futurehouse-client 0.3.19.dev129__py3-none-any.whl → 0.3.19.dev133__py3-none-any.whl

futurehouse_client/clients/rest_client.py

@@ -1053,24 +1053,11 @@ class RestClient:
         status_url = None
 
         try:
-            # Upload all chunks except the last one in parallel
-            if total_chunks > 1:
-                self._upload_chunks_parallel(
-                    job_name,
-                    file_path,
-                    file_name,
-                    upload_id,
-                    total_chunks - 1,
-                    total_chunks,
-                )
-
-            # Upload the last chunk separately (handles assembly)
-            status_url = self._upload_final_chunk(
+            status_url = self._upload_chunks_parallel(
                 job_name,
                 file_path,
                 file_name,
                 upload_id,
-                total_chunks - 1,
                 total_chunks,
             )
 
@@ -1086,149 +1073,74 @@ class RestClient:
         file_path: Path,
         file_name: str,
         upload_id: str,
-        num_regular_chunks: int,
         total_chunks: int,
-    ) -> None:
-        """Upload chunks in parallel batches.
+    ) -> str | None:
+        """Upload all chunks in parallel batches, including the final chunk.
 
         Args:
             job_name: The key of the crow to upload to.
             file_path: The path to the file to upload.
             file_name: The name to use for the file.
             upload_id: The upload ID to use.
-            num_regular_chunks: Number of regular chunks (excluding final chunk).
             total_chunks: Total number of chunks.
 
-        Raises:
-            FileUploadError: If there's an error uploading any chunk.
-        """
-        if num_regular_chunks <= 0:
-            return
-
-        # Process chunks in batches
-        for batch_start in range(0, num_regular_chunks, self.MAX_CONCURRENT_CHUNKS):
-            batch_end = min(
-                batch_start + self.MAX_CONCURRENT_CHUNKS, num_regular_chunks
-            )
-
-            # Upload chunks in this batch concurrently
-            with ThreadPoolExecutor(max_workers=self.MAX_CONCURRENT_CHUNKS) as executor:
-                futures = {
-                    executor.submit(
-                        self._upload_single_chunk,
-                        job_name,
-                        file_path,
-                        file_name,
-                        upload_id,
-                        chunk_index,
-                        total_chunks,
-                    ): chunk_index
-                    for chunk_index in range(batch_start, batch_end)
-                }
-
-                for future in as_completed(futures):
-                    chunk_index = futures[future]
-                    try:
-                        future.result()
-                        logger.debug(
-                            f"Uploaded chunk {chunk_index + 1}/{total_chunks} of {file_name}"
-                        )
-                    except Exception as e:
-                        logger.error(f"Error uploading chunk {chunk_index}: {e}")
-                        raise FileUploadError(
-                            f"Error uploading chunk {chunk_index} of {file_name}: {e}"
-                        ) from e
-
-    def _upload_single_chunk(
-        self,
-        job_name: str,
-        file_path: Path,
-        file_name: str,
-        upload_id: str,
-        chunk_index: int,
-        total_chunks: int,
-    ) -> None:
-        """Upload a single chunk.
-
-        Args:
-            job_name: The key of the crow to upload to.
-            file_path: The path to the file to upload.
-            file_name: The name to use for the file.
-            upload_id: The upload ID to use.
-            chunk_index: The index of this chunk.
-            total_chunks: Total number of chunks.
+        Returns:
+            The status URL from the final chunk response, or None if no chunks.
 
         Raises:
-            Exception: If there's an error uploading the chunk.
+            FileUploadError: If there's an error uploading any chunk.
         """
-        with open(file_path, "rb") as f:
-            # Read the chunk from the file
-            f.seek(chunk_index * self.CHUNK_SIZE)
-            chunk_data = f.read(self.CHUNK_SIZE)
+        if total_chunks <= 0:
+            return None
 
-            # Prepare and send the chunk
-            with tempfile.NamedTemporaryFile() as temp_file:
-                temp_file.write(chunk_data)
-                temp_file.flush()
+        if total_chunks > 1:
+            num_regular_chunks = total_chunks - 1
+            for batch_start in range(0, num_regular_chunks, self.MAX_CONCURRENT_CHUNKS):
+                batch_end = min(
+                    batch_start + self.MAX_CONCURRENT_CHUNKS, num_regular_chunks
+                )
 
-                # Create form data
-                with open(temp_file.name, "rb") as chunk_file_obj:
-                    files = {
-                        "chunk": (
+                # Upload chunks in this batch concurrently
+                with ThreadPoolExecutor(
+                    max_workers=self.MAX_CONCURRENT_CHUNKS
+                ) as executor:
+                    futures = {
+                        executor.submit(
+                            self._upload_single_chunk,
+                            job_name,
+                            file_path,
                             file_name,
-                            chunk_file_obj,
-                            "application/octet-stream",
-                        )
-                    }
-                    data = {
-                        "file_name": file_name,
-                        "chunk_index": chunk_index,
-                        "total_chunks": total_chunks,
-                        "upload_id": upload_id,
+                            upload_id,
+                            chunk_index,
+                            total_chunks,
+                        ): chunk_index
+                        for chunk_index in range(batch_start, batch_end)
                     }
 
-                    # Send the chunk
-                    response = self.multipart_client.post(
-                        f"/v0.1/crows/{job_name}/upload-chunk",
-                        files=files,
-                        data=data,
-                    )
-                    response.raise_for_status()
-
-    def _upload_final_chunk(
-        self,
-        job_name: str,
-        file_path: Path,
-        file_name: str,
-        upload_id: str,
-        chunk_index: int,
-        total_chunks: int,
-    ) -> str | None:
-        """Upload the final chunk with retry logic for missing chunks.
-
-        Args:
-            job_name: The key of the crow to upload to.
-            file_path: The path to the file to upload.
-            file_name: The name to use for the file.
-            upload_id: The upload ID to use.
-            chunk_index: The index of the final chunk.
-            total_chunks: Total number of chunks.
-
-        Returns:
-            The status URL from the response.
-
-        Raises:
-            FileUploadError: If there's an error uploading the final chunk.
-        """
+                    for future in as_completed(futures):
+                        chunk_index = futures[future]
+                        try:
+                            future.result()
+                            logger.debug(
+                                f"Uploaded chunk {chunk_index + 1}/{total_chunks} of {file_name}"
+                            )
+                        except Exception as e:
+                            logger.error(f"Error uploading chunk {chunk_index}: {e}")
+                            raise FileUploadError(
+                                f"Error uploading chunk {chunk_index} of {file_name}: {e}"
+                            ) from e
+
+        # Upload the final chunk with retry logic
+        final_chunk_index = total_chunks - 1
         retries = 0
         max_retries = 3
-        retry_delay = 2.0  # seconds
+        retry_delay = 2.0
 
         while retries < max_retries:
             try:
                 with open(file_path, "rb") as f:
                     # Read the final chunk from the file
-                    f.seek(chunk_index * self.CHUNK_SIZE)
+                    f.seek(final_chunk_index * self.CHUNK_SIZE)
                     chunk_data = f.read(self.CHUNK_SIZE)
 
                     # Prepare and send the chunk
@@ -1247,7 +1159,7 @@ class RestClient:
                             }
                             data = {
                                 "file_name": file_name,
-                                "chunk_index": chunk_index,
+                                "chunk_index": final_chunk_index,
                                 "total_chunks": total_chunks,
                                 "upload_id": upload_id,
                             }
@@ -1274,7 +1186,7 @@ class RestClient:
                             status_url = response_data.get("status_url")
 
                             logger.debug(
-                                f"Uploaded final chunk {chunk_index + 1}/{total_chunks} of {file_name}"
+                                f"Uploaded final chunk {final_chunk_index + 1}/{total_chunks} of {file_name}"
                             )
                             return status_url
 
@@ -1293,6 +1205,62 @@ class RestClient:
             f"Failed to upload final chunk of {file_name} after {max_retries} retries"
         )
 
+    def _upload_single_chunk(
+        self,
+        job_name: str,
+        file_path: Path,
+        file_name: str,
+        upload_id: str,
+        chunk_index: int,
+        total_chunks: int,
+    ) -> None:
+        """Upload a single chunk.
+
+        Args:
+            job_name: The key of the crow to upload to.
+            file_path: The path to the file to upload.
+            file_name: The name to use for the file.
+            upload_id: The upload ID to use.
+            chunk_index: The index of this chunk.
+            total_chunks: Total number of chunks.
+
+        Raises:
+            Exception: If there's an error uploading the chunk.
+        """
+        with open(file_path, "rb") as f:
+            # Read the chunk from the file
+            f.seek(chunk_index * self.CHUNK_SIZE)
+            chunk_data = f.read(self.CHUNK_SIZE)
+
+            # Prepare and send the chunk
+            with tempfile.NamedTemporaryFile() as temp_file:
+                temp_file.write(chunk_data)
+                temp_file.flush()
+
+                # Create form data
+                with open(temp_file.name, "rb") as chunk_file_obj:
+                    files = {
+                        "chunk": (
+                            file_name,
+                            chunk_file_obj,
+                            "application/octet-stream",
+                        )
+                    }
+                    data = {
+                        "file_name": file_name,
+                        "chunk_index": chunk_index,
+                        "total_chunks": total_chunks,
+                        "upload_id": upload_id,
+                    }
+
+                    # Send the chunk
+                    response = self.multipart_client.post(
+                        f"/v0.1/crows/{job_name}/upload-chunk",
+                        files=files,
+                        data=data,
+                    )
+                    response.raise_for_status()
+
     @retry(
         stop=stop_after_attempt(MAX_RETRY_ATTEMPTS),
         wait=wait_exponential(multiplier=RETRY_MULTIPLIER, max=MAX_RETRY_WAIT),

{futurehouse_client-0.3.19.dev129.dist-info → futurehouse_client-0.3.19.dev133.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: futurehouse-client
-Version: 0.3.19.dev129
+Version: 0.3.19.dev133
 Summary: A client for interacting with endpoints of the FutureHouse service.
 Author-email: FutureHouse technical staff <hello@futurehouse.org>
 Classifier: Operating System :: OS Independent

{futurehouse_client-0.3.19.dev129.dist-info → futurehouse_client-0.3.19.dev133.dist-info}/RECORD

@@ -1,7 +1,8 @@
 futurehouse_client/__init__.py,sha256=OzGDkVm5UTUzd4n8yOmRjMF73YrK0FaIQX5gS3Dk8Zo,304
+futurehouse_client/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 futurehouse_client/clients/__init__.py,sha256=-HXNj-XJ3LRO5XM6MZ709iPs29YpApss0Q2YYg1qMZw,280
 futurehouse_client/clients/job_client.py,sha256=JgB5IUAyCmnhGRsYc3bgKldA-lkM1JLwHRwwUeOCdus,11944
-futurehouse_client/clients/rest_client.py,sha256=CiHUYJFZrfRr5mrkt41hxJMesZ2zkCBxqOFJb0t0LGo,55465
+futurehouse_client/clients/rest_client.py,sha256=3wfVz6d2KuRQUr_nms7P25yVR6aTjsRrSkqmVs55soA,54552
 futurehouse_client/models/__init__.py,sha256=5x-f9AoM1hGzJBEHcHAXSt7tPeImST5oZLuMdwp0mXc,554
 futurehouse_client/models/app.py,sha256=VCtg0ygd-TSrR6DtfljTBt9jnl1eBNal8UXHFdkDg88,28587
 futurehouse_client/models/client.py,sha256=n4HD0KStKLm6Ek9nL9ylP-bkK10yzAaD1uIDF83Qp_A,1828
@@ -11,7 +12,7 @@ futurehouse_client/utils/auth.py,sha256=tgWELjKfg8eWme_qdcRmc8TjQN9DVZuHHaVXZNHL
 futurehouse_client/utils/general.py,sha256=A_rtTiYW30ELGEZlWCIArO7q1nEmqi8hUlmBRYkMQ_c,767
 futurehouse_client/utils/module_utils.py,sha256=aFyd-X-pDARXz9GWpn8SSViUVYdSbuy9vSkrzcVIaGI,4955
 futurehouse_client/utils/monitoring.py,sha256=UjRlufe67kI3VxRHOd5fLtJmlCbVA2Wqwpd4uZhXkQM,8728
-futurehouse_client-0.3.19.dev129.dist-info/METADATA,sha256=Nd9KxyuyzbrgLFESjeo8-DwXkiK2-xtN-rfoyqaX-ys,12767
-futurehouse_client-0.3.19.dev129.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-futurehouse_client-0.3.19.dev129.dist-info/top_level.txt,sha256=TRuLUCt_qBnggdFHCX4O_BoCu1j2X43lKfIZC-ElwWY,19
-futurehouse_client-0.3.19.dev129.dist-info/RECORD,,
+futurehouse_client-0.3.19.dev133.dist-info/METADATA,sha256=fthzi_rIC5z8zWuUHtLeLOpS_hRR6Hn_6TmGlq0FOEY,12767
+futurehouse_client-0.3.19.dev133.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+futurehouse_client-0.3.19.dev133.dist-info/top_level.txt,sha256=TRuLUCt_qBnggdFHCX4O_BoCu1j2X43lKfIZC-ElwWY,19
+futurehouse_client-0.3.19.dev133.dist-info/RECORD,,