structifyai 1.174.0__py3-none-any.whl → 1.175.0__py3-none-any.whl

structify/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "structify"
-__version__ = "1.174.0"  # x-release-please-version
+__version__ = "1.175.0"  # x-release-please-version

structify/resources/polars.py

@@ -30,7 +30,7 @@ from ..lib.cost_confirmation import request_cost_confirmation_if_needed
 from .external_dataframe_proxy import ServicesProxy
 from ..types.save_requirement_param import RequiredEntity, RequiredProperty
 from ..types.dataset_descriptor_param import DatasetDescriptorParam
-from ..types.structure_run_async_params import SourcePdf, SourceWebWeb
+from ..types.structure_run_async_params import SourceWebWeb
 
 __all__ = ["PolarsResource"]
 
@@ -847,6 +847,16 @@ class PolarsResource(SyncAPIResource):
 
         table_param = as_table_param(table_name, schema)
 
+        # Create dataset for this PDF
+        dataset_name = f"structure_pdfs_{table_name}_{uuid.uuid4().hex}"
+        self._client.datasets.create(
+            name=dataset_name,
+            description="",
+            tables=[table_param],
+            relationships=[],
+            ephemeral=True,
+        )
+
         node_id = get_node_id()
 
         # Validate model format if provided as string
@@ -857,9 +867,7 @@ class PolarsResource(SyncAPIResource):
                     "model must be in format 'provider.model_name' (e.g. 'bedrock.claude-sonnet-4-bedrock')"
                 )
 
-        # Build lookups for per-row instructions by index (not by path, since the same PDF may appear multiple times)
         paths_df = document_paths.collect()
-
         instructions_list: list[str | None] = []
 
         if instructions is not None and not isinstance(instructions, str):
@@ -868,108 +876,74 @@ class PolarsResource(SyncAPIResource):
                 raise ValueError(f"instructions shape {instr_df.shape} != document_paths shape {paths_df.shape}")
             instructions_list = cast(List[Optional[str]], instr_df[instr_df.columns[0]].to_list())
 
-        def structure_batch(batch_df: pl.DataFrame) -> pl.DataFrame:
-            # Track by row index since the same PDF may appear multiple times with different instructions
-            batch_rows = batch_df.with_row_index("__row_idx__").to_dicts()
-            valid_rows = [row for row in batch_rows if row.get(path_column) is not None]
-
-            if not valid_rows:
-                return pl.DataFrame(schema=polars_schema)
+        # Request cost confirmation before dispatching costly PDF extraction jobs
+        if not request_cost_confirmation_if_needed(self._client, paths_df.shape[0]):
+            raise Exception(f"User cancelled PDF extraction for {table_name}")
 
-            # Request cost confirmation before dispatching costly PDF extraction jobs
-            if not request_cost_confirmation_if_needed(self._client, len(valid_rows)):
-                raise Exception(f"User cancelled PDF extraction for {table_name}")
+        job_to_pdf_path: dict[str, str] = {}
 
-            # Process each PDF document
-            job_ids: list[str] = []
-            idx_to_dataset: dict[int, str] = {}
-
-            def process_pdf(row: dict[str, Any]) -> Tuple[str, int, str]:
-                row_idx = row["__row_idx__"]
-                pdf_path = row[path_column]
-                dataset_name = f"structure_pdfs_{table_name}_{uuid.uuid4().hex}"
-
-                # Create dataset for this PDF
-                self._client.datasets.create(
-                    name=dataset_name,
-                    description="",
-                    tables=[table_param],
-                    relationships=[],
-                    ephemeral=True,
-                )
-
-                # Upload the PDF document
-                with open(pdf_path, "rb") as pdf_file:
+        # Process each PDF document
+        def process_pdf(pdf_path: str, instructions: str | None) -> Tuple[List[str], str]:
+            # Upload the PDF document
+            unique_pdf_name = f"{uuid.uuid4().hex}.pdf"
+            with open(pdf_path, "rb") as pdf_file:
+                try:
                     self._client.documents.upload(
                         content=pdf_file,
                         file_type="PDF",
                         dataset=dataset_name,
-                        path=f"{dataset_name}.pdf".encode(),
+                        path=unique_pdf_name.encode(),
                     )
-
-                # Get per-row instructions and model
+                except Exception as e:
+                    if "Document already exists" not in str(e):
+                        raise e
+
+            job_ids = self._client.structure.pdf(
+                dataset=dataset_name,
+                path=unique_pdf_name,
+                node_id=node_id,
+                instructions=instructions,
+                mode="Single" if mode == "single" else "Batch",
+                model=model,
+            ).job_ids
+            return job_ids, pdf_path
+
+        with ThreadPoolExecutor(max_workers=MAX_PARALLEL_REQUESTS) as executor:
+            futures: List[Future[Tuple[List[str], str]]] = []
+            for i in range(paths_df.shape[0]):
+                path: str | None = paths_df[path_column][i]
                 pdf_instructions: str | None = None
                 if isinstance(instructions, str):
                     pdf_instructions = instructions
                 elif instructions_list:
-                    pdf_instructions = cast(Optional[str], instructions_list[row_idx])
-                elif conditioning:
+                    pdf_instructions = instructions_list[i]
+                if pdf_instructions is None and conditioning:
                     pdf_instructions = conditioning
+                if path is not None:
+                    futures.append(executor.submit(process_pdf, path, pdf_instructions))
+            for future in tqdm(as_completed(futures), total=len(futures), desc="Preparing PDFs"):
+                job_ids, pdf_path = future.result()
+                for job_id in job_ids:
+                    job_to_pdf_path[job_id] = pdf_path
+
+        # Wait for all PDF processing jobs to complete
+        self._client.jobs.wait_for_jobs(dataset_name=dataset_name, title=f"Parsing PDFs", node_id=node_id)
+
+        # Get all of the entities with their job_ids
+        entities = self._client.datasets.view_table(dataset=dataset_name, name=table_name)
+        structured_results: List[Dict[str, Any]] = [
+            {**entity.properties, path_column: job_to_pdf_path[entity.job_ids[0]]} for entity in entities
+        ]
 
-                job_id = self._client.structure.run_async(
-                    dataset=dataset_name,
-                    source=SourcePdf(pdf={"path": f"{dataset_name}.pdf", "single_agent": mode == "single"}),
-                    node_id=node_id,
-                    instructions=pdf_instructions,
-                    model=model,
-                )
-                return job_id, row_idx, dataset_name
-
-            with ThreadPoolExecutor(max_workers=MAX_PARALLEL_REQUESTS) as executor:
-                futures = [executor.submit(process_pdf, row) for row in valid_rows]
-                for future in tqdm(as_completed(futures), total=len(futures), desc="Preparing PDFs"):
-                    job_id, row_idx, dataset_name = future.result()
-                    job_ids.append(job_id)
-                    idx_to_dataset[row_idx] = dataset_name
-
-            # Wait for all PDF processing jobs to complete
-            self._client.jobs.wait_for_jobs(job_ids=job_ids, title=f"Parsing {table_name} from PDFs", node_id=node_id)
-
-            # Collect results from all processed PDFs - each result is tagged with its source row_idx
-            structured_results: list[dict[str, Any]] = []
-
-            def collect_pdf_results(row_idx: int, dataset_name: str) -> List[Dict[str, Any]]:
-                pdf_path = batch_rows[row_idx][path_column]
-                entities_result = self._client.datasets.view_table(dataset=dataset_name, name=table_name)
-                return [
-                    {**entity.properties, path_column: pdf_path, "__row_idx__": row_idx} for entity in entities_result
-                ]
-
-            with ThreadPoolExecutor(max_workers=MAX_PARALLEL_REQUESTS) as executor:
-                collect_futures = [
-                    executor.submit(collect_pdf_results, row_idx, dataset_name)
-                    for row_idx, dataset_name in idx_to_dataset.items()
-                ]
-                for future in tqdm(
-                    as_completed(collect_futures), total=len(collect_futures), desc="Collecting PDF extractions"
-                ):
-                    results = future.result()
-                    structured_results.extend(results)
-
-            # Ensure all columns are present with None for missing values
-            for result_row in structured_results:
-                for col_name in polars_schema.names():
-                    if col_name not in result_row:
-                        result_row[col_name] = None
-
-            if not structured_results:
-                return pl.DataFrame(schema=polars_schema)
-
-            # Build result dataframe directly from structured_results without joining
-            # Each entity is already tagged with path_column from its source PDF
-            return pl.DataFrame(structured_results, schema=polars_schema)
+        # Ensure all columns are present with None for missing values
+        for result_row in structured_results:
+            for col_name in polars_schema.names():
+                if col_name not in result_row:
+                    result_row[col_name] = None
 
-        return document_paths.map_batches(structure_batch, schema=polars_schema, no_optimizations=True)
+        # Build result dataframe directly from structured_results without joining
+        # Each entity is already tagged with path_column from its source PDF
+        return pl.DataFrame(structured_results, schema=polars_schema).lazy()
 
     def tag(
         self,

structify/resources/structure.py

@@ -3,10 +3,12 @@
 from __future__ import annotations
 
 from typing import Iterable, Optional
+from typing_extensions import Literal
 
 import httpx
 
 from ..types import (
+    structure_pdf_params,
     structure_run_async_params,
     structure_job_status_params,
     structure_enhance_property_params,
@@ -26,6 +28,7 @@ from .._response import (
 from .._base_client import make_request_options
 from ..types.knowledge_graph_param import KnowledgeGraphParam
 from ..types.save_requirement_param import SaveRequirementParam
+from ..types.structure_pdf_response import StructurePdfResponse
 from ..types.structure_job_status_response import StructureJobStatusResponse
 
 __all__ = ["StructureResource", "AsyncStructureResource"]
@@ -267,6 +270,55 @@ class StructureResource(SyncAPIResource):
             cast_to=StructureJobStatusResponse,
         )
 
+    def pdf(
+        self,
+        *,
+        dataset: str,
+        path: str,
+        instructions: Optional[str] | Omit = omit,
+        mode: Literal["Single", "Batch"] | Omit = omit,
+        model: Optional[str] | Omit = omit,
+        node_id: Optional[str] | Omit = omit,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = not_given,
+    ) -> StructurePdfResponse:
+        """For single mode, creates one job for the entire PDF.
+
+        For batch mode, creates one
+        job per PDF page.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._post(
+            "/structure/pdf",
+            body=maybe_transform(
+                {
+                    "dataset": dataset,
+                    "path": path,
+                    "instructions": instructions,
+                    "mode": mode,
+                    "model": model,
+                    "node_id": node_id,
+                },
+                structure_pdf_params.StructurePdfParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=StructurePdfResponse,
+        )
+
     def run_async(
         self,
         *,
@@ -560,6 +612,55 @@ class AsyncStructureResource(AsyncAPIResource):
             cast_to=StructureJobStatusResponse,
         )
 
+    async def pdf(
+        self,
+        *,
+        dataset: str,
+        path: str,
+        instructions: Optional[str] | Omit = omit,
+        mode: Literal["Single", "Batch"] | Omit = omit,
+        model: Optional[str] | Omit = omit,
+        node_id: Optional[str] | Omit = omit,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = not_given,
+    ) -> StructurePdfResponse:
+        """For single mode, creates one job for the entire PDF.
+
+        For batch mode, creates one
+        job per PDF page.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return await self._post(
+            "/structure/pdf",
+            body=await async_maybe_transform(
+                {
+                    "dataset": dataset,
+                    "path": path,
+                    "instructions": instructions,
+                    "mode": mode,
+                    "model": model,
+                    "node_id": node_id,
+                },
+                structure_pdf_params.StructurePdfParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=StructurePdfResponse,
+        )
+
     async def run_async(
         self,
         *,
@@ -636,6 +737,9 @@ class StructureResourceWithRawResponse:
         self.job_status = to_raw_response_wrapper(
             structure.job_status,
         )
+        self.pdf = to_raw_response_wrapper(
+            structure.pdf,
+        )
         self.run_async = to_raw_response_wrapper(
             structure.run_async,
         )
@@ -660,6 +764,9 @@ class AsyncStructureResourceWithRawResponse:
         self.job_status = async_to_raw_response_wrapper(
             structure.job_status,
         )
+        self.pdf = async_to_raw_response_wrapper(
+            structure.pdf,
+        )
         self.run_async = async_to_raw_response_wrapper(
             structure.run_async,
         )
@@ -684,6 +791,9 @@ class StructureResourceWithStreamingResponse:
         self.job_status = to_streamed_response_wrapper(
             structure.job_status,
         )
+        self.pdf = to_streamed_response_wrapper(
+            structure.pdf,
+        )
         self.run_async = to_streamed_response_wrapper(
             structure.run_async,
         )
@@ -708,6 +818,9 @@ class AsyncStructureResourceWithStreamingResponse:
         self.job_status = async_to_streamed_response_wrapper(
             structure.job_status,
         )
+        self.pdf = async_to_streamed_response_wrapper(
+            structure.pdf,
+        )
         self.run_async = async_to_streamed_response_wrapper(
             structure.run_async,
         )

structify/resources/whitelabel.py

@@ -108,6 +108,41 @@ class WhitelabelResource(SyncAPIResource):
             cast_to=NoneType,
         )
 
+    def proxy_post(
+        self,
+        path: str,
+        *,
+        service: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = not_given,
+    ) -> None:
+        """
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not service:
+            raise ValueError(f"Expected a non-empty value for `service` but received {service!r}")
+        if not path:
+            raise ValueError(f"Expected a non-empty value for `path` but received {path!r}")
+        extra_headers = {"Accept": "*/*", **(extra_headers or {})}
+        return self._post(
+            f"/whitelabel/{service}/{path}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=NoneType,
+        )
+
 
 class AsyncWhitelabelResource(AsyncAPIResource):
     @cached_property
@@ -198,6 +233,41 @@ class AsyncWhitelabelResource(AsyncAPIResource):
             cast_to=NoneType,
         )
 
+    async def proxy_post(
+        self,
+        path: str,
+        *,
+        service: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = not_given,
+    ) -> None:
+        """
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not service:
+            raise ValueError(f"Expected a non-empty value for `service` but received {service!r}")
+        if not path:
+            raise ValueError(f"Expected a non-empty value for `path` but received {path!r}")
+        extra_headers = {"Accept": "*/*", **(extra_headers or {})}
+        return await self._post(
+            f"/whitelabel/{service}/{path}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=NoneType,
+        )
+
 
 class WhitelabelResourceWithRawResponse:
     def __init__(self, whitelabel: WhitelabelResource) -> None:
@@ -209,6 +279,9 @@ class WhitelabelResourceWithRawResponse:
         self.proxy_get = to_raw_response_wrapper(
             whitelabel.proxy_get,
         )
+        self.proxy_post = to_raw_response_wrapper(
+            whitelabel.proxy_post,
+        )
 
 
 class AsyncWhitelabelResourceWithRawResponse:
@@ -221,6 +294,9 @@ class AsyncWhitelabelResourceWithRawResponse:
         self.proxy_get = async_to_raw_response_wrapper(
             whitelabel.proxy_get,
         )
+        self.proxy_post = async_to_raw_response_wrapper(
+            whitelabel.proxy_post,
+        )
 
 
 class WhitelabelResourceWithStreamingResponse:
@@ -233,6 +309,9 @@ class WhitelabelResourceWithStreamingResponse:
         self.proxy_get = to_streamed_response_wrapper(
             whitelabel.proxy_get,
         )
+        self.proxy_post = to_streamed_response_wrapper(
+            whitelabel.proxy_post,
+        )
 
 
 class AsyncWhitelabelResourceWithStreamingResponse:
@@ -245,3 +324,6 @@ class AsyncWhitelabelResourceWithStreamingResponse:
         self.proxy_get = async_to_streamed_response_wrapper(
             whitelabel.proxy_get,
         )
+        self.proxy_post = async_to_streamed_response_wrapper(
+            whitelabel.proxy_post,
+        )

structify/types/__init__.py

@@ -111,6 +111,7 @@ from .scrape_list_response import ScrapeListResponse as ScrapeListResponse
 from .scrape_scrape_params import ScrapeScrapeParams as ScrapeScrapeParams
 from .select_team_response import SelectTeamResponse as SelectTeamResponse
 from .source_list_response import SourceListResponse as SourceListResponse
+from .structure_pdf_params import StructurePdfParams as StructurePdfParams
 from .update_team_response import UpdateTeamResponse as UpdateTeamResponse
 from .workflow_stop_params import WorkflowStopParams as WorkflowStopParams
 from .connector_auth_method import ConnectorAuthMethod as ConnectorAuthMethod
@@ -148,6 +149,7 @@ from .list_projects_response import ListProjectsResponse as ListProjectsResponse
 from .remove_member_response import RemoveMemberResponse as RemoveMemberResponse
 from .save_requirement_param import SaveRequirementParam as SaveRequirementParam
 from .scrape_scrape_response import ScrapeScrapeResponse as ScrapeScrapeResponse
+from .structure_pdf_response import StructurePdfResponse as StructurePdfResponse
 from .team_add_member_params import TeamAddMemberParams as TeamAddMemberParams
 from .workflow_schedule_info import WorkflowScheduleInfo as WorkflowScheduleInfo
 from .connector_create_params import ConnectorCreateParams as ConnectorCreateParams

structify/types/admin/admin_list_jobs_response.py

@@ -35,7 +35,7 @@ class ParametersStructuringInputAgentAgentPdfPdf(BaseModel):
 
     path: str
 
-    single_agent: Optional[bool] = None
+    page: Optional[int] = None
 
 
 class ParametersStructuringInputAgentAgentPdf(BaseModel):

structify/types/dataset_view_table_response.py

@@ -1,6 +1,6 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import Dict, Union, Optional
+from typing import Dict, List, Union, Optional
 from datetime import datetime
 from typing_extensions import Literal, TypeAlias
 
@@ -103,6 +103,8 @@ class DatasetViewTableResponse(BaseModel):
 
     dataset_id: str
 
+    job_ids: List[str]
+
     label: str
 
     properties: Dict[str, Properties]

structify/types/job_list_response.py

@@ -36,7 +36,7 @@ class ParametersStructuringInputAgentAgentPdfPdf(BaseModel):
 
     path: str
 
-    single_agent: Optional[bool] = None
+    page: Optional[int] = None
 
 
 class ParametersStructuringInputAgentAgentPdf(BaseModel):

structify/types/structure_pdf_params.py

@@ -0,0 +1,22 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Optional
+from typing_extensions import Literal, Required, TypedDict
+
+__all__ = ["StructurePdfParams"]
+
+
+class StructurePdfParams(TypedDict, total=False):
+    dataset: Required[str]
+
+    path: Required[str]
+
+    instructions: Optional[str]
+
+    mode: Literal["Single", "Batch"]
+
+    model: Optional[str]
+
+    node_id: Optional[str]

structify/types/structure_pdf_response.py

@@ -0,0 +1,11 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import List
+
+from .._models import BaseModel
+
+__all__ = ["StructurePdfResponse"]
+
+
+class StructurePdfResponse(BaseModel):
+    job_ids: List[str]

structify/types/structure_run_async_params.py

@@ -40,7 +40,7 @@ class SourcePdfPdf(TypedDict, total=False):
 
     path: Required[str]
 
-    single_agent: bool
+    page: Optional[int]
 
 
 class SourcePdf(TypedDict, total=False):

structify/types/usage_group_key.py

@@ -5,5 +5,16 @@ from typing_extensions import Literal, TypeAlias
 __all__ = ["UsageGroupKey"]
 
 UsageGroupKey: TypeAlias = Literal[
-    "web", "pdf", "derive", "scrape", "apollo", "searchapi", "newsapi", "match", "connectorexplore", "other"
+    "web",
+    "pdf",
+    "derive",
+    "scrape",
+    "apollo",
+    "searchapi",
+    "newsapi",
+    "secapi",
+    "cufinder",
+    "match",
+    "connectorexplore",
+    "other",
 ]

{structifyai-1.174.0.dist-info → structifyai-1.175.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: structifyai
-Version: 1.174.0
+Version: 1.175.0
 Summary: The official Python library for the structify API
 Project-URL: Homepage, https://github.com/StructifyAI/structify-python
 Project-URL: Repository, https://github.com/StructifyAI/structify-python

{structifyai-1.174.0.dist-info → structifyai-1.175.0.dist-info}/RECORD

@@ -11,7 +11,7 @@ structify/_resource.py,sha256=tJi4pDQooQZ_zJwEwrLj-U-ye2hC-cbmr1GzIwCT10Y,1118
 structify/_response.py,sha256=RuNhMDiZUdPqEbmFJHDVI4FMPDszk8QjK9LVWm1Fagk,28806
 structify/_streaming.py,sha256=n4C9M7ITmANYn9LaWHNoqJdIIyF7svLco2qst7u3M7U,10233
 structify/_types.py,sha256=jj4p-m3vpUma0AdhPWIaljHZXeb4RKnrAusjVdpDy5Y,7597
-structify/_version.py,sha256=jAkquXwE8e7XiNAPLSl0tNkX5WyP7kZ2giJmBdnOmlM,163
+structify/_version.py,sha256=dFWqiyi95rQIGV0JregKsPLWuKqIKL7smQzi9sUXYwo,163
 structify/pagination.py,sha256=ycybhWcpKk4ztsMcCA6C0WZiJejGrSx6bSr8LLskJUY,4346
 structify/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 structify/_utils/__init__.py,sha256=7fch0GT9zpNnErbciSpUNa-SjTxxjY6kxHxKMOM4AGs,2305
@@ -39,7 +39,7 @@ structify/resources/external_dataframe_proxy.py,sha256=DSn0YwWIembR__ZtDxVCJtyY3
 structify/resources/jobs.py,sha256=gO1aSByi1dMvW90UDsMmNhLHFCOY4ENLkZcAx4gbLHY,30108
 structify/resources/match.py,sha256=gDWEWnKwEoLbheQAMFltJCk2ysa_L9AuJMOaauM7c4Y,12248
 structify/resources/nango.py,sha256=Zl0M1XhlVe61jHVd-SdYI9uEbEhIRmskvlk7Xp0Lh8g,9166
-structify/resources/polars.py,sha256=--kVVUoJ7AvGirnc462r0TRsK7lcXXDQD0XWhqtG3K8,61552
+structify/resources/polars.py,sha256=KFeiRPCVGIssbFdTnGtTUEXNW46j6nHzmHN5rPeqhRA,59995
 structify/resources/projects.py,sha256=YDikBDB9D1EXyZ2GyRx4GlpQ83snw51YlNuU1sLHqho,14117
 structify/resources/public_sessions.py,sha256=_JmssE0MMjeGdxT0FWtrkcceSV4skeEkVGYeO2FkJes,9976
 structify/resources/sandbox.py,sha256=Gc7uxZAOmbXA477UHvS244BokI1Tte_34xxIj0dC5PA,16984
@@ -48,9 +48,9 @@ structify/resources/server.py,sha256=39G3yhj3NYWgnLq0PdS86lwDvXK6MIWAIl7G0ZLMJro
 structify/resources/sessions.py,sha256=xNKjg_H9JoZenCnMJJOURBa2bmXquwweqehCF4hGMlY,65863
 structify/resources/slack.py,sha256=CSMW3Eo-XRjISvNOgAcGyBdnmI9SdkQplE0P0lqTnjM,13082
 structify/resources/sources.py,sha256=K5jLPKu3LNH5vKL9V1NQlHaG8UI982H74enN_cjdf-0,13181
-structify/resources/structure.py,sha256=VMk10L7npByPFtSrddvVNPcGP-iE8PPO_4i2JctYp3A,28974
+structify/resources/structure.py,sha256=jXcVdUuNc6Q-HwUhSDcsJONbBqRf_FcPAjNtjXYxrdo,33124
 structify/resources/teams.py,sha256=W1Isf2lXDLB-xdzgR8qJwCrziElOY25OcsQfIBKQ1Sc,55575
-structify/resources/whitelabel.py,sha256=owhs0tsBtuR9yS7H33Kf3dvbiB7kRlOIXVbLt-7Wx2Q,9579
+structify/resources/whitelabel.py,sha256=0TQ3HJccdflpAVcHqydizT4RznkRvZBRP0--0szjDps,12865
 structify/resources/whitelabel_service.py,sha256=cqGBpRegtJRjt5WFNPCXTwyGKHTAbBUrhJP1UroEhVQ,9439
 structify/resources/wiki.py,sha256=0j5qrgQ72LE7GsuLyy4nylnEuU0VDSjTMRejaHlqa-8,19490
 structify/resources/workflow.py,sha256=D8gnt22usKoY9iA3HTNlbQyWTykNUXl2m3rFsPAKZZc,9273
@@ -77,7 +77,7 @@ structify/resources/user/__init__.py,sha256=vkGsM7uGlBXVVwuEZlICVLlIY3F75n_MIMpm
 structify/resources/user/api_keys.py,sha256=LgKvDGkHPGIumWmEsPCA1XYAKNuRN48nqG6anv8AVHE,14331
 structify/resources/user/stripe.py,sha256=MazkGQ5VuNY7TxMPXxJULBVerMGmOXaEcXpfQsUYoNs,13574
 structify/resources/user/user.py,sha256=Hk8D5KB8in-A9eJSYFIosILz4HgSWDwDDAVIQeQ8Sew,25411
-structify/types/__init__.py,sha256=-lfSCQdu7PvLm8aF4T-si-zkz6wxQsDFAmAu_GFxanQ,28089
+structify/types/__init__.py,sha256=-qTlAuUWBfkvyS4M2A3X4CzLDOps2JQa_x5PLIwTcHk,28245
 structify/types/accept_invitation_response.py,sha256=hWIn0Sscg7U0uOqoZA5SG8ENZQuM0tsbgOlrJbcW37s,260
 structify/types/add_member_response.py,sha256=_YjqwGpsfdR9Dw6SAQ99TveFSD8Hov1IjVPMH2KC3CM,882
 structify/types/admin_grant_access_response.py,sha256=xCPdmejgZs1oXOsfgbW-cYKZa8getHRotRnsCav5QPU,390
@@ -181,7 +181,7 @@ structify/types/dataset_update_relationship_params.py,sha256=vSCxeVJM03JM4DlXmKp
 structify/types/dataset_view_relationships_params.py,sha256=Pptxz8SRA6-qFriarOLylrvIdbQyV7rqO_MVGvP4FjA,1001
 structify/types/dataset_view_relationships_response.py,sha256=IC5Ot90K_F7wSPBSz7rfIkw6nF8paVz1tIL-C37IiUw,1866
 structify/types/dataset_view_table_params.py,sha256=vD_qKTv13_mbBP4zIorOSO--RL2qp5pQRxXPdwPBuio,985
-structify/types/dataset_view_table_response.py,sha256=DsEctnRL0w2pzzDFHU18bIeBhD5bIalXVoLRe3M9Hro,1816
+structify/types/dataset_view_table_response.py,sha256=Ultse5-ichgV1vnD0UJmfslMpEqQlnqPmRmQYId0gkA,1846
 structify/types/dataset_view_tables_with_relationships_params.py,sha256=0qCpvobbsnHIij0Z4nAJfDPl_bf8zGoER-3Cmve7Yxw,1021
 structify/types/dataset_view_tables_with_relationships_response.py,sha256=4ICbKqXwWfYrq7N-ZvlFY4kZCwVdo9k65NMaIL4KWZA,5670
 structify/types/delete_chat_session_response.py,sha256=HuoT7Z2lgj4Elha7SBWf7dBamo-vjoGY5kedaoAYgUw,292
@@ -263,7 +263,7 @@ structify/types/job_event_body.py,sha256=pO11fTm5sGvQ6cx7NvWTitUaunSqdRo-d5tkyh4
 structify/types/job_get_scrapers_response.py,sha256=-E9DaG9YCkrJeuffZ3RiI6MW7ZsbPjmlVU3jUQdggSY,694
 structify/types/job_get_source_entities_response.py,sha256=vlGKFkMKjII48EexIEBbls6-PAVYDDkmR1__aKjuiUo,3379
 structify/types/job_list_params.py,sha256=i_MYi4vIFC6Dq5I4E4htYwZejqRE5qAf2lgg_SRBuKg,1012
-structify/types/job_list_response.py,sha256=OMDUmcntraqTS8B-h7WqREgqbRVClK2a8ah4tFZnEgQ,5143
+structify/types/job_list_response.py,sha256=mKWo7HlbMFtsGMjQHLwT8SCuV_Ktmf6pIAqoPx9a5HU,5134
 structify/types/job_status_params.py,sha256=wx7NFv1gJaIsy_fmCdPeEsVblIFRN5P1cUP1q8210HA,412
 structify/types/job_status_response.py,sha256=xcfeCXfxVOFQStG5W0Xm6Cszv2k6u-VKjb2UDfz_McE,277
 structify/types/knowledge_graph.py,sha256=HskWSN4Kw6nhwmIN6vZTNihKuYgztuDuAs2sZMhih90,607
@@ -347,7 +347,9 @@ structify/types/structure_is_complete_params.py,sha256=foIrU7dW5seHipqVl2f-Leb7n
 structify/types/structure_is_complete_response.py,sha256=YWZrs4YK7EyjdZk1GAfb9cFwJS9HCJ0mW52uxesGqu8,216
 structify/types/structure_job_status_params.py,sha256=uQmvkdZQ8c3XSAZSmshm0X7iDfyAKztzofziXSveWIE,479
 structify/types/structure_job_status_response.py,sha256=jZVwojhMJB7w7BwaMQH5W_lqlzb-fKd2SKo6v1xrW-A,806
-structify/types/structure_run_async_params.py,sha256=s3NONQF-Bwl9Iu8YKoiHJfNcoM_Z5C_byKHCAMfYiI8,1818
+structify/types/structure_pdf_params.py,sha256=wzVxFvV0g-CZYtAjtqN5TIqLq6cUMf0B5sAzEENRvkg,473
+structify/types/structure_pdf_response.py,sha256=WZMCq0HhYSbxkB8yLbfMJ0urYuWTPyTAiQdUNcLbLck,244
+structify/types/structure_run_async_params.py,sha256=w7BwDOjGMkSRZTf5f84nQNFQ5kKM_RZjXomS1wvfjXI,1819
 structify/types/structure_run_async_response.py,sha256=RTK4Vh2A-51560FjLrRJiqKNgBjnuq3hd671_hi5lHU,212
 structify/types/survey_submission_response.py,sha256=OTjhVLOGQWATs7VEsn1_OVNxu5oKOS_WwSu82UhmtbE,240
 structify/types/table.py,sha256=Zx9V6jSUeCYAOn3ghA5xejTXVaplVm6Mew_qOVHxkIY,915
@@ -374,7 +376,7 @@ structify/types/update_member_role_response.py,sha256=q1jTM0lFmzvH8ki7GcSQAh-5cN
 structify/types/update_table_response.py,sha256=ypJEzW8JcLOmLpoZq46i7ADFUkHqBGeo0AhtiSq8HAk,1156
 structify/types/update_team_response.py,sha256=AOm2-jzpjF3UVRh3msyc8W00l-hNpk_SqmMI3oK_YBE,230
 structify/types/update_visibility_response.py,sha256=veHlq90QKDCiY__DLHGQQloyZhxy6wSJ_6bPn18bisY,279
-structify/types/usage_group_key.py,sha256=a0Fc4_raXvnASGPL9XSbMTsTq9uQxQEYFDJ-Jn2SMu0,313
+structify/types/usage_group_key.py,sha256=vOoCW73KNK4BG6KXLPCF9D7J6JjJBssFvhDdDPlkHpo,380
 structify/types/user_enrich_params.py,sha256=bXOLnyPxG-iqFk1KZCjRGpcy8osRs6QYEVLtN_7HXf8,280
 structify/types/user_info.py,sha256=L2FckSvAPuRpgwNgwOoYF2WYwHrHmuDDAehq3-EeUWQ,1562
 structify/types/user_refresh_params.py,sha256=_YdNJ1SZg78JkhfLGiuMq-NsN7zAlF0DG3GvDaaMOpY,324
@@ -405,7 +407,7 @@ structify/types/workflow_stop_params.py,sha256=DLTGDSENmcdNQoV_LnsL0lvf3i3Ntupkp
 structify/types/admin/__init__.py,sha256=sI2YUoXg3jsKmOp7cu3j_z2PoGhwQ5x7aTf822d192k,3540
 structify/types/admin/admin_dataset_return.py,sha256=hRSCIBTiaCb1O1lcxsKyxnzGiYuIUEvnMxXkNbyj4_U,495
 structify/types/admin/admin_delete_jobs_response.py,sha256=c3IO7bCqAoEN9PHYM88gO0JT0e6hCaHyIBtxC84Tw38,225
-structify/types/admin/admin_list_jobs_response.py,sha256=2V64gFWFH3scKfj0P_Lo3d_YxoORoLdrm9sUPSwI6S0,4876
+structify/types/admin/admin_list_jobs_response.py,sha256=27IoyuBSaBJPnSSyiZt3q5dHfeyBzrGtnq7nGvJcapM,4867
 structify/types/admin/admin_sandbox.py,sha256=ADrK4ugEA85vNCzKe7OYr8tzkf5nQWnnACfzmyBATvI,679
 structify/types/admin/admin_teams_list_response.py,sha256=FDrXxx3dQKuHuLfByYzaS5EzZYkZAtyJH7E7t6UoUX4,1167
 structify/types/admin/cancel_subscription_response.py,sha256=ph43hCKHtpjp2eSV9X9eitJA-mpnrRdjwe8KxPLJdKY,246
@@ -489,7 +491,7 @@ structify/types/user/stripe_create_portal_session_params.py,sha256=5AYRC8z_SlKmd
 structify/types/user/stripe_create_session_params.py,sha256=DFcNLNzEWeupkGQ9J5PafsuL_bIU9cLEIhAmFPsRlfo,387
 structify/types/user/stripe_create_subscription_params.py,sha256=d8HfiC94gJbG-cC_WvBz6xYCvxKJO_EP2yyVmVvufrU,424
 structify/types/user/subscription_plan.py,sha256=qKJMM-zPpYolYC1DlypOwPpxlyJBLkQqFK_0VpwktJs,222
-structifyai-1.174.0.dist-info/METADATA,sha256=_cBl4jOw0K_TtGioyYQoTH0ezXr8x4NPfipk2wWlzVk,16399
-structifyai-1.174.0.dist-info/WHEEL,sha256=C2FUgwZgiLbznR-k0b_5k3Ai_1aASOXDss3lzCUsUug,87
-structifyai-1.174.0.dist-info/licenses/LICENSE,sha256=9CwgrmGz3rZSTT-KqGc1gua-7g8B4ThTgMtUgPALh5c,11339
-structifyai-1.174.0.dist-info/RECORD,,
+structifyai-1.175.0.dist-info/METADATA,sha256=tnYJNNHg3u1mib1jjY1qNCerum6ifMs8NTMJuymRUBo,16399
+structifyai-1.175.0.dist-info/WHEEL,sha256=C2FUgwZgiLbznR-k0b_5k3Ai_1aASOXDss3lzCUsUug,87
+structifyai-1.175.0.dist-info/licenses/LICENSE,sha256=9CwgrmGz3rZSTT-KqGc1gua-7g8B4ThTgMtUgPALh5c,11339
+structifyai-1.175.0.dist-info/RECORD,,