knowhere-python-sdk 0.3.0__py3-none-any.whl → 0.3.1__py3-none-any.whl

knowhere/__init__.py

@@ -39,6 +39,9 @@ from knowhere.types.document import Document, DocumentListResponse
 from knowhere.types.job import Job, JobError, JobProgress, JobResult
 from knowhere.types.params import ParsingParams, WebhookConfig
 from knowhere.types.retrieval import (
+    RetrievalChannel,
+    RetrievalFilterMode,
+    RetrievalSectionExclusion,
     RetrievalSource,
     RetrievalQueryResponse,
     RetrievalResult,
@@ -97,6 +100,9 @@ __all__: list[str] = [
     "Document",
     "DocumentListResponse",
     # Retrieval types
+    "RetrievalChannel",
+    "RetrievalFilterMode",
+    "RetrievalSectionExclusion",
     "RetrievalSource",
     "RetrievalQueryResponse",
     "RetrievalResult",

knowhere/_version.py

@@ -1 +1 @@
-__version__ = "0.3.0"  # x-release-please-version
+__version__ = "0.3.1"  # x-release-please-version

knowhere/resources/jobs.py

@@ -145,8 +145,12 @@ class Jobs(SyncAPIResource):
             if not job_result.result_url:
                 raise InvalidStateError("JobResult does not have a result_url.")
             result_url: str = job_result.result_url
+            namespace: Optional[str] = job_result.namespace
+            document_id: Optional[str] = job_result.document_id
         else:
             result_url = job_result
+            namespace = None
+            document_id = None
 
         response: httpx.Response = self._client._client.get(
             result_url, timeout=self._client.upload_timeout
@@ -154,7 +158,10 @@ class Jobs(SyncAPIResource):
         response.raise_for_status()
         zip_bytes: bytes = response.content
 
-        return parseResultZip(zip_bytes, verify_checksum=verify_checksum)
+        parsed_result = parseResultZip(zip_bytes, verify_checksum=verify_checksum)
+        parsed_result.namespace = namespace
+        parsed_result.document_id = document_id
+        return parsed_result
 
 
 class AsyncJobs(AsyncAPIResource):
@@ -251,8 +258,12 @@ class AsyncJobs(AsyncAPIResource):
             if not job_result.result_url:
                 raise InvalidStateError("JobResult does not have a result_url.")
             result_url: str = job_result.result_url
+            namespace: Optional[str] = job_result.namespace
+            document_id: Optional[str] = job_result.document_id
         else:
             result_url = job_result
+            namespace = None
+            document_id = None
 
         response: httpx.Response = await self._client._client.get(
             result_url, timeout=self._client.upload_timeout
@@ -260,4 +271,7 @@ class AsyncJobs(AsyncAPIResource):
         response.raise_for_status()
         zip_bytes: bytes = response.content
 
-        return parseResultZip(zip_bytes, verify_checksum=verify_checksum)
+        parsed_result = parseResultZip(zip_bytes, verify_checksum=verify_checksum)
+        parsed_result.namespace = namespace
+        parsed_result.document_id = document_id
+        return parsed_result

knowhere/resources/retrieval.py

@@ -5,7 +5,12 @@ from __future__ import annotations
 from typing import Any, Dict, Optional
 
 from knowhere.resources._base import AsyncAPIResource, SyncAPIResource
-from knowhere.types.retrieval import RetrievalQueryResponse
+from knowhere.types.retrieval import (
+    RetrievalChannel,
+    RetrievalFilterMode,
+    RetrievalQueryResponse,
+    RetrievalSectionExclusion,
+)
 
 
 class Retrieval(SyncAPIResource):
@@ -17,8 +22,16 @@ class Retrieval(SyncAPIResource):
         query: str,
         namespace: Optional[str] = None,
         top_k: Optional[int] = None,
+        data_type: Optional[int] = None,
+        signal_paths: Optional[list[str]] = None,
+        filter_mode: Optional[RetrievalFilterMode] = None,
+        channels: Optional[list[RetrievalChannel]] = None,
+        channel_weights: Optional[dict[RetrievalChannel, float]] = None,
+        rerank: Optional[bool] = None,
+        threshold: Optional[float] = None,
+        internal_recall_k: Optional[int] = None,
         exclude_document_ids: Optional[list[str]] = None,
-        exclude_sections: Optional[list[dict[str, str]]] = None,
+        exclude_sections: Optional[list[RetrievalSectionExclusion]] = None,
     ) -> RetrievalQueryResponse:
         """Query published documents in a namespace."""
         body: Dict[str, Any] = {"query": query}
@@ -26,6 +39,22 @@ class Retrieval(SyncAPIResource):
             body["namespace"] = namespace
         if top_k is not None:
             body["top_k"] = top_k
+        if data_type is not None:
+            body["data_type"] = data_type
+        if signal_paths is not None:
+            body["signal_paths"] = signal_paths
+        if filter_mode is not None:
+            body["filter_mode"] = filter_mode
+        if channels is not None:
+            body["channels"] = channels
+        if channel_weights is not None:
+            body["channel_weights"] = channel_weights
+        if rerank is not None:
+            body["rerank"] = rerank
+        if threshold is not None:
+            body["threshold"] = threshold
+        if internal_recall_k is not None:
+            body["internal_recall_k"] = internal_recall_k
         if exclude_document_ids is not None:
             body["exclude_document_ids"] = exclude_document_ids
         if exclude_sections is not None:
@@ -48,8 +77,16 @@ class AsyncRetrieval(AsyncAPIResource):
         query: str,
         namespace: Optional[str] = None,
         top_k: Optional[int] = None,
+        data_type: Optional[int] = None,
+        signal_paths: Optional[list[str]] = None,
+        filter_mode: Optional[RetrievalFilterMode] = None,
+        channels: Optional[list[RetrievalChannel]] = None,
+        channel_weights: Optional[dict[RetrievalChannel, float]] = None,
+        rerank: Optional[bool] = None,
+        threshold: Optional[float] = None,
+        internal_recall_k: Optional[int] = None,
         exclude_document_ids: Optional[list[str]] = None,
-        exclude_sections: Optional[list[dict[str, str]]] = None,
+        exclude_sections: Optional[list[RetrievalSectionExclusion]] = None,
     ) -> RetrievalQueryResponse:
         """Query published documents in a namespace."""
         body: Dict[str, Any] = {"query": query}
@@ -57,6 +94,22 @@ class AsyncRetrieval(AsyncAPIResource):
             body["namespace"] = namespace
         if top_k is not None:
             body["top_k"] = top_k
+        if data_type is not None:
+            body["data_type"] = data_type
+        if signal_paths is not None:
+            body["signal_paths"] = signal_paths
+        if filter_mode is not None:
+            body["filter_mode"] = filter_mode
+        if channels is not None:
+            body["channels"] = channels
+        if channel_weights is not None:
+            body["channel_weights"] = channel_weights
+        if rerank is not None:
+            body["rerank"] = rerank
+        if threshold is not None:
+            body["threshold"] = threshold
+        if internal_recall_k is not None:
+            body["internal_recall_k"] = internal_recall_k
         if exclude_document_ids is not None:
             body["exclude_document_ids"] = exclude_document_ids
         if exclude_sections is not None:

knowhere/types/__init__.py

@@ -6,6 +6,9 @@ from knowhere.types.document import Document, DocumentListResponse
 from knowhere.types.job import Job, JobError, JobResult
 from knowhere.types.params import ParsingParams, WebhookConfig
 from knowhere.types.retrieval import (
+    RetrievalChannel,
+    RetrievalFilterMode,
+    RetrievalSectionExclusion,
     RetrievalSource,
     RetrievalQueryResponse,
     RetrievalResult,
@@ -38,6 +41,9 @@ __all__: list[str] = [
     "Document",
     "DocumentListResponse",
     # retrieval
+    "RetrievalChannel",
+    "RetrievalFilterMode",
+    "RetrievalSectionExclusion",
     "RetrievalSource",
     "RetrievalQueryResponse",
     "RetrievalResult",

knowhere/types/job.py

@@ -41,7 +41,6 @@ class Job(BaseModel):
     status: str
     source_type: str
     namespace: Optional[str] = None
-    document_id: Optional[str] = None
     data_id: Optional[str] = None
     created_at: Optional[datetime] = None
     upload_url: Optional[str] = None

knowhere/types/result.py

@@ -272,6 +272,8 @@ class ParseResult:
     kb_csv: Optional[str]
     hierarchy_view_html: Optional[str]
     raw_zip: bytes
+    namespace: Optional[str]
+    document_id: Optional[str]
 
     def __init__(
         self,
@@ -285,6 +287,8 @@ class ParseResult:
         kb_csv: Optional[str],
         hierarchy_view_html: Optional[str],
         raw_zip: bytes,
+        namespace: Optional[str] = None,
+        document_id: Optional[str] = None,
     ) -> None:
         self.manifest = manifest
         self.chunks = chunks
@@ -295,6 +299,8 @@ class ParseResult:
         self.kb_csv = kb_csv
         self.hierarchy_view_html = hierarchy_view_html
         self.raw_zip = raw_zip
+        self.namespace = namespace
+        self.document_id = document_id
 
     # -- convenience properties --
 

knowhere/types/retrieval.py

@@ -2,11 +2,22 @@
 
 from __future__ import annotations
 
-from typing import Optional
+from typing import Literal, Optional, TypedDict
 
 from pydantic import BaseModel
 
 
+RetrievalChannel = Literal["path", "content", "term"]
+RetrievalFilterMode = Literal["delete", "keep"]
+
+
+class RetrievalSectionExclusion(TypedDict):
+    """Section exclusion for follow-up retrieval queries."""
+
+    document_id: str
+    section_path: str
+
+
 class RetrievalSource(BaseModel):
     """Caller-facing source reference attached to a retrieval result."""
 
@@ -30,4 +41,5 @@ class RetrievalQueryResponse(BaseModel):
 
     namespace: str
     query: str
+    router_used: Optional[str] = None
     results: list[RetrievalResult]

{knowhere_python_sdk-0.3.0.dist-info → knowhere_python_sdk-0.3.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: knowhere-python-sdk
-Version: 0.3.0
+Version: 0.3.1
 Summary: Official Python SDK for the Knowhere document parsing API
 Project-URL: Homepage, https://knowhereto.ai
 Project-URL: Documentation, https://docs.knowhereto.ai
@@ -67,8 +67,9 @@ for chunk in result.text_chunks:
 ## Retrieval and document lifecycle
 
 New documents are published into a retrieval namespace. The server returns a
-stable `document_id` when you create a job; persist that value if you need to
-update or archive the same document later.
+stable `document_id` after the job is published. `client.jobs.create(...)`
+does not return a usable `document_id`; persist `job_result.document_id` if you
+need to update or archive the same document later.
 
 ```python
 job = client.jobs.create(
@@ -77,7 +78,11 @@ job = client.jobs.create(
     namespace="support-center",
 )
 
-print(job.document_id)  # "doc_..."
+job_result = client.jobs.wait(job.job_id)
+document_id = job_result.document_id
+
+if document_id is None:
+    raise RuntimeError("Expected document_id after successful publication.")
 ```
 
 After the job is done and published, query the canonical document content:
@@ -87,8 +92,13 @@ response = client.retrieval.query(
     namespace="support-center",
     query="How do I reset Bluetooth pairing?",
     top_k=5,
+    channels=["path", "term"],
+    filter_mode="keep",
+    signal_paths=["Bluetooth", "Pairing"],
 )
 
+print(response.router_used)
+
 for result in response.results:
     print(result.content)
     print(result.score)
@@ -101,13 +111,13 @@ Use `document_id` to update or archive a document:
 update_job = client.jobs.create(
     source_type="url",
     source_url="https://example.com/manual-v2.pdf",
-    document_id=job.document_id,
+    document_id=document_id,
 )
 
-document = client.documents.get(job.document_id)
+document = client.documents.get(document_id)
 print(document.status)
 
-client.documents.archive(job.document_id)
+client.documents.archive(document_id)
 ```
 
 You can also list documents in a namespace:
@@ -146,6 +156,8 @@ result = client.parse(
 
 print(result.manifest.source_file_name)  # "report.pdf"
 print(len(result.chunks))                # 152
+print(result.namespace)                  # "default" or your explicit namespace
+print(result.document_id)                # Published canonical document id
 ```
 
 ### Access different chunk types
@@ -209,14 +221,14 @@ job = client.jobs.create(
     parsing_params={"model": "advanced", "ocr_enabled": True},
 )
 
-print(job.document_id)  # Persist this to update/archive the document later.
-
 # Step 2: Upload file to presigned URL
 client.jobs.upload(job, file=Path("report.pdf"))
 
 # Step 3: Poll until done (adaptive backoff)
 job_result = client.jobs.wait(job.job_id, poll_interval=10.0, poll_timeout=1800.0)
 
+print(job_result.document_id)  # Persist this to update/archive the document later.
+
 # Step 4: Download and parse results
 result = client.jobs.load(job_result)
 print(result.statistics)

{knowhere_python_sdk-0.3.0.dist-info → knowhere_python_sdk-0.3.1.dist-info}/RECORD

@@ -1,4 +1,4 @@
-knowhere/__init__.py,sha256=FLKrentC0o9j1GZTSTlx7A1S_mWmXWceomBScdPbXg8,2854
+knowhere/__init__.py,sha256=wicVid8SW7a3AqabHmHI6iIxpY5Tm732eMyQgBQ7zDM,3016
 knowhere/_base_client.py,sha256=ddeRR1lWLhes5ipvYX6-TMEecjjiEBGfQdPw_vnSNqA,17978
 knowhere/_client.py,sha256=WYb-Fhi3x3nQYNfQG9eCgOpLc_wVyAawfPZWdZhFESg,9586
 knowhere/_constants.py,sha256=ZNCFQC00NpUZIyc_XZ0uemjJE-E8uKAbv3BDa3po9cg,885
@@ -6,7 +6,7 @@ knowhere/_exceptions.py,sha256=NflH7phh_bNFOJmQ758V4mZCAFQskpGXACMz2JIfFNU,11896
 knowhere/_logging.py,sha256=tNqEA1dLv-adTT6qRq5RBeO35FoWrnS3gwt7gKChLTA,1376
 knowhere/_response.py,sha256=EsrM794qxCykvl82UkszeqjJzm9_OSq7nsyzaSCnx0I,1415
 knowhere/_types.py,sha256=8-JFaRcxgBJbw2mV9BwnmCktFVph41a1mduwtXlYidI,1775
-knowhere/_version.py,sha256=BW_DctcKYzNRp1g4_DgZOvYCUcP3tNHyQKvZG3uopBM,50
+knowhere/_version.py,sha256=ma0Xv9k49qOL337sii6xfWylMGz1MNXbCb4rszXekbo,50
 knowhere/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 knowhere/lib/__init__.py,sha256=e953V5ny3VmDtCw7y_4uPwdTkwwNpe_Y6o4AEgz3ujw,50
 knowhere/lib/polling.py,sha256=s0EPHozAvNhXLqr5uwU8YXkkwAdF0ji_nIN0QfR6avY,4500
@@ -15,15 +15,15 @@ knowhere/lib/upload.py,sha256=eT-O9_wB2WkWUAsUd7VzaKY6DVfNeA6WMHRdwm0HM0o,7849
 knowhere/resources/__init__.py,sha256=ClsR-yn_0E4KOopD_Yq13wbPHHjl9s15XpydN-d2Rzo,393
 knowhere/resources/_base.py,sha256=tgKphNTsgMhktWp6_rhyVOZyee4CYlDmD5O1_jWVvYo,1829
 knowhere/resources/documents.py,sha256=u_gmrElvpMOABaHkEuTyaYvh4D_CG4pHZt23r8tivaY,2314
-knowhere/resources/jobs.py,sha256=IhcJIQ_jho6dSsdJLSS0VRB6xuWw12BRJrjO_4NjEMs,9099
-knowhere/resources/retrieval.py,sha256=yVCUWlOg6_ZJhXfiy5_AjqLZZm2Zx8ltqhj1kJ1gKIM,2302
-knowhere/types/__init__.py,sha256=fKMA0NA2lZ-eag1FIeScnwz2ImV6LD-T3YJVfUBsA98,1290
+knowhere/resources/jobs.py,sha256=xYhgYP3Vz7SgGEckmXOvZocNru_4nsS4BoqquojncNw,9727
+knowhere/resources/retrieval.py,sha256=t_jFY-7wYfYVSH6e3WYgn0IaoaPcABXaeZoqcs-pUIo,4543
+knowhere/types/__init__.py,sha256=-T1Rx90y1W3kSW63v6QbXDgTO9aE097vx98xvRaYejU,1452
 knowhere/types/document.py,sha256=LbFleglvm538vSDDho82j7fVxvgMXdIVm9wrWemLShY,711
-knowhere/types/job.py,sha256=_ORhgn_tnvQm_gyrCS39EsDV3dOKImBeJXGjEq3JLag,2510
+knowhere/types/job.py,sha256=VsLUFuELZo8rRemuekTbliTIwaD6CR_dAjgdSriPmw4,2472
 knowhere/types/params.py,sha256=7DyBd4xMxtLPch-A1130-gI0ajKOv2G5tbSMkE8n6-E,543
-knowhere/types/result.py,sha256=UmoxaFmxt2bhrP-2O6jYL89C2WuwZh2xcyyHl46Q1_Y,12925
-knowhere/types/retrieval.py,sha256=-YzsKyusajVdGx4v1lR9Kts-Fh5D41uXf17lSL4ZyJM,777
+knowhere/types/result.py,sha256=uSpvOadmKOF5-n_uBTkmWAho2eDsOAUZoK_W96X2jeU,13143
+knowhere/types/retrieval.py,sha256=EopqmAx2DeO9AmEbd50emdu2mTbTxrhGoJ6DwvvoUCI,1090
 knowhere/types/shared.py,sha256=K5ezX212othxgCviiE2WnwWFY2MS08pXKJ8Km1ZWmjc,104
-knowhere_python_sdk-0.3.0.dist-info/METADATA,sha256=T7MT_NBl2sqb_FcBuxU97Eacm8YDXn8jcP3DLRnLQH0,7922
-knowhere_python_sdk-0.3.0.dist-info/WHEEL,sha256=QccIxa26bgl1E6uMy58deGWi-0aeIkkangHcxk2kWfw,87
-knowhere_python_sdk-0.3.0.dist-info/RECORD,,
+knowhere_python_sdk-0.3.1.dist-info/METADATA,sha256=VSSYe-vr9NLen7NEf-BZc4lT3OJF4bGtL3ivqYbItDk,8429
+knowhere_python_sdk-0.3.1.dist-info/WHEEL,sha256=QccIxa26bgl1E6uMy58deGWi-0aeIkkangHcxk2kWfw,87
+knowhere_python_sdk-0.3.1.dist-info/RECORD,,