Relieve dependency constraints

.github/workflows/release.yml

@@ -15,17 +15,19 @@ on:
 
 jobs:
   build_wheels:
-    name: Build wheels on ${{ matrix.os }}
+    name: Build wheels on ${{ matrix.os }}/${{ matrix.arch || '*' }}
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [ubuntu-20.04, windows-2019, macos-latest]
+        os: [ubuntu-latest, windows-latest, macos-latest]
 
     steps:
       - uses: actions/checkout@v2
 
       - name: Build wheels
-        uses: pypa/[email protected]
+        uses: pypa/[email protected]
+        env:
+          CIBW_ARCHS_MACOS: "x86_64 arm64"
 
       - uses: actions/upload-artifact@v2
         with:

.github/workflows/test-build.yml

@@ -8,22 +8,25 @@ on:
   workflow_dispatch:
   pull_request:
     branches:
-      - v*.*.*
-      - build-*
+    - v*.*.*
+    - build-*
 
 jobs:
   build_wheels:
     name: Build wheels on ${{ matrix.os }}
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [ubuntu-20.04, windows-2019, macos-10.15]
+        os: [ubuntu-latest, windows-latest, macos-latest]
 
     steps:
       - uses: actions/checkout@v2
 
       - name: Build wheels
-        uses: pypa/[email protected]
+        # 2.4 is too low (can't build for macos, 2.16 is too high (OpenSSL issues)
+        uses: pypa/[email protected]
+        env:
+          CIBW_ARCHS_MACOS: "x86_64 arm64"
 
   build_sdist:
     name: Build source distribution

edsnlp/data/converters.py

@@ -259,7 +259,7 @@ class StandoffDoc2DictConverter:
     def __init__(
         self,
         *,
-        span_getter: SpanGetterArg = None,
+        span_getter: Optional[SpanGetterArg] = None,
         span_attributes: AttributesMappingArg = {},
     ):
         self.span_getter = span_getter

edsnlp/pipes/core/terminology/terminology.py

@@ -114,7 +114,7 @@ def __init__(
         ignore_excluded: bool = False,
         ignore_space_tokens: bool = False,
         term_matcher: Literal["exact", "simstring"] = "exact",
-        term_matcher_config: Dict[str, Any] = None,
+        term_matcher_config: Optional[Dict[str, Any]] = None,
         label,
         span_setter: SpanSetterArg = {"ents": True},
     ):

edsnlp/pipes/misc/dates/dates.py

@@ -169,7 +169,7 @@ def __init__(
         relative: Optional[List[str]] = None,
         duration: Optional[List[str]] = None,
         false_positive: Optional[List[str]] = None,
-        on_ents_only: Union[bool, str, Iterable[str]] = None,
+        on_ents_only: Optional[Union[bool, str, Iterable[str]]] = None,
         span_getter: Optional[SpanGetterArg] = None,
         merge_mode: Literal["intersect", "align"] = "intersect",
         detect_periods: bool = False,

edsnlp/pipes/qualifiers/family/family.py

@@ -117,8 +117,8 @@ def __init__(
         family: Optional[List[str]] = None,
         termination: Optional[List[str]] = None,
         use_sections: bool = True,
-        span_getter: SpanGetterArg = None,
-        on_ents_only: Union[bool, str, List[str], Set[str]] = None,
+        span_getter: Optional[SpanGetterArg] = None,
+        on_ents_only: Optional[Union[bool, str, List[str], Set[str]]] = None,
         explain: bool = False,
     ):
         terms = dict(

edsnlp/pipes/qualifiers/history/history.py

@@ -187,8 +187,8 @@ def __init__(
         history_limit: int = 14,
         closest_dates_only: bool = True,
         exclude_birthdate: bool = True,
-        span_getter: SpanGetterArg = None,
-        on_ents_only: Union[bool, str, List[str], Set[str]] = None,
+        span_getter: Optional[SpanGetterArg] = None,
+        on_ents_only: Optional[Union[bool, str, List[str], Set[str]]] = None,
         explain: bool = False,
     ):
         terms = dict(

edsnlp/pipes/qualifiers/hypothesis/hypothesis.py

@@ -157,8 +157,8 @@ def __init__(
         verbs_hyp: Optional[List[str]] = None,
         termination: Optional[List[str]] = None,
         attr: str = "NORM",
-        span_getter: SpanGetterArg = None,
-        on_ents_only: Union[bool, str, List[str], Set[str]] = None,
+        span_getter: Optional[SpanGetterArg] = None,
+        on_ents_only: Optional[Union[bool, str, List[str], Set[str]]] = None,
         within_ents: bool = False,
         explain: bool = False,
     ):

edsnlp/pipes/qualifiers/negation/negation.py

@@ -159,8 +159,8 @@ def __init__(
         verbs: Optional[List[str]] = None,
         termination: Optional[List[str]] = None,
         attr: str = "NORM",
-        span_getter: SpanGetterArg = None,
-        on_ents_only: Union[bool, str, List[str], Set[str]] = None,
+        span_getter: Optional[SpanGetterArg] = None,
+        on_ents_only: Optional[Union[bool, str, List[str], Set[str]]] = None,
         within_ents: bool = False,
         explain: bool = False,
     ):

edsnlp/pipes/qualifiers/reported_speech/reported_speech.py

@@ -122,8 +122,8 @@ def __init__(
         quotation: Optional[List[str]] = None,
         verbs: Optional[List[str]] = None,
         attr: str = "NORM",
-        span_getter: SpanGetterArg = None,
-        on_ents_only: Union[bool, str, List[str], Set[str]] = None,
+        span_getter: Optional[SpanGetterArg] = None,
+        on_ents_only: Optional[Union[bool, str, List[str], Set[str]]] = None,
         within_ents: bool = False,
         explain: bool = False,
     ):

pyproject.toml

@@ -25,13 +25,15 @@ dependencies = [
     "numpy>=1.15.0; python_version>='3.8'",
     "pandas>=1.1.0,<2.0.0; python_version<'3.8'",
     "pandas>=1.4.0,<2.0.0; python_version>='3.8'",
-    "typing_extensions!=4.7.0,>=4.0.0; python_version>='3.8'",  # https://github.com/explosion/spaCy/issues/12659
-    "typing_extensions>=4.0.0; python_version<'3.8'",
+    "typing-extensions>=4.0.0",
     "dill",
     # Packaging
     "build>=1.0.0",
     "toml",
     "pyarrow>=3.0.0",  # support for fragment.metadata.num_rows
+    # this is only to avoid backtracking issues with spacy's capping
+    "pydantic<2.0.0; python_version<'3.8'",
+    "pydantic-core<2.0.0; python_version<'3.8'",
 ]
 [project.optional-dependencies]
 dev = [

tests/conftest.py

@@ -1,6 +1,7 @@
 from datetime import datetime
 
 import pandas as pd
+import pytest
 import spacy
 from pytest import fixture
 
@@ -151,7 +152,6 @@ def df_notes():
 
 
 def make_df_note(text, module):
-    import databricks.koalas  # noqa F401
     from pyspark.sql import types as T
     from pyspark.sql.session import SparkSession
 
@@ -180,6 +180,10 @@ def make_df_note(text, module):
         return notes
 
     if module == "koalas":
+        try:
+            import databricks.koalas  # noqa F401
+        except ImportError:
+            pytest.skip("Koalas not installed")
         return notes.to_koalas()