Isort

spacy/__init__.py

@@ -17,9 +17,9 @@ from .cli.info import info  # noqa: F401
 from .errors import Errors
 from .glossary import explain  # noqa: F401
 from .language import Language
+from .registrations import REGISTRY_POPULATED, populate_registry
 from .util import logger, registry  # noqa: F401
 from .vocab import Vocab
-from .registrations import populate_registry, REGISTRY_POPULATED
 
 if sys.maxunicode == 65535:
     raise SystemError(Errors.E130)

spacy/pipeline/attributeruler.py

@@ -1,7 +1,7 @@
+import importlib
+import sys
 from pathlib import Path
 from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple, Union
-import sys
-import importlib
 
 import srsly
 

spacy/pipeline/dep_parser.pyx

@@ -1,8 +1,8 @@
 # cython: infer_types=True, binding=True
-from collections import defaultdict
-from typing import Callable, Optional
 import importlib
 import sys
+from collections import defaultdict
+from typing import Callable, Optional
 
 from thinc.api import Config, Model
 

spacy/pipeline/edit_tree_lemmatizer.py

@@ -1,7 +1,7 @@
-from collections import Counter
-from itertools import islice
 import importlib
 import sys
+from collections import Counter
+from itertools import islice
 from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple, cast
 
 import numpy as np

spacy/pipeline/entity_linker.py

@@ -1,9 +1,9 @@
+import importlib
 import random
+import sys
 from itertools import islice
 from pathlib import Path
 from typing import Any, Callable, Dict, Iterable, List, Optional, Union
-import sys
-import importlib
 
 import srsly
 from thinc.api import Config, CosineDistance, Model, Optimizer, set_dropout_rate

spacy/pipeline/entityruler.py

@@ -1,8 +1,8 @@
+import importlib
+import sys
 import warnings
 from collections import defaultdict
 from pathlib import Path
-import importlib
-import sys
 from typing import Any, Callable, Dict, Iterable, List, Optional, Sequence, Tuple, Union
 
 import srsly

spacy/pipeline/factories.py

@@ -1,58 +1,52 @@
-from typing import Dict, Any, Callable, Iterable, List, Optional, Union, Tuple
+from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple, Union
+
 from thinc.api import Model
 from thinc.types import Floats2d, Ragged
-from ..tokens.doc import Doc
+
-from ..tokens.span import Span
+from ..kb import Candidate, KnowledgeBase
-from ..kb import KnowledgeBase, Candidate
-from ..vocab import Vocab
-from ..pipeline.textcat import TextCategorizer
-from ..pipeline.tok2vec import Tok2Vec
-from ..pipeline.spancat import SpanCategorizer, Suggester
-from ..pipeline.textcat_multilabel import MultiLabel_TextCategorizer
-from ..pipeline.entityruler import EntityRuler
-from ..pipeline.span_finder import SpanFinder
-from ..pipeline.ner import EntityRecognizer
-from ..pipeline._parser_internals.transition_system import TransitionSystem
-from ..pipeline.dep_parser import DependencyParser
-from ..pipeline.tagger import Tagger
-from ..pipeline.multitask import MultitaskObjective
-from ..pipeline.senter import SentenceRecognizer
 from ..language import Language
-from ..pipeline.sentencizer import Sentencizer
+from ..pipeline._parser_internals.transition_system import TransitionSystem
+from ..pipeline.attributeruler import AttributeRuler
+from ..pipeline.dep_parser import DEFAULT_PARSER_MODEL, DependencyParser
+from ..pipeline.edit_tree_lemmatizer import (
+    DEFAULT_EDIT_TREE_LEMMATIZER_MODEL,
+    EditTreeLemmatizer,
+)
 
 # Import factory default configurations
-from ..pipeline.entity_linker import DEFAULT_NEL_MODEL
+from ..pipeline.entity_linker import DEFAULT_NEL_MODEL, EntityLinker, EntityLinker_v1
-from ..pipeline.entityruler import DEFAULT_ENT_ID_SEP
+from ..pipeline.entityruler import DEFAULT_ENT_ID_SEP, EntityRuler
-from ..pipeline.tok2vec import DEFAULT_TOK2VEC_MODEL
+from ..pipeline.functions import DocCleaner, TokenSplitter
-from ..pipeline.senter import DEFAULT_SENTER_MODEL
+from ..pipeline.lemmatizer import Lemmatizer
-from ..pipeline.morphologizer import DEFAULT_MORPH_MODEL
+from ..pipeline.morphologizer import DEFAULT_MORPH_MODEL, Morphologizer
+from ..pipeline.multitask import DEFAULT_MT_MODEL, MultitaskObjective
+from ..pipeline.ner import DEFAULT_NER_MODEL, EntityRecognizer
+from ..pipeline.sentencizer import Sentencizer
+from ..pipeline.senter import DEFAULT_SENTER_MODEL, SentenceRecognizer
+from ..pipeline.span_finder import DEFAULT_SPAN_FINDER_MODEL, SpanFinder
+from ..pipeline.span_ruler import DEFAULT_SPANS_KEY as SPAN_RULER_DEFAULT_SPANS_KEY
+from ..pipeline.span_ruler import (
+    SpanRuler,
+    prioritize_existing_ents_filter,
+    prioritize_new_ents_filter,
+)
 from ..pipeline.spancat import (
     DEFAULT_SPANCAT_MODEL,
     DEFAULT_SPANCAT_SINGLELABEL_MODEL,
     DEFAULT_SPANS_KEY,
+    SpanCategorizer,
+    Suggester,
 )
-from ..pipeline.span_ruler import DEFAULT_SPANS_KEY as SPAN_RULER_DEFAULT_SPANS_KEY
+from ..pipeline.tagger import DEFAULT_TAGGER_MODEL, Tagger
-from ..pipeline.edit_tree_lemmatizer import DEFAULT_EDIT_TREE_LEMMATIZER_MODEL
+from ..pipeline.textcat import DEFAULT_SINGLE_TEXTCAT_MODEL, TextCategorizer
-from ..pipeline.textcat_multilabel import DEFAULT_MULTI_TEXTCAT_MODEL
+from ..pipeline.textcat_multilabel import (
-from ..pipeline.span_finder import DEFAULT_SPAN_FINDER_MODEL
+    DEFAULT_MULTI_TEXTCAT_MODEL,
-from ..pipeline.ner import DEFAULT_NER_MODEL
+    MultiLabel_TextCategorizer,
-from ..pipeline.dep_parser import DEFAULT_PARSER_MODEL
-from ..pipeline.tagger import DEFAULT_TAGGER_MODEL
-from ..pipeline.multitask import DEFAULT_MT_MODEL
-from ..pipeline.textcat import DEFAULT_SINGLE_TEXTCAT_MODEL
-from ..pipeline.entity_linker import EntityLinker, EntityLinker_v1
-from ..pipeline.attributeruler import AttributeRuler
-from ..pipeline.lemmatizer import Lemmatizer
-from ..pipeline.functions import TokenSplitter
-from ..pipeline.functions import DocCleaner
-from ..pipeline.span_ruler import (
-    SpanRuler,
-    prioritize_new_ents_filter,
-    prioritize_existing_ents_filter,
 )
-from ..pipeline.edit_tree_lemmatizer import EditTreeLemmatizer
+from ..pipeline.tok2vec import DEFAULT_TOK2VEC_MODEL, Tok2Vec
-from ..pipeline.morphologizer import Morphologizer
+from ..tokens.doc import Doc
-
+from ..tokens.span import Span
+from ..vocab import Vocab
 
 # Global flag to track if factories have been registered
 FACTORIES_REGISTERED = False

spacy/pipeline/functions.py

@@ -1,7 +1,7 @@
+import importlib
+import sys
 import warnings
 from typing import Any, Dict
-import sys
-import importlib
 
 import srsly
 

spacy/pipeline/lemmatizer.py

@@ -1,7 +1,7 @@
-import warnings
-from pathlib import Path
 import importlib
 import sys
+import warnings
+from pathlib import Path
 from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple, Union
 
 from thinc.api import Model

spacy/pipeline/morphologizer.pyx

@@ -1,8 +1,8 @@
 # cython: infer_types=True, binding=True
+import importlib
+import sys
 from itertools import islice
 from typing import Callable, Dict, Optional, Union
-import sys
-import importlib
 
 from thinc.api import Config, Model, SequenceCategoricalCrossentropy
 

spacy/pipeline/multitask.pyx

@@ -1,7 +1,7 @@
 # cython: infer_types=True, binding=True
-from typing import Optional
-import sys
 import importlib
+import sys
+from typing import Optional
 
 import numpy
 from thinc.api import Config, CosineDistance, Model, set_dropout_rate, to_categorical

spacy/pipeline/ner.pyx

@@ -1,8 +1,8 @@
 # cython: infer_types=True, binding=True
-from collections import defaultdict
-from typing import Callable, Optional
 import importlib
 import sys
+from collections import defaultdict
+from typing import Callable, Optional
 
 from thinc.api import Config, Model
 

spacy/pipeline/sentencizer.pyx

@@ -1,7 +1,7 @@
 # cython: infer_types=True, binding=True
-from typing import Callable, List, Optional
 import importlib
 import sys
+from typing import Callable, List, Optional
 
 import srsly
 

spacy/pipeline/senter.pyx

@@ -1,8 +1,8 @@
 # cython: infer_types=True, binding=True
+import importlib
+import sys
 from itertools import islice
 from typing import Callable, Optional
-import sys
-import importlib
 
 from thinc.api import Config, Model, SequenceCategoricalCrossentropy
 

spacy/pipeline/span_finder.py

@@ -1,6 +1,6 @@
-from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple
-import sys
 import importlib
+import sys
+from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple
 
 from thinc.api import Config, Model, Optimizer, set_dropout_rate
 from thinc.types import Floats2d

spacy/pipeline/span_ruler.py

@@ -1,8 +1,8 @@
+import importlib
+import sys
 import warnings
 from functools import partial
 from pathlib import Path
-import importlib
-import sys
 from typing import (
     Any,
     Callable,

spacy/pipeline/spancat.py

@@ -1,7 +1,7 @@
-from dataclasses import dataclass
-from functools import partial
 import importlib
 import sys
+from dataclasses import dataclass
+from functools import partial
 from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple, Union, cast
 
 import numpy

spacy/pipeline/tagger.pyx

@@ -1,8 +1,8 @@
 # cython: infer_types=True, binding=True
-from itertools import islice
-from typing import Callable, Optional
 import importlib
 import sys
+from itertools import islice
+from typing import Callable, Optional
 
 import numpy
 from thinc.api import Config, Model, SequenceCategoricalCrossentropy, set_dropout_rate

spacy/pipeline/textcat.py

@@ -1,11 +1,11 @@
+import importlib
+import sys
 from itertools import islice
 from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple
 
 import numpy
 from thinc.api import Config, Model, Optimizer, get_array_module, set_dropout_rate
 from thinc.types import Floats2d
-import sys
-import importlib
 
 from ..errors import Errors
 from ..language import Language

spacy/pipeline/textcat_multilabel.py

@@ -1,7 +1,7 @@
+import importlib
+import sys
 from itertools import islice
 from typing import Any, Callable, Dict, Iterable, List, Optional
-import sys
-import importlib
 
 from thinc.api import Config, Model
 from thinc.types import Floats2d

spacy/pipeline/tok2vec.py

@@ -1,6 +1,6 @@
-from itertools import islice
 import importlib
 import sys
+from itertools import islice
 from typing import Any, Callable, Dict, Iterable, List, Optional, Sequence
 
 from thinc.api import Config, Model, Optimizer, set_dropout_rate

spacy/registrations.py

@@ -21,26 +21,26 @@ def populate_registry() -> None:
         return
 
     # Import all necessary modules
-    from .util import registry, make_first_longest_spans_filter
-
-    # Import all pipeline components that were using registry decorators
-    from .pipeline.tagger import make_tagger_scorer
-    from .pipeline.ner import make_ner_scorer
-    from .pipeline.lemmatizer import make_lemmatizer_scorer
-    from .pipeline.span_finder import make_span_finder_scorer
-    from .pipeline.spancat import (
-        make_spancat_scorer,
-        build_ngram_suggester,
-        build_ngram_range_suggester,
-        build_preset_spans_suggester,
-    )
     from .pipeline.entityruler import (
         make_entity_ruler_scorer as make_entityruler_scorer,
     )
+    from .pipeline.lemmatizer import make_lemmatizer_scorer
+    from .pipeline.ner import make_ner_scorer
     from .pipeline.sentencizer import senter_score as make_sentencizer_scorer
     from .pipeline.senter import make_senter_scorer
+    from .pipeline.span_finder import make_span_finder_scorer
+    from .pipeline.spancat import (
+        build_ngram_range_suggester,
+        build_ngram_suggester,
+        build_preset_spans_suggester,
+        make_spancat_scorer,
+    )
+
+    # Import all pipeline components that were using registry decorators
+    from .pipeline.tagger import make_tagger_scorer
     from .pipeline.textcat import make_textcat_scorer
     from .pipeline.textcat_multilabel import make_textcat_multilabel_scorer
+    from .util import make_first_longest_spans_filter, registry
 
     # Register miscellaneous components
     registry.misc("spacy.first_longest_spans_filter.v1")(
@@ -55,14 +55,14 @@ def populate_registry() -> None:
 
     # Import ML components that use registry
     from .ml.models.tok2vec import (
-        tok2vec_listener_v1,
+        BiLSTMEncoder,
-        build_hash_embed_cnn_tok2vec,
-        build_Tok2Vec_model,
-        MultiHashEmbed,
         CharacterEmbed,
         MaxoutWindowEncoder,
         MishWindowEncoder,
-        BiLSTMEncoder,
+        MultiHashEmbed,
+        build_hash_embed_cnn_tok2vec,
+        build_Tok2Vec_model,
+        tok2vec_listener_v1,
     )
 
     # Register scorers

spacy/tests/test_factory_imports.py

@@ -1,9 +1,10 @@
 # coding: utf-8
 """Test factory import compatibility from original and new locations."""
 
-import pytest
 import importlib
 
+import pytest
+
 
 @pytest.mark.parametrize(
     "factory_name,original_module,compat_module",

spacy/tests/test_factory_registrations.py

@@ -1,7 +1,9 @@
-import json
 import inspect
-import pytest
+import json
 from pathlib import Path
+
+import pytest
+
 from spacy.language import Language
 from spacy.util import registry
 
@@ -10,6 +12,7 @@ REFERENCE_FILE = Path(__file__).parent / "factory_registrations.json"
 
 # Monkey patch the util.is_same_func to handle Cython functions
 import inspect
+
 from spacy import util
 
 original_is_same_func = util.is_same_func

spacy/tests/test_registry_population.py

@@ -1,7 +1,9 @@
 import json
 import os
-import pytest
 from pathlib import Path
+
+import pytest
+
 from spacy.util import registry
 
 # Path to the reference registry contents, relative to this file

spacy/util.py

@@ -135,7 +135,7 @@ class registry(thinc.registry):
     @classmethod
     def ensure_populated(cls) -> None:
         """Ensure the registry is populated with all necessary components."""
-        from .registrations import populate_registry, REGISTRY_POPULATED
+        from .registrations import REGISTRY_POPULATED, populate_registry
 
         if not REGISTRY_POPULATED:
             populate_registry()