradients/dummy:0", "start_idx": 0, "length": 1}'
-        "]}"
-    )
-
-    step = json.loads(static_step_data)
-    step["meta"]["mode"] = mode
-    step["meta"]["mode_step"] = mode_step
-
-    index_file_location = IndexFileLocationUtils.get_index_key_for_step(
-        trial_dir, global_step, worker_name
-    )
-    Path(os.path.dirname(index_file_location)).mkdir(parents=True, exist_ok=True)
-    with open(index_file_location, "w") as f:
-        json.dump(step, f)
+# Local
+from ..utils import dummy_step_creator, dummy_trial_creator
 
 
 @pytest.mark.slow

tests/analysis/utils.py

@@ -1,5 +1,7 @@
 # Standard Library
+import json
 import os
+from pathlib import Path
 
 # Third Party
 import numpy as np
@@ -8,6 +10,7 @@
 from smdebug.core.access_layer.s3handler import DeleteRequest, S3Handler
 from smdebug.core.collection_manager import CollectionManager
 from smdebug.core.config_constants import DEFAULT_COLLECTIONS_FILE_NAME
+from smdebug.core.locations import IndexFileLocationUtils
 from smdebug.core.writer import FileWriter
 
 
@@ -51,3 +54,33 @@ def check_trial(trial_obj, num_steps, num_tensors):
 
 def delete_s3_prefix(bucket, prefix):
     S3Handler.delete_prefix(delete_request=DeleteRequest(Bucket=bucket, Prefix=prefix))
+
+
+def dummy_trial_creator(trial_dir, num_workers, job_ended):
+    Path(trial_dir).mkdir(parents=True, exist_ok=True)
+    cm = CollectionManager()
+    for i in range(num_workers):
+        collection_file_name = f"worker_{i}_collections.json"
+        cm.export(trial_dir, collection_file_name)
+    if job_ended:
+        Path(os.path.join(trial_dir, "training_job_end.ts")).touch()
+
+
+def dummy_step_creator(trial_dir, global_step, mode, mode_step, worker_name):
+    static_step_data = (
+        '{"meta": {"mode": "TRAIN", "mode_step": 0, "event_file_name": ""}, '
+        '"tensor_payload": ['
+        '{"tensorname": "gradients/dummy:0", "start_idx": 0, "length": 1}'
+        "]}"
+    )
+
+    step = json.loads(static_step_data)
+    step["meta"]["mode"] = mode
+    step["meta"]["mode_step"] = mode_step
+
+    index_file_location = IndexFileLocationUtils.get_index_key_for_step(
+        trial_dir, global_step, worker_name
+    )
+    Path(os.path.dirname(index_file_location)).mkdir(parents=True, exist_ok=True)
+    with open(index_file_location, "w") as f:
+        json.dump(step, f)

tests/core/test_paths.py

@@ -13,6 +13,22 @@
 from smdebug.core.access_layer.utils import training_has_ended
 from smdebug.core.hook_utils import verify_and_get_out_dir
 from smdebug.core.utils import SagemakerSimulator, ScriptSimulator
+from smdebug.trials import create_trial
+
+# Local
+from ..analysis.utils import dummy_trial_creator
+
+
+def test_whitespace_handling_in_path_str():
+    _id = str(uuid.uuid4())
+    path = os.path.join("ts_output/train/", _id)
+    dummy_trial_creator(trial_dir=path, num_workers=1, job_ended=True)
+
+    # Test Leading Whitespace Handling
+    create_trial("   " + path)
+
+    # Test Trailing Whitespace Handling
+    create_trial(path + "  ")
 
 
 def test_outdir_non_sagemaker():