From 18a1b556580ae71c93f40b42050782d63e9d4883 Mon Sep 17 00:00:00 2001
From: EvaBardou <bardou@teklia.com>
Date: Thu, 5 Oct 2023 11:34:42 +0200
Subject: [PATCH] Blast Reporter class and usage
---
arkindex_worker/reporting.py | 236 --------------
arkindex_worker/worker/__init__.py | 16 +-
arkindex_worker/worker/classification.py | 3 -
arkindex_worker/worker/element.py | 4 -
arkindex_worker/worker/entity.py | 31 +-
arkindex_worker/worker/metadata.py | 4 -
arkindex_worker/worker/transcription.py | 7 -
docs/contents/implem/configure.md | 17 -
docs/contents/implem/index.md | 1 -
docs/ref/reporting.md | 3 -
mkdocs.yml | 1 -
.../test_classifications.py | 13 -
tests/test_elements_worker/test_entities.py | 105 +------
tests/test_reporting.py | 296 ------------------
14 files changed, 9 insertions(+), 728 deletions(-)
delete mode 100644 arkindex_worker/reporting.py
delete mode 100644 docs/ref/reporting.md
delete mode 100644 tests/test_reporting.py
diff --git a/arkindex_worker/reporting.py b/arkindex_worker/reporting.py
deleted file mode 100644
index b9ee03ef..00000000
--- a/arkindex_worker/reporting.py
+++ /dev/null
@@ -1,236 +0,0 @@
-# -*- coding: utf-8 -*-
-"""
-Generator for the ``ml_report.json`` file, to report created worker results and exceptions.
-"""
-
-import json
-import traceback
-from collections import Counter
-from datetime import datetime
-from pathlib import Path
-from typing import Dict, List, Optional, Union
-from uuid import UUID
-
-from apistar.exceptions import ErrorResponse
-
-from arkindex_worker import logger
-from arkindex_worker.models import Transcription
-
-
-class Reporter(object):
- """
- Helper to generate an ``ml_report.json`` artifact.
- """
-
- def __init__(
- self,
- name: Optional[str] = "Unknown worker",
- slug: Optional[str] = "unknown-slug",
- version: Optional[str] = None,
- **kwargs,
- ):
- self.report_data = {
- "name": name,
- "slug": slug,
- "version": version,
- "started": datetime.utcnow().isoformat(),
- "elements": {},
- }
- logger.info(f"Starting ML report for {name}")
-
- def __repr__(self):
- return "{}({})".format(self.__class__.__name__, self.report_data["slug"])
-
- def _get_element(self, element_id):
- return self.report_data["elements"].setdefault(
- str(element_id),
- {
- "started": datetime.utcnow().isoformat(),
- # Created element counts, by type slug
- "elements": {},
- # Created transcriptions count
- "transcriptions": 0,
- # Created classification counts, by class
- "classifications": {},
- # Created entities ({"id": "", "type": "", "name": ""}) from this element
- "entities": [],
- # Created transcription entities ({"transcription_id": "", "entity_id": ""}) from this element
- "transcription_entities": [],
- # Created metadata ({"id": "", "type": "", "name": ""}) from this element
- "metadata": [],
- "errors": [],
- },
- )
-
- def process(self, element_id: Union[str, UUID]):
- """
- Report that a specific element ID is being processed.
-
- :param element_id: ID of the element being processed.
- """
- # Just call the element initializer
- self._get_element(element_id)
-
- def add_element(self, parent_id: Union[str, UUID], type: str, type_count: int = 1):
- """
- Report creating an element as a child of another.
-
- :param parent_id: ID of the parent element.
- :param type: Slug of the type of the child element.
- :param type_count: How many elements of this type were created.
- """
- elements = self._get_element(parent_id)["elements"]
- elements.setdefault(type, 0)
- elements[type] += type_count
-
- def add_classification(self, element_id: Union[str, UUID], class_name: str):
- """
- Report creating a classification on an element.
-
- :param element_id: ID of the element.
- :param class_name: Name of the ML class of the new classification.
- """
- classifications = self._get_element(element_id)["classifications"]
- classifications.setdefault(class_name, 0)
- classifications[class_name] += 1
-
- def add_classifications(
- self, element_id: Union[str, UUID], classifications: List[Dict[str, str]]
- ):
- """
- Report creating one or more classifications at once on an element.
-
- :param element_id: ID of the element.
- :param classifications: List of classifications.
- Each classification is represented as a ``dict`` with a ``class_name`` key
- holding the name of the ML class being used.
- """
- assert isinstance(
- classifications, list
- ), "A list is required for classifications"
- element = self._get_element(element_id)
- # Retrieve the previous existing classification counts, if any
- counter = Counter(**element["classifications"])
- # Add the new ones
- counter.update(
- [classification["class_name"] for classification in classifications]
- )
- element["classifications"] = dict(counter)
-
- def add_transcription(self, element_id: Union[str, UUID], count=1):
- """
- Report creating a transcription on an element.
-
- :param element_id: ID of the element.
- :param count: Number of transcriptions created at once
- """
- self._get_element(element_id)["transcriptions"] += count
-
- def add_entity(
- self,
- element_id: Union[str, UUID],
- entity_id: Union[str, UUID],
- type: str,
- name: str,
- ):
- """
- Report creating an entity on an element.
-
- :param element_id: ID of the element.
- :param entity_id: ID of the new entity.
- :param type: Type of the entity.
- :param name: Name of the entity.
- """
- entities = self._get_element(element_id)["entities"]
- entities.append({"id": entity_id, "type": type, "name": name})
-
- def add_transcription_entity(
- self,
- entity_id: Union[str, UUID],
- transcription: Transcription,
- transcription_entity_id: Union[str, UUID],
- ):
- """
- Report creating a transcription entity on an element.
-
- :param entity_id: ID of the entity element.
- :param transcription: Transcription to add the entity on
- :param transcription_entity_id: ID of the transcription entity that is created.
- """
- transcription_entities = self._get_element(transcription.element.id)[
- "transcription_entities"
- ]
- transcription_entities.append(
- {
- "transcription_id": transcription.id,
- "entity_id": entity_id,
- "transcription_entity_id": transcription_entity_id,
- }
- )
-
- def add_entity_link(self, *args, **kwargs):
- """
- Report creating an entity link. Not currently supported.
-
- :raises NotImplementedError:
- """
- raise NotImplementedError
-
- def add_entity_role(self, *args, **kwargs):
- """
- Report creating an entity role. Not currently supported.
-
- :raises NotImplementedError:
- """
- raise NotImplementedError
-
- def add_metadata(
- self,
- element_id: Union[str, UUID],
- metadata_id: Union[str, UUID],
- type: str,
- name: str,
- ):
- """
- Report creating a metadata from an element.
-
- :param element_id: ID of the element.
- :param metadata_id: ID of the new metadata.
- :param type: Type of the metadata.
- :param name: Name of the metadata.
- """
- metadata = self._get_element(element_id)["metadata"]
- metadata.append({"id": metadata_id, "type": type, "name": name})
-
- def error(self, element_id: Union[str, UUID], exception: Exception):
- """
- Report that a Python exception occurred when processing an element.
-
- :param element_id: ID of the element.
- :param exception: A Python exception.
- """
- error_data = {
- "class": exception.__class__.__name__,
- "message": str(exception),
- }
- if exception.__traceback__ is not None:
- error_data["traceback"] = "\n".join(
- traceback.format_tb(exception.__traceback__)
- )
-
- if isinstance(exception, ErrorResponse):
- error_data["message"] = exception.title
- error_data["status_code"] = exception.status_code
- error_data["content"] = exception.content
-
- self._get_element(element_id)["errors"].append(error_data)
-
- def save(self, path: Union[str, Path]):
- """
- Save the ML report to the specified path.
-
- :param path: Path to save the ML report to.
- """
- logger.info(f"Saving ML report to {path}")
- with open(path, "w") as f:
- json.dump(self.report_data, f)
diff --git a/arkindex_worker/worker/__init__.py b/arkindex_worker/worker/__init__.py
index 7ff2f70e..e2a6aa8e 100644
--- a/arkindex_worker/worker/__init__.py
+++ b/arkindex_worker/worker/__init__.py
@@ -15,7 +15,6 @@ from apistar.exceptions import ErrorResponse
from arkindex_worker import logger
from arkindex_worker.cache import CachedElement
from arkindex_worker.models import Element
-from arkindex_worker.reporting import Reporter
from arkindex_worker.worker.base import BaseWorker
from arkindex_worker.worker.classification import ClassificationMixin
from arkindex_worker.worker.element import ElementMixin
@@ -156,17 +155,12 @@ class ElementsWorker(
super().configure()
super().configure_cache()
- # Add report concerning elements
- self.report = Reporter(
- **self.worker_details, version=getattr(self, "worker_version_id", None)
- )
-
def run(self):
"""
Implements an Arkindex worker that goes through each element returned by
- [list_elements][arkindex_worker.worker.ElementsWorker.list_elements]. It calls [process_element][arkindex_worker.worker.ElementsWorker.process_element], catching exceptions
- and reporting them using the [Reporter][arkindex_worker.reporting.Reporter], and handles saving the report
- once the process is complete as well as WorkerActivity updates when enabled.
+ [list_elements][arkindex_worker.worker.ElementsWorker.list_elements].
+ It calls [process_element][arkindex_worker.worker.ElementsWorker.process_element],
+ catching exceptions, and handles saving WorkerActivity updates when enabled.
"""
self.configure()
@@ -232,10 +226,6 @@ class ElementsWorker(
self.update_activity(element.id, ActivityState.Error)
except Exception:
pass
- self.report.error(element_id, e)
-
- # Save report as local artifact
- self.report.save(self.work_dir / "ml_report.json")
if failed:
logger.error(
diff --git a/arkindex_worker/worker/classification.py b/arkindex_worker/worker/classification.py
index 98cc6eae..e28cefe7 100644
--- a/arkindex_worker/worker/classification.py
+++ b/arkindex_worker/worker/classification.py
@@ -176,8 +176,6 @@ class ClassificationMixin(object):
# Propagate any other API error
raise
- self.report.add_classification(element.id, ml_class)
-
return created
def create_classifications(
@@ -248,7 +246,6 @@ class ClassificationMixin(object):
for created_cl in created_cls:
created_cl["class_name"] = self.retrieve_ml_class(created_cl["ml_class"])
- self.report.add_classification(element.id, created_cl["class_name"])
if self.use_cache:
# Store classifications in local cache
diff --git a/arkindex_worker/worker/element.py b/arkindex_worker/worker/element.py
index c9ed5697..2dee2fed 100644
--- a/arkindex_worker/worker/element.py
+++ b/arkindex_worker/worker/element.py
@@ -142,7 +142,6 @@ class ElementMixin(object):
"confidence": confidence,
},
)
- self.report.add_element(element.id, type)
return sub_element["id"] if slim_output else sub_element
@@ -238,9 +237,6 @@ class ElementMixin(object):
},
)
- for element in elements:
- self.report.add_element(parent.id, element["type"])
-
if self.use_cache:
# Create the image as needed and handle both an Element and a CachedElement
if isinstance(parent, CachedElement):
diff --git a/arkindex_worker/worker/entity.py b/arkindex_worker/worker/entity.py
index 32416263..ccdf2267 100644
--- a/arkindex_worker/worker/entity.py
+++ b/arkindex_worker/worker/entity.py
@@ -8,7 +8,7 @@ from typing import Dict, List, Optional, TypedDict, Union
from peewee import IntegrityError
from arkindex_worker import logger
-from arkindex_worker.cache import CachedElement, CachedEntity, CachedTranscriptionEntity
+from arkindex_worker.cache import CachedEntity, CachedTranscriptionEntity
from arkindex_worker.models import Element, Transcription
Entity = TypedDict(
@@ -68,7 +68,6 @@ class EntityMixin(object):
def create_entity(
self,
- element: Union[Element, CachedElement],
name: str,
type: str,
metas=dict(),
@@ -78,14 +77,9 @@ class EntityMixin(object):
Create an entity on the given corpus.
If cache support is enabled, a [CachedEntity][arkindex_worker.cache.CachedEntity] will also be created.
- :param element: An element on which the entity will be reported with the [Reporter][arkindex_worker.reporting.Reporter].
- This does not have any effect on the entity itself.
:param name: Name of the entity.
:param type: Type of the entity.
"""
- assert element and isinstance(
- element, (Element, CachedElement)
- ), "element shouldn't be null and should be an Element or CachedElement"
assert name and isinstance(
name, str
), "name shouldn't be null and should be of type str"
@@ -119,7 +113,6 @@ class EntityMixin(object):
"worker_run_id": self.worker_run_id,
},
)
- self.report.add_entity(element.id, entity["id"], entity_type_id, name)
if self.use_cache:
# Store entity in local cache
@@ -196,7 +189,6 @@ class EntityMixin(object):
id=transcription.id,
body=body,
)
- self.report.add_transcription_entity(entity, transcription, transcription_ent)
if self.use_cache:
# Store transcription entity in local cache
@@ -247,11 +239,6 @@ class EntityMixin(object):
transcription, Transcription
), "transcription shouldn't be null and should be of type Transcription"
- # Needed for MLreport
- assert (
- hasattr(transcription, "element") and transcription.element
- ), f"No element linked to {transcription}"
-
assert entities and isinstance(
entities, list
), "entities shouldn't be null and should be of type list"
@@ -301,22 +288,6 @@ class EntityMixin(object):
},
)
- for entity, created_objects in zip(entities, created_ids["entities"]):
- # Report entity creation
- self.report.add_entity(
- transcription.element.id,
- created_objects["entity_id"],
- entity.get("type_id"),
- entity.get("name"),
- )
-
- # Report transcription entity creation
- self.report.add_transcription_entity(
- created_objects["entity_id"],
- transcription,
- created_objects["transcription_entity_id"],
- )
-
return created_ids["entities"]
def list_transcription_entities(
diff --git a/arkindex_worker/worker/metadata.py b/arkindex_worker/worker/metadata.py
index fca1e92b..c6bae33c 100644
--- a/arkindex_worker/worker/metadata.py
+++ b/arkindex_worker/worker/metadata.py
@@ -105,7 +105,6 @@ class MetaDataMixin(object):
"worker_run_id": self.worker_run_id,
},
)
- self.report.add_metadata(element.id, metadata["id"], type.value, name)
return metadata["id"]
@@ -182,9 +181,6 @@ class MetaDataMixin(object):
},
)["metadata_list"]
- for meta in created_metadatas:
- self.report.add_metadata(element.id, meta["id"], meta["type"], meta["name"])
-
return created_metadatas
def list_element_metadata(
diff --git a/arkindex_worker/worker/transcription.py b/arkindex_worker/worker/transcription.py
index 5ddab34c..7ce96c68 100644
--- a/arkindex_worker/worker/transcription.py
+++ b/arkindex_worker/worker/transcription.py
@@ -88,8 +88,6 @@ class TranscriptionMixin(object):
},
)
- self.report.add_transcription(element.id)
-
if self.use_cache:
# Store transcription in local cache
try:
@@ -181,9 +179,6 @@ class TranscriptionMixin(object):
},
)["transcriptions"]
- for created_tr in created_trs:
- self.report.add_transcription(created_tr["element_id"])
-
if self.use_cache:
# Store transcriptions in local cache
try:
@@ -308,8 +303,6 @@ class TranscriptionMixin(object):
logger.debug(
f"A sub_element of {element.id} with type {sub_element_type} was created during transcriptions bulk creation"
)
- self.report.add_element(element.id, sub_element_type)
- self.report.add_transcription(annotation["element_id"])
if self.use_cache:
# Store transcriptions and their associated element (if created) in local cache
diff --git a/docs/contents/implem/configure.md b/docs/contents/implem/configure.md
index 59865d18..4901dbe1 100644
--- a/docs/contents/implem/configure.md
+++ b/docs/contents/implem/configure.md
@@ -83,20 +83,6 @@ The multiple configuration sources from the Arkindex-mode are merged into a uniq
One information cannot be retrieved directly from the configuration file and is required in some cases: the ID of the Arkindex corpus which the elements processed belong to. This is retrieved via the `ARKINDEX_CORPUS_ID` environment variable.
-## Worker reporter
-At the end of a worker execution, a report about the publication done by the worker is generated in JSON-format. This lists
-
-- the starting time,
-- the number of elements created, grouped by type,
-- the number of transcription created,
-- the number of classifications created, grouped by class,
-- the number of entities created,
-- the number of entities created on transcriptions,
-- the number of metadatas created,
-- the encountered errors' logs.
-
-This is done by the many helper described in the [reporting module](../../ref/reporting.md). They use the `report` attribute initialized at the configuration stage.
-
## Setting Debug logging level
There are three ways to activate the debug mode:
@@ -132,9 +118,6 @@ Many attributes are set on the worker during at the configuration stage. Here is
`process_information`
: The details about the process parent to this worker execution. Only set in Arkindex mode.
-`reporter`
-: The `Reporter` instance that will generate the `ml_report.json` artifacts which sums up the publication done during this execution and the errors encountered.
-
`secrets`
: A dictionary mapping the secret name to their parsed content.
diff --git a/docs/contents/implem/index.md b/docs/contents/implem/index.md
index df755b4a..03c4c8f1 100644
--- a/docs/contents/implem/index.md
+++ b/docs/contents/implem/index.md
@@ -28,7 +28,6 @@ flowchart LR
subgraph id3[Loop over each element]
element_processing --> element_processing
end
- element_processing -- Save ML report to disk --> reporting
end
init --> run
end
diff --git a/docs/ref/reporting.md b/docs/ref/reporting.md
deleted file mode 100644
index 039553de..00000000
--- a/docs/ref/reporting.md
+++ /dev/null
@@ -1,3 +0,0 @@
-# Reporting
-
-::: arkindex_worker.reporting
diff --git a/mkdocs.yml b/mkdocs.yml
index 59b40f39..994f2c0b 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -88,7 +88,6 @@ nav:
- Models: ref/models.md
- Git & Gitlab support: ref/git.md
- Image utilities: ref/image.md
- - Reporting: ref/reporting.md
- Cache: ref/cache.md
- Utils: ref/utils.md
- Releases: releases.md
diff --git a/tests/test_elements_worker/test_classifications.py b/tests/test_elements_worker/test_classifications.py
index 23db7426..457f3dc7 100644
--- a/tests/test_elements_worker/test_classifications.py
+++ b/tests/test_elements_worker/test_classifications.py
@@ -423,11 +423,6 @@ def test_create_classification(responses, mock_elements_worker):
"high_confidence": True,
}
- # Classification has been created and reported
- assert mock_elements_worker.report.report_data["elements"][elt.id][
- "classifications"
- ] == {"a_class": 1}
-
def test_create_classification_with_cache(responses, mock_elements_worker_with_cache):
mock_elements_worker_with_cache.classes = {"a_class": "0000"}
@@ -470,11 +465,6 @@ def test_create_classification_with_cache(responses, mock_elements_worker_with_c
"high_confidence": True,
}
- # Classification has been created and reported
- assert mock_elements_worker_with_cache.report.report_data["elements"][elt.id][
- "classifications"
- ] == {"a_class": 1}
-
# Check that created classification was properly stored in SQLite cache
assert list(CachedClassification.select()) == [
CachedClassification(
@@ -524,9 +514,6 @@ def test_create_classification_duplicate_worker_run(responses, mock_elements_wor
"high_confidence": True,
}
- # Classification has NOT been created
- assert mock_elements_worker.report.report_data["elements"] == {}
-
def test_create_classifications_wrong_element(mock_elements_worker):
with pytest.raises(AssertionError) as e:
diff --git a/tests/test_elements_worker/test_entities.py b/tests/test_elements_worker/test_entities.py
index b989f3b3..f64b8865 100644
--- a/tests/test_elements_worker/test_entities.py
+++ b/tests/test_elements_worker/test_entities.py
@@ -12,43 +12,16 @@ from arkindex_worker.cache import (
CachedTranscription,
CachedTranscriptionEntity,
)
-from arkindex_worker.models import Element, Transcription
+from arkindex_worker.models import Transcription
from arkindex_worker.worker.entity import MissingEntityType
from arkindex_worker.worker.transcription import TextOrientation
from . import BASE_API_CALLS
-def test_create_entity_wrong_element(mock_elements_worker):
- with pytest.raises(AssertionError) as e:
- mock_elements_worker.create_entity(
- element=None,
- name="Bob Bob",
- type="person",
- )
- assert (
- str(e.value)
- == "element shouldn't be null and should be an Element or CachedElement"
- )
-
- with pytest.raises(AssertionError) as e:
- mock_elements_worker.create_entity(
- element="not element type",
- name="Bob Bob",
- type="person",
- )
- assert (
- str(e.value)
- == "element shouldn't be null and should be an Element or CachedElement"
- )
-
-
def test_create_entity_wrong_name(mock_elements_worker):
- elt = Element({"id": "12341234-1234-1234-1234-123412341234"})
-
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_entity(
- element=elt,
name=None,
type="person",
)
@@ -56,7 +29,6 @@ def test_create_entity_wrong_name(mock_elements_worker):
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_entity(
- element=elt,
name=1234,
type="person",
)
@@ -64,11 +36,8 @@ def test_create_entity_wrong_name(mock_elements_worker):
def test_create_entity_wrong_type(mock_elements_worker):
- elt = Element({"id": "12341234-1234-1234-1234-123412341234"})
-
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_entity(
- element=elt,
name="Bob Bob",
type=None,
)
@@ -76,7 +45,6 @@ def test_create_entity_wrong_type(mock_elements_worker):
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_entity(
- element=elt,
name="Bob Bob",
type=1234,
)
@@ -84,13 +52,10 @@ def test_create_entity_wrong_type(mock_elements_worker):
def test_create_entity_wrong_corpus(monkeypatch, mock_elements_worker):
- elt = Element({"id": "12341234-1234-1234-1234-123412341234"})
-
# Triggering an error on metas param, not giving corpus should work since
# ARKINDEX_CORPUS_ID environment variable is set on mock_elements_worker
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_entity(
- element=elt,
name="Bob Bob",
type="person",
metas="wrong metas",
@@ -99,11 +64,8 @@ def test_create_entity_wrong_corpus(monkeypatch, mock_elements_worker):
def test_create_entity_wrong_metas(mock_elements_worker):
- elt = Element({"id": "12341234-1234-1234-1234-123412341234"})
-
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_entity(
- element=elt,
name="Bob Bob",
type="person",
metas="wrong metas",
@@ -112,11 +74,8 @@ def test_create_entity_wrong_metas(mock_elements_worker):
def test_create_entity_wrong_validated(mock_elements_worker):
- elt = Element({"id": "12341234-1234-1234-1234-123412341234"})
-
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_entity(
- element=elt,
name="Bob Bob",
type="person",
validated="wrong validated",
@@ -127,7 +86,6 @@ def test_create_entity_wrong_validated(mock_elements_worker):
def test_create_entity_api_error(responses, mock_elements_worker):
# Set one entity type
mock_elements_worker.entity_types = {"person": "person-entity-type-id"}
- elt = Element({"id": "12341234-1234-1234-1234-123412341234"})
responses.add(
responses.POST,
"http://testserver/api/v1/entity/",
@@ -136,7 +94,6 @@ def test_create_entity_api_error(responses, mock_elements_worker):
with pytest.raises(ErrorResponse):
mock_elements_worker.create_entity(
- element=elt,
name="Bob Bob",
type="person",
)
@@ -158,7 +115,6 @@ def test_create_entity(responses, mock_elements_worker):
# Set one entity type
mock_elements_worker.entity_types = {"person": "person-entity-type-id"}
- elt = Element({"id": "12341234-1234-1234-1234-123412341234"})
responses.add(
responses.POST,
"http://testserver/api/v1/entity/",
@@ -167,7 +123,6 @@ def test_create_entity(responses, mock_elements_worker):
)
entity_id = mock_elements_worker.create_entity(
- element=elt,
name="Bob Bob",
type="person",
)
@@ -193,8 +148,6 @@ def test_create_entity_missing_type(responses, mock_elements_worker):
"""
Create entity with an unknown type will fail.
"""
- elt = Element({"id": "12341234-1234-1234-1234-123412341234"})
-
# Call to list entity types
responses.add(
responses.GET,
@@ -213,7 +166,6 @@ def test_create_entity_missing_type(responses, mock_elements_worker):
AssertionError, match="Entity type `new-entity` not found in the corpus."
):
mock_elements_worker.create_entity(
- element=elt,
name="Bob Bob",
type="new-entity",
)
@@ -232,7 +184,6 @@ def test_create_entity_missing_type(responses, mock_elements_worker):
def test_create_entity_with_cache(responses, mock_elements_worker_with_cache):
# Set one entity type
mock_elements_worker_with_cache.entity_types = {"person": "person-entity-type-id"}
- elt = CachedElement.create(id="12341234-1234-1234-1234-123412341234", type="thing")
responses.add(
responses.POST,
"http://testserver/api/v1/entity/",
@@ -241,7 +192,6 @@ def test_create_entity_with_cache(responses, mock_elements_worker_with_cache):
)
entity_id = mock_elements_worker_with_cache.create_entity(
- element=elt,
name="Bob Bob",
type="person",
)
@@ -910,22 +860,11 @@ def test_create_transcription_entities_wrong_transcription(
)
-def test_create_transcription_entities_no_transcription_element(mock_elements_worker):
- with pytest.raises(AssertionError) as e:
- mock_elements_worker.create_transcription_entities(
- transcription=Transcription(id="transcription_id"),
- entities=[],
- )
- assert str(e.value) == "No element linked to Transcription (transcription_id)"
-
-
@pytest.mark.parametrize("entities", (None, "not a list of entities", 1))
def test_create_transcription_entities_wrong_entities(mock_elements_worker, entities):
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_transcription_entities(
- transcription=Transcription(
- id="transcription_id", element={"id": "element_id"}
- ),
+ transcription=Transcription(id="transcription_id"),
entities=entities,
)
assert str(e.value) == "entities shouldn't be null and should be of type list"
@@ -934,9 +873,7 @@ def test_create_transcription_entities_wrong_entities(mock_elements_worker, enti
def test_create_transcription_entities_wrong_entities_subtype(mock_elements_worker):
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_transcription_entities(
- transcription=Transcription(
- id="transcription_id", element={"id": "element_id"}
- ),
+ transcription=Transcription(id="transcription_id"),
entities=["not a dict"],
)
assert str(e.value) == "Entity at index 0 in entities: Should be of type dict"
@@ -1030,17 +967,14 @@ def test_create_transcription_entities_wrong_entity(
):
with pytest.raises(AssertionError) as e:
mock_elements_worker.create_transcription_entities(
- transcription=Transcription(
- id="transcription_id", element={"id": "element_id"}
- ),
+ transcription=Transcription(id="transcription_id"),
entities=[entity],
)
assert str(e.value) == error
def test_create_transcription_entities(responses, mock_elements_worker):
- element_id = "element_id"
- transcription = Transcription(id="transcription-id", element={"id": element_id})
+ transcription = Transcription(id="transcription-id")
# Call to Transcription entities creation in bulk
responses.add(
responses.POST,
@@ -1091,35 +1025,6 @@ def test_create_transcription_entities(responses, mock_elements_worker):
assert len(created_objects) == 1
- assert element_id in mock_elements_worker.report.report_data["elements"]
- ml_report = mock_elements_worker.report.report_data["elements"][element_id]
-
- assert "started" in ml_report
- del ml_report["started"]
-
- # Check reporting
- assert ml_report == {
- "elements": {},
- "transcriptions": 0,
- "classifications": {},
- "entities": [
- {
- "id": "entity-id",
- "type": "22222222-2222-2222-2222-222222222222",
- "name": "Teklia",
- }
- ],
- "transcription_entities": [
- {
- "transcription_id": "transcription-id",
- "entity_id": "entity-id",
- "transcription_entity_id": "transc-entity-id",
- }
- ],
- "metadata": [],
- "errors": [],
- }
-
assert len(responses.calls) == len(BASE_API_CALLS) + 1
assert [
(call.request.method, call.request.url) for call in responses.calls
diff --git a/tests/test_reporting.py b/tests/test_reporting.py
deleted file mode 100644
index af214f01..00000000
--- a/tests/test_reporting.py
+++ /dev/null
@@ -1,296 +0,0 @@
-# -*- coding: utf-8 -*-
-import json
-import uuid
-from datetime import datetime
-from tempfile import NamedTemporaryFile
-
-import pytest
-from apistar.exceptions import ErrorResponse
-
-from arkindex_worker.models import Transcription
-from arkindex_worker.reporting import Reporter
-
-
-def test_init():
- version_id = str(uuid.uuid4())
- reporter = Reporter(name="Worker", slug="worker-slug", version=version_id)
- assert "started" in reporter.report_data
- del reporter.report_data["started"]
- assert reporter.report_data == {
- "name": "Worker",
- "slug": "worker-slug",
- "version": version_id,
- "elements": {},
- }
-
-
-def test_process():
- reporter = Reporter("worker")
- reporter.process("myelement")
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- assert "started" in element_data
- del element_data["started"]
- assert element_data == {
- "elements": {},
- "transcriptions": 0,
- "classifications": {},
- "entities": [],
- "transcription_entities": [],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_element():
- reporter = Reporter("worker")
- reporter.add_element("myelement", type="text_line")
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {"text_line": 1},
- "transcriptions": 0,
- "classifications": {},
- "entities": [],
- "transcription_entities": [],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_element_count():
- """
- Report multiple elements with the same parent and type
- """
- reporter = Reporter("worker")
- reporter.add_element("myelement", type="text_line", type_count=42)
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {"text_line": 42},
- "transcriptions": 0,
- "classifications": {},
- "entities": [],
- "transcription_entities": [],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_classification():
- reporter = Reporter("worker")
- reporter.add_classification("myelement", "three")
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {},
- "transcriptions": 0,
- "classifications": {"three": 1},
- "entities": [],
- "transcription_entities": [],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_classifications():
- reporter = Reporter("worker")
- with pytest.raises(AssertionError):
- reporter.add_classifications("myelement", {"not": "a list"})
-
- reporter.add_classifications(
- "myelement", [{"class_name": "three"}, {"class_name": "two"}]
- )
- reporter.add_classifications(
- "myelement",
- [
- {"class_name": "three"},
- {"class_name": "two", "high_confidence": True},
- {"class_name": "three", "confidence": 0.42},
- ],
- )
-
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {},
- "transcriptions": 0,
- "classifications": {"three": 3, "two": 2},
- "entities": [],
- "transcription_entities": [],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_transcription():
- reporter = Reporter("worker")
- reporter.add_transcription("myelement")
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {},
- "transcriptions": 1,
- "classifications": {},
- "entities": [],
- "transcription_entities": [],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_transcription_count():
- """
- Report multiple transcriptions with the same element and type
- """
- reporter = Reporter("worker")
- reporter.add_transcription("myelement", 1337)
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {},
- "transcriptions": 1337,
- "classifications": {},
- "entities": [],
- "transcription_entities": [],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_entity():
- reporter = Reporter("worker")
- reporter.add_entity(
- "myelement",
- "12341234-1234-1234-1234-123412341234",
- "person-entity-type-id",
- "Bob Bob",
- )
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {},
- "transcriptions": 0,
- "classifications": {},
- "entities": [
- {
- "id": "12341234-1234-1234-1234-123412341234",
- "type": "person-entity-type-id",
- "name": "Bob Bob",
- }
- ],
- "transcription_entities": [],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_transcription_entity():
- reporter = Reporter("worker")
- reporter.add_transcription_entity(
- "5678",
- Transcription({"id": "1234-5678", "element": {"id": "myelement"}}),
- "1234",
- )
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {},
- "transcriptions": 0,
- "classifications": {},
- "entities": [],
- "transcription_entities": [
- {
- "transcription_id": "1234-5678",
- "entity_id": "5678",
- "transcription_entity_id": "1234",
- }
- ],
- "metadata": [],
- "errors": [],
- }
-
-
-def test_add_metadata():
- reporter = Reporter("worker")
- reporter.add_metadata(
- "myelement", "12341234-1234-1234-1234-123412341234", "location", "Teklia"
- )
- assert "myelement" in reporter.report_data["elements"]
- element_data = reporter.report_data["elements"]["myelement"]
- del element_data["started"]
- assert element_data == {
- "elements": {},
- "transcriptions": 0,
- "classifications": {},
- "entities": [],
- "transcription_entities": [],
- "metadata": [
- {
- "id": "12341234-1234-1234-1234-123412341234",
- "type": "location",
- "name": "Teklia",
- }
- ],
- "errors": [],
- }
-
-
-def test_error():
- reporter = Reporter("worker")
- reporter.error("myelement", ZeroDivisionError("What have you done"))
- reporter.error(
- "myelement",
- ErrorResponse(
- title="I'm a teapot",
- status_code=418,
- content='{"coffee": "Can\'t touch this"}',
- ),
- )
- assert reporter.report_data["elements"]["myelement"]["errors"] == [
- {"class": "ZeroDivisionError", "message": "What have you done"},
- {
- "class": "ErrorResponse",
- "message": "I'm a teapot",
- "status_code": 418,
- "content": '{"coffee": "Can\'t touch this"}',
- },
- ]
-
-
-def test_reporter_save(mocker):
- datetime_mock = mocker.MagicMock()
- datetime_mock.utcnow.return_value = datetime(2000, 1, 1)
- mocker.patch("arkindex_worker.reporting.datetime", datetime_mock)
- version_id = str(uuid.uuid4())
- reporter = Reporter(name="Worker", slug="worker-slug", version=version_id)
- reporter.add_element("myelement", type="text_line", type_count=4)
- with NamedTemporaryFile() as f:
- reporter.save(f.name)
- exported_data = json.load(f)
- assert exported_data == {
- "name": "Worker",
- "slug": "worker-slug",
- "started": "2000-01-01T00:00:00",
- "version": version_id,
- "elements": {
- "myelement": {
- "classifications": {},
- "elements": {"text_line": 4},
- "entities": [],
- "transcription_entities": [],
- "errors": [],
- "metadata": [],
- "started": "2000-01-01T00:00:00",
- "transcriptions": 0,
- }
- },
- }
--
GitLab