invenio-vocabularies 2.3.1__py2.py3-none-any.whl → 6.3.1__py2.py3-none-any.whl

invenio_vocabularies/resources/config.py

@@ -0,0 +1,105 @@
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2024 CERN.
+# Copyright (C) 2024 University of Münster.
+#
+# Invenio-Vocabularies is free software; you can redistringibute it and/or modify it
+# under the terms of the MIT License; see LICENSE file for more details.
+
+"""Resources config."""
+
+import marshmallow as ma
+from flask_resources import (
+    BaseListSchema,
+    HTTPJSONException,
+    JSONSerializer,
+    MarshmallowSerializer,
+    ResourceConfig,
+    ResponseHandler,
+    create_error_handler,
+)
+from invenio_records_resources.resources import (
+    RecordResource,
+    RecordResourceConfig,
+    SearchRequestArgsSchema,
+)
+from invenio_records_resources.resources.errors import ErrorHandlersMixin
+from invenio_records_resources.resources.records.args import SearchRequestArgsSchema
+from invenio_records_resources.resources.records.headers import etag_headers
+from invenio_records_resources.services.base.config import ConfiguratorMixin
+
+from .serializer import VocabularyL10NItemSchema
+
+
+class VocabularySearchRequestArgsSchema(SearchRequestArgsSchema):
+    """Vocabularies search request parameters."""
+
+    tags = ma.fields.Str()
+    active = ma.fields.Boolean()
+    status = ma.fields.Boolean()
+
+
+class VocabulariesResourceConfig(RecordResourceConfig):
+    """Vocabulary resource configuration."""
+
+    blueprint_name = "vocabularies"
+    url_prefix = "/vocabularies"
+    routes = {
+        "list": "/<type>",
+        "item": "/<type>/<pid_value>",
+        "tasks": "/tasks",
+    }
+
+    request_view_args = {
+        "pid_value": ma.fields.Str(),
+        "type": ma.fields.Str(required=True),
+    }
+
+    request_search_args = VocabularySearchRequestArgsSchema
+
+    response_handlers = {
+        "application/json": ResponseHandler(JSONSerializer(), headers=etag_headers),
+        "application/vnd.inveniordm.v1+json": ResponseHandler(
+            MarshmallowSerializer(
+                format_serializer_cls=JSONSerializer,
+                object_schema_cls=VocabularyL10NItemSchema,
+                list_schema_cls=BaseListSchema,
+            ),
+            headers=etag_headers,
+        ),
+    }
+
+
+class VocabularyTypeResourceConfig(ResourceConfig, ConfiguratorMixin):
+    """Vocabulary list resource config."""
+
+    # /vocabulary/vocabulary_id
+    # Blueprint configuration
+    blueprint_name = "vocabulary_types"
+    url_prefix = "/vocabularies"
+
+    routes = {"list": "/"}
+
+    # Request parsing
+    request_read_args = {}
+    request_view_args = {
+        "pid_value": ma.fields.String,
+        "type": ma.fields.String,
+    }
+    request_search_args = VocabularySearchRequestArgsSchema
+
+    error_handlers = {
+        **ErrorHandlersMixin.error_handlers,
+        # TODO: Add custom error handlers here
+    }
+    response_handlers = {
+        "application/json": ResponseHandler(JSONSerializer(), headers=etag_headers),
+        "application/vnd.inveniordm.v1+json": ResponseHandler(
+            MarshmallowSerializer(
+                format_serializer_cls=JSONSerializer,
+                object_schema_cls=VocabularyL10NItemSchema,
+                list_schema_cls=BaseListSchema,
+            ),
+            headers=etag_headers,
+        ),
+    }

invenio_vocabularies/resources/resource.py

@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright (C) 2020-2021 CERN.
+# Copyright (C) 2020-2024 CERN.
+# Copyright (C) 2024 Uni Münster.
 #
 # Invenio-Vocabularies is free software; you can redistribute it and/or
 # modify it under the terms of the MIT License; see LICENSE file for more
@@ -8,6 +9,8 @@
 
 """Vocabulary resource."""
 
+import json
+
 import marshmallow as ma
 from flask import g
 from flask_resources import (
@@ -18,6 +21,7 @@ from flask_resources import (
     resource_requestctx,
     response_handler,
 )
+from invenio_access.permissions import system_identity
 from invenio_records_resources.resources import (
     RecordResource,
     RecordResourceConfig,
@@ -37,55 +41,20 @@ from marshmallow import fields
 from .serializer import VocabularyL10NItemSchema
 
 
-#
-# Request args
-#
-class VocabularySearchRequestArgsSchema(SearchRequestArgsSchema):
-    """Add parameter to parse tags."""
-
-    tags = fields.Str()
-
-
-#
-# Resource config
-#
-class VocabulariesResourceConfig(RecordResourceConfig):
-    """Vocabulary resource configuration."""
-
-    blueprint_name = "vocabularies"
-    url_prefix = "/vocabularies"
-    routes = {"list": "/<type>", "item": "/<type>/<pid_value>", "tasks": "/tasks"}
-
-    request_view_args = {
-        "pid_value": ma.fields.Str(),
-        "type": ma.fields.Str(required=True),
-    }
-
-    request_search_args = VocabularySearchRequestArgsSchema
-
-    response_handlers = {
-        "application/json": ResponseHandler(JSONSerializer(), headers=etag_headers),
-        "application/vnd.inveniordm.v1+json": ResponseHandler(
-            MarshmallowSerializer(
-                format_serializer_cls=JSONSerializer,
-                object_schema_cls=VocabularyL10NItemSchema,
-                list_schema_cls=BaseListSchema,
-            ),
-            headers=etag_headers,
-        ),
-    }
-
-
 #
 # Resource
 #
 class VocabulariesResource(RecordResource):
-    """Resource for generic vocabularies."""
+    """Resource for generic vocabularies.
+
+    Provide the API /api/vocabularies/
+    """
 
     def create_url_rules(self):
         """Create the URL rules for the record resource."""
         routes = self.config.routes
         rules = super().create_url_rules()
+
         rules.append(
             route("POST", routes["tasks"], self.launch),
         )
@@ -164,3 +133,24 @@ class VocabulariesResource(RecordResource):
         """Create a task."""
         self.service.launch(g.identity, resource_requestctx.data or {})
         return "", 202
+
+
+class VocabulariesAdminResource(RecordResource):
+    """Resource for vocabularies admin interface."""
+
+    def create_url_rules(self):
+        """Create the URL rules for the record resource."""
+        routes = self.config.routes
+
+        rules = [route("GET", routes["list"], self.search)]
+
+        return rules
+
+    @request_search_args
+    @response_handler(many=True)
+    def search(self):
+        """Return information about _all_ vocabularies."""
+        identity = g.identity
+        hits = self.service.search(identity, params=resource_requestctx.args)
+
+        return hits.to_dict(), 200

invenio_vocabularies/resources/schema.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright (C) 2020-2021 CERN.
+# Copyright (C) 2020-2024 CERN.
 # Copyright (C) 2021 Northwestern University.
 #
 # Invenio-Vocabularies is free software; you can redistribute it and/or
@@ -8,6 +8,7 @@
 # details.
 
 """Vocabulary resource schema."""
+
 from marshmallow import Schema, fields
 
 from invenio_vocabularies.resources.serializer import L10NString

invenio_vocabularies/services/__init__.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright (C) 2020-2021 CERN.
+# Copyright (C) 2020-2024 CERN.
 #
 # Invenio-Vocabularies is free software; you can redistribute it and/or
 # modify it under the terms of the MIT License; see LICENSE file for more
@@ -8,9 +8,12 @@
 
 """Services module."""
 
-from .service import VocabulariesService, VocabulariesServiceConfig
+from .config import VocabulariesServiceConfig, VocabularyTypesServiceConfig
+from .service import VocabulariesService, VocabularyTypeService
 
 __all__ = (
     "VocabulariesService",
+    "VocabularyTypeService",
     "VocabulariesServiceConfig",
+    "VocabularyTypesServiceConfig",
 )

invenio_vocabularies/services/config.py

@@ -0,0 +1,179 @@
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2024 CERN.
+# Copyright (C) 2024 University of Münster.
+#
+# Invenio-Vocabularies is free software; you can redistribute it and/or
+# modify it under the terms of the MIT License; see LICENSE file for more
+# details.
+
+"""Vocabulary services configs."""
+
+import sqlalchemy as sa
+from flask import current_app
+from invenio_i18n import lazy_gettext as _
+from invenio_records_resources.services import (
+    Link,
+    LinksTemplate,
+    RecordService,
+    RecordServiceConfig,
+    SearchOptions,
+    pagination_links,
+)
+from invenio_records_resources.services.base import (
+    ConditionalLink,
+    Service,
+    ServiceListResult,
+)
+from invenio_records_resources.services.records.components import DataComponent
+from invenio_records_resources.services.records.params import (
+    FilterParam,
+    SuggestQueryParser,
+)
+
+from ..records.api import Vocabulary
+from ..records.models import VocabularyType
+from . import results
+from .components import PIDComponent, VocabularyTypeComponent
+from .permissions import PermissionPolicy
+from .schema import TaskSchema, VocabularySchema
+
+
+def is_custom_vocabulary_type(vocabulary_type, context):
+    """Check if the vocabulary type is a custom vocabulary type."""
+    return vocabulary_type["id"] in current_app.config.get(
+        "VOCABULARIES_CUSTOM_VOCABULARY_TYPES", []
+    )
+
+
+class VocabularySearchOptions(SearchOptions):
+    """Search options for vocabularies."""
+
+    params_interpreters_cls = [
+        FilterParam.factory(param="tags", field="tags"),
+    ] + SearchOptions.params_interpreters_cls
+
+    suggest_parser_cls = SuggestQueryParser.factory(
+        fields=[
+            "id.text^100",
+            "id.text._2gram",
+            "id.text._3gram",
+            "title.en^5",
+            "title.en._2gram",
+            "title.en._3gram",
+        ],
+    )
+
+    sort_default = "bestmatch"
+
+    sort_default_no_query = "title"
+
+    sort_options = {
+        "bestmatch": dict(
+            title=_("Best match"),
+            fields=["_score"],  # ES defaults to desc on `_score` field
+        ),
+        "title": dict(
+            title=_("Title"),
+            fields=["title_sort"],
+        ),
+        "newest": dict(
+            title=_("Newest"),
+            fields=["-created"],
+        ),
+        "oldest": dict(
+            title=_("Oldest"),
+            fields=["created"],
+        ),
+    }
+
+
+class VocabularyTypeSearchOptions(SearchOptions):
+    """Search options for vocabulary types."""
+
+    sort_options = {
+        "id": dict(
+            title=_("ID"),
+            fields=["id"],
+        ),
+    }
+
+    sort_default = "id"
+
+    sort_default_no_query = "id"
+
+    sort_direction_options = {
+        "asc": dict(title=_("Ascending"), fn=sa.asc),
+        "desc": dict(title=_("Descending"), fn=sa.desc),
+    }
+
+    sort_direction_default = "asc"
+
+
+class VocabulariesServiceConfig(RecordServiceConfig):
+    """Vocabulary service configuration."""
+
+    service_id = "vocabularies"
+    indexer_queue_name = "vocabularies"
+    permission_policy_cls = PermissionPolicy
+    record_cls = Vocabulary
+    schema = VocabularySchema
+    task_schema = TaskSchema
+
+    search = VocabularySearchOptions
+
+    components = [
+        # Order of components are important!
+        VocabularyTypeComponent,
+        DataComponent,
+        PIDComponent,
+    ]
+
+    links_item = {
+        "self": Link(
+            "{+api}/vocabularies/{type}/{id}",
+            vars=lambda record, vars: vars.update(
+                {
+                    "id": record.pid.pid_value,
+                    "type": record.type.id,
+                }
+            ),
+        ),
+    }
+
+    links_search = pagination_links("{+api}/vocabularies/{type}{?args*}")
+
+
+class VocabularyTypesServiceConfig(RecordServiceConfig):
+    """Vocabulary types service configuration."""
+
+    service_id = "vocabulary_types"
+    permission_policy_cls = PermissionPolicy
+    record_cls = VocabularyType
+    schema = VocabularySchema  # Works but should be VocabularyTypeSchema if this is defined at some point
+    result_list_cls = results.VocabularyTypeList
+
+    links_item = {
+        "self": ConditionalLink(
+            cond=is_custom_vocabulary_type,
+            if_=Link(
+                "{+api}/{id}",
+                vars=lambda vocab_type, vars: vars.update({"id": vocab_type["id"]}),
+            ),
+            else_=Link(
+                "{+api}/vocabularies/{id}",
+                vars=lambda vocab_type, vars: vars.update({"id": vocab_type["id"]}),
+            ),
+        )
+    }
+
+    search = VocabularyTypeSearchOptions
+
+    components = [
+        # Order of components are important!
+        VocabularyTypeComponent,
+        DataComponent,
+        PIDComponent,
+    ]
+
+    links_search = pagination_links("{+api}/vocabularies/{type}{?args*}")

invenio_vocabularies/services/custom_fields/__init__.py

@@ -1,12 +1,16 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright (C) 2022 CERN.
+# Copyright (C) 2022-2024 CERN.
 #
 # Invenio-RDM-Records is free software; you can redistribute it and/or modify
 # it under the terms of the MIT License; see LICENSE file for more details.
 
 """Custom Fields for InvenioRDM."""
 
+from .subject import SUBJECT_FIELDS, SUBJECT_FIELDS_UI
 from .vocabulary import VocabularyCF
 
-__all__ = "VocabularyCF"
+__all__ = [
+    "VocabularyCF",
+    "SUBJECT_FIELDS_UI" "SUBJECT_FIELDS",
+]

invenio_vocabularies/services/custom_fields/subject.py

@@ -0,0 +1,82 @@
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2024 CERN.
+#
+# Invenio-RDM-Records is free software; you can redistribute it and/or modify
+# it under the terms of the MIT License; see LICENSE file for more details.
+
+
+"""Custom fields."""
+
+from invenio_i18n import lazy_gettext as _
+
+from ...contrib.subjects.api import Subject
+from ...contrib.subjects.schema import SubjectRelationSchema
+from .vocabulary import VocabularyCF
+
+
+class SubjectCF(VocabularyCF):
+    """Custom field for subjects."""
+
+    field_keys = ["id", "subject"]
+
+    def __init__(self, **kwargs):
+        """Constructor."""
+        super().__init__(
+            vocabulary_id="subjects",
+            schema=SubjectRelationSchema,
+            ui_schema=SubjectRelationSchema,
+            **kwargs,
+        )
+        self.pid_field = Subject.pid
+
+    @property
+    def mapping(self):
+        """Return the mapping."""
+        _mapping = {
+            "type": "object",
+            "properties": {
+                "@v": {"type": "keyword"},
+                "id": {"type": "keyword"},
+                "subject": {"type": "keyword"},
+            },
+        }
+
+        return _mapping
+
+
+SUBJECT_FIELDS_UI = [
+    {
+        "section": _("Subjects"),
+        "fields": [
+            dict(
+                field="subjects",
+                ui_widget="SubjectAutocompleteDropdown",
+                isGenericVocabulary=False,
+                props=dict(
+                    label="Keywords and subjects",
+                    icon="tag",
+                    description="The subjects related to the community",
+                    placeholder="Search for a subject by name e.g. Psychology ...",
+                    autocompleteFrom="api/subjects",
+                    noQueryMessage="Search for subjects...",
+                    autocompleteFromAcceptHeader="application/vnd.inveniordm.v1+json",
+                    required=False,
+                    multiple=True,
+                    clearable=True,
+                    allowAdditions=False,
+                ),
+                template="invenio_vocabularies/subjects.html",
+            )
+        ],
+    }
+]
+
+
+SUBJECT_FIELDS = {
+    SubjectCF(
+        name="subjects",
+        multiple=True,
+        dump_options=False,
+    )
+}

invenio_vocabularies/services/custom_fields/vocabulary.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright (C) 2022 CERN.
+# Copyright (C) 2022-2024 CERN.
 #
 # Invenio-RDM-Records is free software; you can redistribute it and/or modify
 # it under the terms of the MIT License; see LICENSE file for more details.
@@ -12,6 +12,7 @@ from invenio_records_resources.services.custom_fields.base import BaseCF
 from marshmallow import fields
 
 from ...proxies import current_service
+from ...records.api import Vocabulary
 from ...resources.serializer import VocabularyL10NItemSchema
 from ...services.schema import VocabularyRelationSchema
 
@@ -38,7 +39,7 @@ class VocabularyCF(BaseCF):
         sort_by=None,
         schema=VocabularyRelationSchema,
         ui_schema=VocabularyL10NItemSchema,
-        **kwargs
+        **kwargs,
     ):
         """Constructor."""
         super().__init__(name, **kwargs)
@@ -49,6 +50,7 @@ class VocabularyCF(BaseCF):
         self.sort_by = sort_by
         self.schema = schema
         self.ui_schema = ui_schema
+        self.pid_field = Vocabulary.pid.with_type_ctx(self.vocabulary_id)
 
     @property
     def mapping(self):
@@ -58,7 +60,7 @@ class VocabularyCF(BaseCF):
             "properties": {
                 "@v": {"type": "keyword"},
                 "id": {"type": "keyword"},
-                "title": {"type": "object", "dynamic": True},
+                "title": {"type": "object", "dynamic": "true"},
             },
         }
 

invenio_vocabularies/services/permissions.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 #
-# Copyright (C) 2020-2021 CERN.
+# Copyright (C) 2020-2024 CERN.
 #
 # Invenio-Vocabularies is free software; you can redistribute it and/or
 # modify it under the terms of the MIT License; see LICENSE file for more
@@ -21,3 +21,5 @@ class PermissionPolicy(RecordPermissionPolicy):
     can_update = [SystemProcess()]
     can_delete = [SystemProcess()]
     can_manage = [SystemProcess()]
+    # this permission is needed for the /api/vocabularies/ endpoint
+    can_list_vocabularies = [SystemProcess(), AnyUser()]

invenio_vocabularies/services/results.py

@@ -0,0 +1,110 @@
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2024 CERN.
+# Copyright (C) 2024 Uni Münster.
+#
+# Invenio-Vocabularies is free software; you can redistribute it and/or
+# modify it under the terms of the MIT License; see LICENSE file for more
+# details.
+
+"""Vocabulary results."""
+
+from flask import current_app
+from invenio_records_resources.proxies import current_service_registry
+from invenio_records_resources.services import RecordServiceConfig
+from invenio_records_resources.services.records.results import RecordList
+from invenio_search import current_search_client
+
+from ..proxies import current_service
+
+
+class VocabularyTypeList(RecordList):
+    """Ensures that vocabulary type metadata is returned in the proper format."""
+
+    @property
+    def total(self):
+        """Get total number of hits."""
+        return self._results.total
+
+    @property
+    def custom_vocabulary_names(self):
+        """Checks whether vocabulary is a custom vocabulary."""
+        return current_app.config.get("VOCABULARIES_CUSTOM_VOCABULARY_TYPES", [])
+
+    def to_dict(self):
+        """Formats result to a dict of hits."""
+        config_vocab_types = current_app.config.get(
+            "INVENIO_VOCABULARY_TYPE_METADATA", {}
+        )
+
+        count_terms_agg = {}
+        generic_stats = self._generic_vocabulary_statistics()
+        custom_stats = self._custom_vocabulary_statistics()
+
+        for k in generic_stats.keys() | custom_stats.keys():
+            count_terms_agg[k] = generic_stats.get(k, 0) + custom_stats.get(k, 0)
+
+        hits = self._results.items
+
+        # Extend database data with configuration & aggregation data.
+        results = []
+        for db_vocab_type in hits:
+            result = {
+                "id": db_vocab_type.id,
+                "pid_type": db_vocab_type.pid_type,
+                "count": count_terms_agg.get(db_vocab_type.id, 0),
+                "is_custom_vocabulary": db_vocab_type.id
+                in self.custom_vocabulary_names,
+            }
+
+            if db_vocab_type.id in config_vocab_types:
+                for k, v in config_vocab_types[db_vocab_type.id].items():
+                    result[k] = v
+
+            results.append(result)
+
+        for hit in results:
+            if self._links_item_tpl:
+                hit["links"] = self._links_item_tpl.expand(self._identity, hit)
+
+        res = {
+            "hits": {
+                "hits": results,
+                "total": self.total,
+            }
+        }
+
+        if self._params:
+            if self._links_tpl:
+                res["links"] = self._links_tpl.expand(self._identity, self.pagination)
+
+        return res
+
+    def _custom_vocabulary_statistics(self):
+        # query database for count of terms in custom vocabularies
+        returndict = {}
+        for vocab_type in self.custom_vocabulary_names:
+            custom_service = current_service_registry.get(vocab_type)
+            record_cls = custom_service.config.record_cls
+            returndict[vocab_type] = record_cls.model_cls.query.count()
+
+        return returndict
+
+    def _generic_vocabulary_statistics(self):
+        # Opensearch query for generic vocabularies
+        config: RecordServiceConfig = (
+            current_service.config
+        )  # TODO: Where to get the config from here? current_service is None
+        search_opts = config.search
+
+        search = search_opts.search_cls(
+            using=current_search_client,
+            index=config.record_cls.index.search_alias,
+        )
+
+        search.aggs.bucket("vocabularies", {"terms": {"field": "type.id", "size": 100}})
+
+        search_result = search.execute()
+        buckets = search_result.aggs.to_dict()["vocabularies"]["buckets"]
+
+        return {bucket["key"]: bucket["doc_count"] for bucket in buckets}