opensource/galaxy
3
0
Fork 0
mirror of https://github.com/galaxyproject/galaxy.git synced 2026-09-24 16:30:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

Improve error message for __EXTRACT_DATASETS__ tool

Browse Source 
Provides a reasonable error message to users, instead of the internal
error https://sentry.galaxyproject.org/share/issue/a197159192b64b9db065a762ee5dbbfc/:
```
KeyError: 'Dataset collection has no element_identifier with key 2.'
  File "galaxy/tools/__init__.py", line 1972, in handle_single_execution
    rval = self.execute(
  File "galaxy/tools/__init__.py", line 2069, in execute
    return self.tool_action.execute(
  File "galaxy/tools/actions/model_operations.py", line 88, in execute
    self._produce_outputs(
  File "galaxy/tools/actions/model_operations.py", line 119, in _produce_outputs
    tool.produce_outputs(
  File "galaxy/tools/__init__.py", line 3351, in produce_outputs
    extracted_element = collection[incoming["which"]["identifier"]]
  File "galaxy/model/__init__.py", line 6434, in __getitem__
    raise KeyError(error_message)
```
Also makes the `identifier` parameter explicitly required (it is now
inferred to be optional because text parameters are by default optional
if no validator raises an exception upon validating an empty string).

That prevents
```
KeyError: 'Dataset collection has no element_identifier with key None.'
  File "galaxy/tools/__init__.py", line 1972, in handle_single_execution
    rval = self.execute(
  File "galaxy/tools/__init__.py", line 2069, in execute
    return self.tool_action.execute(
  File "galaxy/tools/actions/model_operations.py", line 88, in execute
    self._produce_outputs(
  File "galaxy/tools/actions/model_operations.py", line 119, in _produce_outputs
    tool.produce_outputs(
  File "galaxy/tools/__init__.py", line 3351, in produce_outputs
    extracted_element = collection[incoming["which"]["identifier"]]
  File "galaxy/model/__init__.py", line 6434, in __getitem__
    raise KeyError(error_message)
```
This commit is contained in:
mvdbeek
2024-05-02 12:40:52 +02:00
parent 8f4968f02a
commit acec4f6de3
3 changed files with 32 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lib/galaxy/tools/__init__.py
+8 -2
View File
@@ -3356,9 +3356,15 @@ class ExtractDatasetCollectionTool(DatabaseOperationTool):
if how == "first":
extracted_element = collection.first_dataset_element
elif how == "by_identifier":
extracted_element = collection[incoming["which"]["identifier"]]
try:
extracted_element = collection[incoming["which"]["identifier"]]
except KeyError as e:
raise exceptions.MessageException(e.args[0])
elif how == "by_index":
extracted_element = collection[int(incoming["which"]["index"])]
try:
extracted_element = collection[int(incoming["which"]["index"])]
except KeyError as e:
raise exceptions.MessageException(e.args[0])
else:
raise exceptions.MessageException("Invalid tool parameters.")
extracted = extracted_element.element_object
lib/galaxy/tools/extract_dataset.xml
+3 -3
View File
@@ -19,7 +19,7 @@
</param>
<when value="first" />
<when value="by_identifier">
<param name="identifier" label="Element identifier:" type="text">
<param name="identifier" label="Element identifier:" type="text" optional="false">
<sanitizer invalid_char="">
<valid initial="string.ascii_letters,string.digits">
<add value="_" />
@@ -52,9 +52,9 @@ Description
The tool allow extracting datasets based on position (**The first dataset** and **Select by index** options) or name (**Select by element identifier** option). This tool effectively collapses the inner-most collection into a dataset. For nested collections (e.g a list of lists of lists: outer:middle:inner, extracting the inner dataset element) a new list is created where the selected element takes the position of the inner-most collection (so outer:middle, where middle is not a collection but the inner dataset element).
.. class:: warningmark
.. class:: warningmark
**Note**: Dataset index (numbering) begins with 0 (zero).
**Note**: Dataset index (numbering) begins with 0 (zero).
.. class:: infomark
lib/galaxy_test/api/test_tools.py
+21 -2
View File
@@ -701,7 +701,7 @@ class TestToolsApi(ApiTestCase, TestsTools):
hdca1_id = self.dataset_collection_populator.create_list_in_history(
history_id, contents=["a\nb\nc\nd", "e\nf\ng\nh"], wait=True
).json()["outputs"][0]["id"]
self.dataset_populator.run_tool(
run_response = self.dataset_populator.run_tool(
tool_id="cat_data_and_sleep",
inputs={
"sleep_time": 15,
@@ -709,15 +709,34 @@ class TestToolsApi(ApiTestCase, TestsTools):
},
history_id=history_id,
)
output_hdca_id = run_response["implicit_collections"][0]["id"]
run_response = self.dataset_populator.run_tool(
tool_id="__EXTRACT_DATASET__",
inputs={
"data_collection": {"src": "hdca", "id": hdca1_id},
"data_collection": {"src": "hdca", "id": output_hdca_id},
},
history_id=history_id,
)
assert run_response["outputs"][0]["state"] != "ok"
@skip_without_tool("__EXTRACT_DATASET__")
def test_extract_dataset_invalid_element_identifier(self):
with self.dataset_populator.test_history(require_new=False) as history_id:
hdca1_id = self.dataset_collection_populator.create_list_in_history(
history_id, contents=["a\nb\nc\nd", "e\nf\ng\nh"], wait=True
).json()["outputs"][0]["id"]
run_response = self.dataset_populator.run_tool_raw(
tool_id="__EXTRACT_DATASET__",
inputs={
"data_collection": {"src": "hdca", "id": hdca1_id},
"which": {"which_dataset": "by_index", "index": 100},
},
history_id=history_id,
input_format="21.01",
)
assert run_response.status_code == 400
assert run_response.json()["err_msg"] == "Dataset collection has no element_index with key 100."
@skip_without_tool("__FILTER_FAILED_DATASETS__")
def test_filter_failed_list(self):
with self.dataset_populator.test_history(require_new=False) as history_id: