futurehouse-client 0.3.19.dev129__tar.gz → 0.3.19.dev133__tar.gz

{futurehouse_client-0.3.19.dev129 → futurehouse_client-0.3.19.dev133}/PKG-INFO

@@ -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 → futurehouse_client-0.3.19.dev133}/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 → futurehouse_client-0.3.19.dev133}/futurehouse_client.egg-info/PKG-INFO

@@ -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 → futurehouse_client-0.3.19.dev133}/futurehouse_client.egg-info/SOURCES.txt

@@ -5,6 +5,7 @@ uv.lock
 docs/__init__.py
 docs/client_notebook.ipynb
 futurehouse_client/__init__.py
+futurehouse_client/py.typed
 futurehouse_client.egg-info/PKG-INFO
 futurehouse_client.egg-info/SOURCES.txt
 futurehouse_client.egg-info/dependency_links.txt

{futurehouse_client-0.3.19.dev129 → futurehouse_client-0.3.19.dev133}/tests/test_rest.py

@@ -264,9 +264,6 @@ class TestParallelChunking:
         client._upload_single_chunk = types.MethodType(
             RestClient._upload_single_chunk, client
         )
-        client._upload_final_chunk = types.MethodType(
-            RestClient._upload_final_chunk, client
-        )
         client._upload_single_file = types.MethodType(
             RestClient._upload_single_file, client
         )
@@ -372,19 +369,34 @@ class TestParallelChunking:
         num_regular_chunks = 5  # Smaller number for easier testing
         total_chunks = 6
 
-        # Use patch to mock the _upload_single_chunk method
+        # Mock file content for final chunk
+        chunk_content = b"A" * 1000
+
+        # Mock final chunk response
+        mock_response = MagicMock()
+        mock_response.status_code = 200
+        mock_response.raise_for_status.return_value = None
+        mock_response.json.return_value = {"status_url": "http://test.com/status"}
+        mock_client.multipart_client.post.return_value = mock_response
+
+        # Use patch to mock the _upload_single_chunk method and file operations
         with patch.object(mock_client, "_upload_single_chunk") as mock_upload_chunk:
-            # Call the method - it should use ThreadPoolExecutor internally
-            mock_client._upload_chunks_parallel(
-                job_name,
-                file_path,
-                file_name,
-                upload_id,
-                num_regular_chunks,
-                total_chunks,
-            )
-
-            # Verify all chunks were processed by checking the call count
+            with patch("builtins.open", mock_open(read_data=chunk_content)):
+                with patch("tempfile.NamedTemporaryFile") as mock_temp_file:
+                    mock_temp_file.return_value.__enter__.return_value.name = (
+                        "temp_chunk"
+                    )
+
+                    # Call the method - it should use ThreadPoolExecutor internally
+                    mock_client._upload_chunks_parallel(
+                        job_name,
+                        file_path,
+                        file_name,
+                        upload_id,
+                        total_chunks,
+                    )
+
+            # Verify all regular chunks were processed by checking the call count
             assert mock_upload_chunk.call_count == num_regular_chunks
 
             # Verify the calls were made with correct parameters
@@ -437,14 +449,13 @@ class TestParallelChunking:
                 assert data["total_chunks"] == total_chunks
                 assert data["upload_id"] == upload_id
 
-    def test_upload_final_chunk_with_retry_on_conflict(self, mock_client):
+    def test_upload_chunks_parallel_retry_on_conflict(self, mock_client):
         """Test final chunk upload with retry logic for missing chunks (409 conflict)."""
         job_name = "test-job"
         file_path = Path("test_file.txt")
         file_name = "test_file.txt"
         upload_id = "test-upload-id"
-        chunk_index = 2
-        total_chunks = 3
+        total_chunks = 1
 
         # Mock file content
         chunk_content = b"A" * 1000
@@ -473,12 +484,11 @@ class TestParallelChunking:
                         "temp_chunk"
                     )
 
-                    status_url = mock_client._upload_final_chunk(
+                    status_url = mock_client._upload_chunks_parallel(
                         job_name,
                         file_path,
                         file_name,
                         upload_id,
-                        chunk_index,
                         total_chunks,
                     )
 
@@ -487,29 +497,24 @@ class TestParallelChunking:
                     assert status_url == "http://test.com/status"
                     mock_sleep.assert_called_once()  # Verify sleep was called for retry
 
-    def test_upload_final_chunk_max_retries_exceeded(self, mock_client):
+    def test_upload_chunks_parallel_final_chunk_max_retries_exceeded(self, mock_client):
         """Test final chunk upload fails after max retries."""
         job_name = "test-job"
         file_path = Path("test_file.txt")
         file_name = "test_file.txt"
         upload_id = "test-upload-id"
-        chunk_index = 2
-        total_chunks = 3
+        total_chunks = 1
 
         # Mock file content
         chunk_content = b"A" * 1000
 
-        # Mock response that always returns 409 (conflict) and raises an exception on raise_for_status
-        mock_response = MagicMock()
-        mock_response.status_code = 409
-        # Make raise_for_status raise an exception after the retries are exhausted
-        from httpx import HTTPStatusError, Request, codes
+        # Create a side effect that simulates an exception on every attempt
+        def side_effect(*args, **kwargs):
+            raise Exception("Simulated upload failure")  # noqa: TRY002
 
-        mock_request = MagicMock(spec=Request)
-        mock_response.raise_for_status.side_effect = HTTPStatusError(
-            "409 Conflict", request=mock_request, response=mock_response
-        )
-        mock_client.multipart_client.post.return_value = mock_response
+        mock_client.multipart_client.post.side_effect = side_effect
+
+        from httpx import codes
 
         with patch("builtins.open", mock_open(read_data=chunk_content)):
             with patch("tempfile.NamedTemporaryFile") as mock_temp_file:
@@ -523,17 +528,16 @@ class TestParallelChunking:
                         with pytest.raises(
                             FileUploadError, match="Error uploading final chunk"
                         ):
-                            mock_client._upload_final_chunk(
+                            mock_client._upload_chunks_parallel(
                                 job_name,
                                 file_path,
                                 file_name,
                                 upload_id,
-                                chunk_index,
                                 total_chunks,
                             )
 
-                        # Verify that retries were attempted (should be 3 attempts total)
-                        assert mock_client.multipart_client.post.call_count == 3
+                # Verify that retries were attempted (should be 3 attempts total)
+                assert mock_client.multipart_client.post.call_count == 3
 
     def test_upload_directory_recursive(self, mock_client):
         """Test uploading a directory with nested files."""