indico_toolkit 6.14.0__tar.gz → 6.14.2__tar.gz

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: indico-toolkit
-Version: 6.14.0
+Version: 6.14.2
 Summary: Classes, functions, and abstractions for Indico IPA
 License: MIT
 Author: Indico Solutions Engineering
@@ -102,10 +102,10 @@ Formatting, linting, type checking, and tests are defined as
 [Poe](https://poethepoet.natn.io/) tasks in `pyproject.toml`.
 
 ```bash
-poetry run poe {format,check,test,all}
+poetry poe {format,check,test,all}
 ```
 
-Code changes or additions should pass `poetry run poe all` before opening a PR.
+Code changes or additions should pass `poetry poe all` before opening a PR.
 
 
 ### Tests
@@ -117,7 +117,7 @@ By default, only required unit tests are executed. Extra unit tests and integrat
 tests are skipped.
 
 ```bash
-poetry run poe {test,all}
+poetry poe {test,all}
 ```
 
 Extra unit tests are skipped when their dependencies are not installed. To execute extra
@@ -125,7 +125,7 @@ unit tests, install one or more extras and run the tests.
 
 ```bash
 poetry install --all-extras
-poetry run poe {test,all}
+poetry poe {test,all}
 ```
 
 Integration tests make API calls to an IPA environment and require a host and API token
@@ -133,7 +133,7 @@ to execute. These tests create datasets, setup workflows, and train models. **Ex
 them to take tens of minutes to run.**
 
 ```bash
-poetry run poe test-integration \
+poetry poe test-integration \
     --host try.indico.io \
     --token indico_api_token.txt
 ```

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/README.md

@@ -68,10 +68,10 @@ Formatting, linting, type checking, and tests are defined as
 [Poe](https://poethepoet.natn.io/) tasks in `pyproject.toml`.
 
 ```bash
-poetry run poe {format,check,test,all}
+poetry poe {format,check,test,all}
 ```
 
-Code changes or additions should pass `poetry run poe all` before opening a PR.
+Code changes or additions should pass `poetry poe all` before opening a PR.
 
 
 ### Tests
@@ -83,7 +83,7 @@ By default, only required unit tests are executed. Extra unit tests and integrat
 tests are skipped.
 
 ```bash
-poetry run poe {test,all}
+poetry poe {test,all}
 ```
 
 Extra unit tests are skipped when their dependencies are not installed. To execute extra
@@ -91,7 +91,7 @@ unit tests, install one or more extras and run the tests.
 
 ```bash
 poetry install --all-extras
-poetry run poe {test,all}
+poetry poe {test,all}
 ```
 
 Integration tests make API calls to an IPA environment and require a host and API token
@@ -99,7 +99,7 @@ to execute. These tests create datasets, setup workflows, and train models. **Ex
 them to take tens of minutes to run.**
 
 ```bash
-poetry run poe test-integration \
+poetry poe test-integration \
     --host try.indico.io \
     --token indico_api_token.txt
 ```

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/__init__.py

@@ -21,4 +21,4 @@ __all__ = (
     "ToolkitStaggeredLoopError",
     "ToolkitStatusError",
 )
-__version__ = "6.14.0"
+__version__ = "6.14.2"

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/client.py

@@ -8,7 +8,7 @@ from .retry import retry
 
 
 @retry(IndicoRequestError, ConnectionError)
-def create_client(
+def create_client(  # type: ignore[no-any-unimported]
     host: str,
     api_token_path: "str | None" = None,
     api_token_string: "str | None" = None,

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/etloutput/__init__.py

@@ -1,7 +1,7 @@
 from typing import TYPE_CHECKING
 
 from ..results import NULL_BOX, NULL_SPAN, Box, Span
-from ..results.utilities import get, has
+from ..results.utils import get, has
 from .cell import Cell, CellType
 from .errors import EtlOutputError, TableCellNotFoundError, TokenNotFoundError
 from .etloutput import EtlOutput

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/etloutput/cell.py

@@ -2,7 +2,7 @@ from dataclasses import dataclass
 from enum import Enum
 
 from ..results import NULL_SPAN, Box, Span
-from ..results.utilities import get
+from ..results.utils import get
 from .range import Range
 
 

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/etloutput/range.py

@@ -1,6 +1,6 @@
 from dataclasses import dataclass
 
-from ..results.utilities import get
+from ..results.utils import get
 
 
 @dataclass(order=True, frozen=True)

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/etloutput/table.py

@@ -2,7 +2,7 @@ from dataclasses import dataclass
 from operator import attrgetter
 
 from ..results import Box
-from ..results.utilities import get
+from ..results.utils import get
 from .cell import Cell
 
 

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/etloutput/token.py

@@ -1,7 +1,7 @@
 from dataclasses import dataclass
 
 from ..results import Box, Span
-from ..results.utilities import get
+from ..results.utils import get
 
 
 @dataclass(frozen=True)

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/indico_wrapper/__init__.py

@@ -1,15 +1,9 @@
-from .dataset import Datasets
-from .doc_extraction import DocExtraction
 from .download import Download
 from .indico_wrapper import IndicoWrapper
-from .reviewer import Reviewer
 from .workflow import Workflow
 
 __all__ = (
-    "Datasets",
-    "DocExtraction",
     "Download",
     "IndicoWrapper",
-    "Reviewer",
     "Workflow",
 )

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/ocr/__init__.py

@@ -1,9 +1,7 @@
-from .customocr_object import CustomOcr
 from .ondoc_object import OnDoc
 from .standard_object import StandardOcr
 
 __all__ = (
-    "CustomOcr",
     "OnDoc",
     "StandardOcr",
 )

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/polling/autoreview.py

@@ -3,8 +3,8 @@ import logging
 from dataclasses import dataclass
 from typing import TYPE_CHECKING
 
-from indico import AsyncIndicoClient, IndicoConfig  # type: ignore[import-untyped]
-from indico.queries import (  # type: ignore[import-untyped]
+from indico import AsyncIndicoClient, IndicoConfig
+from indico.queries import (
     GetSubmission,
     JobStatus,
     RetrieveStorageObject,
@@ -41,7 +41,7 @@ class AutoReviewPoller:
     and submits the review results concurrently.
     """
 
-    def __init__(
+    def __init__(  # type: ignore[no-any-unimported]
         self,
         config: IndicoConfig,
         workflow_id: int,

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/polling/downstream.py

@@ -2,12 +2,12 @@ import asyncio
 import logging
 from typing import TYPE_CHECKING
 
-from indico import AsyncIndicoClient, IndicoConfig  # type: ignore[import-untyped]
-from indico.queries import (  # type: ignore[import-untyped]
+from indico import AsyncIndicoClient, IndicoConfig
+from indico.queries import (
     GetSubmission,
     UpdateSubmission,
 )
-from indico.types import Submission  # type: ignore[import-untyped]
+from indico.types import Submission
 
 from ..retry import retry
 from .queries import SubmissionIdsPendingDownstream
@@ -29,7 +29,7 @@ class DownstreamPoller:
     them concurrently, and marks them as retrieved.
     """
 
-    def __init__(
+    def __init__(  # type: ignore[no-any-unimported]
         self,
         config: IndicoConfig,
         workflow_id: int,

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/polling/queries.py

@@ -1,12 +1,12 @@
 from typing import TYPE_CHECKING
 
-from indico.queries import GraphQLRequest  # type: ignore[import-untyped]
+from indico.queries import GraphQLRequest
 
 if TYPE_CHECKING:
     from typing import Any
 
 
-class SubmissionIdsPendingAutoReview(GraphQLRequest):  # type: ignore[misc]
+class SubmissionIdsPendingAutoReview(GraphQLRequest):  # type: ignore[misc, no-any-unimported]
     QUERY = """
     query SubmissionIdsPendingAutoReview($workflowIds: [Int]) {
         submissions(
@@ -33,7 +33,7 @@ class SubmissionIdsPendingAutoReview(GraphQLRequest):  # type: ignore[misc]
         }
 
 
-class SubmissionIdsPendingDownstream(GraphQLRequest):  # type: ignore[misc]
+class SubmissionIdsPendingDownstream(GraphQLRequest):  # type: ignore[misc, no-any-unimported]
     QUERY = """
     query SubmissionIdsPendingDownstream($workflowIds: [Int]) {
         submissions(

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/__init__.py

@@ -23,7 +23,7 @@ from .predictions import (
 )
 from .result import Result
 from .review import Review, ReviewType
-from .utilities import get
+from .utils import get
 
 if TYPE_CHECKING:
     from collections.abc import Awaitable, Callable
@@ -101,4 +101,4 @@ def _load(result: object) -> Result:
     elif file_version == 3:
         return Result.from_v3_dict(result)
     else:
-        raise ResultError(f"unsupported file version `{file_version!r}`")
+        raise ResultError(f"unsupported file version `{file_version}`")

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/document.py

@@ -1,6 +1,6 @@
 from dataclasses import dataclass
 
-from .utilities import get
+from .utils import get
 
 
 @dataclass(frozen=True, order=True)
@@ -12,13 +12,14 @@ class Document:
     error: str
     traceback: str
 
-    # Auto review changes must reproduce all model sections that were present in the
-    # original result file. This may not be possible from the predictions alone--if a
-    # model had an empty section because it didn't produce predictions or if all of
-    # the predictions were removed to reject them. As such, the models seen when
-    # parsing result files are tracked per-document so that the empty sections can be
-    # reproduced later.
+    # Auto review changes must reproduce all model and component sections that were
+    # present in the original result file. This may not be possible from the
+    # predictions alone--if a model or component had an empty section because it didn't
+    # produce predictions or if all of the predictions for that section were dropped.
+    # As such, the models and components seen when parsing a result file are tracked
+    # per-document so that the empty sections can be reproduced later.
     _model_sections: "frozenset[str]"
+    _component_sections: "frozenset[str]"
 
     @staticmethod
     def from_v1_dict(result: object) -> "Document":
@@ -38,6 +39,7 @@ class Document:
             error="",
             traceback="",
             _model_sections=model_names,
+            _component_sections=frozenset(),
         )
 
     @staticmethod
@@ -46,7 +48,9 @@ class Document:
         Create a `Document` from a v3 document dictionary.
         """
         model_results = get(document, dict, "model_results", "ORIGINAL")
+        component_results = get(document, dict, "component_results", "ORIGINAL")
         model_ids = frozenset(model_results.keys())
+        component_ids = frozenset(component_results.keys())
         etl_output_uri = get(document, str, "etl_output")
 
         return Document(
@@ -57,6 +61,7 @@ class Document:
             error="",
             traceback="",
             _model_sections=model_ids,
+            _component_sections=component_ids,
         )
 
     @staticmethod
@@ -75,4 +80,5 @@ class Document:
             error=error,
             traceback=traceback,
             _model_sections=frozenset(),
+            _component_sections=frozenset(),
         )

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/model.py

@@ -1,7 +1,7 @@
 from dataclasses import dataclass
 from enum import Enum
 
-from .utilities import get, has
+from .utils import get, has
 
 
 class ModelGroupType(Enum):

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/normalization.py

@@ -1,7 +1,8 @@
 import re
+from itertools import chain
 from typing import TYPE_CHECKING
 
-from .utilities import get, has
+from .utils import get, has
 
 if TYPE_CHECKING:
     from collections.abc import Iterator
@@ -88,12 +89,18 @@ def normalize_v3_result(result: "Any") -> None:
     """
     task_type_by_model_group_id = {
         model_group_id: model_group["task_type"]
-        for model_group_id, model_group in result["modelgroup_metadata"].items()
+        for model_group_id, model_group in chain(
+            result["modelgroup_metadata"].items(),
+            result.get("component_metadata", {}).items(),
+        )
     }
     predictions_with_task_type: "Iterator[tuple[Any, str]]" = (
-        (prediction, task_type_by_model_group_id[model_group_id])
+        (prediction, task_type_by_model_group_id.get(model_group_id, ""))
         for submission_result in get(result, list, "submission_results")
-        for review_result in get(submission_result, dict, "model_results").values()
+        for review_result in chain(
+            get(submission_result, dict, "model_results").values(),
+            get(submission_result, dict, "component_results").values(),
+        )
         for model_group_id, model_results in review_result.items()
         for prediction in model_results
     )
@@ -139,6 +146,10 @@ def normalize_v3_result(result: "Any") -> None:
         if task_type == "summarization" and "citations" not in prediction:
             prediction["citations"] = []
 
+    # Prior to 7.2, v3 result files don't include a `component_metadata` section.
+    if not has(result, dict, "component_metadata"):
+        result["component_metadata"] = {}
+
     # Prior to 6.8, v3 result files don't include a `reviews` section.
     if not has(result, dict, "reviews"):
         result["reviews"] = {}

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictionlist.py

@@ -13,7 +13,7 @@ from .predictions import (
     Unbundling,
 )
 from .review import Review, ReviewType
-from .utilities import nfilter
+from .utils import nfilter
 
 if TYPE_CHECKING:
     from collections.abc import Callable, Collection, Container, Iterable
@@ -337,7 +337,7 @@ class PredictionList(List[PredictionType]):
         elif result.version == 3:
             return self.to_v3_changes(result.documents)
         else:
-            raise ValueError(f"unsupported file version `{result.version!r}`")
+            raise ValueError(f"unsupported file version `{result.version}`")
 
     def to_v1_changes(self, document: "Document") -> "dict[str, Any]":
         """
@@ -370,13 +370,8 @@ class PredictionList(List[PredictionType]):
                 continue
 
             model_results: "dict[str, Any]" = {}
-            changes.append(
-                {
-                    "submissionfile_id": document.id,
-                    "model_results": model_results,
-                    "component_results": {},
-                }
-            )
+            component_results: "dict[str, Any]" = {}
+
             predictions_by_model = self.where(
                 document=document,
             ).groupby(
@@ -384,12 +379,30 @@ class PredictionList(List[PredictionType]):
             )
 
             for model, predictions in predictions_by_model.items():
-                model_results[str(model.id)] = [
+                model_id = str(model.id)
+                prediction_dicts = [
                     prediction.to_v3_dict() for prediction in predictions
                 ]
 
+                if model_id in document._model_sections:
+                    model_results[model_id] = prediction_dicts
+                elif model_id in document._component_sections:
+                    component_results[model_id] = prediction_dicts
+
             for model_id in document._model_sections:
                 if model_id not in model_results:
                     model_results[model_id] = []
 
+            for component_id in document._component_sections:
+                if component_id not in component_results:
+                    component_results[component_id] = []
+
+            changes.append(
+                {
+                    "submissionfile_id": document.id,
+                    "model_results": model_results,
+                    "component_results": component_results,
+                }
+            )
+
         return changes

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/__init__.py

@@ -62,7 +62,7 @@ def from_v1_dict(
     elif model.type == FORM_EXTRACTION:
         return FormExtraction.from_v1_dict(document, model, review, prediction)
     else:
-        raise ResultError(f"unsupported v1 model type `{model.type!r}`")
+        raise ResultError(f"unsupported v1 model type {model.type!r}")
 
 
 def from_v3_dict(
@@ -85,4 +85,4 @@ def from_v3_dict(
     elif model.type == UNBUNDLING:
         return Unbundling.from_v3_dict(document, model, review, prediction)
     else:
-        raise ResultError(f"unsupported v3 model type `{model.type!r}`")
+        raise ResultError(f"unsupported v3 model type {model.type!r}")

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/box.py

@@ -1,7 +1,7 @@
 from dataclasses import dataclass
 from typing import TYPE_CHECKING
 
-from ..utilities import get
+from ..utils import get
 
 if TYPE_CHECKING:
     from typing import Final

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/citation.py

@@ -1,7 +1,7 @@
 from dataclasses import dataclass
 from typing import TYPE_CHECKING
 
-from ..utilities import get
+from ..utils import get
 from .span import NULL_SPAN, Span
 
 if TYPE_CHECKING:

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/classification.py

@@ -2,7 +2,7 @@ from dataclasses import dataclass
 from typing import TYPE_CHECKING
 
 from ..review import Review
-from ..utilities import get, omit
+from ..utils import get, omit
 from .prediction import Prediction
 
 if TYPE_CHECKING:

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/documentextraction.py

@@ -2,7 +2,7 @@ from dataclasses import dataclass
 from typing import TYPE_CHECKING
 
 from ..review import Review
-from ..utilities import get, has, omit
+from ..utils import get, has, omit
 from .extraction import Extraction
 from .group import Group
 from .span import NULL_SPAN, Span
@@ -127,8 +127,10 @@ class DocumentExtraction(Extraction):
             "end": self.span.end,
         }
 
-        prediction["normalized"]["formatted"] = self.text
-        prediction["text"] = self.text  # 6.10 sometimes reverts to raw text in review.
+        if self.text != get(prediction, str, "normalized", "formatted"):
+            prediction["normalized"]["formatted"] = self.text
+            prediction["normalized"]["text"] = self.text
+            prediction["text"] = self.text
 
         if self.accepted:
             prediction["accepted"] = True
@@ -149,8 +151,10 @@ class DocumentExtraction(Extraction):
             "spans": [span.to_dict() for span in self.spans],
         }
 
-        prediction["normalized"]["formatted"] = self.text
-        prediction["text"] = self.text  # 6.10 sometimes reverts to raw text in review.
+        if self.text != get(prediction, str, "normalized", "formatted"):
+            prediction["normalized"]["formatted"] = self.text
+            prediction["normalized"]["text"] = self.text
+            prediction["text"] = self.text
 
         if self.accepted:
             prediction["accepted"] = True

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/formextraction.py

@@ -3,7 +3,7 @@ from enum import Enum
 from typing import TYPE_CHECKING
 
 from ..review import Review
-from ..utilities import get, has, omit
+from ..utils import get, has, omit
 from .box import Box
 from .extraction import Extraction
 
@@ -97,18 +97,23 @@ class FormExtraction(Extraction):
         }
 
         if self.type == FormExtractionType.CHECKBOX:
-            prediction["normalized"]["structured"]["checked"] = self.checked
-            prediction["normalized"]["formatted"] = (
-                "Checked" if self.checked else "Unchecked"
-            )
+            prediction["normalized"]["structured"] = {"checked": self.checked}
+            text = "Checked" if self.checked else "Unchecked"
+            prediction["normalized"]["formatted"] = text
+            prediction["normalized"]["text"] = self.text
+            prediction["text"] = self.text
         elif self.type == FormExtractionType.SIGNATURE:
-            prediction["normalized"]["structured"]["signed"] = self.signed
-            prediction["normalized"]["formatted"] = (
-                "Signed" if self.signed else "Unsigned"
-            )
+            prediction["normalized"]["structured"] = {"signed": self.signed}
+            text = "Signed" if self.signed else "Unsigned"
+            prediction["normalized"]["formatted"] = text
+            # Don't overwrite the text of the signature stored in these attributes.
+            # prediction["normalized"]["text"] = self.text
+            # prediction["text"] = self.text
         elif self.type == FormExtractionType.TEXT:
-            prediction["normalized"]["formatted"] = self.text
-            prediction["text"] = self.text  # 6.10 sometimes reverts to text in review.
+            if self.text != get(prediction, str, "normalized", "formatted"):
+                prediction["normalized"]["formatted"] = self.text
+                prediction["normalized"]["text"] = self.text
+                prediction["text"] = self.text
 
         if self.accepted:
             prediction["accepted"] = True

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/group.py

@@ -1,7 +1,7 @@
-from dataclasses import dataclass
+from dataclasses import dataclass, replace
 from typing import TYPE_CHECKING
 
-from ..utilities import get
+from ..utils import get
 
 if TYPE_CHECKING:
     from typing import Any
@@ -13,6 +13,14 @@ class Group:
     name: str
     index: int
 
+    def __next__(self) -> "Group":
+        """
+        Return the `Group` with the next index.
+
+        Supports `group = next(group)`.
+        """
+        return replace(self, index=self.index + 1)
+
     @staticmethod
     def from_dict(group: object) -> "Group":
         return Group(

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/span.py

@@ -1,7 +1,7 @@
 from dataclasses import dataclass
 from typing import TYPE_CHECKING
 
-from ..utilities import get
+from ..utils import get
 
 if TYPE_CHECKING:
     from typing import Any, Final

{indico_toolkit-6.14.0 → indico_toolkit-6.14.2}/indico_toolkit/results/predictions/summarization.py

@@ -2,7 +2,7 @@ from dataclasses import dataclass
 from typing import TYPE_CHECKING
 
 from ..review import Review
-from ..utilities import get, has, omit
+from ..utils import get, has, omit
 from .citation import NULL_CITATION, Citation
 from .extraction import Extraction