Merge pull request #16 from hydroshare/metadata-in

Metadata in

hsmodels/schemas/__init__.py

@@ -1,11 +1,10 @@
 import inspect
-from datetime import datetime
 from enum import Enum
 
 from pydantic import AnyUrl, BaseModel
 from rdflib import Graph, Literal, URIRef
 
-from hsmodels.namespaces import DC, HSTERMS, ORE, RDF, RDFS1, XSD
+from hsmodels.namespaces import DC, HSTERMS, ORE, RDF, RDFS1
 from hsmodels.schemas.aggregations import (
     FileSetMetadata,
     GeographicFeatureMetadata,
@@ -84,7 +83,7 @@ def parse_file(schema, file, file_format='xml', subject=None):
 def rdf_graph(schema):
     for rdf_schema, user_schema in user_schemas.items():
         if isinstance(schema, user_schema):
-            return _rdf_graph(rdf_schema(**schema.dict()), Graph())
+            return _rdf_graph(rdf_schema(**schema.dict(to_rdf=True)), Graph())
     return _rdf_graph(schema, Graph())
 
 

hsmodels/schemas/aggregations.py

@@ -24,14 +24,14 @@
 )
 from hsmodels.schemas.rdf.validators import language_constraint, subjects_constraint
 from hsmodels.schemas.root_validators import (
+    normalize_additional_metadata,
     parse_abstract,
     parse_additional_metadata,
     parse_file_types,
     parse_url,
     split_coverages,
 )
 from hsmodels.schemas.validators import (
-    normalize_additional_metadata,
     parse_multidimensional_spatial_reference,
     parse_spatial_coverage,
     parse_spatial_reference,
@@ -84,9 +84,7 @@ class BaseAggregationMetadata(BaseMetadata):
     _subjects_constraint = validator('subjects', allow_reuse=True)(subjects_constraint)
     _language_constraint = validator('language', allow_reuse=True)(language_constraint)
     _parse_spatial_coverage = validator("spatial_coverage", allow_reuse=True, pre=True)(parse_spatial_coverage)
-    _normalize_additional_metadata = validator("additional_metadata", allow_reuse=True, pre=True)(
-        normalize_additional_metadata
-    )
+    _normalize_additional_metadata = root_validator(allow_reuse=True, pre=True)(normalize_additional_metadata)
 
 
 class GeographicRasterMetadata(BaseAggregationMetadata):

hsmodels/schemas/base_models.py

@@ -15,6 +15,7 @@ def dict(
         exclude_unset: bool = False,
         exclude_defaults: bool = False,
         exclude_none: bool = True,
+        to_rdf: bool = False
     ) -> 'DictStrAny':
         """
         Generate a dictionary representation of the model, optionally specifying which fields to include or exclude.
@@ -35,7 +36,7 @@ def dict(
             exclude_none=exclude_none,
         )
 
-        if hasattr(self.Config, "schema_config"):
+        if to_rdf and hasattr(self.Config, "schema_config"):
             schema_config = self.Config.schema_config
             if "dictionary_field" in schema_config:
                 for field in schema_config["dictionary_field"]:

hsmodels/schemas/rdf/root_validators.py

@@ -84,8 +84,16 @@ def rdf_parse_file_types(cls, values):
 
 def parse_rdf_extended_metadata(cls, values):
     if "additional_metadata" in values:
-        values["extended_metadata"] = values["additional_metadata"]
+        if type(values["additional_metadata"] is list):
+            values["extended_metadata"] = values["additional_metadata"]
+        else:
+            assert type(values["additional_metadata"]) is dict
+            em = []
+            for key, value in values["additional_metadata"].items():
+                em.append({"key": key, "value": value})
+            values["extended_metadata"] = em
         del values["additional_metadata"]
+
     return values
 
 

hsmodels/schemas/resource.py

@@ -17,49 +17,28 @@
 )
 from hsmodels.schemas.rdf.validators import language_constraint, subjects_constraint
 from hsmodels.schemas.root_validators import (
+    normalize_additional_metadata,
     parse_abstract,
     parse_additional_metadata,
     parse_url,
     split_coverages,
     split_dates,
 )
-from hsmodels.schemas.validators import (
-    list_not_empty,
-    normalize_additional_metadata,
-    parse_identifier,
-    parse_sources,
-    parse_spatial_coverage,
-)
+from hsmodels.schemas.validators import list_not_empty, parse_identifier, parse_sources, parse_spatial_coverage
 
 
-class ResourceMetadata(BaseMetadata):
+class ResourceMetadataIn(BaseMetadata):
     """
-    A class used to represent the metadata for a resource
+    A class used to represent the metadata for a resource that can be modified
     """
 
     class Config:
         title = 'Resource Metadata'
 
         schema_config = {
-            'read_only': ['type', 'identifier', 'created', 'modified', 'published', 'url'],
             'dictionary_field': ['additional_metadata'],
         }
 
-    type: str = Field(
-        const=True,
-        default="CompositeResource",
-        title="Resource Type",
-        description="An object containing a URL that points to the HydroShare resource type selected from the hsterms namespace",
-        allow_mutation=False,
-    )
-
-    url: AnyUrl = Field(title="URL", description="An object containing the URL for a resource", allow_mutation=False)
-
-    identifier: AnyUrl = Field(
-        title="Identifier",
-        description="An object containing the URL-encoded unique identifier for a resource",
-        allow_mutation=False,
-    )
     title: str = Field(
         max_length=300, default=None, title="Title", description="A string containing the name given to a resource"
     )
@@ -99,24 +78,6 @@ class Config:
     rights: Rights = Field(
         title="Rights", description="An object containing information about rights held in an over a resource"
     )
-    created: datetime = Field(
-        default_factory=datetime.now,
-        title="Creation date",
-        description="A datetime object containing the instant associated with when a resource was created",
-        allow_mutation=False,
-    )
-    modified: datetime = Field(
-        default_factory=datetime.now,
-        title="Modified date",
-        description="A datetime object containing the instant associated with when a resource was last modified",
-        allow_mutation=False,
-    )
-    published: datetime = Field(
-        default=None,
-        title="Published date",
-        description="A datetime object containing the instant associated with when a resource was published",
-        allow_mutation=False,
-    )
     awards: List[AwardInfo] = Field(
         default=[],
         title="Funding agency information",
@@ -142,18 +103,66 @@ class Config:
     )
 
     _parse_coverages = root_validator(pre=True, allow_reuse=True)(split_coverages)
-    _parse_dates = root_validator(pre=True, allow_reuse=True)(split_dates)
     _parse_additional_metadata = root_validator(pre=True, allow_reuse=True)(parse_additional_metadata)
     _parse_abstract = root_validator(pre=True)(parse_abstract)
-    _parse_url = root_validator(pre=True, allow_reuse=True)(parse_url)
-
-    _parse_identifier = validator("identifier", pre=True)(parse_identifier)
     _parse_sources = validator("sources", pre=True)(parse_sources)
     _parse_spatial_coverage = validator("spatial_coverage", allow_reuse=True, pre=True)(parse_spatial_coverage)
-    _normalize_additional_metadata = validator("additional_metadata", allow_reuse=True, pre=True)(
-        normalize_additional_metadata
-    )
+
+    _normalize_additional_metadata = root_validator(allow_reuse=True, pre=True)(normalize_additional_metadata)
 
     _subjects_constraint = validator('subjects', allow_reuse=True)(subjects_constraint)
     _language_constraint = validator('language', allow_reuse=True)(language_constraint)
     _creators_constraint = validator('creators')(list_not_empty)
+
+
+class ResourceMetadata(ResourceMetadataIn):
+    """
+    A class used to represent the metadata for a resource
+    """
+
+    class Config:
+        title = 'Resource Metadata'
+
+        schema_config = {
+            'read_only': ['type', 'identifier', 'created', 'modified', 'published', 'url'],
+            'dictionary_field': ['additional_metadata'],
+        }
+
+    type: str = Field(
+        const=True,
+        default="CompositeResource",
+        title="Resource Type",
+        description="An object containing a URL that points to the HydroShare resource type selected from the hsterms namespace",
+        allow_mutation=False,
+    )
+
+    url: AnyUrl = Field(title="URL", description="An object containing the URL for a resource", allow_mutation=False)
+
+    identifier: AnyUrl = Field(
+        title="Identifier",
+        description="An object containing the URL-encoded unique identifier for a resource",
+        allow_mutation=False,
+    )
+    created: datetime = Field(
+        default_factory=datetime.now,
+        title="Creation date",
+        description="A datetime object containing the instant associated with when a resource was created",
+        allow_mutation=False,
+    )
+    modified: datetime = Field(
+        default_factory=datetime.now,
+        title="Modified date",
+        description="A datetime object containing the instant associated with when a resource was last modified",
+        allow_mutation=False,
+    )
+    published: datetime = Field(
+        default=None,
+        title="Published date",
+        description="A datetime object containing the instant associated with when a resource was published",
+        allow_mutation=False,
+    )
+
+    _parse_dates = root_validator(pre=True, allow_reuse=True)(split_dates)
+    _parse_url = root_validator(pre=True, allow_reuse=True)(parse_url)
+
+    _parse_identifier = validator("identifier", pre=True)(parse_identifier)

hsmodels/schemas/root_validators.py

@@ -100,3 +100,22 @@ def parse_file_types(cls, values):
                 del values[file_type.name]
     values['file_types'] = file_types_list
     return values
+
+
+def normalize_additional_metadata(cls, values):
+    if "additional_metadata" in values:
+        value = values["additional_metadata"]
+        if isinstance(value, list):
+            as_dict = {}
+            for val in value:
+                if not isinstance(val, dict):
+                    raise ValueError(f"List entry {val} must be a dict")
+                if "key" not in val:
+                    raise ValueError(f"Missing the 'key' key in {val}")
+                if "value" not in val:
+                    raise ValueError(f"Missing the 'value' key in {val}")
+                if val["key"] in as_dict:
+                    raise ValueError(f"Found a duplicate key {val['key']}")
+                as_dict[val["key"]] = val["value"]
+            values["additional_metadata"] = as_dict
+    return values

hsmodels/schemas/validators.py

@@ -52,23 +52,6 @@ def parse_spatial_coverage(cls, value):
     return value
 
 
-def normalize_additional_metadata(cls, value):
-    if isinstance(value, list):
-        as_dict = {}
-        for val in value:
-            if not isinstance(val, dict):
-                raise ValueError(f"List entry {val} must be a dict")
-            if "key" not in val:
-                raise ValueError(f"Missing the 'key' key in {val}")
-            if "value" not in val:
-                raise ValueError(f"Missing the 'value' key in {val}")
-            if val["key"] in as_dict:
-                raise ValueError(f"Found a duplicate key {val['key']}")
-            as_dict[val["key"]] = val["value"]
-        return as_dict
-    return value
-
-
 def list_not_empty(cls, l):
     if len(l) == 0:
         raise ValueError("list must contain at least one entry")

setup.py

@@ -4,7 +4,7 @@
 
 setup(
     name='hsmodels',
-    version='0.3.0',
+    version='0.3.1',
     packages=find_packages(include=['hsmodels', 'hsmodels.*', 'hsmodels.schemas.*', 'hsmodels.schemas.rdf.*'],
                            exclude=("tests",)),
     install_requires=[

tests/conftest.py

@@ -1,9 +1,10 @@
 import os
+
 import pytest
 
 
 @pytest.fixture
 def change_test_dir(request):
     os.chdir(request.fspath.dirname)
     yield
-    os.chdir(request.config.invocation_dir)
+    os.chdir(request.config.invocation_dir)