nucliadb-models 6.9.5.post5452__py3-none-any.whl → 6.10.0.post5694__py3-none-any.whl

nucliadb_models/hydration.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 #
-from typing import Annotated, Optional, Union
+from typing import Annotated
 
 from pydantic import BaseModel, Field, StringConstraints
 
@@ -105,23 +105,23 @@ class GenericFieldHydration(BaseModel, extra="forbid"):
 
 
 class FieldHydration(BaseModel, extra="forbid"):
-    text: Optional[TextFieldHydration] = Field(
+    text: TextFieldHydration | None = Field(
         default_factory=TextFieldHydration,
         description="Text fields hydration options",
     )
-    file: Optional[FileFieldHydration] = Field(
+    file: FileFieldHydration | None = Field(
         default_factory=FileFieldHydration,
         description="File fields hydration options",
     )
-    link: Optional[LinkFieldHydration] = Field(
+    link: LinkFieldHydration | None = Field(
         default_factory=LinkFieldHydration,
         description="Link fields hydration options",
     )
-    conversation: Optional[ConversationFieldHydration] = Field(
+    conversation: ConversationFieldHydration | None = Field(
         default_factory=ConversationFieldHydration,
         description="Conversation fields hydration options",
     )
-    generic: Optional[GenericFieldHydration] = Field(
+    generic: GenericFieldHydration | None = Field(
         default_factory=GenericFieldHydration,
         description="Generic fields hydration options",
     )
@@ -141,7 +141,7 @@ class NeighbourParagraphHydration(BaseModel, extra="forbid"):
 
 
 class RelatedParagraphHydration(BaseModel, extra="forbid"):
-    neighbours: Optional[NeighbourParagraphHydration] = Field(
+    neighbours: NeighbourParagraphHydration | None = Field(
         default=None,
         description="Hydrate extra paragraphs that surround the original one",
     )
@@ -205,11 +205,11 @@ class ParagraphHydration(BaseModel, extra="forbid"):
         default=True,
         description="Hydrate paragraph text",
     )
-    image: Optional[ImageParagraphHydration] = Field(
+    image: ImageParagraphHydration | None = Field(
         default=None,
         description="Hydrate options for paragraphs extracted from images (using OCR, inception...)",
     )
-    table: Optional[TableParagraphHydration] = Field(
+    table: TableParagraphHydration | None = Field(
         default=None,
         description="Hydrate options for paragraphs extracted from tables",
     )
@@ -217,19 +217,19 @@ class ParagraphHydration(BaseModel, extra="forbid"):
     # TODO: at some point, we should add hydration options for paragraphs from
     # audio and video
 
-    page: Optional[ParagraphPageHydration] = Field(
+    page: ParagraphPageHydration | None = Field(
         default=None,
         description="Hydrte options for paragraphs within a page. This applies to paragraphs in fields with pages",
     )
 
-    related: Optional[RelatedParagraphHydration] = Field(
+    related: RelatedParagraphHydration | None = Field(
         default=None,
         description="Hydration options for related paragraphs. For example, neighbours or sibling paragraphs",
     )
 
 
 class Hydration(BaseModel, extra="forbid"):
-    resource: Optional[ResourceHydration] = Field(
+    resource: ResourceHydration | None = Field(
         default_factory=ResourceHydration,
         description="Resource hydration options",
     )
@@ -270,22 +270,22 @@ class HydratedResource(BaseModel, extra="forbid"):
     id: str = Field(description="Unique resource id")
     slug: str = Field(description="Resource slug")
 
-    title: Optional[str] = None
-    summary: Optional[str] = None
+    title: str | None = None
+    summary: str | None = None
 
-    origin: Optional[Origin] = None
+    origin: Origin | None = None
 
-    security: Optional[ResourceSecurity] = None
+    security: ResourceSecurity | None = None
 
     # TODO: add resource labels to hydrated resources
 
 
 class FieldExtractedData(BaseModel, extra="forbid"):
-    text: Optional[str] = None
+    text: str | None = None
 
 
 class SplitFieldExtractedData(BaseModel, extra="forbid"):
-    texts: Optional[dict[str, str]] = None
+    texts: dict[str, str] | None = None
 
 
 class HydratedTextField(BaseModel, extra="forbid"):
@@ -293,8 +293,8 @@ class HydratedTextField(BaseModel, extra="forbid"):
     resource: str = Field("Field resource id")
     field_type: FieldTypeName = FieldTypeName.TEXT
 
-    value: Optional[FieldText] = None
-    extracted: Optional[FieldExtractedData] = None
+    value: FieldText | None = None
+    extracted: FieldExtractedData | None = None
 
 
 class HydratedFileField(BaseModel, extra="forbid"):
@@ -302,10 +302,10 @@ class HydratedFileField(BaseModel, extra="forbid"):
     resource: str = Field("Field resource id")
     field_type: FieldTypeName = FieldTypeName.FILE
 
-    value: Optional[FieldFile] = None
-    extracted: Optional[FieldExtractedData] = None
+    value: FieldFile | None = None
+    extracted: FieldExtractedData | None = None
 
-    previews: Optional[dict[str, Image]] = Field(
+    previews: dict[str, Image] | None = Field(
         default=None,
         title="Previews of specific parts of the field",
         description=(
@@ -323,8 +323,8 @@ class HydratedLinkField(BaseModel, extra="forbid"):
     resource: str = Field("Field resource id")
     field_type: FieldTypeName = FieldTypeName.LINK
 
-    value: Optional[FieldLink] = None
-    extracted: Optional[FieldExtractedData] = None
+    value: FieldLink | None = None
+    extracted: FieldExtractedData | None = None
 
 
 class HydratedConversationField(BaseModel, extra="forbid"):
@@ -332,8 +332,8 @@ class HydratedConversationField(BaseModel, extra="forbid"):
     resource: str = Field("Field resource id")
     field_type: FieldTypeName = FieldTypeName.CONVERSATION
 
-    value: Optional[FieldConversation] = None
-    extracted: Optional[FieldExtractedData] = None
+    value: FieldConversation | None = None
+    extracted: FieldExtractedData | None = None
 
 
 class HydratedGenericField(BaseModel, extra="forbid"):
@@ -341,24 +341,24 @@ class HydratedGenericField(BaseModel, extra="forbid"):
     resource: str = Field("Field resource id")
     field_type: FieldTypeName = FieldTypeName.TEXT
 
-    value: Optional[str] = None
-    extracted: Optional[FieldExtractedData] = None
+    value: str | None = None
+    extracted: FieldExtractedData | None = None
 
 
 class RelatedNeighbourParagraphRefs(BaseModel, extra="forbid"):
-    before: Optional[list[str]] = None
-    after: Optional[list[str]] = None
+    before: list[str] | None = None
+    after: list[str] | None = None
 
 
 class RelatedParagraphRefs(BaseModel, extra="forbid"):
-    neighbours: Optional[RelatedNeighbourParagraphRefs] = None
-    parents: Optional[list[str]] = None
-    siblings: Optional[list[str]] = None
-    replacements: Optional[list[str]] = None
+    neighbours: RelatedNeighbourParagraphRefs | None = None
+    parents: list[str] | None = None
+    siblings: list[str] | None = None
+    replacements: list[str] | None = None
 
 
 class HydratedParagraphImage(BaseModel, extra="forbid"):
-    source_image: Optional[Image] = Field(
+    source_image: Image | None = Field(
         default=None,
         description=(
             "Source image for this paragraph. This only applies to paragraphs "
@@ -369,7 +369,7 @@ class HydratedParagraphImage(BaseModel, extra="forbid"):
 
 
 class HydratedParagraphTable(BaseModel, extra="forbid"):
-    page_preview_ref: Optional[str] = Field(
+    page_preview_ref: str | None = Field(
         default=None,
         description=(
             "Referento to the page preview for this paragraph. The actual "
@@ -381,7 +381,7 @@ class HydratedParagraphTable(BaseModel, extra="forbid"):
 
 
 class HydratedParagraphPage(BaseModel, extra="forbid"):
-    page_preview_ref: Optional[str] = Field(
+    page_preview_ref: str | None = Field(
         default=None,
         description=(
             "Reference to the page preview for this paragraph. The actual "
@@ -398,28 +398,26 @@ class HydratedParagraph(BaseModel, extra="forbid"):
     field: str = Field(description="Paragraph field id")
     resource: str = Field(description="Paragraph resource id")
 
-    text: Optional[str] = None
+    text: str | None = None
 
     # TODO: add labels to hydrated paragraphs
     # labels: Optional[list[str]] = None
 
-    related: Optional[RelatedParagraphRefs] = None
+    related: RelatedParagraphRefs | None = None
 
-    image: Optional[HydratedParagraphImage] = None
-    table: Optional[HydratedParagraphTable] = None
-    page: Optional[HydratedParagraphPage] = None
+    image: HydratedParagraphImage | None = None
+    table: HydratedParagraphTable | None = None
+    page: HydratedParagraphPage | None = None
 
 
 class Hydrated(BaseModel, extra="forbid"):
     resources: dict[str, HydratedResource]
     fields: dict[
         str,
-        Union[
-            HydratedTextField,
-            HydratedFileField,
-            HydratedLinkField,
-            HydratedConversationField,
-            HydratedGenericField,
-        ],
+        HydratedTextField
+        | HydratedFileField
+        | HydratedLinkField
+        | HydratedConversationField
+        | HydratedGenericField,
     ]
     paragraphs: dict[str, HydratedParagraph]

nucliadb_models/internal/predict.py

@@ -19,13 +19,11 @@ Models for Predict API v1.
 ATENTION! Keep these models in sync with models on Predict API
 """
 
-from typing import List, Optional
-
 from pydantic import BaseModel, Field
 
 
 class SentenceSearch(BaseModel):
-    vectors: dict[str, List[float]] = Field(
+    vectors: dict[str, list[float]] = Field(
         default_factory=dict,
         description="Sentence vectors for each semantic model",
         min_length=1,
@@ -45,14 +43,14 @@ class Ner(BaseModel):
 
 
 class TokenSearch(BaseModel):
-    tokens: List[Ner] = []
+    tokens: list[Ner] = []
     time: float
     input_tokens: int = 0
 
 
 class QueryInfo(BaseModel):
-    language: Optional[str]
-    stop_words: List[str] = Field(default_factory=list)
+    language: str | None
+    stop_words: list[str] = Field(default_factory=list)
     semantic_thresholds: dict[str, float] = Field(
         default_factory=dict,
         description="Semantic threshold for each semantic model",
@@ -60,10 +58,10 @@ class QueryInfo(BaseModel):
     )
     visual_llm: bool
     max_context: int
-    entities: Optional[TokenSearch]
-    sentence: Optional[SentenceSearch]
+    entities: TokenSearch | None
+    sentence: SentenceSearch | None
     query: str
-    rephrased_query: Optional[str] = None
+    rephrased_query: str | None = None
 
 
 class RerankModel(BaseModel):

nucliadb_models/internal/shards.py

@@ -13,7 +13,6 @@
 # limitations under the License.
 #
 from enum import Enum
-from typing import List, Optional
 
 from pydantic import BaseModel
 
@@ -58,9 +57,9 @@ class ShardReplica(BaseModel):
 
 class ShardObject(BaseModel):
     shard: str
-    nidx_shard_id: Optional[str]
+    nidx_shard_id: str | None
 
 
 class KnowledgeboxShards(BaseModel):
     kbid: str
-    shards: List[ShardObject]
+    shards: list[ShardObject]

nucliadb_models/labels.py

@@ -14,9 +14,8 @@
 #
 
 from enum import Enum
-from typing import Dict, List, Optional
 
-from pydantic import BaseModel, model_validator
+from pydantic import BaseModel, Field, model_validator
 from typing_extensions import Self
 
 BASE_LABELS: dict[str, set[str]] = {
@@ -96,18 +95,26 @@ class LabelSetKind(str, Enum):
 
 
 class Label(BaseModel):
-    title: str
-    related: Optional[str] = None
-    text: Optional[str] = None
-    uri: Optional[str] = None
+    title: str = Field(
+        description="Title of the label. This is the display name for the label shown in the UI and also used for searching."
+    )
+    related: str | None = None
+    text: str | None = None
+    uri: str | None = None
 
 
 class LabelSet(BaseModel):
-    title: Optional[str] = None
-    color: Optional[str] = "blue"
+    title: str | None = Field(
+        default=None,
+        description="Title of the labelset. It is a prettier display name for the labelset shown in the UI but it is not intended to be used for searching.",
+    )
+    color: str | None = "blue"
     multiple: bool = True
-    kind: List[LabelSetKind] = []
-    labels: List[Label] = []
+    kind: list[LabelSetKind] = []
+    labels: list[Label] = Field(
+        default_factory=list,
+        description="List of labels in the labelset. The titles of the labels must be unique within the labelset.",
+    )
 
     @model_validator(mode="after")
     def check_unique_labels(self) -> Self:
@@ -123,4 +130,4 @@ class LabelSet(BaseModel):
 
 class KnowledgeBoxLabels(BaseModel):
     uuid: str
-    labelsets: Dict[str, LabelSet] = {}
+    labelsets: dict[str, LabelSet] = {}

nucliadb_models/link.py

@@ -13,7 +13,6 @@
 # limitations under the License.
 #
 from datetime import datetime
-from typing import Dict, Optional
 
 from pydantic import BaseModel, Field
 
@@ -25,19 +24,19 @@ from pydantic import BaseModel, Field
 
 
 class FieldLink(BaseModel):
-    added: Optional[datetime] = None
-    headers: Optional[Dict[str, str]] = None
-    cookies: Optional[Dict[str, str]] = None
-    uri: Optional[str] = None
-    language: Optional[str] = None
-    localstorage: Optional[Dict[str, str]] = None
-    css_selector: Optional[str] = None
-    xpath: Optional[str] = None
-    extract_strategy: Optional[str] = Field(
+    added: datetime | None = None
+    headers: dict[str, str] | None = None
+    cookies: dict[str, str] | None = None
+    uri: str | None = None
+    language: str | None = None
+    localstorage: dict[str, str] | None = None
+    css_selector: str | None = None
+    xpath: str | None = None
+    extract_strategy: str | None = Field(
         default=None,
         description="Id of the Nuclia extract strategy used at processing time. If not set, the default strategy was used. Extract strategies are defined at the learning configuration api.",
     )
-    split_strategy: Optional[str] = Field(
+    split_strategy: str | None = Field(
         default=None,
         description="Id of the Nuclia split strategy used at processing time. If not set, the default strategy was used. Split strategies are defined at the learning configuration api.",
     )
@@ -47,18 +46,18 @@ class FieldLink(BaseModel):
 
 
 class LinkField(BaseModel):
-    headers: Optional[Dict[str, str]] = {}
-    cookies: Optional[Dict[str, str]] = {}
+    headers: dict[str, str] | None = {}
+    cookies: dict[str, str] | None = {}
     uri: str
-    language: Optional[str] = None
-    localstorage: Optional[Dict[str, str]] = {}
-    css_selector: Optional[str] = None
-    xpath: Optional[str] = None
-    extract_strategy: Optional[str] = Field(
+    language: str | None = None
+    localstorage: dict[str, str] | None = {}
+    css_selector: str | None = None
+    xpath: str | None = None
+    extract_strategy: str | None = Field(
         default=None,
         description="Id of the Nuclia extract strategy to use at processing time. If not set, the default strategy will be used. Extract strategies are defined at the learning configuration api.",
     )
-    split_strategy: Optional[str] = Field(
+    split_strategy: str | None = Field(
         default=None,
         description="Id of the Nuclia split strategy used at processing time. If not set, the default strategy was used. Split strategies are defined at the learning configuration api.",
     )

nucliadb_models/metadata.py

@@ -15,7 +15,7 @@
 import warnings
 from datetime import datetime
 from enum import Enum
-from typing import Any, Dict, List, Optional
+from typing import Any
 
 from pydantic import BaseModel, Field, field_validator, model_validator
 from typing_extensions import Self
@@ -49,7 +49,7 @@ class RelationNodeType(str, Enum):
 class RelationEntity(BaseModel):
     value: str
     type: RelationNodeType
-    group: Optional[str] = None
+    group: str | None = None
 
     @model_validator(mode="after")
     def check_relation_is_valid(self) -> Self:
@@ -60,20 +60,20 @@ class RelationEntity(BaseModel):
 
 
 class RelationMetadata(BaseModel):
-    paragraph_id: Optional[str] = None
-    source_start: Optional[int] = None
-    source_end: Optional[int] = None
-    to_start: Optional[int] = None
-    to_end: Optional[int] = None
-    data_augmentation_task_id: Optional[str] = None
+    paragraph_id: str | None = None
+    source_start: int | None = None
+    source_end: int | None = None
+    to_start: int | None = None
+    to_end: int | None = None
+    data_augmentation_task_id: str | None = None
 
 
 class Relation(BaseModel):
     relation: RelationType
-    label: Optional[str] = None
-    metadata: Optional[RelationMetadata] = None
+    label: str | None = None
+    metadata: RelationMetadata | None = None
 
-    from_: Optional[RelationEntity] = Field(default=None, alias="from")
+    from_: RelationEntity | None = Field(default=None, alias="from")
     to: RelationEntity
 
     @model_validator(mode="after")
@@ -100,9 +100,9 @@ class Relation(BaseModel):
 
 
 class InputMetadata(BaseModel):
-    metadata: Dict[str, str] = {}
-    language: Optional[str] = None
-    languages: Optional[List[str]] = Field(default=None, max_length=1024)
+    metadata: dict[str, str] = {}
+    language: str | None = None
+    languages: list[str] | None = Field(default=None, max_length=1024)
 
 
 class ResourceProcessingStatus(Enum):
@@ -120,7 +120,7 @@ class Metadata(InputMetadata):
 
 class FieldClassification(BaseModel):
     field: FieldID
-    classifications: List[Classification] = []
+    classifications: list[Classification] = []
 
 
 class ComputedMetadata(BaseModel):
@@ -129,12 +129,12 @@ class ComputedMetadata(BaseModel):
     without having to load the whole computed metadata field.
     """
 
-    field_classifications: List[FieldClassification] = []
+    field_classifications: list[FieldClassification] = []
 
 
 class UserMetadata(BaseModel):
-    classifications: List[UserClassification] = []
-    relations: List[Relation] = []
+    classifications: list[UserClassification] = []
+    relations: list[Relation] = []
 
 
 class TokenSplit(BaseModel):
@@ -154,7 +154,7 @@ class TokenSplit(BaseModel):
 
 
 class ParagraphAnnotation(BaseModel):
-    classifications: List[UserClassification] = []
+    classifications: list[UserClassification] = []
     key: str
 
 
@@ -169,12 +169,12 @@ class VisualSelection(BaseModel):
     left: float
     right: float
     bottom: float
-    token_ids: List[int]
+    token_ids: list[int]
 
 
 class PageSelections(BaseModel):
     page: int
-    visual: List[VisualSelection]
+    visual: list[VisualSelection]
 
     def __init__(self, **data):
         warnings.warn(
@@ -190,57 +190,69 @@ class UserFieldMetadata(BaseModel):
     Field-level metadata set by the user via the rest api
     """
 
-    paragraphs: List[ParagraphAnnotation] = []
-    question_answers: List[QuestionAnswerAnnotation] = []
+    paragraphs: list[ParagraphAnnotation] = []
+    question_answers: list[QuestionAnswerAnnotation] = []
     field: FieldID
 
 
 class Basic(BaseModel):
-    icon: Optional[str] = None
-    title: Optional[str] = None
-    summary: Optional[str] = None
-    thumbnail: Optional[str] = None
-    created: Optional[datetime] = None
-    modified: Optional[datetime] = None
-    metadata: Optional[Metadata] = None
-    usermetadata: Optional[UserMetadata] = None
-    fieldmetadata: Optional[List[UserFieldMetadata]] = None
-    computedmetadata: Optional[ComputedMetadata] = None
-    uuid: Optional[str] = None
-    last_seqid: Optional[int] = None
-    last_account_seq: Optional[int] = None
+    icon: str | None = None
+    title: str | None = None
+    summary: str | None = None
+    thumbnail: str | None = None
+    created: datetime | None = None
+    modified: datetime | None = None
+    metadata: Metadata | None = None
+    usermetadata: UserMetadata | None = None
+    fieldmetadata: list[UserFieldMetadata] | None = None
+    computedmetadata: ComputedMetadata | None = None
+    uuid: str | None = None
+    last_seqid: int | None = None
+    last_account_seq: int | None = None
+
+
+class SyncMetadata(BaseModel):
+    file_id: str = Field(description="Identifier of the file in the origin cloud storage system")
+    auth_provider: str = Field(
+        description="Authentication provider used to access the origin cloud storage system"
+    )
 
 
 class InputOrigin(BaseModel):
-    source_id: Optional[str] = None
-    url: Optional[str] = None
-    created: Optional[DateTime] = Field(
+    source_id: str | None = None
+    url: str | None = None
+    created: DateTime | None = Field(
         default=None,
         description="Creation date of the resource at the origin system. This can be later used for date range filtering on search endpoints. Have a look at the advanced search documentation page: https://docs.nuclia.dev/docs/rag/advanced/search/#date-filtering",
     )
-    modified: Optional[DateTime] = Field(
+    modified: DateTime | None = Field(
         default=None,
         description="Modification date of the resource at the origin system. This can be later used for date range filtering on search endpoints.  Have a look at the advanced search documentation page: https://docs.nuclia.dev/docs/rag/advanced/search/#date-filtering",
     )
-    metadata: Dict[str, str] = Field(
+    metadata: dict[str, str] = Field(
         default={},
         title="Metadata",
         description="Generic metadata from the resource at the origin system. It can later be used for filtering on search endpoints with '/origin.metadata/{key}/{value}'",
     )
-    tags: List[str] = Field(
+    tags: list[str] = Field(
         default=[],
         title="Tags",
         description="Resource tags about the origin system. It can later be used for filtering on search endpoints with '/origin.tags/{tag}'",
         max_length=300,
     )
-    collaborators: List[str] = Field(default=[], max_length=100)
-    filename: Optional[str] = None
-    related: List[str] = Field(default=[], max_length=100)
-    path: Optional[str] = Field(
+    collaborators: list[str] = Field(default=[], max_length=100)
+    filename: str | None = None
+    related: list[str] = Field(default=[], max_length=100)
+    path: str | None = Field(
         default=None,
         description="Path of the original resource. Typically used to store folder structure information of the resource at the origin system. It can be later used for filtering on search endpoints with '/origin.path/{path}'",
         max_length=2048,
     )
+    sync_metadata: SyncMetadata | None = Field(
+        default=None,
+        title="Sync Metadata",
+        description="Metadata related to the resource from the origin system fetched by the Progress Agentic RAG's Cloud Storage Sync service.",
+    )
 
     @field_validator("tags")
     def validate_tag_length(cls, tags):
@@ -253,10 +265,10 @@ class InputOrigin(BaseModel):
 class Origin(InputOrigin):
     # Created and modified are redefined to
     # use native datetime objects and skip validation
-    created: Optional[datetime] = None
-    modified: Optional[datetime] = None
+    created: datetime | None = None
+    modified: datetime | None = None
 
-    tags: List[str] = Field(
+    tags: list[str] = Field(
         default=[],
         title="Tags",
         description="Resource tags about the origin system. It can later be used for filtering on search endpoints with '/origin.tags/{tag}'",
@@ -268,16 +280,16 @@ class Origin(InputOrigin):
         API = "API"
         PYSDK = "PYSDK"
 
-    source: Optional[Source] = Source.API
+    source: Source | None = Source.API
 
 
 class Extra(BaseModel):
-    metadata: Dict[Any, Any] = Field(
+    metadata: dict[Any, Any] = Field(
         ...,
         title="Metadata",
-        description="Arbitrary JSON metadata provided by the user that is not meant to be searchable, but can be serialized on results.",  # noqa
+        description="Arbitrary JSON metadata provided by the user that is not meant to be searchable, but can be serialized on results.",
     )
 
 
 class Relations(BaseModel):
-    relations: Optional[List[Relation]] = None
+    relations: list[Relation] | None = None