Improve download experience

.pre-commit-config.yaml

@@ -45,13 +45,13 @@ repos:
       - id: no-commit-to-branch
         args: [--branch, dev, --branch, int, --branch, main]
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.6.3
+    rev: v0.8.2
     hooks:
       - id: ruff
         args: [--fix, --exit-non-zero-on-fix]
       - id: ruff-format
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.11.2
+    rev: v1.13.0
     hooks:
       - id: mypy
         args: [--no-warn-unused-ignores]

.pyproject_generation/pyproject_custom.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ghga_connector"
-version = "1.5.1"
+version = "1.5.2"
 description = "GHGA Connector - A CLI client application for interacting with the GHGA system."
 dependencies = [
     "typer~=0.12",

README.md

@@ -26,21 +26,21 @@ We recommend using the provided Docker container.
 
 A pre-build version is available at [docker hub](https://hub.docker.com/repository/docker/ghga/ghga-connector):
 ```bash
-docker pull ghga/ghga-connector:1.5.1
+docker pull ghga/ghga-connector:1.5.2
 ```
 
 Or you can build the container yourself from the [`./Dockerfile`](./Dockerfile):
 ```bash
 # Execute in the repo's root dir:
-docker build -t ghga/ghga-connector:1.5.1 .
+docker build -t ghga/ghga-connector:1.5.2 .
 ```
 
 For production-ready deployment, we recommend using Kubernetes, however,
 for simple use cases, you could execute the service using docker
 on a single server:
 ```bash
 # The entrypoint is preconfigured:
-docker run -p 8080:8080 ghga/ghga-connector:1.5.1 --help
+docker run -p 8080:8080 ghga/ghga-connector:1.5.2 --help
 ```
 
 If you prefer not to use containers, you may install the service from source:

pyproject.toml

@@ -21,7 +21,7 @@ classifiers = [
     "Intended Audience :: Developers",
 ]
 name = "ghga_connector"
-version = "1.5.1"
+version = "1.5.2"
 description = "GHGA Connector - A CLI client application for interacting with the GHGA system."
 dependencies = [
     "typer~=0.12",

src/ghga_connector/cli.py

@@ -32,7 +32,6 @@
 from ghga_connector.config import CONFIG
 from ghga_connector.core import (
     AbstractMessageDisplay,
-    HttpxClientConfigurator,
     MessageColors,
     WorkPackageAccessor,
     async_client,
@@ -42,7 +41,7 @@
 from ghga_connector.core.downloading.batch_processing import FileStager
 from ghga_connector.core.main import (
     decrypt_file,
-    download_files,
+    download_file,
     get_wps_token,
     upload_file,
 )
@@ -112,7 +111,7 @@ def exception_hook(
     )
 
     if value.args:
-        message = value.args[0]
+        message += f"\n{value.args[0]}"
 
     message_display.failure(message)
 
@@ -215,11 +214,6 @@ async def async_upload(
 ):
     """Upload a file asynchronously"""
     message_display = init_message_display(debug=debug)
-    HttpxClientConfigurator.configure(
-        exponential_backoff_max=CONFIG.exponential_backoff_max,
-        max_retries=CONFIG.max_retries,
-        retry_status_codes=CONFIG.retry_status_codes,
-    )
     async with async_client() as client:
         parameters = await retrieve_upload_parameters(client)
         await upload_file(
@@ -260,10 +254,16 @@ def download(
     debug: bool = typer.Option(
         False, help="Set this option in order to view traceback for errors."
     ),
+    overwrite: bool = typer.Option(
+        False,
+        help="Set to true to overwrite already existing files in the output directory.",
+    ),
 ):
     """Wrapper for the async download function"""
     asyncio.run(
-        async_download(output_dir, my_public_key_path, my_private_key_path, debug)
+        async_download(
+            output_dir, my_public_key_path, my_private_key_path, debug, overwrite
+        )
     )
 
 
@@ -272,6 +272,7 @@ async def async_download(
     my_public_key_path: Path,
     my_private_key_path: Path,
     debug: bool = False,
+    overwrite: bool = False,
 ):
     """Download files asynchronously"""
     if not my_public_key_path.is_file():
@@ -286,11 +287,6 @@ async def async_download(
     )
 
     message_display = init_message_display(debug=debug)
-    HttpxClientConfigurator.configure(
-        exponential_backoff_max=CONFIG.exponential_backoff_max,
-        max_retries=CONFIG.max_retries,
-        retry_status_codes=CONFIG.retry_status_codes,
-    )
     message_display.display("\nFetching work package token...")
     work_package_information = get_work_package_information(
         my_private_key=my_private_key, message_display=message_display
@@ -305,6 +301,7 @@ async def async_download(
             work_package_information=work_package_information,
         )
 
+        message_display.display("Preparing files for download...")
         stager = FileStager(
             wanted_file_ids=list(parameters.file_ids_with_extension),
             dcs_api_url=parameters.dcs_api_url,
@@ -318,7 +315,7 @@ async def async_download(
             staged_files = await stager.get_staged_files()
             for file_id in staged_files:
                 message_display.display(f"Downloading file with id '{file_id}'...")
-                await download_files(
+                await download_file(
                     api_url=parameters.dcs_api_url,
                     client=client,
                     file_id=file_id,
@@ -329,6 +326,7 @@ async def async_download(
                     part_size=CONFIG.part_size,
                     message_display=message_display,
                     work_package_accessor=parameters.work_package_accessor,
+                    overwrite=overwrite,
                 )
             staged_files.clear()
 

src/ghga_connector/core/__init__.py

@@ -18,7 +18,7 @@
 It should not contain any service API-related code.
 """
 
-from .client import HttpxClientConfigurator, async_client, httpx_client  # noqa: F401
+from .client import async_client, httpx_client, retry_handler  # noqa: F401
 from .file_operations import (  # noqa: F401
     calc_part_ranges,
     get_segments,

src/ghga_connector/core/client.py

@@ -15,6 +15,7 @@
 """Handling session initialization for httpx"""
 
 from contextlib import asynccontextmanager, contextmanager
+from functools import cached_property
 
 import httpx
 from tenacity import (
@@ -25,23 +26,17 @@
     wait_exponential_jitter,
 )
 
+from ghga_connector.config import CONFIG
 from ghga_connector.constants import TIMEOUT
 
 
 class HttpxClientConfigurator:
     """Helper class to make max_retries user configurable"""
 
-    retry_handler: AsyncRetrying
-
-    @classmethod
-    def configure(
-        cls,
-        exponential_backoff_max: int,
-        max_retries: int,
-        retry_status_codes: list[int],
-    ):
+    @cached_property
+    def retry_handler(self):
         """Configure client retry handler with exponential backoff"""
-        cls.retry_handler = AsyncRetrying(
+        return AsyncRetrying(
             reraise=True,
             retry=(
                 retry_if_exception_type(
@@ -52,23 +47,38 @@ def configure(
                     )
                 )
                 | retry_if_result(
-                    lambda response: response.status_code in retry_status_codes
+                    lambda response: response.status_code in CONFIG.retry_status_codes
                 )
             ),
-            stop=stop_after_attempt(max_retries),
-            wait=wait_exponential_jitter(max=exponential_backoff_max),
+            stop=stop_after_attempt(CONFIG.max_retries),
+            wait=wait_exponential_jitter(max=CONFIG.exponential_backoff_max),
         )
 
 
+retry_handler = HttpxClientConfigurator().retry_handler
+
+
 @contextmanager
 def httpx_client():
     """Yields a context manager httpx client and closes it afterward"""
-    with httpx.Client(timeout=TIMEOUT) as client:
+    with httpx.Client(
+        timeout=TIMEOUT,
+        limits=httpx.Limits(
+            max_connections=CONFIG.max_concurrent_downloads,
+            max_keepalive_connections=CONFIG.max_concurrent_downloads,
+        ),
+    ) as client:
         yield client
 
 
 @asynccontextmanager
 async def async_client():
     """Yields a context manager async httpx client and closes it afterward"""
-    async with httpx.AsyncClient(timeout=TIMEOUT) as client:
+    async with httpx.AsyncClient(
+        timeout=TIMEOUT,
+        limits=httpx.Limits(
+            max_connections=CONFIG.max_concurrent_downloads,
+            max_keepalive_connections=CONFIG.max_concurrent_downloads,
+        ),
+    ) as client:
         yield client

src/ghga_connector/core/downloading/abstract_downloader.py

@@ -22,6 +22,7 @@
 from typing import Any
 
 from ghga_connector.core import PartRange
+from ghga_connector.core.downloading.progress_bar import ProgressBar
 
 from .structs import URLResponse
 
@@ -34,17 +35,13 @@ def download_file(self, *, output_path: Path, part_size: int):
         """Download file to the specified location and manage lower level details."""
 
     @abstractmethod
-    def await_download_url(self) -> Coroutine[URLResponse, Any, Any]:
+    def fetch_download_url(self) -> Coroutine[URLResponse, Any, Any]:
         """Wait until download URL can be generated.
         Returns a URLResponse containing two elements:
             1. the download url
             2. the file size in bytes
         """
 
-    @abstractmethod
-    def get_download_url(self) -> Coroutine[URLResponse, Any, Any]:
-        """Fetch a presigned URL from which file data can be downloaded."""
-
     @abstractmethod
     def get_file_header_envelope(self) -> Coroutine[bytes, Any, Any]:
         """
@@ -54,7 +51,7 @@ def get_file_header_envelope(self) -> Coroutine[bytes, Any, Any]:
         """
 
     @abstractmethod
-    async def download_to_queue(self, *, part_range: PartRange) -> None:
+    async def download_to_queue(self, *, url: str, part_range: PartRange) -> None:
         """
         Start downloading file parts in parallel into a queue.
         This should be wrapped into asyncio.task and is guarded by a semaphore to limit
@@ -65,14 +62,20 @@ async def download_to_queue(self, *, part_range: PartRange) -> None:
     async def download_content_range(
         self,
         *,
+        url: str,
         start: int,
         end: int,
     ) -> None:
         """Download a specific range of a file's content using a presigned url."""
 
     @abstractmethod
     async def drain_queue_to_file(
-        self, *, file_name: str, file: BufferedWriter, file_size: int, offset: int
+        self,
+        *,
+        file: BufferedWriter,
+        file_size: int,
+        offset: int,
+        progress_bar: ProgressBar,
     ) -> None:
         """Write downloaded file bytes from queue.
         This should be started as asyncio.Task and awaited after the download_to_queue

src/ghga_connector/core/downloading/batch_processing.py

@@ -85,7 +85,7 @@ def get_input(self, *, message: str) -> str:
 
     def handle_response(self, *, response: str):
         """Handle response from get_input."""
-        if not response.lower() == "yes":
+        if not (response.lower() == "yes" or response.lower() == "y"):
             raise exceptions.AbortBatchProcessError()
 
 
@@ -158,7 +158,7 @@ def __init__(  # noqa: PLR0913
         self.work_package_accessor = work_package_accessor
         self.max_wait_time = config.max_wait_time
         self.client = client
-        self.time_started = now = time()
+        self.started_waiting = now = time()
 
         # Successfully staged files with their download URLs and sizes
         # in the beginning, consider all files as staged with a retry time of 0
@@ -180,10 +180,14 @@ async def get_staged_files(self) -> dict[str, URLResponse]:
         These values contain the download URLs and file sizes.
         The dict should cleared after these files have been downloaded.
         """
+        self.message_display.display("Updating list of staged files...")
         staging_items = list(self.unstaged_retry_times.items())
         for file_id, retry_time in staging_items:
             if time() >= retry_time:
                 await self._check_file(file_id=file_id)
+            if len(self.staged_urls.items()) > 0:
+                self.started_waiting = time()  # reset wait timer
+                break
         if not self.staged_urls and not self._handle_failures():
             sleep(1)
         self._check_timeout()
@@ -217,8 +221,10 @@ async def _check_file(self, file_id: str) -> None:
         if isinstance(response, URLResponse):
             del self.unstaged_retry_times[file_id]
             self.staged_urls[file_id] = response
+            self.message_display.display(f"File {file_id} is ready for download.")
         elif isinstance(response, RetryResponse):
             self.unstaged_retry_times[file_id] = time() + response.retry_after
+            self.message_display.display(f"File {file_id} is (still) being staged.")
         else:
             self.missing_files.append(file_id)
 
@@ -227,7 +233,7 @@ def _check_timeout(self):
 
         In that cases, a MaxWaitTimeExceededError is raised.
         """
-        if time() - self.time_started >= self.max_wait_time:
+        if time() - self.started_waiting >= self.max_wait_time:
             raise exceptions.MaxWaitTimeExceededError(max_wait_time=self.max_wait_time)
 
     def _handle_failures(self) -> bool:
@@ -238,8 +244,8 @@ def _handle_failures(self) -> bool:
         """
         if not self.missing_files or self.ignore_failed:
             return False
-        failed = ", ".join(self.missing_files)
-        message = f"No download exists for the following file IDs: {failed}"
+        missing = ", ".join(self.missing_files)
+        message = f"No download exists for the following file IDs: {missing}"
         self.message_display.failure(message)
         if self.finished:
             return False
@@ -250,5 +256,6 @@ def _handle_failures(self) -> bool:
         response = self.io_handler.get_input(message=unknown_ids_present)
         self.io_handler.handle_response(response=response)
         self.message_display.display("Downloading remaining files")
-        self.time_started = time()  # reset the timer
+        self.started_waiting = time()  # reset the timer
+        self.missing_files = []  # reset list of missing files
         return True