dcicutils 8.5.0.1b4__tar.gz → 8.6.0.0b0__tar.gz

{dcicutils-8.5.0.1b4 → dcicutils-8.6.0.0b0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: dcicutils
-Version: 8.5.0.1b4
+Version: 8.6.0.0b0
 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-8.5.0.1b4 → dcicutils-8.6.0.0b0}/dcicutils/misc_utils.py

@@ -982,11 +982,7 @@ def to_integer(value: str, fallback: Optional[Any] = None) -> Optional[Any]:
     try:
         return int(value)
     except Exception:
-        try:
-            return int(float(value))
-        except Exception:
-            pass
-    return fallback
+        return fallback
 
 
 def to_float(value: str, fallback: Optional[Any] = None) -> Optional[Any]:
@@ -1469,33 +1465,28 @@ def string_list(s):
     return [p for p in [part.strip() for part in s.split(",")] if p]
 
 
-def split_string(value: str, delimiter: str, escape: Optional[str] = None, unique: bool = False) -> List[str]:
+def split_string(value: str, delimiter: str, escape: Optional[str] = None) -> List[str]:
     """
     Splits the given string into an array of string based on the given delimiter, and an optional escape character.
     """
     if not isinstance(value, str) or not (value := value.strip()):
         return []
-    result = []
     if not isinstance(escape, str) or not escape:
-        for item in value.split(delimiter):
-            if (item := item.strip()) and (unique is not True or item not in result):
-                result.append(item)
-        return result
+        return [item.strip() for item in value.split(delimiter)]
+    result = []
     item = r""
     escaped = False
     for c in value:
         if c == delimiter and not escaped:
-            if (item := item.strip()) and (unique is not True or item not in result):
-                result.append(item)
+            result.append(item.strip())
             item = r""
         elif c == escape and not escaped:
             escaped = True
         else:
             item += c
             escaped = False
-    if (item := item.strip()) and (unique is not True or item not in result):
-        result.append(item)
-    return result
+    result.append(item.strip())
+    return [item for item in result if item]
 
 
 def right_trim(list_or_tuple: Union[List[Any], Tuple[Any]],

{dcicutils-8.5.0.1b4 → dcicutils-8.6.0.0b0}/dcicutils/portal_utils.py

@@ -23,14 +23,15 @@ class Portal:
     2. From a key dictionary, containing "key" and "secret" property values.
     3. From a key tuple, containing (in order) a key and secret values.
     4. From a keys file assumed to reside in ~/.{app}-keys.json where the given "app" value is either "smaht", "cgap",
-       or "fourfront"; where is assumed to contain a dictionary with a key for the given "env" value, e.g. smaht-local;
-       and with a dictionary value containing "key" and "secret" property values, and an optional "server" property;
-       if an "app" value is not specified but the given "env" value begins with one of the app values then that value
-       will be used, i.e. e.g. if "env" is "smaht-local" and app is unspecified than it is assumed to be "smaht".
+       or "fourfront"; and where this file is assumed to contain a dictionary with a key equal to the given "env"
+       value (e.g. smaht-localhost) and with a dictionary value containing "key" and "secret" property values; if
+       an "app" value is not specified but the given "env" value begins with one of the app values then that value
+       will be used, i.e. e.g. if env is "smaht-localhost" and app is unspecified than it is assumed to be "smaht".
     5. From a keys file as described above (#4) but rather than be identified by the given "env" value it
-       is looked up via the given "server" name and the "server" key dictionary value in the key file.
+       is looked up by the given "server" name and the "server" key dictionary value in the key file.
     6. From a given "vapp" value (which is assumed to be a TestApp or VirtualApp).
-    7. From another Portal object; or from a a pyramid Router object.
+    7. From another Portal object.
+    8. From a a pyramid Router object.
     """
     def __init__(self,
                  arg: Optional[Union[VirtualApp, TestApp, Router, Portal, dict, tuple, str]] = None,
@@ -61,7 +62,6 @@ class Portal:
             self._server = portal._server
             self._key = portal._key
             self._key_pair = portal._key_pair
-            self._key_id = portal._key_id
             self._key_file = portal._key_file
             return
         self._vapp = None
@@ -70,7 +70,6 @@ class Portal:
         self._server = server
         self._key = None
         self._key_pair = None
-        self._key_id = None
         self._key_file = None
         if isinstance(portal, (VirtualApp, TestApp)):
             self._vapp = portal
@@ -96,10 +95,6 @@ class Portal:
                 self._key = key_manager.get_keydict_for_server(self._server)
             self._key_pair = key_manager.keydict_to_keypair(self._key) if self._key else None
             self._key_file = key_manager.keys_file
-        if self._key and (key_id := self._key.get("key")):
-            self._key_id = key_id
-        elif self._key_pair and (key_id := self._key_pair[1]):
-            self._key_id = key_id
 
     @property
     def env(self):
@@ -121,10 +116,6 @@ class Portal:
     def key_pair(self):
         return self._key_pair
 
-    @property
-    def key_id(self):
-        return self._key_id
-
     @property
     def key_file(self):
         return self._key_file
@@ -215,19 +206,13 @@ class Portal:
             super_type_map_flattened[super_type_name] = breadth_first(super_type_map, super_type_name)
         return super_type_map_flattened
 
-    def ping(self) -> bool:
-        try:
-            return self.get("/health").status_code == 200
-        except Exception:
-            return False
-
     def _uri(self, uri: str) -> str:
         if not isinstance(uri, str) or not uri:
             return "/"
         if uri.lower().startswith("http://") or uri.lower().startswith("https://"):
             return uri
         uri = re.sub(r"/+", "/", uri)
-        return (self._server + ("/" if not uri.startswith("/") else "") + uri) if self._server else uri
+        return (self._server + ("/" if uri.startswith("/") else "") + uri) if self._server else uri
 
     def _kwargs(self, **kwargs) -> dict:
         result_kwargs = {"headers":

dcicutils-8.6.0.0b0/dcicutils/schema_utils.py

@@ -0,0 +1,142 @@
+from typing import Any, Dict, List
+
+
+class JsonSchemaConstants:
+    ANY_OF = "anyOf"
+    ARRAY = "array"
+    BOOLEAN = "boolean"
+    ENUM = "enum"
+    INTEGER = "integer"
+    ITEMS = "items"
+    NUMBER = "number"
+    OBJECT = "object"
+    ONE_OF = "oneOf"
+    PATTERN = "pattern"
+    PROPERTIES = "properties"
+    REQUIRED = "required"
+    STRING = "string"
+    TYPE = "type"
+
+
+class EncodedSchemaConstants:
+    DEFAULT = "default"
+    FORMAT = "format"
+    IDENTIFYING_PROPERTIES = "identifyingProperties"
+    LINK_TO = "linkTo"
+    MERGE_REF = "$merge"
+    MIXIN_PROPERTIES = "mixinProperties"
+    REF = "$ref"
+    UNIQUE_KEY = "uniqueKey"
+
+
+class SchemaConstants(JsonSchemaConstants, EncodedSchemaConstants):
+    pass
+
+
+def get_properties(schema: Dict[str, Any]) -> Dict[str, Any]:
+    """Return the properties of a schema."""
+    return schema.get(SchemaConstants.PROPERTIES, {})
+
+
+def get_required(schema: Dict[str, Any]) -> List[str]:
+    """Return the required properties of a schema."""
+    return schema.get(SchemaConstants.REQUIRED, [])
+
+
+def get_any_of(schema: Dict[str, Any]) -> List[Dict[str, Any]]:
+    """Return the anyOf properties of a schema."""
+    return schema.get(SchemaConstants.ANY_OF, [])
+
+
+def get_one_of(schema: Dict[str, Any]) -> List[Dict[str, Any]]:
+    """Return the oneOf properties of a schema."""
+    return schema.get(SchemaConstants.ONE_OF, [])
+
+
+def get_conditionally_required_properties(schema: Dict[str, Any]) -> List[str]:
+    """Get required + possibly required properties.
+
+    Using heuristics here; update as needed.
+    """
+    return sorted(
+        list(
+            set(
+                get_required(schema)
+                + get_any_of_required_properties(schema)
+                + get_one_of_required_properties(schema)
+            )
+        )
+    )
+
+
+def get_any_of_required_properties(schema: Dict[str, Any]) -> List[str]:
+    """Get required properties from anyOf."""
+    return [
+        property_name
+        for any_of_schema in get_any_of(schema)
+        for property_name in get_required(any_of_schema)
+    ]
+
+
+def get_one_of_required_properties(schema: Dict[str, Any]) -> List[str]:
+    """Get required properties from oneOf."""
+    return [
+        property_name
+        for one_of_schema in get_one_of(schema)
+        for property_name in get_required(one_of_schema)
+    ]
+
+
+def get_mixin_properties(schema: Dict[str, Any]) -> List[Dict[str, Any]]:
+    """Return the mixin properties of a schema."""
+    return schema.get(EncodedSchemaConstants.MIXIN_PROPERTIES, [])
+
+
+def get_identifying_properties(schema: Dict[str, Any]) -> List[str]:
+    """Return the identifying properties of a schema."""
+    return schema.get(EncodedSchemaConstants.IDENTIFYING_PROPERTIES, [])
+
+
+def get_schema_type(schema: Dict[str, Any]) -> str:
+    """Return the type of a schema."""
+    return schema.get(SchemaConstants.TYPE, "")
+
+
+def is_array_schema(schema: Dict[str, Any]) -> bool:
+    """Return True if the schema is an array."""
+    return get_schema_type(schema) == SchemaConstants.ARRAY
+
+
+def is_object_schema(schema: Dict[str, Any]) -> bool:
+    """Return True if the schema is an object."""
+    return get_schema_type(schema) == SchemaConstants.OBJECT
+
+
+def is_string_schema(schema: Dict[str, Any]) -> bool:
+    """Return True if the schema is a string."""
+    return get_schema_type(schema) == SchemaConstants.STRING
+
+
+def is_number_schema(schema: Dict[str, Any]) -> bool:
+    """Return True if the schema is a number."""
+    return get_schema_type(schema) == SchemaConstants.NUMBER
+
+
+def is_integer_schema(schema: Dict[str, Any]) -> bool:
+    """Return True if the schema is an integer."""
+    return get_schema_type(schema) == SchemaConstants.INTEGER
+
+
+def is_boolean_schema(schema: Dict[str, Any]) -> bool:
+    """Return True if the schema is a boolean."""
+    return get_schema_type(schema) == SchemaConstants.BOOLEAN
+
+
+def get_items(schema: Dict[str, Any]) -> Dict[str, Any]:
+    """Return the items of a schema."""
+    return schema.get(SchemaConstants.ITEMS, {})
+
+
+def has_property(schema: Dict[str, Any], property_name: str) -> bool:
+    """Return True if the schema has the given property."""
+    return property_name in get_properties(schema)

{dcicutils-8.5.0.1b4 → dcicutils-8.6.0.0b0}/dcicutils/structured_data.py

@@ -42,24 +42,23 @@ StructuredDataSet = Type["StructuredDataSet"]
 class StructuredDataSet:
 
     def __init__(self, file: Optional[str] = None, portal: Optional[Union[VirtualApp, TestApp, Portal]] = None,
-                 schemas: Optional[List[dict]] = None, autoadd: Optional[dict] = None,
-                 data: Optional[List[dict]] = None, order: Optional[List[str]] = None, prune: bool = True) -> None:
+                 schemas: Optional[List[dict]] = None, data: Optional[List[dict]] = None,
+                 order: Optional[List[str]] = None, prune: bool = True) -> None:
         self.data = {} if not data else data  # If portal is None then no schemas nor refs.
         self._portal = Portal(portal, data=self.data, schemas=schemas) if portal else None
         self._order = order
         self._prune = prune
         self._warnings = {}
         self._errors = {}
-        self._resolved_refs = set()
+        self._resolved_refs = []
         self._validated = False
-        self._autoadd_properties = autoadd if isinstance(autoadd, dict) and autoadd else None
         self._load_file(file) if file else None
 
     @staticmethod
     def load(file: str, portal: Optional[Union[VirtualApp, TestApp, Portal]] = None,
-             schemas: Optional[List[dict]] = None, autoadd: Optional[dict] = None,
+             schemas: Optional[List[dict]] = None,
              order: Optional[List[str]] = None, prune: bool = True) -> StructuredDataSet:
-        return StructuredDataSet(file=file, portal=portal, schemas=schemas, autoadd=autoadd, order=order, prune=prune)
+        return StructuredDataSet(file=file, portal=portal, schemas=schemas, order=order, prune=prune)
 
     def validate(self, force: bool = False) -> None:
         if self._validated and not force:
@@ -97,7 +96,7 @@ class StructuredDataSet:
 
     @property
     def resolved_refs(self) -> List[str]:
-        return list(self._resolved_refs)
+        return self._resolved_refs
 
     @property
     def upload_files(self) -> List[str]:
@@ -164,13 +163,11 @@ class StructuredDataSet:
             structured_row = structured_row_template.create_row()
             for column_name, value in row.items():
                 structured_row_template.set_value(structured_row, column_name, value, reader.file, reader.row_number)
-                if self._autoadd_properties:
-                    self._add_properties(structured_row, self._autoadd_properties, schema)
             self._add(type_name, structured_row)
         self._note_warning(reader.warnings, "reader")
         if schema:
             self._note_error(schema._unresolved_refs, "ref")
-            self._resolved_refs.update(schema._resolved_refs)
+            self._resolved_refs = schema._resolved_refs
 
     def _add(self, type_name: str, data: Union[dict, List[dict]]) -> None:
         if self._prune:
@@ -180,11 +177,6 @@ class StructuredDataSet:
         else:
             self.data[type_name] = [data] if isinstance(data, dict) else data
 
-    def _add_properties(self, structured_row: dict, properties: dict, schema: Optional[dict] = None) -> None:
-        for name in properties:
-            if name not in structured_row and (not schema or schema.data.get("properties", {}).get(name)):
-                structured_row[name] = properties[name]
-
     def _note_warning(self, item: Optional[Union[dict, List[dict]]], group: str) -> None:
         self._note_issue(self._warnings, item, group)
 
@@ -245,7 +237,7 @@ class _StructuredRowTemplate:
             return {array_name: array} if array_name else {column_component: value}
 
         def set_value_internal(data: Union[dict, list], value: Optional[Any], src: Optional[str],
-                               path: List[Union[str, int]], typeinfo: Optional[dict], mapv: Optional[Callable]) -> None:
+                               path: List[Union[str, int]], mapv: Optional[Callable]) -> None:
 
             def set_value_backtrack_object(path_index: int, path_element: str) -> None:
                 nonlocal data, path, original_data
@@ -265,7 +257,7 @@ class _StructuredRowTemplate:
                     set_value_backtrack_object(i, p)
                 data = data[p]
             if (p := path[-1]) == -1 and isinstance(value, str):
-                values = _split_array_string(value, unique=typeinfo.get("unique") if typeinfo else False)
+                values = _split_array_string(value)
                 if mapv:
                     values = [mapv(value, src) for value in values]
                 merge_objects(data, values)
@@ -296,13 +288,11 @@ class _StructuredRowTemplate:
         for column_name in column_names or []:
             ensure_column_consistency(column_name)
             rational_column_name = self._schema.rationalize_column_name(column_name) if self._schema else column_name
-            column_typeinfo = self._schema.get_typeinfo(rational_column_name) if self._schema else None
-            map_value_function = column_typeinfo.get("map") if column_typeinfo else None
+            map_value_function = self._schema.get_map_value_function(rational_column_name) if self._schema else None
             if (column_components := _split_dotted_string(rational_column_name)):
                 merge_objects(structured_row_template, parse_components(column_components, path := []), True)
-                self._set_value_functions[column_name] = (
-                    lambda data, value, src, path=path, typeinfo=column_typeinfo, mapv=map_value_function:
-                        set_value_internal(data, value, src, path, typeinfo, mapv))
+                self._set_value_functions[column_name] = (lambda data, value, src, path=path, mapv=map_value_function:
+                                                          set_value_internal(data, value, src, path, mapv))
         return structured_row_template
 
 
@@ -325,8 +315,7 @@ class Schema:
 
     @staticmethod
     def load_by_name(name: str, portal: Portal) -> Optional[dict]:
-        schema_json = portal.get_schema(Schema.type_name(name)) if portal else None
-        return Schema(schema_json, portal) if schema_json else None
+        return Schema(portal.get_schema(Schema.type_name(name)), portal) if portal else None
 
     def validate(self, data: dict) -> List[str]:
         errors = []
@@ -342,7 +331,10 @@ class Schema:
     def resolved_refs(self) -> List[str]:
         return list(self._resolved_refs)
 
-    def get_typeinfo(self, column_name: str) -> Optional[dict]:
+    def get_map_value_function(self, column_name: str) -> Optional[Any]:
+        return (self._get_typeinfo(column_name) or {}).get("map")
+
+    def _get_typeinfo(self, column_name: str) -> Optional[dict]:
         if isinstance(info := self._typeinfo.get(column_name), str):
             info = self._typeinfo.get(info)
         if not info and isinstance(info := self._typeinfo.get(self.unadorn_column_name(column_name)), str):
@@ -475,15 +467,9 @@ class Schema:
                             raise Exception(f"Array of undefined or multiple types in JSON schema NOT supported: {key}")
                         raise Exception(f"Invalid array type specifier in JSON schema: {key}")
                     key = key + ARRAY_NAME_SUFFIX_CHAR
-                    if unique := (property_value.get("uniqueItems") is True):
-                        pass
                     property_value = array_property_items
                     property_value_type = property_value.get("type")
-                typeinfo = self._create_typeinfo(array_property_items, parent_key=key)
-                if unique:
-                    typeinfo[key]["unique"] = True
-                result.update(typeinfo)
-#               result.update(self._create_typeinfo(array_property_items, parent_key=key))
+                result.update(self._create_typeinfo(array_property_items, parent_key=key))
                 continue
             result[key] = {"type": property_value_type, "map": self._map_function({**property_value, "column": key})}
             if ARRAY_NAME_SUFFIX_CHAR in key:
@@ -574,9 +560,7 @@ class Portal(PortalBase):
 
     @lru_cache(maxsize=256)
     def get_schema(self, schema_name: str) -> Optional[dict]:
-        if not (schemas := self.get_schemas()):
-            return None
-        if schema := schemas.get(schema_name := Schema.type_name(schema_name)):
+        if (schemas := self.get_schemas()) and (schema := schemas.get(schema_name := Schema.type_name(schema_name))):
             return schema
         if schema_name == schema_name.upper() and (schema := schemas.get(schema_name.lower().title())):
             return schema
@@ -584,9 +568,8 @@ class Portal(PortalBase):
             return schema
 
     @lru_cache(maxsize=1)
-    def get_schemas(self) -> Optional[dict]:
-        if not (schemas := super().get_schemas()) or (schemas.get("status") == "error"):
-            return None
+    def get_schemas(self) -> dict:
+        schemas = super().get_schemas()
         if self._schemas:
             schemas = copy.deepcopy(schemas)
             for user_specified_schema in self._schemas:
@@ -632,5 +615,5 @@ def _split_dotted_string(value: str):
     return split_string(value, DOTTED_NAME_DELIMITER_CHAR)
 
 
-def _split_array_string(value: str, unique: bool = False):
-    return split_string(value, ARRAY_VALUE_DELIMITER_CHAR, ARRAY_VALUE_DELIMITER_ESCAPE_CHAR, unique=unique)
+def _split_array_string(value: str):
+    return split_string(value, ARRAY_VALUE_DELIMITER_CHAR, ARRAY_VALUE_DELIMITER_ESCAPE_CHAR)

{dcicutils-8.5.0.1b4 → dcicutils-8.6.0.0b0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "dcicutils"
-version = "8.5.0.1b4"  # TODO: To become 8.6.0
+version = "8.6.0.0b0"
 description = "Utility package for interacting with the 4DN Data Portal and other 4DN resources"
 authors = ["4DN-DCIC Team <support@4dnucleome.org>"]
 license = "MIT"