nucliadb-models 6.9.2.post5254__py3-none-any.whl → 6.9.3.post5346__py3-none-any.whl

nucliadb_models/conversation.py

@@ -86,7 +86,7 @@ class FieldConversation(BaseModel):
 
 
 class InputMessageContent(BaseModel):
-    text: str = Field(max_length=10 * 1024)
+    text: str = Field()
     format: MessageFormat = MessageFormat.PLAIN
     attachments: List[FileB64] = Field(default=[], max_length=50)
     attachments_fields: List[FieldRef] = Field(default=[], max_length=50)
@@ -129,7 +129,6 @@ class InputConversationField(BaseModel):
     messages: List[InputMessage] = Field(
         default_factory=list,
         description="List of messages in the conversation field. Each message must have a unique ident. A single conversation can contain up to 51,200 messages. You can add up to 2,048 messages per request.",
-        max_length=2048,
     )
     extract_strategy: Optional[str] = Field(
         default=None,

nucliadb_models/hydration.py

@@ -246,10 +246,11 @@ class Hydration(BaseModel, extra="forbid"):
 ParagraphId = Annotated[
     str,
     StringConstraints(
-        pattern=r"^[0-9a-f]{32}/[acftu]/[a-zA-Z0-9:_-]+/[0-9]+-[0-9]+$",
-        min_length=32 + 1 + 1 + 1 + 1 + 1 + 3,
+        # resource-uuid/field-type/field-id/[split-id/]paragraph-id
+        pattern=r"^[0-9a-f]{32}/[acftu]/[a-zA-Z0-9:_-]+/([^/]{1,128}/)?[0-9]+-[0-9]+$",
+        min_length=32 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 3,
         # max field id of 250 and 10 digit paragraphs. More than enough
-        max_length=32 + 1 + 1 + 1 + 250 + 1 + 21,
+        max_length=32 + 1 + 1 + 1 + 250 + 1 + 128 + 1 + 21,
     ),
 ]
 

nucliadb_models/search.py

@@ -79,8 +79,9 @@ ANSWER_JSON_SCHEMA_EXAMPLE = {
 class ModelParamDefaults:
     applied_autofilters = ParamDefault(
         default=[],
-        title="Autofilters",
-        description="List of filters automatically applied to the search query",
+        title="Applied autofilters",
+        description="[deprecated] list of filters automatically applied to the search query",
+        deprecated=True,
     )
 
 
@@ -296,6 +297,8 @@ class KnowledgeboxSearchResults(JsonBaseModel):
     relations: Optional[Relations] = None
     nodes: Optional[list[dict[str, str]]] = None
     shards: Optional[list[str]] = None
+
+    # TODO: remove on a future major release
     autofilters: list[str] = ModelParamDefaults.applied_autofilters.to_pydantic_field()
 
 
@@ -481,12 +484,6 @@ class SearchParamDefaults:
         description="The list of facets to calculate. The facets follow the same syntax as filters: https://docs.nuclia.dev/docs/rag/advanced/search-filters",  # noqa: E501
         max_items=50,
     )
-    autofilter = ParamDefault(
-        default=False,
-        title="Automatic search filtering",
-        description="[Deprecated] If set to true, the search will automatically add filters to the query. For example, it will filter results containing the entities detected in the query",  # noqa: E501
-        deprecated=True,
-    )
     chat_query = ParamDefault(
         default=...,
         title="Query",
@@ -877,7 +874,9 @@ class BaseSearchRequest(AuditMetadataBase):
     vectorset: Optional[str] = SearchParamDefaults.vectorset.to_pydantic_field()
     with_duplicates: bool = SearchParamDefaults.with_duplicates.to_pydantic_field()
     with_synonyms: bool = SearchParamDefaults.with_synonyms.to_pydantic_field()
-    autofilter: bool = SearchParamDefaults.autofilter.to_pydantic_field()
+    # autofilter is deprecated and its logic was removed. We're just keeping it in the model definition to
+    # avoid breaking changes in the python sdks. Please remove on a future major release.
+    autofilter: SkipJsonSchema[bool] = False
     resource_filters: list[str] = SearchParamDefaults.resource_filters.to_pydantic_field()
     security: Optional[RequestSecurity] = SearchParamDefaults.security.to_pydantic_field()
     show_hidden: bool = SearchParamDefaults.show_hidden.to_pydantic_field()
@@ -1171,7 +1170,7 @@ ALLOWED_FIELD_TYPES: dict[str, str] = {
     "t": "text",
     "f": "file",
     "u": "link",
-    "d": "datetime",
+    "c": "conversation",
     "a": "generic",
 }
 
@@ -1179,16 +1178,19 @@ ALLOWED_FIELD_TYPES: dict[str, str] = {
 class FieldExtensionStrategy(RagStrategy):
     name: Literal["field_extension"] = "field_extension"
     fields: list[str] = Field(
+        default=[],
         title="Fields",
-        description="List of field ids to extend the context with. It will try to extend the retrieval context with the specified fields in the matching resources. The field ids have to be in the format `{field_type}/{field_name}`, like 'a/title', 'a/summary' for title and summary fields or 't/amend' for a text field named 'amend'.",  # noqa: E501
-        min_length=1,
+        description="List of field ids to extend the context with. It will try to extend the retrieval context with the specified fields in the matching resources. The field ids have to be in the format `{field_type}/{field_name}`, like 'a/title', 'a/summary' for title and summary fields or 't/amend' for a text field named 'amend'.",
+    )
+    data_augmentation_field_prefixes: list[str] = Field(
+        default=[],
+        description="List of prefixes for data augmentation added fields to extend the context with. For example, if the prefix is 'simpson', all fields that are a result of data augmentation with that prefix will be used to extend the context.",
     )
 
-    @field_validator("fields", mode="after")
-    @classmethod
-    def fields_validator(cls, fields) -> Self:
+    @model_validator(mode="after")
+    def field_extension_strategy_validator(self) -> Self:
         # Check that the fields are in the format {field_type}/{field_name}
-        for field in fields:
+        for field in self.fields:
             try:
                 field_type, _ = field.strip("/").split("/")
             except ValueError:
@@ -1201,8 +1203,7 @@ class FieldExtensionStrategy(RagStrategy):
                     f"Field '{field}' does not have a valid field type. "
                     f"Valid field types are: {allowed_field_types_part}."
                 )
-
-        return fields
+        return self
 
 
 class FullResourceApplyTo(BaseModel):
@@ -1620,7 +1621,11 @@ class AskRequest(AuditMetadataBase):
         description="Image that will be used together with the query text for retrieval and then sent to the LLM as part of the context. "
         "If a query image is provided, the `extra_context_images` and `rag_images_strategies` will be disabled.",
     )
-    autofilter: bool = SearchParamDefaults.autofilter.to_pydantic_field()
+
+    # autofilter is deprecated and its logic was removed. We're just keeping it in the model definition to
+    # avoid breaking changes in the python sdks. Please remove on a future major release.
+    autofilter: SkipJsonSchema[bool] = False
+
     highlight: bool = SearchParamDefaults.highlight.to_pydantic_field()
     resource_filters: list[str] = SearchParamDefaults.resource_filters.to_pydantic_field()
     prompt: Optional[Union[str, CustomPrompt]] = Field(

{nucliadb_models-6.9.2.post5254.dist-info → nucliadb_models-6.9.3.post5346.dist-info}/METADATA

@@ -1,19 +1,18 @@
 Metadata-Version: 2.4
 Name: nucliadb_models
-Version: 6.9.2.post5254
+Version: 6.9.3.post5346
 Author-email: Nuclia <nucliadb@nuclia.com>
 License-Expression: Apache-2.0
 Project-URL: Homepage, https://nuclia.com
 Project-URL: Repository, https://github.com/nuclia/nucliadb
 Classifier: Development Status :: 4 - Beta
 Classifier: Programming Language :: Python
-Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3 :: Only
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
-Requires-Python: <4,>=3.9
+Requires-Python: <4,>=3.10
 Description-Content-Type: text/markdown
 Requires-Dist: pydantic!=2.11.5,!=2.11.6,>=2.6
 

{nucliadb_models-6.9.2.post5254.dist-info → nucliadb_models-6.9.3.post5346.dist-info}/RECORD

@@ -2,14 +2,14 @@ nucliadb_models/__init__.py,sha256=3y8-htogKuCZcbhaUZdSjTeEjUSeec9aRWyL8AlKCyM,1
 nucliadb_models/common.py,sha256=2dtKG4ZNi9p-yoNY76Uvyu1SlMeNYpH-MnuU3Q6w9Js,8169
 nucliadb_models/configuration.py,sha256=BBrJsNjP324Cw_5J3dBrGwvpkHQYbXEo3TUaI9IqAOg,2449
 nucliadb_models/content_types.py,sha256=36Ga-iGf4ivCqgtXC7imFgegrwHB117s9eqP62JtGv0,3456
-nucliadb_models/conversation.py,sha256=Ts-h7gMf18rw9tbjF2X__vNtMshHRDqbLua_VXb1qOE,5126
+nucliadb_models/conversation.py,sha256=k9bKhkDiqhqmdrDfDPNoUfG7-2H_-KAyuOnETd8zV0E,5081
 nucliadb_models/entities.py,sha256=i-7Y8qmFRRTih5zw0ajv1U_iiXexe66M3TK8hUikQZk,2356
 nucliadb_models/export_import.py,sha256=mNm9IArOLnC6TLupkwqVFhxD5d08mpIVOVFneECv8UA,1073
 nucliadb_models/external_index_providers.py,sha256=pL3leo4MkuJOnKlU1Sg6GT_mnK_VUBxGui-RPmDYVWU,1126
 nucliadb_models/extracted.py,sha256=Owz7LC3le3Dvau3TtRiO8NY84meOf6IxN-RrOqqpMPs,5593
 nucliadb_models/file.py,sha256=tXtgB9c7i2ADsnJ7HdbXyroAmXadGvOeA49htBh7BZo,2263
 nucliadb_models/filters.py,sha256=NQI2-4AFzzJuZy8NeY3jXlTbbU5wxiwMCP-5DrD-7lE,14759
-nucliadb_models/hydration.py,sha256=7SFnAcTQRE9etVccpph6aA1AUqsHVwkzT4YF6Uzl0Gs,14262
+nucliadb_models/hydration.py,sha256=Yo8eM99q9wcaCJj8stL5z3k63ouC3LVANZIZLULZxRQ,14363
 nucliadb_models/labels.py,sha256=9zqRgkpZuX3kUPwsTTgCH7JyOWK7dM5pwyuHJR86YdU,3949
 nucliadb_models/link.py,sha256=PF5hHLwdOed5TMBTxtokkgWtMh1bFnORZjybh0NwVCw,2526
 nucliadb_models/metadata.py,sha256=OOKGy_83NtlG1QKQZEwMuwu4wbVEe7P30Y2QvnGSDto,8933
@@ -17,7 +17,7 @@ nucliadb_models/notifications.py,sha256=mna8-AoD_29Wds0Thl0AF0zpERnJmYGLZX1w1fUo
 nucliadb_models/processing.py,sha256=nhKuHQjqCdb9zJVkYGPTLub23tK9e_lwL5OCDVymZjY,719
 nucliadb_models/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nucliadb_models/resource.py,sha256=RzCos0QRgSMkaV-p7EoceSmt7UTzt9G9be5BKF-iGrQ,9021
-nucliadb_models/search.py,sha256=tUedQpOY1OKnhCybnU-0dae3f6QjoN3rDVhwgDgLj5Y,96008
+nucliadb_models/search.py,sha256=gQEXJ9bXXcxswr7aOzvBeGIQlrq5TgRWIqTxKEbSoCE,96409
 nucliadb_models/security.py,sha256=opxaDLfvk3aU0sjesK0jGrYLx5h4YCwlKKN0moYs_ig,1150
 nucliadb_models/synonyms.py,sha256=afbaVqSQSxGLwi2PusVaLSRpkOtA5AZmWOKd1f4nl2E,690
 nucliadb_models/text.py,sha256=60bxZnOjRHnDdezR8VfR3AZsXTOwePFPs2BKB8wxBak,3414
@@ -33,7 +33,7 @@ nucliadb_models/graph/responses.py,sha256=Sdq8OgFAL1YT-1lJyLLrkqcScvj7YTEqAUwQ-k
 nucliadb_models/internal/__init__.py,sha256=zG33bUz1rHFPtvqQPWn4rDwBJt3FJodGuQYD45quiQg,583
 nucliadb_models/internal/predict.py,sha256=Pnx6MmLfK65eExe1XnVxqmSlvMwdowewwks9BOEoqMw,2029
 nucliadb_models/internal/shards.py,sha256=__y1OZtWGiNcPQEWfSFOj8yw458WGi7mM4vZe0K-L1Y,1691
-nucliadb_models-6.9.2.post5254.dist-info/METADATA,sha256=tm8qgHEYjjOdfu1xJeFlvKgEKhqRrBwKSBEfOQqZEpg,794
-nucliadb_models-6.9.2.post5254.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-nucliadb_models-6.9.2.post5254.dist-info/top_level.txt,sha256=UrY1I8oeovIRwkXLYplssTrxQdUjhSEFDFbnwaIV3tA,16
-nucliadb_models-6.9.2.post5254.dist-info/RECORD,,
+nucliadb_models-6.9.3.post5346.dist-info/METADATA,sha256=BOTbOylsqvvrFfcED7AD3qnYGEBib8Q66Q0amWDcXDw,745
+nucliadb_models-6.9.3.post5346.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+nucliadb_models-6.9.3.post5346.dist-info/top_level.txt,sha256=UrY1I8oeovIRwkXLYplssTrxQdUjhSEFDFbnwaIV3tA,16
+nucliadb_models-6.9.3.post5346.dist-info/RECORD,,