Port quickfix from exploratory branch

Author: mephenor

.pre-commit-config.yaml

@@ -45,7 +45,7 @@ repos:
       - id: no-commit-to-branch
         args: [--branch, dev, --branch, int, --branch, main]
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.9.9
+    rev: v0.11.0
     hooks:
       - id: ruff
         args: [--fix, --exit-non-zero-on-fix]

.pyproject_generation/pyproject_custom.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ghga_connector"
-version = "1.7.0"
+version = "1.7.1"
 description = "GHGA Connector - A CLI client application for interacting with the GHGA system."
 dependencies = [
     "typer~=0.15",

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:1.7.0
+docker pull ghga/ghga-connector:1.7.1
 ```
 
 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:1.7.0 .
+docker build -t ghga/ghga-connector:1.7.1 .
 ```
 
 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:1.7.0 --help
+docker run -p 8080:8080 ghga/ghga-connector:1.7.1 --help
 ```
 
 If you prefer not to use containers, you may install the service from source:

lock/requirements-dev.txt

@@ -4,9 +4,9 @@ annotated-types==0.7.0 \
     # via
     #   -c lock/requirements-dev.txt
     #   pydantic
-anyio==4.8.0 \
-    --hash=sha256:1d9fe889df5212298c0c0723fa20479d1b94883a2df44bd3897aa91083316f7a \
-    --hash=sha256:b5011f270ab5eb0abf13385f851315585cc37ef330dd88e27ec3d34d651fd47a
+anyio==4.9.0 \
+    --hash=sha256:673c0c244e15788651a4ff38710fea9675823028a6f08a5eda409e0c9840a028 \
+    --hash=sha256:9f76d541cad6e36af7beb62e978876f3b41e3e04f2c1fbf0884604c0a9c4d93c
     # via
     #   -c lock/requirements-dev.txt
     #   httpx
@@ -65,14 +65,15 @@ bcrypt==4.3.0 \
     # via
     #   -c lock/requirements-dev.txt
     #   crypt4gh
-boto3==1.37.6 \
-    --hash=sha256:4c661389e68437a3fbc1f63decea24b88f7175e022c68622848d47fdf6e0144f
+boto3==1.37.14 \
+    --hash=sha256:56b4d1e084dbca43d5fdd070f633a84de61a6ce592655b4d239d263d1a0097fc \
+    --hash=sha256:cf2e5e6d56efd5850db8ce3d9094132e4759cf2d4b5fd8200d69456bf61a20f3
     # via
     #   -c lock/requirements-dev.txt
     #   hexkit
-botocore==1.37.6 \
-    --hash=sha256:2cb121a403cbec047d76e2401a402a6b2efd3309169037fbac588e8f7125aec4 \
-    --hash=sha256:cd282fe9c8adbb55a08c7290982a98ac6cc4507fa1c493f48bc43fd6c8376a57
+botocore==1.37.14 \
+    --hash=sha256:709a1796f436f8e378e52170e58501c1f3b5f2d1308238cf1d6a3bdba2e32851 \
+    --hash=sha256:b0adce3f0fb42b914eb05079f50cf368cb9cf9745fdd206bd91fe6ac67b29aca
     # via
     #   -c lock/requirements-dev.txt
     #   boto3
@@ -220,9 +221,9 @@ exceptiongroup==1.2.2 \
     # via
     #   -c lock/requirements-dev.txt
     #   anyio
-ghga-service-commons==3.4.0 \
-    --hash=sha256:4301e60d5aff771bee6d9851a5cd5989ddef9bceab95bda108dd20a21d64c469 \
-    --hash=sha256:f525693b66fad3e779d8c873252f21fedf480a255ebc557ce6bbeba09e718d14
+ghga-service-commons==3.4.1 \
+    --hash=sha256:2ced19b2b0e07d7471b68a73eacb3272473eae74b7e74b7b101b0c90ea4a0170 \
+    --hash=sha256:37d9cf122e776a70fd15d3dfa355dd386d5b3426514a3b3245207875fcbce9eb
     # via
     #   -c lock/requirements-dev.txt
     #   ghga-connector (pyproject.toml)

lock/requirements.txt

@@ -21,7 +21,7 @@ classifiers = [
     "Intended Audience :: Developers",
 ]
 name = "ghga_connector"
-version = "1.7.0"
+version = "1.7.1"
 description = "GHGA Connector - A CLI client application for interacting with the GHGA system."
 dependencies = [
     "typer~=0.15",

pyproject.toml

@@ -16,6 +16,7 @@
 """CLI-specific wrappers around core functions."""
 
 import asyncio
+import logging
 import os
 import sys
 from dataclasses import dataclass
@@ -301,6 +302,10 @@ async def async_download(  # noqa: PLR0913
     overwrite: bool = False,
 ):
     """Download files asynchronously"""
+    # enable debug logging
+    if debug:
+        logging.basicConfig(level=logging.DEBUG)
+
     if not my_public_key_path.is_file():
         raise exceptions.PubKeyFileDoesNotExistError(public_key_path=my_public_key_path)
 

src/ghga_connector/cli.py

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

src/ghga_connector/core/__init__.py

@@ -14,8 +14,7 @@
 # limitations under the License.
 """Handling session initialization for httpx"""
 
-from contextlib import asynccontextmanager, contextmanager
-from functools import cached_property
+from contextlib import asynccontextmanager
 from typing import Union
 
 import hishel
@@ -33,11 +32,11 @@
 from ghga_connector.constants import TIMEOUT
 
 
-class HttpxClientConfigurator:
+class RetryHandler:
     """Helper class to make max_retries user configurable"""
 
-    @cached_property
-    def retry_handler(self):
+    @classmethod
+    def basic(cls):
         """Configure client retry handler with exponential backoff"""
         return AsyncRetrying(
             reraise=True,
@@ -58,22 +57,6 @@ def retry_handler(self):
         )
 
 
-retry_handler = HttpxClientConfigurator().retry_handler
-
-
-@contextmanager
-def httpx_client():
-    """Yields a context manager httpx client and closes it afterward"""
-    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
-
-
 def get_cache_transport(
     wrapped_transport: Union[httpx.AsyncBaseTransport, None] = None,
 ) -> hishel.AsyncCacheTransport:
@@ -84,7 +67,8 @@ def get_cache_transport(
     """
     cache_transport = hishel.AsyncCacheTransport(
         transport=wrapped_transport or httpx.AsyncHTTPTransport(),
-        storage=hishel.AsyncInMemoryStorage(ttl=1800),  # persist for 30 minutes
+        # set ttl to expected lifetime of presigned URL - min-fresh
+        storage=hishel.AsyncInMemoryStorage(ttl=57, capacity=512),
         controller=hishel.Controller(
             cacheable_methods=["POST", "GET"],
             cacheable_status_codes=[200, 201],

src/ghga_connector/core/client.py

@@ -35,7 +35,7 @@ def download_file(self, *, output_path: Path, part_size: int):
         """Download file to the specified location and manage lower level details."""
 
     @abstractmethod
-    def fetch_download_url(self) -> Coroutine[URLResponse, Any, Any]:
+    def fetch_download_url(self, bust_cache: bool) -> Coroutine[URLResponse, Any, Any]:
         """Wait until download URL can be generated.
         Returns a URLResponse containing two elements:
             1. the download url