futurehouse-client 0.3.16__py3-none-any.whl → 0.3.17.dev56__py3-none-any.whl

futurehouse_client/clients/job_client.py

@@ -29,6 +29,7 @@ class JobNames(StrEnum):
     FALCON = "job-futurehouse-paperqa2-deep"
     OWL = "job-futurehouse-hasanyone"
     DUMMY = "job-futurehouse-dummy-env"
+    PHOENIX = "job-futurehouse-phoenix"
 
     @classmethod
     def from_stage(cls, job_name: str, stage: Stage | None = None) -> str:

futurehouse_client/clients/rest_client.py

@@ -6,6 +6,8 @@ import inspect
 import json
 import logging
 import os
+import tempfile
+import uuid
 from collections.abc import Mapping
 from datetime import datetime
 from pathlib import Path
@@ -200,11 +202,16 @@ class TaskResponseVerbose(TaskResponse):
     shared_with: list[SimpleOrganization] | None = None
 
 
+class FileUploadError(RestClientError):
+    """Raised when there's an error uploading a file."""
+
+
 class RestClient:
     REQUEST_TIMEOUT: ClassVar[float] = 30.0  # sec
     MAX_RETRY_ATTEMPTS: ClassVar[int] = 3
     RETRY_MULTIPLIER: ClassVar[int] = 1
     MAX_RETRY_WAIT: ClassVar[int] = 10
+    CHUNK_SIZE: ClassVar[int] = 16 * 1024 * 1024  # 16MB chunks
 
     def __init__(
         self,
@@ -644,6 +651,243 @@ class RestClient:
             raise JobCreationError(f"Error generating docker image: {e!s}") from e
         return build_context
 
+    @retry(
+        stop=stop_after_attempt(MAX_RETRY_ATTEMPTS),
+        wait=wait_exponential(multiplier=RETRY_MULTIPLIER, max=MAX_RETRY_WAIT),
+        retry=retry_if_connection_error,
+    )
+    def upload_file(
+        self,
+        job_name: str,
+        file_path: str | os.PathLike,
+        folder_name: str | None = None,
+    ) -> str:
+        """Upload a file or directory to a futurehouse job bucket.
+
+        Args:
+            job_name: The name of the futurehouse job to upload to.
+            file_path: The local path to the file or directory to upload.
+            folder_name: Optional folder name to use for the upload. If not provided, a random UUID will be used.
+
+        Returns:
+            The upload ID used for the upload.
+
+        Raises:
+            FileUploadError: If there's an error uploading the file.
+        """
+        file_path = Path(file_path)
+        if not file_path.exists():
+            raise FileNotFoundError(f"File or directory not found: {file_path}")
+
+        upload_id = folder_name or str(uuid.uuid4())
+
+        if file_path.is_dir():
+            # Process directory recursively
+            self._upload_directory(job_name, file_path, upload_id)
+        else:
+            # Process single file
+            self._upload_single_file(job_name, file_path, upload_id)
+        logger.info(f"Successfully uploaded {file_path} to {upload_id}")
+        return upload_id
+
+    def _upload_directory(self, job_name: str, dir_path: Path, upload_id: str) -> None:
+        """Upload all files in a directory recursively.
+
+        Args:
+            job_name: The key of the crow to upload to.
+            dir_path: The path to the directory to upload.
+            upload_id: The upload ID to use.
+
+        Raises:
+            FileUploadError: If there's an error uploading any file.
+        """
+        # Skip common directories that shouldn't be uploaded
+        if any(ignore in dir_path.parts for ignore in FILE_UPLOAD_IGNORE_PARTS):
+            return
+
+        try:
+            # Upload all files in the directory recursively
+            for path in dir_path.rglob("*"):
+                if path.is_file() and not any(
+                    ignore in path.parts for ignore in FILE_UPLOAD_IGNORE_PARTS
+                ):
+                    # Use path relative to the original directory as file name
+                    rel_path = path.relative_to(dir_path)
+                    self._upload_single_file(
+                        job_name,
+                        path,
+                        upload_id,
+                        file_name=str(rel_path),
+                    )
+        except Exception as e:
+            raise FileUploadError(f"Error uploading directory {dir_path}: {e}") from e
+
+    def _upload_single_file(
+        self,
+        job_name: str,
+        file_path: Path,
+        upload_id: str,
+        file_name: str | None = None,
+    ) -> None:
+        """Upload a single file in chunks.
+
+        Args:
+            job_name: The key of the crow to upload to.
+            file_path: The path to the file to upload.
+            upload_id: The upload ID to use.
+            file_name: Optional name to use for the file. If not provided, the file's name will be used.
+
+        Raises:
+            FileUploadError: If there's an error uploading the file.
+        """
+        file_name = file_name or file_path.name
+        file_size = file_path.stat().st_size
+        total_chunks = (file_size + self.CHUNK_SIZE - 1) // self.CHUNK_SIZE
+
+        logger.info(f"Uploading {file_path} as {file_name} ({total_chunks} chunks)")
+
+        try:
+            with open(file_path, "rb") as f:
+                for chunk_index in range(total_chunks):
+                    # 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()
+
+                        # Call progress callback if provided
+
+                        logger.debug(
+                            f"Uploaded chunk {chunk_index + 1}/{total_chunks} of {file_name}"
+                        )
+
+            logger.info(f"Successfully uploaded {file_name}")
+
+        except Exception as e:
+            logger.exception(f"Error uploading file {file_path}")
+            raise FileUploadError(f"Error uploading file {file_path}: {e}") from e
+
+    @retry(
+        stop=stop_after_attempt(MAX_RETRY_ATTEMPTS),
+        wait=wait_exponential(multiplier=RETRY_MULTIPLIER, max=MAX_RETRY_WAIT),
+        retry=retry_if_connection_error,
+    )
+    def list_files(self, job_name: str, folder_name: str) -> dict[str, list[str]]:
+        """List files and directories in a GCS location for a given job_name and upload_id.
+
+        Args:
+            job_name: The name of the futurehouse job.
+            folder_name: The specific folder name (upload_id) to list files from.
+
+        Returns:
+            A list of files in the GCS folder.
+
+        Raises:
+            RestClientError: If there is an error listing the files.
+        """
+        try:
+            url = f"/v0.1/crows/{job_name}/list-files"
+            params = {"upload_id": folder_name}
+            response = self.client.get(url, params=params)
+            response.raise_for_status()
+            return response.json()
+        except HTTPStatusError as e:
+            logger.exception(
+                f"Error listing files for job {job_name}, folder {folder_name}: {e.response.text}"
+            )
+            raise RestClientError(
+                f"Error listing files: {e.response.status_code} - {e.response.text}"
+            ) from e
+        except Exception as e:
+            logger.exception(
+                f"Error listing files for job {job_name}, folder {folder_name}"
+            )
+            raise RestClientError(f"Error listing files: {e!s}") from e
+
+    @retry(
+        stop=stop_after_attempt(MAX_RETRY_ATTEMPTS),
+        wait=wait_exponential(multiplier=RETRY_MULTIPLIER, max=MAX_RETRY_WAIT),
+        retry=retry_if_connection_error,
+    )
+    def download_file(
+        self,
+        job_name: str,
+        folder_name: str,
+        file_path: str,
+        destination_path: str | os.PathLike,
+    ) -> None:
+        """Download a file from GCS to a local path.
+
+        Args:
+            job_name: The name of the futurehouse job.
+            folder_name: The specific folder name (upload_id) the file belongs to.
+            file_path: The relative path of the file to download
+                       (e.g., 'data/my_file.csv' or 'my_image.png').
+            destination_path: The local path where the file should be saved.
+
+        Raises:
+            RestClientError: If there is an error downloading the file.
+            FileNotFoundError: If the destination directory does not exist.
+        """
+        destination_path = Path(destination_path)
+        # Ensure the destination directory exists
+        destination_path.parent.mkdir(parents=True, exist_ok=True)
+
+        try:
+            url = f"/v0.1/crows/{job_name}/download-file"
+            params = {"upload_id": folder_name, "file_path": file_path}
+
+            with self.client.stream("GET", url, params=params) as response:
+                response.raise_for_status()  # Check for HTTP errors before streaming
+                with open(destination_path, "wb") as f:
+                    for chunk in response.iter_bytes(chunk_size=8192):
+                        f.write(chunk)
+            logger.info(f"File {file_path} downloaded to {destination_path}")
+        except HTTPStatusError as e:
+            logger.exception(
+                f"Error downloading file {file_path} for job {job_name}, folder {folder_name}: {e.response.text}"
+            )
+            # Clean up partially downloaded file if an error occurs
+            if destination_path.exists():
+                destination_path.unlink()
+            raise RestClientError(
+                f"Error downloading file: {e.response.status_code} - {e.response.text}"
+            ) from e
+        except Exception as e:
+            logger.exception(
+                f"Error downloading file {file_path} for job {job_name}, folder {folder_name}"
+            )
+            if destination_path.exists():
+                destination_path.unlink()  # Clean up partial file
+            raise RestClientError(f"Error downloading file: {e!s}") from e
+
 
 def get_installed_packages() -> dict[str, str]:
     """Returns a dictionary of installed packages and their versions."""

{futurehouse_client-0.3.16.dist-info → futurehouse_client-0.3.17.dev56.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: futurehouse-client
-Version: 0.3.16
+Version: 0.3.17.dev56
 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-0.3.16.dist-info → futurehouse_client-0.3.17.dev56.dist-info}/RECORD

@@ -1,7 +1,7 @@
 futurehouse_client/__init__.py,sha256=ddxO7JE97c6bt7LjNglZZ2Ql8bYCGI9laSFeh9MP6VU,344
 futurehouse_client/clients/__init__.py,sha256=tFWqwIAY5PvwfOVsCje4imjTpf6xXNRMh_UHIKVI1_0,320
-futurehouse_client/clients/job_client.py,sha256=yBFKDNcFnuZDNgoK2d5037rbuzQ7TlSK6MmklEKV8EA,11056
-futurehouse_client/clients/rest_client.py,sha256=Dc29QRNZMO4uxaXNGKyx18Tn-vLaJ6P5fCbM_0u-Z3I,26379
+futurehouse_client/clients/job_client.py,sha256=Fi3YvN4k82AuXCe8vlwxhkK8CXS164NQrs7paj9qIek,11096
+futurehouse_client/clients/rest_client.py,sha256=OBJeRSQezd2BSJGHKQZ4Cg1uhThtOKwgBOSEDI4n0go,36181
 futurehouse_client/models/__init__.py,sha256=ta3jFLM_LsDz1rKDmx8rja8sT7WtSKoFvMgLF0yFpvA,342
 futurehouse_client/models/app.py,sha256=yfZ9tyw4VATVAfYrU7aTdCNPSljLEho09_nIbh8oZDY,23174
 futurehouse_client/models/client.py,sha256=n4HD0KStKLm6Ek9nL9ylP-bkK10yzAaD1uIDF83Qp_A,1828
@@ -9,7 +9,7 @@ futurehouse_client/models/rest.py,sha256=W-wNFTN7HALYFFphw-RQYRMm6_TSa1cl4T-mZ1m
 futurehouse_client/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 futurehouse_client/utils/module_utils.py,sha256=aFyd-X-pDARXz9GWpn8SSViUVYdSbuy9vSkrzcVIaGI,4955
 futurehouse_client/utils/monitoring.py,sha256=UjRlufe67kI3VxRHOd5fLtJmlCbVA2Wqwpd4uZhXkQM,8728
-futurehouse_client-0.3.16.dist-info/METADATA,sha256=uCvzXKeI6i8PRvike8YKVa7-IJQAwL8G2ILjjOf6xIo,8175
-futurehouse_client-0.3.16.dist-info/WHEEL,sha256=0CuiUZ_p9E4cD6NyLD6UG80LBXYyiSYZOKDm5lp32xk,91
-futurehouse_client-0.3.16.dist-info/top_level.txt,sha256=TRuLUCt_qBnggdFHCX4O_BoCu1j2X43lKfIZC-ElwWY,19
-futurehouse_client-0.3.16.dist-info/RECORD,,
+futurehouse_client-0.3.17.dev56.dist-info/METADATA,sha256=KZWJ9eKHsx4-VjrG_O-dB__TqrIK-jeU5kcNHfKPoaI,8181
+futurehouse_client-0.3.17.dev56.dist-info/WHEEL,sha256=0CuiUZ_p9E4cD6NyLD6UG80LBXYyiSYZOKDm5lp32xk,91
+futurehouse_client-0.3.17.dev56.dist-info/top_level.txt,sha256=TRuLUCt_qBnggdFHCX4O_BoCu1j2X43lKfIZC-ElwWY,19
+futurehouse_client-0.3.17.dev56.dist-info/RECORD,,