Changes to restore Python 3.8 compatibility (#1)

* Changes to add Python 3.8 compatability

1. Revert to old-style typing importing to add Python 3.8 backwards (capitalisation of List and Dict)
2. Correct parynthesis use around file opening command for python 3.8 compatability

* Update github workflow to include python 3.8

* Changes for haslib compatability

Remove references to usedforsecurity for hashlib compatiablity

* Correct syntax

.github/workflows/main.yaml

@@ -12,7 +12,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: [ '3.9', '3.10' ]
+        python-version: [ '3.8', '3.9', '3.10' ]
       max-parallel: 2
 
     steps:

app/data/anno_dataset.py

@@ -45,7 +45,7 @@ def _generate_examples(self, filepaths: List[Path]) -> Iterable[Tuple[str, Dict]
 def generate_examples(filepaths: List[Path]) -> Iterable[Tuple[str, Dict]]:
     id_ = 1
     for filepath in filepaths:
-        with (open(str(filepath), "r") as f):
+        with open(str(filepath), "r") as f:
             annotations = json.load(f)
             filtered = filter_by_concept_ids(annotations)
             for project in filtered["projects"]:

app/domain.py

@@ -56,8 +56,8 @@ class Annotation(BaseModel):
     categories: Optional[List[str]] = Field(default=None, description="The categories to which the annotation concept belongs")
     accuracy: Optional[float] = Field(default=None, description="The confidence score of the annotation")
     text: Optional[str] = Field(default=None, description="The string literal of the annotation span")
-    meta_anns: Optional[dict] = Field(default=None, description="The meta annotations")
-    athena_ids: Optional[list[dict]] = Field(default=None, description="The OHDSI Athena concept IDs")
+    meta_anns: Optional[Dict] = Field(default=None, description="The meta annotations")
+    athena_ids: Optional[List[Dict]] = Field(default=None, description="The OHDSI Athena concept IDs")
 
     @root_validator()
     def _validate(cls, values: Dict[str, Any]) -> Dict[str, Any]:

app/processors/metrics_collector.py

@@ -11,8 +11,8 @@
 
 ANCHOR_DELIMITER = ";"
 DOC_SPAN_DELIMITER = "_"
-STATE_MISSING = hashlib.sha1("MISSING".encode("utf-8"), usedforsecurity=False).hexdigest()
-META_STATE_MISSING = hashlib.sha1("{}".encode("utf-8"), usedforsecurity=False).hexdigest()
+STATE_MISSING = hashlib.sha1("MISSING".encode("utf-8")).hexdigest()
+META_STATE_MISSING = hashlib.sha1("{}".encode("utf-8")).hexdigest()
 
 
 def sanity_check_model_with_trainer_export(trainer_export: Union[str, TextIO, Dict],
@@ -425,12 +425,12 @@ def _filter_docspan_by_value(docspan2value: Dict, value: str) -> Dict:
 
 
 def _get_hashed_annotation_state(annotation: Dict, state_keys: Set[str]) -> str:
-    return hashlib.sha1("_".join([str(annotation.get(key)) for key in state_keys]).encode("utf-8"), usedforsecurity=False).hexdigest()
+    return hashlib.sha1("_".join([str(annotation.get(key)) for key in state_keys]).encode("utf-8")).hexdigest()
 
 
 def _get_hashed_meta_annotation_state(meta_anno: Dict) -> str:
     meta_anno = {key: val for key, val in sorted(meta_anno.items(), key=lambda item: item[0])}  # may not be necessary
-    return hashlib.sha1(str(meta_anno).encode("utf=8"), usedforsecurity=False).hexdigest()
+    return hashlib.sha1(str(meta_anno).encode("utf=8")).hexdigest()
 
 
 def _get_cohens_kappa_coefficient(y1_labels: List, y2_labels: List) -> float: