dcicutils 8.8.0.1b23__tar.gz → 8.8.0.1b24__tar.gz

{dcicutils-8.8.0.1b23 → dcicutils-8.8.0.1b24}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dcicutils
-Version: 8.8.0.1b23
+Version: 8.8.0.1b24
 Summary: Utility package for interacting with the 4DN Data Portal and other 4DN resources
 Home-page: https://github.com/4dn-dcic/utils
 License: MIT

{dcicutils-8.8.0.1b23 → dcicutils-8.8.0.1b24}/dcicutils/portal_object_utils.py

@@ -99,7 +99,8 @@ class PortalObject:
     def compare(self, value: Union[dict, PortalObject],
                 consider_refs: bool = False, resolved_refs: List[dict] = None) -> Tuple[dict, int]:
         if consider_refs and isinstance(resolved_refs, list):
-            this_data, nlookups = self._normalized_refs(refs=resolved_refs).data
+            normlized_portal_object, nlookups = self._normalized_refs(refs=resolved_refs)
+            this_data = normlized_portal_object.data
         else:
             this_data = self.data
             nlookups = 0
@@ -213,7 +214,7 @@ class PortalObject:
                     if Portal.is_lookup_root(lookup_options) and not Portal.is_lookup_root_first(lookup_options):
                         identifying_paths.append(f"/{identifying_value}")
                     if Portal.is_lookup_subtypes(lookup_options):
-                        for subtype_name in self._portal.get_schemas_subtype_names():
+                        for subtype_name in self._portal.get_schema_subtype_names(self.type):
                             identifying_paths.append(f"/{subtype_name}/{identifying_value}")
         return identifying_paths or None
 
@@ -249,13 +250,13 @@ class PortalObject:
         if isinstance(value, dict):
             for key in value:
                 path = f"{_path}.{key}" if _path else key
-                value[key] = PortalObject._normalize_data_refs(value[key], refs=refs,
-                                                               schema=schema, portal=portal, _path=path)
+                value[key], nlookups = PortalObject._normalize_data_refs(value[key], refs=refs,
+                                                                         schema=schema, portal=portal, _path=path)
         elif isinstance(value, list):
             for index in range(len(value)):
                 path = f"{_path or ''}#{index}"
-                value[index] = PortalObject._normalize_data_refs(value[index], refs=refs,
-                                                                 schema=schema, portal=portal, _path=path)
+                value[index], nlookups = PortalObject._normalize_data_refs(value[index], refs=refs,
+                                                                           schema=schema, portal=portal, _path=path)
         elif value_type := Schema.get_property_by_path(schema, _path):
             if link_to := value_type.get("linkTo"):
                 ref_path = f"/{link_to}/{value}"

{dcicutils-8.8.0.1b23 → dcicutils-8.8.0.1b24}/dcicutils/portal_utils.py

@@ -398,7 +398,7 @@ class Portal:
         return super_type_map_flattened
 
     @lru_cache(maxsize=64)
-    def get_schemas_subtype_names(self, type_name: str) -> List[str]:
+    def get_schema_subtype_names(self, type_name: str) -> List[str]:
         if not (schemas_super_type_map := self.get_schemas_super_type_map()):
             return []
         return schemas_super_type_map.get(type_name, [])

{dcicutils-8.8.0.1b23 → dcicutils-8.8.0.1b24}/dcicutils/structured_data.py

@@ -222,6 +222,9 @@ class StructuredDataSet:
                         diffs[type_name].append(create_readonly_object(path=identifying_path, uuid=None, diffs=None))
                         if callable(progress):
                             progress({"create": True, "lookups": nlookups})
+                    else:
+                        if callable(progress):
+                            progress({"lookups": nlookups})
         if callable(progress):
             progress({"finish": True})
         return diffs
@@ -257,7 +260,7 @@ class StructuredDataSet:
         self._load_reader(CsvReader(file), type_name=Schema.type_name(file))
 
     def _load_excel_file(self, file: str) -> None:
-        def calculate_total_rows_to_process() -> Tuple[int, int]:
+        def get_counts() -> Tuple[int, int]:
             nonlocal file
             excel = Excel(file)
             nrows = 0
@@ -266,18 +269,27 @@ class StructuredDataSet:
                     nrows += 1
             return nrows, len(excel.sheet_names)
         if self._progress:
-            self._progress_update(calculate_total_rows_to_process)
+            nrows, nsheets = get_counts()
+            self._progress({"start": True, "sheets": nsheets, "rows": nrows})
+        """
+        if self._progress:
+            self._progress_update(get_counts)
+        """
         excel = Excel(file)  # Order the sheet names by any specified ordering (e.g. ala snovault.loadxl).
         order = {Schema.type_name(key): index for index, key in enumerate(self._order)} if self._order else {}
         for sheet_name in sorted(excel.sheet_names, key=lambda key: order.get(Schema.type_name(key), sys.maxsize)):
             self._load_reader(excel.sheet_reader(sheet_name), type_name=Schema.type_name(sheet_name))
+        if self._progress:
+            self._progress({"finish": True})
+        # TODO: Do we really need progress reporting for the below?
         # Check for unresolved reference errors which really are not because of ordering.
         # Yes such internal references will be handled correctly on actual database update via snovault.loadxl.
         if ref_errors := self.ref_errors:
             ref_errors_actual = []
             for ref_error in ref_errors:
                 if not (resolved := self.portal.ref_exists(ref := ref_error["error"])):
-                    # if not (resolved := self.portal.ref_exists_internally(ref := ref_error["error"])):  # TODO
+                    # TODO: Probably do this instead; and if so then no progress needed (per question above).
+                    # if not (resolved := self.portal.ref_exists_internally(ref := ref_error["error"])):
                     ref_errors_actual.append(ref_error)
                 else:
                     self._resolved_refs.add((ref, resolved.get("uuid")))
@@ -310,13 +322,15 @@ class StructuredDataSet:
                     self._add_properties(structured_row, self._autoadd_properties, schema)
             self._add(type_name, structured_row)
             if self._progress:
-                self._progress_update(-1,
-                                      self.ref_total_count,
-                                      self.ref_total_found_count,
-                                      self.ref_total_notfound_count,
-                                      self.ref_lookup_count,
-                                      self.ref_lookup_cache_hit_count,
-                                      self.ref_invalid_identifying_property_count)
+                self._progress({
+                    "parse": True,
+                    "refs": self.ref_total_count,
+                    "refs_found": self.ref_total_found_count,
+                    "refs_not_found": self.ref_total_notfound_count,
+                    "refs_lookup": self.ref_lookup_count,
+                    "refs_cache_hit": self.ref_lookup_cache_hit_count,
+                    "refs_invalid": self.ref_invalid_identifying_property_count
+                })
         self._note_warning(reader.warnings, "reader")
         if schema:
             self._note_error(schema._unresolved_refs, "ref")
@@ -850,7 +864,7 @@ class Portal(PortalBase):
         return schemas
 
     @lru_cache(maxsize=64)
-    def _get_schema_subtypes_names(self, type_name: str) -> List[str]:
+    def _get_schema_subtype_names(self, type_name: str) -> List[str]:
         if not (schemas_super_type_map := self.get_schemas_super_type_map()):
             return []
         return schemas_super_type_map.get(type_name, [])
@@ -913,7 +927,7 @@ class Portal(PortalBase):
             lookup_paths.append(f"/{type_name}/{value}")
         if is_ref_lookup_root and not is_ref_lookup_root_first:
             lookup_paths.append(f"/{value}")
-        subtype_names = self._get_schema_subtypes_names(type_name) if is_ref_lookup_subtypes else []
+        subtype_names = self._get_schema_subtype_names(type_name) if is_ref_lookup_subtypes else []
         for subtype_name in subtype_names:
             lookup_paths.append(f"/{subtype_name}/{value}")
         if not lookup_paths:
@@ -952,7 +966,7 @@ class Portal(PortalBase):
         ref_lookup_strategy, ref_validator = (
             self._ref_lookup_strategy(type_name, self.get_schema(type_name), value))
         is_ref_lookup_subtypes = StructuredDataSet._is_ref_lookup_subtypes(ref_lookup_strategy)
-        subtype_names = self._get_schema_subtypes_names(type_name) if is_ref_lookup_subtypes else []
+        subtype_names = self._get_schema_subtype_names(type_name) if is_ref_lookup_subtypes else []
         for type_name in [type_name] + subtype_names:
             is_resolved, resolved_item = self._ref_exists_single_internally(type_name, value)
             if is_resolved:
@@ -1014,7 +1028,7 @@ class Portal(PortalBase):
                         if not is_uuid(property_value):
                             return False
             return True
-        for schema_name in [type_name] + self._get_schema_subtypes_names(type_name):
+        for schema_name in [type_name] + self._get_schema_subtype_names(type_name):
             if schema := self.get_schema(schema_name):
                 if identifying_properties := schema.get("identifyingProperties"):
                     for identifying_property in identifying_properties:
@@ -1039,7 +1053,7 @@ class Portal(PortalBase):
 
     def _cache_ref(self, type_name: str, value: str, resolved: List[str]) -> None:
         if self._ref_cache is not None:
-            subtype_names = self._get_schema_subtypes_names(type_name)
+            subtype_names = self._get_schema_subtype_names(type_name)
             for type_name in [type_name] + subtype_names:
                 self._ref_cache[f"/{type_name}/{value}"] = resolved
 

{dcicutils-8.8.0.1b23 → dcicutils-8.8.0.1b24}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "dcicutils"
-version = "8.8.0.1b23"  # TODO: To become 8.8.1
+version = "8.8.0.1b24"  # TODO: To become 8.8.1
 description = "Utility package for interacting with the 4DN Data Portal and other 4DN resources"
 authors = ["4DN-DCIC Team <support@4dnucleome.org>"]
 license = "MIT"