nucliadb 2.46.1.post382__py3-none-any.whl → 6.2.1.post2777__py3-none-any.whl

nucliadb/reader/api/models.py

@@ -22,15 +22,12 @@ from typing import TYPE_CHECKING, Any, Optional, Union
 from pydantic import BaseModel
 
 import nucliadb_models as models
-from nucliadb_models.common import FIELD_TYPES_MAP, FieldTypeName
+from nucliadb_models.common import FieldTypeName
 from nucliadb_models.resource import (
     ConversationFieldExtractedData,
-    DatetimeFieldExtractedData,
     Error,
     ExtractedDataType,
     FileFieldExtractedData,
-    KeywordsetFieldExtractedData,
-    LayoutFieldExtractedData,
     LinkFieldExtractedData,
     TextFieldExtractedData,
 )
@@ -41,10 +38,7 @@ if TYPE_CHECKING:  # pragma: no cover
             models.FieldText,
             models.FieldFile,
             models.FieldLink,
-            models.FieldLayout,
             models.Conversation,
-            models.FieldKeywordset,
-            models.FieldDatetime,
         ]
     ]
 else:
@@ -55,19 +49,14 @@ else:
 class ResourceField(BaseModel):
     field_type: FieldTypeName
     field_id: str
-    value: ValueType
-    extracted: ExtractedDataType
+    value: ValueType = None
+    extracted: Optional[ExtractedDataType] = None
     error: Optional[Error] = None
 
 
-FIELD_NAMES_TO_PB_TYPE_MAP = {v: k for k, v in FIELD_TYPES_MAP.items()}
-
 FIELD_NAME_TO_EXTRACTED_DATA_FIELD_MAP: dict[FieldTypeName, Any] = {
     FieldTypeName.TEXT: TextFieldExtractedData,
     FieldTypeName.FILE: FileFieldExtractedData,
     FieldTypeName.LINK: LinkFieldExtractedData,
-    FieldTypeName.DATETIME: DatetimeFieldExtractedData,
-    FieldTypeName.KEYWORDSET: KeywordsetFieldExtractedData,
-    FieldTypeName.LAYOUT: LayoutFieldExtractedData,
     FieldTypeName.CONVERSATION: ConversationFieldExtractedData,
 }

nucliadb/reader/api/v1/__init__.py

@@ -20,7 +20,6 @@
 from . import download  # noqa
 from . import export_import  # noqa
 from . import knowledgebox  # noqa
-from . import learning_collector  # noqa
 from . import learning_config  # noqa
 from . import resource  # noqa
 from . import services  # noqa

nucliadb/reader/api/v1/download.py

@@ -28,14 +28,14 @@ from fastapi_versioning import version
 from starlette.datastructures import Headers
 from starlette.responses import StreamingResponse
 
-from nucliadb.ingest.orm.resource import FIELD_TYPE_TO_ID
+from nucliadb.common.ids import FIELD_TYPE_PB_TO_STR
+from nucliadb.common.models_utils import to_proto
 from nucliadb.ingest.serialize import get_resource_uuid_by_slug
 from nucliadb.reader import SERVICE_NAME, logger
-from nucliadb.reader.api.models import FIELD_NAMES_TO_PB_TYPE_MAP
 from nucliadb_models.common import FieldTypeName
 from nucliadb_models.resource import NucliaDBRoles
 from nucliadb_utils.authentication import requires_one
-from nucliadb_utils.storages.storage import StorageField  # type: ignore
+from nucliadb_utils.storages.storage import ObjectMetadata, Range, StorageField
 from nucliadb_utils.utilities import get_storage
 
 from .router import KB_PREFIX, RESOURCE_PREFIX, RSLUG_PREFIX, api
@@ -50,7 +50,7 @@ class DownloadType(Enum):
     f"/{KB_PREFIX}/{{kbid}}/{RSLUG_PREFIX}/{{rslug}}/{{field_type}}/{{field_id}}/download/extracted/{{download_field:path}}",  # noqa
     tags=["Resource fields"],
     status_code=200,
-    name="Download extracted binary file (by slug)",
+    summary="Download extracted binary file (by slug)",
 )
 @requires_one([NucliaDBRoles.READER])
 @version(1)
@@ -62,16 +62,14 @@ async def download_extract_file_rslug_prefix(
     field_id: str,
     download_field: str,
 ) -> Response:
-    return await _download_extract_file(
-        request, kbid, field_type, field_id, download_field, rslug=rslug
-    )
+    return await _download_extract_file(request, kbid, field_type, field_id, download_field, rslug=rslug)
 
 
 @api.get(
     f"/{KB_PREFIX}/{{kbid}}/{RESOURCE_PREFIX}/{{rid}}/{{field_type}}/{{field_id}}/download/extracted/{{download_field:path}}",  # noqa
     tags=["Resource fields"],
     status_code=200,
-    name="Download extracted binary file (by id)",
+    summary="Download extracted binary file (by id)",
 )
 @requires_one([NucliaDBRoles.READER])
 @version(1)
@@ -83,9 +81,7 @@ async def download_extract_file_rid_prefix(
     field_id: str,
     download_field: str,
 ) -> Response:
-    return await _download_extract_file(
-        request, kbid, field_type, field_id, download_field, rid=rid
-    )
+    return await _download_extract_file(request, kbid, field_type, field_id, download_field, rid=rid)
 
 
 async def _download_extract_file(
@@ -101,8 +97,8 @@ async def _download_extract_file(
 
     storage = await get_storage(service_name=SERVICE_NAME)
 
-    pb_field_type = FIELD_NAMES_TO_PB_TYPE_MAP[field_type]
-    field_type_letter = FIELD_TYPE_TO_ID[pb_field_type]
+    pb_field_type = to_proto.field_type_name(field_type)
+    field_type_letter = FIELD_TYPE_PB_TO_STR[pb_field_type]
 
     sf = storage.file_extracted(kbid, rid, field_type_letter, field_id, download_field)
 
@@ -113,7 +109,7 @@ async def _download_extract_file(
     f"/{KB_PREFIX}/{{kbid}}/{RSLUG_PREFIX}/{{rslug}}/file/{{field_id}}/download/field",
     tags=["Resource fields"],
     status_code=200,
-    name="Download field binary field (by slug)",
+    summary="Download field binary field (by slug)",
 )
 @requires_one([NucliaDBRoles.READER])
 @version(1)
@@ -124,16 +120,14 @@ async def download_field_file_rslug_prefix(
     field_id: str,
     inline: bool = False,
 ) -> Response:
-    return await _download_field_file(
-        request, kbid, field_id, rslug=rslug, inline=inline
-    )
+    return await _download_field_file(request, kbid, field_id, rslug=rslug, inline=inline)
 
 
 @api.get(
     f"/{KB_PREFIX}/{{kbid}}/{RESOURCE_PREFIX}/{{rid}}/file/{{field_id}}/download/field",
     tags=["Resource fields"],
     status_code=200,
-    name="Download field binary field (by id)",
+    summary="Download field binary field (by id)",
 )
 @requires_one([NucliaDBRoles.READER])
 @version(1)
@@ -164,68 +158,11 @@ async def _download_field_file(
     return await download_api(sf, request.headers, inline=inline)
 
 
-@api.get(
-    f"/{KB_PREFIX}/{{kbid}}/{RSLUG_PREFIX}/{{rslug}}/layout/{{field_id}}/download/field/{{download_field}}",
-    tags=["Resource fields"],
-    status_code=200,
-    name="Download layout binary field (by slug)",
-)
-@requires_one([NucliaDBRoles.READER])
-@version(1)
-async def download_field_layout_rslug_prefix(
-    request: Request,
-    kbid: str,
-    rslug: str,
-    field_id: str,
-    download_field: str,
-) -> Response:
-    return await _download_field_layout(
-        request, kbid, field_id, download_field, rslug=rslug
-    )
-
-
-@api.get(
-    f"/{KB_PREFIX}/{{kbid}}/{RESOURCE_PREFIX}/{{rid}}/layout/{{field_id}}/download/field/{{download_field}}",
-    tags=["Resource fields"],
-    status_code=200,
-    name="Download layout binary field (by id)",
-)
-@requires_one([NucliaDBRoles.READER])
-@version(1)
-async def download_field_layout_rid_prefix(
-    request: Request,
-    kbid: str,
-    rid: str,
-    field_id: str,
-    download_field: str,
-) -> Response:
-    return await _download_field_layout(
-        request, kbid, field_id, download_field, rid=rid
-    )
-
-
-async def _download_field_layout(
-    request: Request,
-    kbid: str,
-    field_id: str,
-    download_field: str,
-    rid: Optional[str] = None,
-    rslug: Optional[str] = None,
-) -> Response:
-    rid = await _get_resource_uuid_from_params(kbid, rid, rslug)
-
-    storage = await get_storage(service_name=SERVICE_NAME)
-
-    sf = storage.layout_field(kbid, rid, field_id, download_field)
-
-    return await download_api(sf, request.headers)
-
-
 @api.get(
     f"/{KB_PREFIX}/{{kbid}}/{RSLUG_PREFIX}/{{rslug}}/conversation/{{field_id}}/download/field/{{message_id}}/{{file_num}}",  # noqa
     tags=["Resource fields"],
     status_code=200,
-    name="Download conversation binary field (by slug)",
+    summary="Download conversation binary field (by slug)",
 )
 @requires_one([NucliaDBRoles.READER])
 @version(1)
@@ -237,16 +174,14 @@ async def download_field_conversation_rslug_prefix(
     message_id: str,
     file_num: int,
 ) -> Response:
-    return await _download_field_conversation(
-        request, kbid, field_id, message_id, file_num, rslug=rslug
-    )
+    return await _download_field_conversation(request, kbid, field_id, message_id, file_num, rslug=rslug)
 
 
 @api.get(
     f"/{KB_PREFIX}/{{kbid}}/{RESOURCE_PREFIX}/{{rid}}/conversation/{{field_id}}/download/field/{{message_id}}/{{file_num}}",  # noqa
     tags=["Resource fields"],
     status_code=200,
-    name="Download conversation binary field (by id)",
+    summary="Download conversation binary field (by id)",
 )
 @requires_one([NucliaDBRoles.READER])
 @version(1)
@@ -258,9 +193,7 @@ async def download_field_conversation_rid_prefix(
     message_id: str,
     file_num: int,
 ) -> Response:
-    return await _download_field_conversation(
-        request, kbid, field_id, message_id, file_num, rid=rid
-    )
+    return await _download_field_conversation(request, kbid, field_id, message_id, file_num, rid=rid)
 
 
 async def _download_field_conversation(
@@ -282,13 +215,13 @@ async def _download_field_conversation(
 
 
 async def download_api(sf: StorageField, headers: Headers, inline: bool = False):
-    metadata = await sf.exists()
+    metadata: Optional[ObjectMetadata] = await sf.exists()
     if metadata is None:
         raise HTTPException(status_code=404, detail="Specified file doesn't exist")
 
-    file_size = int(metadata.get("SIZE", -1))
-    content_type = metadata.get("CONTENT_TYPE", "application/octet-stream")
-    filename = metadata.get("FILENAME", "file")
+    file_size = metadata.size or -1
+    content_type = metadata.content_type or "application/octet-stream"
+    filename = metadata.filename or "file"
     filename = safe_http_header_encode(filename)
 
     status_code = 200
@@ -299,8 +232,10 @@ async def download_api(sf: StorageField, headers: Headers, inline: bool = False)
         "Content-Type": content_type,
         "Content-Disposition": content_disposition,
     }
-    download_headers = {}
+
+    range = Range()
     if "range" in headers and file_size > -1:
+        status_code = 206
         range_request = headers["range"]
         try:
             start, end, range_size = parse_media_range(range_request, file_size)
@@ -341,23 +276,21 @@ async def download_api(sf: StorageField, headers: Headers, inline: bool = False)
                 headers={"Content-Range": f"bytes */{file_size}"},
                 status_code=416,
             )
-        status_code = 206
         logger.debug(f"Range request: {range_request}")
         extra_headers["Content-Length"] = f"{range_size}"
         extra_headers["Content-Range"] = f"bytes {start}-{end}/{file_size}"
-        download_headers["Range"] = range_request
+        range.start = start
+        range.end = end
 
     return StreamingResponse(
-        sf.storage.download(sf.bucket, sf.key, headers=download_headers),  # type: ignore
+        sf.storage.download(sf.bucket, sf.key, range=range),
         status_code=status_code,
         media_type=content_type,
         headers=extra_headers,
     )
 
 
-async def _get_resource_uuid_from_params(
-    kbid, rid: Optional[str], rslug: Optional[str]
-) -> str:
+async def _get_resource_uuid_from_params(kbid, rid: Optional[str], rslug: Optional[str]) -> str:
     if not any([rid, rslug]):
         raise ValueError("Either rid or slug must be set")
 

nucliadb/reader/api/v1/export_import.py

@@ -41,7 +41,7 @@ from nucliadb_utils.authentication import requires_one
 @api.get(
     f"/{KB_PREFIX}/{{kbid}}/export/{{export_id}}",
     status_code=200,
-    name="Download a Knowledge Box export",
+    summary="Download a Knowledge Box export",
     tags=["Knowledge Boxes"],
     response_class=StreamingResponse,
 )
@@ -100,7 +100,7 @@ async def download_export_and_delete(
 @api.get(
     f"/{KB_PREFIX}/{{kbid}}/export/{{export_id}}/status",
     status_code=200,
-    name="Get the status of a Knowledge Box Export",
+    summary="Get the status of a Knowledge Box Export",
     response_model=StatusResponse,
     tags=["Knowledge Boxes"],
 )
@@ -119,7 +119,7 @@ async def get_export_status_endpoint(
 @api.get(
     f"/{KB_PREFIX}/{{kbid}}/import/{{import_id}}/status",
     status_code=200,
-    name="Get the status of a Knowledge Box Import",
+    summary="Get the status of a Knowledge Box Import",
     response_model=StatusResponse,
     tags=["Knowledge Boxes"],
 )
@@ -160,5 +160,5 @@ async def _get_status(
 
 
 async def exists_kb(kbid: str) -> bool:
-    async with datamanagers.with_transaction(read_only=True) as txn:
+    async with datamanagers.with_ro_transaction() as txn:
         return await datamanagers.kb.exists_kb(txn, kbid=kbid)

nucliadb/reader/api/v1/knowledgebox.py

@@ -18,14 +18,14 @@
 # along with this program. If not, see <http://www.gnu.org/licenses/>.
 #
 from fastapi import HTTPException
-from fastapi_versioning import version  # type: ignore
+from fastapi_versioning import version
 from starlette.requests import Request
 
 from nucliadb.common import datamanagers
 from nucliadb.common.maindb.utils import get_driver
+from nucliadb.common.models_utils import from_proto
 from nucliadb.reader.api.v1.router import KB_PREFIX, KBS_PREFIX, api
 from nucliadb_models.resource import (
-    KnowledgeBoxConfig,
     KnowledgeBoxList,
     KnowledgeBoxObj,
     KnowledgeBoxObjSummary,
@@ -37,7 +37,7 @@ from nucliadb_utils.authentication import requires, requires_one
 @api.get(
     f"/{KBS_PREFIX}",
     status_code=200,
-    name="List Knowledge Boxes",
+    summary="List Knowledge Boxes",
     response_model=KnowledgeBoxList,
     tags=["Knowledge Boxes"],
     include_in_schema=False,
@@ -46,17 +46,17 @@ from nucliadb_utils.authentication import requires, requires_one
 @version(1)
 async def get_kbs(request: Request, prefix: str = "") -> KnowledgeBoxList:
     driver = get_driver()
-    async with driver.transaction() as txn:
+    async with driver.transaction(read_only=True) as txn:
         response = KnowledgeBoxList()
         async for kbid, slug in datamanagers.kb.get_kbs(txn, prefix=prefix):
-            response.kbs.append(KnowledgeBoxObjSummary(slug=slug or None, uuid=kbid))  # type: ignore
+            response.kbs.append(KnowledgeBoxObjSummary(slug=slug or None, uuid=kbid))
         return response
 
 
 @api.get(
     f"/{KB_PREFIX}/{{kbid}}",
     status_code=200,
-    name="Get Knowledge Box",
+    summary="Get Knowledge Box",
     response_model=KnowledgeBoxObj,
     tags=["Knowledge Boxes"],
 )
@@ -64,22 +64,22 @@ async def get_kbs(request: Request, prefix: str = "") -> KnowledgeBoxList:
 @version(1)
 async def get_kb(request: Request, kbid: str) -> KnowledgeBoxObj:
     driver = get_driver()
-    async with driver.transaction() as txn:
+    async with driver.transaction(read_only=True) as txn:
         kb_config = await datamanagers.kb.get_config(txn, kbid=kbid)
         if kb_config is None:
             raise HTTPException(status_code=404, detail="Knowledge Box does not exist")
 
         return KnowledgeBoxObj(
             uuid=kbid,
-            slug=kb_config.slug,  # type: ignore
-            config=KnowledgeBoxConfig.from_message(kb_config),
+            slug=kb_config.slug,
+            config=from_proto.knowledgebox_config(kb_config),
         )
 
 
 @api.get(
     f"/{KB_PREFIX}/s/{{slug}}",
     status_code=200,
-    name="Get Knowledge Box (by slug)",
+    summary="Get Knowledge Box (by slug)",
     response_model=KnowledgeBoxObj,
     tags=["Knowledge Boxes"],
 )
@@ -87,7 +87,7 @@ async def get_kb(request: Request, kbid: str) -> KnowledgeBoxObj:
 @version(1)
 async def get_kb_by_slug(request: Request, slug: str) -> KnowledgeBoxObj:
     driver = get_driver()
-    async with driver.transaction() as txn:
+    async with driver.transaction(read_only=True) as txn:
         kbid = await datamanagers.kb.get_kb_uuid(txn, slug=slug)
         if kbid is None:
             raise HTTPException(status_code=404, detail="Knowledge Box does not exist")
@@ -98,6 +98,6 @@ async def get_kb_by_slug(request: Request, slug: str) -> KnowledgeBoxObj:
 
         return KnowledgeBoxObj(
             uuid=kbid,
-            slug=kb_config.slug,  # type: ignore
-            config=KnowledgeBoxConfig.from_message(kb_config),
+            slug=kb_config.slug,
+            config=from_proto.knowledgebox_config(kb_config),
         )

nucliadb/reader/api/v1/learning_config.py

@@ -31,7 +31,7 @@ from nucliadb_utils.settings import is_onprem_nucliadb
 @api.get(
     path=f"/{KB_PREFIX}/{{kbid}}/models/{{model_id}}/{{filename:path}}",
     status_code=200,
-    name="Download the Knowledege Box model",
+    summary="Download the Knowledege Box model",
     description="Download the trained model or any other generated file as a result of a training task on a Knowledge Box.",  # noqa
     response_model=None,
     tags=["Models"],
@@ -44,15 +44,13 @@ async def download_model(
     model_id: str,
     filename: str,
 ):
-    return await learning_config_proxy(
-        request, "GET", f"/download/{kbid}/model/{model_id}/{filename}"
-    )
+    return await learning_config_proxy(request, "GET", f"/download/{kbid}/model/{model_id}/{filename}")
 
 
 @api.get(
     path=f"/{KB_PREFIX}/{{kbid}}/configuration",
     status_code=200,
-    name="Get Knowledge Box models configuration",
+    summary="Get Knowledge Box models configuration",
     description="Current configuration of models assigned to a Knowledge Box",
     response_model=None,
     tags=["Models"],
@@ -74,7 +72,7 @@ async def get_configuration(
 @api.get(
     path=f"/{KB_PREFIX}/{{kbid}}/models",
     status_code=200,
-    name="Get available models",
+    summary="Get available models",
     description="Get available models",
     response_model=None,
     tags=["Models"],
@@ -91,7 +89,7 @@ async def get_models(
 @api.get(
     path=f"/{KB_PREFIX}/{{kbid}}/model/{{model_id}}",
     status_code=200,
-    name="Get model metadata",
+    summary="Get model metadata",
     description="Get metadata for a particular model",
     response_model=None,
     tags=["Models"],
@@ -114,7 +112,7 @@ async def get_model(
 @api.get(
     path=f"/{KB_PREFIX}/{{kbid}}/schema",
     status_code=200,
-    name="Learning configuration schema",
+    summary="Learning configuration schema",
     description="Get jsonschema definition to update the `learning_configuration` of your Knowledge Box",
     response_model=None,
     tags=["Models"],
@@ -131,7 +129,7 @@ async def get_schema_for_configuration_updates(
 @api.get(
     path=f"/nua/schema",
     status_code=200,
-    name="Learning configuration schema for Knowledge Box creation",
+    summary="Learning configuration schema for Knowledge Box creation",
     description="Get jsonschema definition for `learning_configuration` field for the Knowledge Box creation payload",
     response_model=None,
     tags=["Models"],
@@ -143,7 +141,5 @@ async def get_schema_for_configuration_creation(
     request: Request,
 ):
     if not is_onprem_nucliadb():
-        return HTTPClientError(
-            status_code=404, detail="Endpoint not available for Hosted NucliaDB"
-        )
+        return HTTPClientError(status_code=404, detail="Endpoint not available for Hosted NucliaDB")
     return await learning_config_proxy(request, "GET", f"/schema")