feat: use top-level key for repo type instead of type field

Replace the flat `remotes:` map (with `type: "remote"/"virtual"/"local"`) with
separate top-level sections — `remote:`, `virtual:`, `local:` — so the repo
type is declared structurally and the `type:` field is no longer needed.

Config loader normalises the new format to the existing internal representation
(injecting `type` into each remote dict), so all handler code is unchanged.
Adds a TestYamlTypeKeys suite covering all three type keys, mixed files, and
field preservation. Includes README migration guide for splitting a single
remotes file into per-type-and-package conf.d files.

.dockerignore

@@ -0,0 +1,15 @@
+.git/
+.venv/
+dist/
+tests/
+remotes.yaml
+ca-bundle.pem
+.env
+*.log
+docker-compose.yml
+.woodpecker/
+.tox/
+.ruff_cache/
+.pytest_cache/
+.pre-commit-cache/
+minio_data/

.gitignore

@@ -35,7 +35,6 @@ env/
 
 # Environment variables
 .env
-remotes.yaml
 
 # Logs
 *.log
@@ -43,5 +42,20 @@ remotes.yaml
 # uv
 uv.lock
 
+# tox
+.tox/
+
+# pytest
+.pytest_cache/
+
+# pre-commit
+.pre-commit-cache/
+
+# ruff
+.ruff_cache/
+
 # Docker volumes
 minio_data/
+
+# Local configuration overrides
+ca-bundle.pem

.pre-commit-config.yaml

@@ -0,0 +1,7 @@
+repos:
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.15.12
+    hooks:
+      - id: ruff
+        args: [--fix, --exit-non-zero-on-fix]
+      - id: ruff-format

.woodpecker/build.yaml

@@ -0,0 +1,9 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: docker-build
+    image: woodpeckerci/plugin-docker-buildx
+    settings:
+      repo: git.unkin.net/unkin/artifactapi
+      dry_run: true

.woodpecker/docker.yaml

@@ -0,0 +1,18 @@
+when:
+  - event: tag
+    ref: refs/tags/v*
+
+steps:
+  - name: docker
+    image: woodpeckerci/plugin-docker-buildx
+    settings:
+      registry: git.unkin.net
+      repo: git.unkin.net/unkin/artifactapi
+      username: droneci
+      password:
+        from_secret: DRONECI_PASSWORD
+      tags:
+        - ${CI_COMMIT_TAG}
+        - latest
+      build_args:
+        - VERSION=${CI_COMMIT_TAG##v}

.woodpecker/pre-commit.yaml

@@ -0,0 +1,9 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: pre-commit
+    image: git.unkin.net/unkin/almalinux9-base:20260308
+    commands:
+      - uvx pre-commit run --all-files
+

.woodpecker/test.yaml

@@ -0,0 +1,8 @@
+when:
+  - event: pull_request
+
+steps:
+  - name: test
+    image: git.unkin.net/unkin/almalinux9-base:20260308
+    commands:
+      - uvx --python 3.11 --with tox-uv tox

Dockerfile

@@ -1,49 +1,23 @@
-# Use Alpine Linux as base image
-FROM python:3.11-alpine
+FROM git.unkin.net/unkin/almalinux9-base:latest
 
-# Set working directory
-WORKDIR /app
+ARG VERSION=0.0.0.dev0
 
-# Install system dependencies
-RUN apk add --no-cache \
-    gcc \
-    musl-dev \
-    libffi-dev \
-    postgresql-dev \
-    curl \
-    wget \
-    tar
+COPY . /build
 
-# Install uv
-ARG PACKAGE_VERSION=0.9.21
-RUN wget -O /app/uv-x86_64-unknown-linux-musl.tar.gz https://github.com/astral-sh/uv/releases/download/${PACKAGE_VERSION}/uv-x86_64-unknown-linux-musl.tar.gz && \
-    tar xf /app/uv-x86_64-unknown-linux-musl.tar.gz -C /app && \
-    mv /app/uv-x86_64-unknown-linux-musl/uv /usr/local/bin/uv && \
-    rm -rf /app/uv-x86_64-unknown-linux-musl* && \
-    chmod +x /usr/local/bin/uv && \
-    uv --version
+RUN HATCH_VCS_PRETEND_VERSION=${VERSION} \
+    SETUPTOOLS_SCM_PRETEND_VERSION=${VERSION} \
+    uv build --wheel --directory /build && \
+    useradd -m -r -s /bin/sh appuser
 
-# Create non-root user first
-RUN adduser -D -s /bin/sh appuser && \
-    chown -R appuser:appuser /app
-
-# Copy dependency files and change ownership
-COPY --chown=appuser:appuser pyproject.toml uv.lock README.md ./
-
-# Switch to appuser and install Python dependencies
 USER appuser
-RUN uv sync --frozen
+RUN uv tool install --from /build/dist/*.whl artifactapi
 
-# Copy application source
-COPY --chown=appuser:appuser src/ ./src/
-COPY --chown=appuser:appuser remotes.yaml ./
+USER root
+RUN rm -rf /build
 
-# Expose port
 EXPOSE 8000
-
-# Health check
-HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
-    CMD curl -f http://localhost:8000/health || exit 1
-
-# Run the application
-CMD ["uv", "run", "python", "-m", "src.artifactapi.main"]
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 CMD curl -f http://localhost:8000/health || exit 1
+USER appuser
+ENV PATH="/home/appuser/.local/bin:$PATH"
+WORKDIR /app
+CMD ["artifactapi"]

Makefile

@@ -1,7 +1,7 @@
-.PHONY: build install dev clean test lint format docker-build docker-up docker-down docker-logs docker-rebuild docker-clean docker-restart
+.PHONY: build install dev clean test lint format pre-commit tox docker-build docker-up docker-down docker-logs docker-rebuild docker-clean docker-restart
 
 build:
-	docker build --no-cache -t artifactapi:latest .
+	docker build -t artifactapi:dev .
 
 install: build
 
@@ -17,7 +17,13 @@ clean:
 	rm -rf *.egg-info/
 
 test:
-	uv run pytest
+	uvx --python 3.11 --with tox-uv tox
+
+tox:
+	uvx --python 3.11 --with tox-uv tox
+
+pre-commit:
+	uvx --python 3.11 pre-commit run --all-files
 
 lint:
 	uv run ruff check --fix .
@@ -45,3 +51,26 @@ docker-clean:
 	docker system prune -f
 
 docker-restart: docker-down docker-up
+
+# Bump helpers — reads the latest semver tag and creates the next one.
+# If no tag exists yet, starts from v0.0.0.
+_LATEST := $(shell git tag --sort=-v:refname | grep -E '^v[0-9]+\.[0-9]+\.[0-9]+$$' | head -1)
+_BASE   := $(if $(_LATEST),$(_LATEST),v0.0.0)
+_MAJ    := $(shell echo $(_BASE) | sed 's/^v//' | cut -d. -f1)
+_MIN    := $(shell echo $(_BASE) | sed 's/^v//' | cut -d. -f2)
+_PAT    := $(shell echo $(_BASE) | sed 's/^v//' | cut -d. -f3)
+
+patch:
+	@NEW=v$(_MAJ).$(_MIN).$(shell expr $(_PAT) + 1); \
+	git tag $$NEW && echo "Tagged $$NEW" && $(MAKE) _tag TAG=$$NEW
+
+minor:
+	@NEW=v$(_MAJ).$(shell expr $(_MIN) + 1).0; \
+	git tag $$NEW && echo "Tagged $$NEW" && $(MAKE) _tag TAG=$$NEW
+
+major:
+	@NEW=v$(shell expr $(_MAJ) + 1).0.0; \
+	git tag $$NEW && echo "Tagged $$NEW" && $(MAKE) _tag TAG=$$NEW
+
+_tag:
+	git push origin $(TAG)

SPEC.md

@@ -0,0 +1,137 @@
+# ArtifactAPI Specification
+
+## Repository model
+
+Every repository entry in `remotes.yaml` has two orthogonal fields:
+
+| field | values | meaning |
+|---|---|---|
+| `type` | `local`, `remote`, `virtual` | repository kind — how the repo is served |
+| `package` | `docker`, `rpm`, `alpine`, `generic` | package format — what protocol and caching rules to apply |
+
+**type**
+
+- `local` — files are uploaded directly to the API and stored in S3; no upstream.
+- `remote` — proxies and caches content from an upstream URL (`base_url`).
+- `virtual` — aggregates multiple repositories (not yet implemented).
+
+**package**
+
+- `docker` — upstream speaks the OCI Distribution API (Bearer auth, manifest/blob paths).
+- `rpm` — upstream is an RPM repository; repodata files are index files.
+- `alpine` — upstream is an Alpine APK repository; `APKINDEX.tar.gz` is an index file.
+- `generic` — plain HTTP file download; no format-specific logic.
+
+---
+
+## Caching
+
+Two cache classes determine retention:
+
+| class | stored | TTL |
+|---|---|---|
+| **file** | S3 object, no Redis entry | `file_ttl` — `0` means indefinite |
+| **index** | S3 object + Redis TTL key | `index_ttl` — when the Redis key expires the S3 object is deleted and re-fetched |
+
+Index files are mutable metadata that must expire. File-class objects are treated as immutable and cached indefinitely (unless `file_ttl` is set).
+
+---
+
+## Docker package rules
+
+### URL construction
+
+Remote URLs are prefixed with `/v2/` for `package: docker` remotes:
+
+```
+{base_url}/v2/{path}
+```
+
+e.g. `library/nginx/manifests/latest` → `https://registry-1.docker.io/v2/library/nginx/manifests/latest`
+
+### Authentication
+
+Docker registries use Bearer token challenges. On a `401 Unauthorized` response, the API:
+
+1. Parses the `WWW-Authenticate: Bearer` header for `realm`, `service`, and `scope`.
+2. Fetches a token from the auth realm, supplying `username`/`password` from the remote config if present.
+3. Retries the request with `Authorization: Bearer <token>`.
+
+Tokens are cached in-memory keyed by `(realm, service, scope, username)` and expire 30 seconds before their stated `expires_in`.
+
+### Cache classification
+
+| path pattern | mutable | class | TTL source |
+|---|---|---|---|
+| `/manifests/<tag>` | yes | index | `index_ttl` |
+| `/tags/list` | yes | index | `index_ttl` |
+| `/manifests/sha256:<digest>` | no | file | `file_ttl` |
+| `/blobs/sha256:<digest>` | no | file | `file_ttl` |
+
+Tag-based manifests and tag lists are mutable and cached as index. Digest-pinned manifests and blobs are content-addressed and cached indefinitely as files.
+
+### Blob deduplication
+
+Blobs are stored under a digest-keyed path shared across all images on the same remote:
+
+```
+{remote_name}/blobs/sha256/{digest}
+```
+
+The same layer pulled by different images is stored once.
+
+### Accept headers
+
+| path | `Accept` header sent upstream |
+|---|---|
+| `/manifests/…` | `application/vnd.docker.distribution.manifest.v2+json`, `application/vnd.oci.image.manifest.v1+json`, `application/vnd.oci.image.index.v1+json`, `application/vnd.docker.distribution.manifest.list.v2+json` |
+| `/blobs/…` | `application/octet-stream` |
+
+---
+
+## OCI Distribution API endpoint
+
+The API exposes a native Docker registry interface so clients can use `docker pull` directly:
+
+```
+GET  /v2/                                   — version ping
+GET  /v2/{remote}/{image}/manifests/{ref}   — fetch manifest
+HEAD /v2/{remote}/{image}/manifests/{ref}   — manifest metadata
+GET  /v2/{remote}/{image}/blobs/{digest}    — fetch blob
+HEAD /v2/{remote}/{image}/blobs/{digest}    — blob metadata
+```
+
+Responses include `Docker-Distribution-Api-Version`, `Docker-Content-Digest`, and the correct OCI `Content-Type` (detected from the manifest `mediaType` field).
+
+Only remotes with `package: docker` are accessible via this endpoint. All other remotes return `400`.
+
+---
+
+## include_patterns
+
+`include_patterns` is a list of Python regexes applied to every request before any upstream fetch or cache lookup.
+
+**Generic remotes (`/api/v1/remote/…`):**
+- Patterns match against the file path and the full path.
+- Index files (mutable metadata) bypass pattern checks and are always allowed.
+
+**Docker remotes (`/v2/…`):**
+- Patterns match against the image name (first two path segments, e.g. `library/nginx`) and the full path.
+- The index-file exemption does **not** apply — patterns restrict whole images, including their manifests and tag lists.
+- No patterns configured → all images allowed.
+
+Returns `403` when a request is blocked.
+
+---
+
+## Versioning
+
+The package version is derived from git tags via `hatch-vcs`. Tags follow the format `v{MAJOR}.{MINOR}.{PATCH}`.
+
+Docker images are built with the version injected at build time:
+
+```
+SETUPTOOLS_SCM_PRETEND_VERSION=<version> uv sync --frozen
+```
+
+The `Makefile` provides `patch`, `minor`, and `major` targets that tag the current commit and rebuild the container image.

conf.d/alpine.yaml

@@ -0,0 +1,11 @@
+remotes:
+  alpine:
+    base_url: "https://dl-cdn.alpinelinux.org"
+    type: "remote"
+    package: "alpine"
+    description: "Alpine Linux APK package repository"
+    immutable_patterns:
+      - ".*/x86_64/.*\\.apk$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 7200

conf.d/github.yaml

@@ -0,0 +1,12 @@
+remotes:
+  github:
+    base_url: "https://github.com"
+    type: "remote"
+    package: "generic"
+    description: "GitHub releases and files"
+    immutable_patterns:
+      - "gruntwork-io/terragrunt/.*terragrunt_linux_amd64.*"
+      - "prometheus/node_exporter/.*/node_exporter-.*\\.linux-amd64\\.tar\\.gz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 0

conf.d/pypi.yaml

@@ -0,0 +1,17 @@
+remotes:
+  pypi:
+    base_url: "https://files.pythonhosted.org"
+    type: "remote"
+    package: "pypi"
+    description: "Python Package Index"
+    check_mutable_updates: true
+    quarantine_new: true
+    quarantine_days: 3
+    immutable_patterns:
+      - "packages/.*\\.whl$"
+      - "packages/.*\\.whl\\.metadata$"
+      - "packages/.*\\.tar\\.gz$"
+      - "packages/.*\\.zip$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 600

docker-compose.yml

@@ -5,9 +5,13 @@ services:
     build:
       context: .
       dockerfile: Dockerfile
-      no_cache: true
+      args:
+        - VERSION=2.2.2.dev0
     ports:
       - "8000:8000"
+    volumes:
+      - ./examples/single-file/remotes.yaml:/app/remotes.yaml:ro,z
+      - ./ca-bundle.pem:/app/ca-bundle.pem:ro,z
     environment:
       - CONFIG_PATH=/app/remotes.yaml
       - DBHOST=postgres
@@ -21,6 +25,7 @@ services:
       - MINIO_SECRET_KEY=minioadmin
       - MINIO_BUCKET=artifacts
       - MINIO_SECURE=false
+      - REQUESTS_CA_BUNDLE=/app/ca-bundle.pem
     depends_on:
       postgres:
         condition: service_healthy

examples/conf.d-method/alpine.yaml

@@ -0,0 +1,10 @@
+remote:
+  alpine:
+    base_url: "https://dl-cdn.alpinelinux.org"
+    package: "alpine"
+    description: "Alpine Linux APK package repository"
+    immutable_patterns:
+      - ".*/x86_64/.*\\.apk$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 7200

examples/conf.d-method/github.yaml

@@ -0,0 +1,11 @@
+remote:
+  github:
+    base_url: "https://github.com"
+    package: "generic"
+    description: "GitHub releases and files"
+    immutable_patterns:
+      - "gruntwork-io/terragrunt/.*terragrunt_linux_amd64.*"
+      - "prometheus/node_exporter/.*/node_exporter-.*\\.linux-amd64\\.tar\\.gz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 0

examples/conf.d-method/pypi.yaml

@@ -0,0 +1,16 @@
+remote:
+  pypi:
+    base_url: "https://files.pythonhosted.org"
+    package: "pypi"
+    description: "Python Package Index"
+    check_mutable_updates: true
+    quarantine_new: true
+    quarantine_days: 3
+    immutable_patterns:
+      - "packages/.*\\.whl$"
+      - "packages/.*\\.whl\\.metadata$"
+      - "packages/.*\\.tar\\.gz$"
+      - "packages/.*\\.zip$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 600

examples/single-file/remotes.yaml

@@ -0,0 +1,481 @@
+# Example remotes configuration — copy and adapt for your environment.
+#
+# Top-level keys determine the repository type:
+#   remote:  — proxy to an upstream URL, cache responses in S3
+#   virtual: — merge indexes from multiple member remotes into one
+#   local:   — store files uploaded via PUT, serve via GET
+#
+# immutable_patterns: artifacts cached forever (e.g. release binaries, versioned tags).
+# mutable_patterns:   artifacts that expire after cache.mutable_ttl seconds and are
+#                     re-fetched from upstream on next request (e.g. index files,
+#                     branch archives). Defaults to the package-type built-ins when
+#                     not set (APKINDEX, repomd.xml, Docker manifests, etc.).
+# cache:
+#   immutable_ttl: TTL for immutable files (0 = forever, rarely needed to change).
+#   mutable_ttl:   TTL in seconds for mutable files. Omit to use the default (3600).
+#
+# quarantine_new:  Set to true to block immutable artifacts published within the last
+#                  quarantine_days days. Requests return 404 until the quarantine period
+#                  expires. Fails open when the publish date cannot be determined.
+# quarantine_days: Number of days to quarantine newly published artifacts (requires
+#                  quarantine_new: true). The upstream Last-Modified header is used as
+#                  the publish date.
+#
+# WARNING: this file may contain credentials — do not commit real values.
+#
+# Global configuration
+#s3:
+#  endpoint: "localhost:9000"
+#  access_key: "minioadmin"
+#  secret_key: "minioadmin"
+#  bucket: "artifacts"
+#  secure: false
+#
+#redis:
+#  url: "redis://localhost:6379/0"
+#
+#database:
+#  url: "postgresql://artifacts:artifacts123@localhost:5432/artifacts"
+#
+
+remote:
+  github:
+    base_url: "https://github.com"
+    package: "generic"
+    description: "GitHub releases and files"
+    immutable_patterns:
+      - "gruntwork-io/terragrunt/.*terragrunt_linux_amd64.*"
+      - "lxc/incus/.*\\.tar\\.gz$"
+      - "prometheus/node_exporter/.*/node_exporter-.*\\.linux-amd64\\.tar\\.gz$"
+      - "VictoriaMetrics/VictoriaMetrics/.*/vmutils-linux-amd64-.*\\.tar\\.gz$"
+      - "VictoriaMetrics/VictoriaMetrics/.*/victoria-metrics-linux-amd64-.*-cluster\\.tar\\.gz$"
+      - "VictoriaMetrics/VictoriaMetrics/.*/victoria-logs-linux-amd64-.*\\.tar\\.gz$"
+      - "VictoriaMetrics/VictoriaMetrics/.*/vlutils-linux-amd64-.*\\.tar\\.gz$"
+      - "prometheus-community/bind_exporter/.*/bind_exporter-.*\\.linux-amd64\\.tar\\.gz$"
+      - "prometheus-community/pgbouncer_exporter/.*/pgbouncer_exporter-.*\\.linux-amd64\\.tar\\.gz$"
+      - "prometheus-community/postgres_exporter/.*/postgres_exporter-.*\\.linux-amd64\\.tar\\.gz$"
+      - "onedr0p/exportarr/.*/exportarr_.*_linux_amd64\\.tar\\.gz$"
+      - "tynany/frr_exporter/.*/frr_exporter-.*\\.linux-amd64\\.tar\\.gz$"
+      - "camptocamp/prometheus-puppetdb-exporter/.*/prometheus-puppetdb-exporter-.*\\.linux-amd64\\.tar\\.gz$"
+      - "grafana/jsonnet-language-server/.*/jsonnet-language-server_.*_linux_amd64$"
+      - "helmfile/helmfile/.*/helmfile_.*_linux_amd64\\.tar\\.gz$"
+      - "helmfile/vals/.*/vals_.*_linux_amd64\\.tar\\.gz$"
+      - "openbao/openbao-plugins/.*/openbao-plugin-secrets-consul_linux_amd64_.*\\.tar\\.gz$"
+      - "openbao/openbao-plugins/.*/openbao-plugin-secrets-nomad_linux_amd64_.*\\.tar\\.gz$"
+      - "apple/foundationdb/.*/libfdb_c\\.x86_64\\.so$"
+      - "stalwartlabs/stalwart/.*/stalwart-cli-x86_64-unknown-linux-gnu\\.tar\\.gz$"
+      - "stalwartlabs/stalwart/.*/stalwart-foundationdb-x86_64-unknown-linux-gnu\\.tar\\.gz$"
+      - "stalwartlabs/stalwart/.*/stalwart-x86_64-unknown-linux-gnu\\.tar\\.gz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 0
+
+  github-archive:
+    base_url: "https://github.com"
+    package: "generic"
+    description: "GitHub repository archive tarballs"
+    immutable_patterns:
+      # Tag archives are immutable — a tag never changes
+      - ".*/archive/refs/tags/.*\\.tar\\.gz$"
+    mutable_patterns:
+      # Branch archives can change on every push
+      - ".*/archive/refs/heads/main\\.tar\\.gz$"
+      - ".*/archive/refs/heads/master\\.tar\\.gz$"
+    # Before re-downloading an expired branch archive, check whether it has
+    # actually changed (304 Not Modified → just refresh the TTL, no transfer).
+    # Only applies to user-defined mutable_patterns, not package-type defaults.
+    check_mutable_updates: true
+    cache:
+      immutable_ttl: 0      # Tag archives cached indefinitely
+      mutable_ttl: 86400    # Branch archives refreshed after 1 day
+
+  gitea-dl:
+    base_url: "https://dl.gitea.com"
+    package: "generic"
+    description: "Gitea download site"
+    immutable_patterns:
+      - "act_runner/.*/act_runner-.*-linux-amd64$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 0
+
+  hashicorp-releases:
+    base_url: "https://releases.hashicorp.com"
+    package: "generic"
+    description: "HashiCorp product releases"
+    immutable_patterns:
+      - "terraform/.*terraform_.*_linux_amd64\\.zip$"
+      - "terraform/.*terraform_.*_windows_amd64\\.zip$"
+      - "terraform/.*terraform_.*_darwin_amd64\\.zip$"
+      - "vault/.*vault_.*_linux_amd64\\.zip$"
+      - "vault/.*vault_.*_windows_amd64\\.zip$"
+      - "vault/.*vault_.*_darwin_amd64\\.zip$"
+      - "consul-cni/.*/consul-cni_.*_linux_amd64\\.zip$"
+      - "consul/.*/consul_.*_linux_amd64\\.zip$"
+      - "nomad-autoscaler/.*/nomad-autoscaler_.*_linux_amd64\\.zip$"
+      - "nomad/.*/nomad_.*_linux_amd64\\.zip$"
+      - "packer/.*/packer_.*_linux_amd64\\.zip$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 0
+
+  alpine:
+    base_url: "https://dl-cdn.alpinelinux.org"
+    package: "alpine"
+    description: "Alpine Linux APK package repository"
+    immutable_patterns:
+      - ".*/x86_64/.*\\.apk$"
+    # check_mutable_updates not set: APKINDEX.tar.gz is a package-type default
+    # and is always re-fetched on expiry — conditional checks are skipped for
+    # built-in mutable patterns regardless of this flag.
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 7200  # Index files (APKINDEX.tar.gz) cached for 2 hours
+
+  almalinux:
+    base_url: "https://gsl-syd.mm.fcix.net/almalinux"
+    package: "rpm"
+    description: "AlmaLinux RPM package repository"
+    immutable_patterns:
+      - ".*/x86_64/.*\\.rpm$"
+      - ".*/noarch/.*\\.rpm$"
+      - ".*/repodata/.*$"
+      - ".*\\.rpm$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 7200  # Metadata files cached for 2 hours
+
+  epel:
+    base_url: "http://mirror.aarnet.edu.au/pub/epel"
+    package: "rpm"
+    description: "EPEL (Extra Packages for Enterprise Linux)"
+    immutable_patterns:
+      - "8/Everything/x86_64/.*\\.rpm$"
+      - "9/Everything/x86_64/.*\\.rpm$"
+      - "10/Everything/x86_64/.*\\.rpm$"
+      - ".*/noarch/.*\\.rpm$"
+      - ".*/repodata/.*$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 7200
+
+  fedora:
+    base_url: "https://gsl-syd.mm.fcix.net/fedora/linux"
+    package: "rpm"
+    description: "Fedora Linux RPM package repository"
+    immutable_patterns:
+      - "releases/.*/Everything/x86_64/.*\\.rpm$"
+      - "updates/.*/Everything/x86_64/.*\\.rpm$"
+      - "development/.*/Everything/x86_64/.*\\.rpm$"
+      - ".*/noarch/.*\\.rpm$"
+      - "updates/.*/Everything/x86_64/repodata/.*$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 300  # Metadata files cached for 5 minutes
+
+  ghcr:
+    base_url: "https://ghcr.io"
+    package: "docker"
+    description: "GitHub Container Registry"
+    # username: "your-github-username"
+    # password: "your-github-pat"  # needs read:packages scope
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 300
+
+  dockerhub:
+    base_url: "https://registry-1.docker.io"
+    package: "docker"
+    description: "Docker Hub registry"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 300
+
+  pypi:
+    base_url: "https://files.pythonhosted.org"
+    package: "pypi"
+    description: "Python Package Index — simple index and package files via a single remote"
+    check_mutable_updates: true
+    quarantine_new: true
+    quarantine_days: 3
+    immutable_patterns:
+      - "packages/.*\\.whl$"
+      - "packages/.*\\.whl\\.metadata$"
+      - "packages/.*\\.tar\\.gz$"
+      - "packages/.*\\.zip$"
+      - "packages/.*\\.egg$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 600  # Simple index pages refreshed after 10 minutes
+
+  pypi-gitea:
+    base_url: "https://gitea.example.com/api/packages/myorg/pypi"
+    package: "pypi"
+    description: "Private Gitea PyPI registry — simple index and files at the same host"
+    # username: "your-gitea-username"
+    # password: "your-personal-access-token"  # needs package:read scope
+    check_mutable_updates: true
+    immutable_patterns:
+      - "files/.*\\.whl$"
+      - "files/.*\\.whl\\.metadata$"
+      - "files/.*\\.tar\\.gz$"
+      - "files/.*\\.zip$"
+      - "files/.*\\.egg$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 600
+
+  npm:
+    base_url: "https://registry.npmjs.org"
+    package: "npm"
+    description: "npm registry — package metadata with tarball URL rewriting"
+    check_mutable_updates: true
+    immutable_patterns:
+      - \.tgz$
+    mutable_patterns:
+      - ^(?!.*\.tgz$).*
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 600  # Package metadata refreshed after 10 minutes
+
+  hashicorp-helm:
+    base_url: "https://helm.releases.hashicorp.com"
+    package: "helm"
+    description: "HashiCorp Helm chart repository (Vault, Consul, Nomad, etc.)"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  metallb:
+    base_url: "https://metallb.github.io/metallb"
+    package: "helm"
+    description: "MetalLB load balancer Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  jetstack:
+    base_url: "https://charts.jetstack.io"
+    package: "helm"
+    description: "Jetstack Helm charts (cert-manager)"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  rancher-stable:
+    base_url: "https://releases.rancher.com/server-charts/stable"
+    package: "helm"
+    description: "Rancher stable Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  purelb:
+    base_url: "https://gitlab.com/api/v4/projects/20400619/packages/helm/stable"
+    package: "helm"
+    description: "PureLB load balancer Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  istio:
+    base_url: "https://istio-release.storage.googleapis.com/charts"
+    package: "helm"
+    description: "Istio service mesh Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  cnpg:
+    base_url: "https://cloudnative-pg.github.io/charts"
+    package: "helm"
+    description: "CloudNativePG operator Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  ceph-csi:
+    base_url: "https://ceph.github.io/csi-charts"
+    package: "helm"
+    description: "Ceph CSI driver Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  external-dns:
+    base_url: "https://kubernetes-sigs.github.io/external-dns/"
+    package: "helm"
+    description: "ExternalDNS Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  intel-helm:
+    base_url: "https://intel.github.io/helm-charts/"
+    package: "helm"
+    description: "Intel Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  elastic:
+    base_url: "https://helm.elastic.co"
+    package: "helm"
+    description: "Elastic stack Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  k8up-io:
+    base_url: "https://k8up-io.github.io/k8up"
+    package: "helm"
+    description: "K8up backup operator Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  victoriametrics:
+    base_url: "https://victoriametrics.github.io/helm-charts/"
+    package: "helm"
+    description: "VictoriaMetrics observability Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  grafana:
+    base_url: "https://grafana.github.io/helm-charts"
+    package: "helm"
+    description: "Grafana observability Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  helm-openldap:
+    base_url: "https://jp-gouin.github.io/helm-openldap/"
+    package: "helm"
+    description: "OpenLDAP Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  woodpecker:
+    base_url: "https://woodpecker-ci.org/"
+    package: "helm"
+    description: "Woodpecker CI Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  stakater:
+    base_url: "https://stakater.github.io/stakater-charts"
+    package: "helm"
+    description: "Stakater Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  jfrog:
+    base_url: "https://charts.jfrog.io/"
+    package: "helm"
+    description: "JFrog Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+  openvox:
+    base_url: "https://openvoxproject.github.io/openvox-helm-chart"
+    package: "helm"
+    description: "OpenVox Helm charts"
+    check_mutable_updates: true
+    immutable_patterns:
+      - "\\.tgz$"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 3600
+
+virtual:
+  helm-all:
+    package: "helm"
+    description: "Virtual repository merging all helm remotes — member order is priority order for duplicate chart+version"
+    members:
+      - hashicorp-helm
+      - metallb
+      - jetstack
+      - rancher-stable
+      - purelb
+      - istio
+      - cnpg
+      - ceph-csi
+      - external-dns
+      - intel-helm
+      - elastic
+      - k8up-io
+      - victoriametrics
+      - grafana
+      - helm-openldap
+      - woodpecker
+      - stakater
+      - jfrog
+      - openvox
+
+local:
+  local-generic:
+    package: "generic"
+    description: "Local generic file repository"
+    cache:
+      immutable_ttl: 0
+      mutable_ttl: 0

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "artifactapi"
-version = "2.0.0"
+dynamic = ["version"]
 description = "Generic artifact caching system with support for various package managers"
 
 dependencies = [
@@ -23,9 +23,12 @@ license = {text = "MIT"}
 artifactapi = "artifactapi.main:main"
 
 [build-system]
-requires = ["hatchling"]
+requires = ["hatchling", "hatch-vcs"]
 build-backend = "hatchling.build"
 
+[tool.hatch.version]
+source = "vcs"
+
 [tool.hatch.metadata]
 allow-direct-references = true
 
@@ -39,5 +42,18 @@ dev = [
     "black>=23.9.0",
     "isort>=5.12.0",
     "mypy>=1.6.0",
-    "ruff>=0.1.0",
+    "ruff>=0.4.0",
+    "tox>=4.0.0",
+    "pre-commit>=3.0.0",
 ]
+
+[tool.pytest.ini_options]
+asyncio_mode = "auto"
+testpaths = ["tests"]
+
+[tool.ruff]
+line-length = 140
+
+[tool.ruff.lint]
+select = ["E", "F", "I", "UP"]
+ignore = ["E501"]

src/artifactapi/artifact/__init__.py

@@ -0,0 +1,3 @@
+from . import discovery, docker, flush, local, proxy
+
+__all__ = ["discovery", "docker", "flush", "local", "proxy"]

src/artifactapi/artifact/discovery.py

@@ -0,0 +1,82 @@
+import logging
+import re
+from typing import Any
+from urllib.parse import urlparse
+
+import httpx
+from fastapi import HTTPException
+
+from .proxy import cache_single_artifact
+
+logger = logging.getLogger(__name__)
+
+
+async def _discover_github_releases(remote: str, include_pattern: str) -> list[str]:
+    match = re.match(r"github\.com/([^/]+)/([^/]+)", remote)
+    if not match:
+        raise HTTPException(status_code=400, detail="Invalid GitHub remote format")
+
+    owner, repo = match.groups()
+
+    async with httpx.AsyncClient(follow_redirects=True) as client:
+        response = await client.get(f"https://api.github.com/repos/{owner}/{repo}/releases")
+        if response.status_code != 200:
+            raise HTTPException(status_code=response.status_code, detail=f"Failed to fetch releases: {response.text}")
+
+        releases = response.json()
+        regex = re.compile(include_pattern.replace("*", ".*"))
+        return [
+            asset["browser_download_url"]
+            for release in releases
+            for asset in release.get("assets", [])
+            if regex.search(asset["browser_download_url"])
+        ]
+
+
+async def _discover(remote: str, include_pattern: str) -> list[str]:
+    if "github.com" in remote:
+        return await _discover_github_releases(remote, include_pattern)
+    raise HTTPException(status_code=400, detail=f"Unsupported remote: {remote}")
+
+
+async def cache_artifacts(remote: str, include_pattern: str, storage) -> dict[str, Any]:
+    try:
+        matching_urls = await _discover(remote, include_pattern)
+
+        if not matching_urls:
+            return {"message": "No matching artifacts found", "cached_count": 0, "artifacts": []}
+
+        cached_artifacts = []
+        for url in matching_urls:
+            result = await cache_single_artifact(url, "", "", storage, {})
+            cached_artifacts.append(result)
+
+        cached_count = sum(1 for a in cached_artifacts if a["status"] in ["cached", "already_cached"])
+        return {
+            "message": f"Processed {len(matching_urls)} artifacts, {cached_count} successfully cached",
+            "cached_count": cached_count,
+            "artifacts": cached_artifacts,
+        }
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+async def list_artifacts(remote: str, include_pattern: str, storage) -> dict[str, Any]:
+    try:
+        matching_urls = await _discover(remote, include_pattern)
+        cached_artifacts = []
+        for url in matching_urls:
+            parsed = urlparse(url)
+            key = storage.get_object_key(remote, parsed.path)
+            if storage.exists(key):
+                cached_artifacts.append({"url": url, "cached_url": storage.get_url(key), "key": key})
+
+        return {
+            "remote": remote,
+            "pattern": include_pattern,
+            "total_found": len(matching_urls),
+            "cached_count": len(cached_artifacts),
+            "artifacts": cached_artifacts,
+        }
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))

src/artifactapi/artifact/docker.py

@@ -0,0 +1,103 @@
+import hashlib
+import json
+import logging
+import re
+
+from fastapi import HTTPException, Request, Response
+
+from . import proxy as _proxy
+
+logger = logging.getLogger(__name__)
+
+
+def ping() -> Response:
+    return Response(
+        content="{}",
+        media_type="application/json",
+        headers={"Docker-Distribution-Api-Version": "registry/2.0"},
+    )
+
+
+async def proxy(request: Request, remote_name: str, path: str, storage, cache, config, metrics) -> Response:
+    remote_config = config.get_remote_config(remote_name)
+    if not remote_config:
+        raise HTTPException(status_code=404, detail=f"Remote '{remote_name}' not configured")
+    if remote_config.get("package") != "docker":
+        raise HTTPException(status_code=400, detail=f"Remote '{remote_name}' is not a docker remote")
+
+    patterns = config.get_immutable_patterns(remote_name, "")
+    if patterns:
+        path_parts = path.split("/")
+        image_name = "/".join(path_parts[:2]) if len(path_parts) >= 2 else path
+        if not any(re.search(p, path) or re.search(p, image_name) for p in patterns):
+            logger.info(f"PATTERN BLOCKED: {remote_name}/{path}")
+            raise HTTPException(status_code=403, detail="Image not allowed by configuration patterns")
+
+    base_url = remote_config.get("base_url", "").rstrip("/")
+    remote_url = f"{base_url}/v2/{path}"
+
+    cached_key = storage.get_object_key(remote_name, path)
+    if not storage.exists(cached_key):
+        cached_key = None
+
+    is_mutable = cache.is_mutable_file(path, config.get_mutable_patterns(remote_name))
+
+    if cached_key and is_mutable:
+        if not cache.is_index_valid(remote_name, path):
+            if not await _proxy.handle_expired_mutable(remote_name, path, remote_url, config, cache, storage):
+                cached_key = None
+
+    if not cached_key:
+        logger.info(f"Cache MISS: {remote_name}/{path} - fetching from remote: {remote_url}")
+        result = await _proxy.cache_single_artifact(remote_url, remote_name, path, storage, remote_config)
+        if result["status"] == "error":
+            raise HTTPException(status_code=502, detail=f"Failed to fetch: {result['error']}")
+        if result["status"] == "cached" and is_mutable:
+            cache_config = config.get_cache_config(remote_name)
+            mutable_ttl = cache_config.get("mutable_ttl", 3600)
+            cache.mark_index_cached(remote_name, path, mutable_ttl)
+            logger.info(f"Mutable file cached with TTL: {remote_name}/{path} (ttl: {mutable_ttl}s)")
+            if result.get("etag") or result.get("last_modified"):
+                cache.store_mutable_meta(remote_name, path, result.get("etag"), result.get("last_modified"))
+        if not is_mutable:
+            published = result.get("last_modified")
+            if published:
+                cache.store_artifact_published(remote_name, path, published)
+            _proxy._check_quarantine(remote_name, published, config)
+    elif not is_mutable:
+        published = cache.get_artifact_published(remote_name, path)
+        if not published:
+            published = await _proxy._fetch_last_modified(remote_url, remote_config)
+            if published:
+                cache.store_artifact_published(remote_name, path, published)
+        _proxy._check_quarantine(remote_name, published, config)
+
+    artifact_data = storage.download_object(storage.get_object_key(remote_name, path))
+
+    is_blob = "/blobs/" in path
+    if is_blob:
+        content_type = "application/octet-stream"
+    else:
+        try:
+            manifest_json = json.loads(artifact_data)
+            content_type = manifest_json.get("mediaType")
+            if not content_type:
+                if "manifests" in manifest_json:
+                    content_type = "application/vnd.oci.image.index.v1+json"
+                else:
+                    content_type = "application/vnd.oci.image.manifest.v1+json"
+        except Exception:
+            content_type = "application/vnd.oci.image.manifest.v1+json"
+
+    digest = f"sha256:{hashlib.sha256(artifact_data).hexdigest()}"
+    headers = {
+        "Docker-Distribution-Api-Version": "registry/2.0",
+        "Docker-Content-Digest": digest,
+        "Content-Length": str(len(artifact_data)),
+    }
+
+    if request.method == "HEAD":
+        return Response(status_code=200, headers=headers, media_type=content_type)
+
+    metrics.record_cache_hit(remote_name, len(artifact_data))
+    return Response(content=artifact_data, media_type=content_type, headers=headers)

src/artifactapi/artifact/flush.py

@@ -0,0 +1,66 @@
+import logging
+
+from fastapi import HTTPException
+
+logger = logging.getLogger(__name__)
+
+
+def handle(remote: str | None, cache_type: str, cache, storage) -> dict:
+    try:
+        result = {"remote": remote, "cache_type": cache_type, "flushed": {"redis_keys": 0, "s3_objects": 0, "operations": []}}
+
+        if cache_type in ["all", "index", "metrics"] and cache.available and cache.client:
+            patterns = []
+
+            if cache_type in ["all", "index"]:
+                if remote:
+                    patterns += [f"index:{remote}:*", f"mutable:meta:{remote}:*"]
+                else:
+                    patterns += ["index:*", "mutable:meta:*"]
+
+            if cache_type in ["all", "metrics"]:
+                patterns.append(f"metrics:*:{remote}" if remote else "metrics:*")
+
+            for pattern in patterns:
+                keys = cache.client.keys(pattern)
+                if keys:
+                    cache.client.delete(*keys)
+                    result["flushed"]["redis_keys"] += len(keys)
+                    logger.info(f"Cache flush: deleted {len(keys)} Redis keys matching '{pattern}'")
+
+            if result["flushed"]["redis_keys"] > 0:
+                result["flushed"]["operations"].append(f"Deleted {result['flushed']['redis_keys']} Redis keys")
+
+        if cache_type in ["all", "files"]:
+            try:
+                list_params = {"Bucket": storage.bucket}
+                if remote:
+                    list_params["Prefix"] = f"{remote}/"
+
+                response = storage.client.list_objects_v2(**list_params)
+                if "Contents" in response:
+                    objects_to_delete = [obj["Key"] for obj in response["Contents"]]
+                    for key in objects_to_delete:
+                        try:
+                            storage.client.delete_object(Bucket=storage.bucket, Key=key)
+                            result["flushed"]["s3_objects"] += 1
+                        except Exception as e:
+                            logger.warning(f"Failed to delete S3 object {key}: {e}")
+
+                    if objects_to_delete:
+                        scope = f" for remote '{remote}'" if remote else ""
+                        result["flushed"]["operations"].append(f"Deleted {len(objects_to_delete)} S3 objects{scope}")
+                        logger.info(f"Cache flush: deleted {len(objects_to_delete)} S3 objects{scope}")
+
+            except Exception as e:
+                result["flushed"]["operations"].append(f"S3 flush failed: {str(e)}")
+                logger.error(f"Cache flush S3 error: {e}")
+
+        if not result["flushed"]["operations"]:
+            result["flushed"]["operations"].append("No cache entries found to flush")
+
+        return result
+
+    except Exception as e:
+        logger.error(f"Cache flush error: {e}")
+        raise HTTPException(status_code=500, detail=f"Cache flush failed: {str(e)}")

src/artifactapi/artifact/local.py

@@ -0,0 +1,108 @@
+import hashlib
+import logging
+
+from fastapi import HTTPException, Response, UploadFile
+from fastapi.responses import JSONResponse
+
+logger = logging.getLogger(__name__)
+
+
+async def upload(remote_name: str, path: str, file: UploadFile, storage, database, config) -> JSONResponse:
+    remote_config = config.get_remote_config(remote_name)
+    if not remote_config:
+        raise HTTPException(status_code=404, detail=f"Remote '{remote_name}' not configured")
+    if remote_config.get("type") != "local":
+        raise HTTPException(status_code=400, detail="Upload only supported for local repositories")
+
+    try:
+        content = await file.read()
+        sha256_sum = hashlib.sha256(content).hexdigest()
+
+        if database.file_exists(remote_name, path):
+            raise HTTPException(status_code=409, detail="File already exists")
+
+        s3_key = f"local/{remote_name}/{path}"
+        content_type = file.content_type or "application/octet-stream"
+
+        try:
+            storage.upload(s3_key, content)
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=f"Upload failed: {e}")
+
+        success = database.add_local_file(
+            repository_name=remote_name,
+            file_path=path,
+            s3_key=s3_key,
+            size_bytes=len(content),
+            sha256_sum=sha256_sum,
+            content_type=content_type,
+        )
+
+        if not success:
+            storage.delete_object(s3_key)
+            raise HTTPException(status_code=500, detail="Failed to save file metadata")
+
+        return JSONResponse(
+            {
+                "message": "File uploaded successfully",
+                "file_path": path,
+                "size_bytes": len(content),
+                "sha256_sum": sha256_sum,
+                "content_type": content_type,
+            }
+        )
+
+    except HTTPException:
+        raise
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Upload failed: {str(e)}")
+
+
+def check_exists(remote_name: str, path: str, database, config) -> Response:
+    remote_config = config.get_remote_config(remote_name)
+    if not remote_config:
+        raise HTTPException(status_code=404, detail=f"Remote '{remote_name}' not configured")
+
+    if remote_config.get("type") != "local":
+        raise HTTPException(status_code=405, detail="HEAD method only supported for local repositories")
+
+    try:
+        metadata = database.get_local_file_metadata(remote_name, path)
+        if not metadata:
+            raise HTTPException(status_code=404, detail="File not found")
+
+        return Response(
+            headers={
+                "Content-Length": str(metadata["size_bytes"]),
+                "Content-Type": metadata.get("content_type", "application/octet-stream"),
+                "X-SHA256": metadata["sha256_sum"],
+                "X-Created-At": metadata["created_at"].isoformat() if metadata["created_at"] else "",
+                "X-Uploaded-At": metadata["uploaded_at"].isoformat() if metadata["uploaded_at"] else "",
+            }
+        )
+    except HTTPException:
+        raise
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Check failed: {str(e)}")
+
+
+def delete(remote_name: str, path: str, storage, database, config) -> JSONResponse:
+    remote_config = config.get_remote_config(remote_name)
+    if not remote_config:
+        raise HTTPException(status_code=404, detail=f"Remote '{remote_name}' not configured")
+    if remote_config.get("type") != "local":
+        raise HTTPException(status_code=400, detail="Delete only supported for local repositories")
+
+    try:
+        s3_key = database.delete_local_file(remote_name, path)
+        if not s3_key:
+            raise HTTPException(status_code=404, detail="File not found")
+
+        if not storage.delete_object(s3_key):
+            logger.warning(f"Failed to delete S3 object {s3_key} after database removal")
+
+        return JSONResponse({"message": "File deleted successfully"})
+    except HTTPException:
+        raise
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Delete failed: {str(e)}")

src/artifactapi/artifact/proxy.py

@@ -0,0 +1,331 @@
+import base64
+import logging
+import os
+import re
+from datetime import UTC, datetime, timedelta
+from email.utils import parsedate_to_datetime
+
+import httpx
+from fastapi import HTTPException, Request, Response
+
+from ..auth import get_docker_token_for_response
+from ..remote import helm as _helm
+from ..remote import npm as _npm
+from ..remote import python as _pypi
+from ..remote.base import get_content_type
+
+logger = logging.getLogger(__name__)
+
+
+class UpstreamUnreachable(Exception):
+    """Raised when the upstream backend cannot be contacted (network or timeout error)."""
+
+
+def _check_quarantine(remote_name: str, last_modified_str: str | None, config) -> None:
+    """Raise HTTP 404 if the artifact is within the per-remote quarantine window.
+
+    Fails open (allows the request) when the publish date cannot be determined.
+    """
+    enabled, days = config.get_quarantine_config(remote_name)
+    if not enabled or not days:
+        return
+    if not last_modified_str:
+        return  # cannot determine age → allow
+    try:
+        publish_date = parsedate_to_datetime(last_modified_str)
+    except Exception:
+        return  # unparseable → allow
+    cutoff = datetime.now(UTC) - timedelta(days=days)
+    if publish_date > cutoff:
+        available_on = (publish_date + timedelta(days=days)).date()
+        raise HTTPException(
+            status_code=404,
+            detail=(
+                f"Package quarantined: published {publish_date.date()}, available after {available_on} ({days}-day new-release quarantine)"
+            ),
+        )
+
+
+async def _fetch_last_modified(remote_url: str, remote_cfg: dict) -> str | None:
+    """HEAD the upstream URL and return the Last-Modified header, or None on any failure."""
+    auth = _basic_auth_header(remote_cfg)
+    try:
+        async with httpx.AsyncClient(follow_redirects=True) as client:
+            response = await client.head(remote_url, headers=auth, timeout=10.0)
+        return response.headers.get("Last-Modified")
+    except Exception:
+        return None
+
+
+def _basic_auth_header(remote_cfg: dict) -> dict[str, str]:
+    username = remote_cfg.get("username")
+    password = remote_cfg.get("password")
+    if username and password:
+        token = base64.b64encode(f"{username}:{password}".encode()).decode()
+        return {"Authorization": f"Basic {token}"}
+    return {}
+
+
+def _resolve_content(
+    data: bytes,
+    path: str,
+    filename: str,
+    remote_config: dict,
+    request: Request,
+    remote_name: str = "",
+) -> tuple[bytes, str]:
+    package = remote_config.get("package")
+    proxy_base = str(request.base_url).rstrip("/")
+    base_url = remote_config.get("base_url", "").rstrip("/")
+
+    if package == "pypi":
+        return _pypi.resolve_content(data, path, filename, remote_config.get("immutable_patterns", []), base_url, proxy_base, remote_name)
+    if package == "npm":
+        return _npm.resolve_content(data, path, filename, remote_config.get("immutable_patterns", []), base_url, proxy_base, remote_name)
+    if package == "helm":
+        return _helm.resolve_content(data, path, filename, base_url, proxy_base, remote_name)
+    return data, get_content_type(filename)
+
+
+def construct_url(remote_config: dict, path: str) -> str:
+    base_url = remote_config.get("base_url", "").rstrip("/")
+    if remote_config.get("package") == "docker":
+        return f"{base_url}/v2/{path}"
+    if remote_config.get("package") == "pypi":
+        return _pypi.construct_url(base_url, path)
+    return f"{base_url}/{path}"
+
+
+async def cache_single_artifact(url: str, remote_name: str, path: str, storage, remote_config: dict) -> dict:
+    key = storage.get_object_key(remote_name, path)
+
+    if storage.exists(key):
+        logger.info(f"Cache ALREADY EXISTS: {url} (key: {key})")
+        return {"url": url, "cached_url": storage.get_url(key), "status": "already_cached"}
+
+    try:
+        is_docker = remote_config.get("package") == "docker" or "/v2/" in url
+        headers = {}
+        username = remote_config.get("username")
+        password = remote_config.get("password")
+
+        if is_docker:
+            if "/manifests/" in url:
+                headers["Accept"] = (
+                    "application/vnd.docker.distribution.manifest.v2+json,"
+                    "application/vnd.oci.image.manifest.v1+json,"
+                    "application/vnd.oci.image.index.v1+json,"
+                    "application/vnd.docker.distribution.manifest.list.v2+json"
+                )
+            elif "/blobs/" in url:
+                headers["Accept"] = "application/octet-stream"
+        elif username and password:
+            headers["Authorization"] = "Basic " + base64.b64encode(f"{username}:{password}".encode()).decode()
+
+        async with httpx.AsyncClient(follow_redirects=True) as client:
+            response = await client.get(url, headers=headers)
+
+            if response.status_code == 401 and is_docker:
+                www_auth = response.headers.get("WWW-Authenticate", "")
+                token = await get_docker_token_for_response(www_auth, username, password)
+                if token:
+                    headers["Authorization"] = f"Bearer {token}"
+                    response = await client.get(url, headers=headers)
+
+            response.raise_for_status()
+            storage.upload(key, response.content)
+            logger.info(f"Cache ADD SUCCESS: {url} (size: {len(response.content)} bytes, key: {key})")
+
+            return {
+                "url": url,
+                "cached_url": storage.get_url(key),
+                "storage_path": f"s3://{storage.bucket}/{key}",
+                "size": len(response.content),
+                "status": "cached",
+                "etag": response.headers.get("ETag"),
+                "last_modified": response.headers.get("Last-Modified"),
+            }
+
+    except Exception as e:
+        return {"url": url, "status": "error", "error": str(e)}
+
+
+async def _upstream_reachable(url: str, auth_headers: dict | None = None) -> bool:
+    try:
+        async with httpx.AsyncClient(follow_redirects=True) as client:
+            await client.head(url, headers=auth_headers or {}, timeout=10.0)
+        return True
+    except (httpx.NetworkError, httpx.TimeoutException):
+        return False
+    except Exception:
+        return True
+
+
+async def check_upstream_changed(remote_url: str, remote_name: str, path: str, cache, auth_headers: dict | None = None) -> bool:
+    meta = cache.get_mutable_meta(remote_name, path)
+    if not meta:
+        return True
+
+    headers = dict(auth_headers or {})
+    if meta.get("etag"):
+        headers["If-None-Match"] = meta["etag"]
+    if meta.get("last_modified"):
+        headers["If-Modified-Since"] = meta["last_modified"]
+    if not (meta.get("etag") or meta.get("last_modified")):
+        return True
+
+    try:
+        async with httpx.AsyncClient(follow_redirects=True) as client:
+            response = await client.head(remote_url, headers=headers)
+        return response.status_code != 304
+    except (httpx.NetworkError, httpx.TimeoutException) as exc:
+        raise UpstreamUnreachable(str(exc)) from exc
+
+
+async def handle_expired_mutable(remote_name: str, path: str, remote_url: str, config, cache, storage) -> bool:
+    """Handle an expired mutable file. Returns True if the cached copy is still valid."""
+    mutable_ttl = config.get_cache_config(remote_name).get("mutable_ttl", 3600)
+    remote_cfg = config.get_remote_config(remote_name) or {}
+    auth = _basic_auth_header(remote_cfg)
+    check_updates = remote_cfg.get("check_mutable_updates", False)
+    user_mutable = check_updates and cache.is_mutable_file(path, config.get_user_mutable_patterns(remote_name))
+
+    if user_mutable:
+        try:
+            changed = await check_upstream_changed(remote_url, remote_name, path, cache, auth)
+        except UpstreamUnreachable:
+            cache.mark_index_cached(remote_name, path, mutable_ttl)
+            logger.warning(f"Mutable STALE (backend unreachable): {remote_name}/{path} - TTL extended ({mutable_ttl}s)")
+            return True
+        if not changed:
+            cache.mark_index_cached(remote_name, path, mutable_ttl)
+            logger.info(f"Mutable file UNCHANGED: {remote_name}/{path} - TTL refreshed ({mutable_ttl}s)")
+            return True
+        logger.info(f"Mutable file CHANGED: {remote_name}/{path} - re-downloading")
+    else:
+        if not await _upstream_reachable(remote_url, auth):
+            cache.mark_index_cached(remote_name, path, mutable_ttl)
+            logger.warning(f"Mutable STALE (backend unreachable): {remote_name}/{path} - TTL extended ({mutable_ttl}s)")
+            return True
+        logger.info(f"Mutable file EXPIRED: {remote_name}/{path} - removing from cache")
+
+    cache.cleanup_expired_index(storage, remote_name, path)
+    return False
+
+
+async def handle(request: Request, remote_name: str, path: str, storage, cache, config, database, metrics) -> Response:
+    remote_config = config.get_remote_config(remote_name)
+    if not remote_config:
+        raise HTTPException(status_code=404, detail=f"Remote '{remote_name}' not configured")
+
+    if remote_config.get("type") == "local":
+        metadata = database.get_local_file_metadata(remote_name, path)
+        if not metadata:
+            raise HTTPException(status_code=404, detail="File not found")
+        content = storage.download_object(metadata["s3_key"])
+        if content is None:
+            raise HTTPException(status_code=500, detail="File not accessible")
+        return Response(
+            content=content,
+            media_type=metadata.get("content_type", "application/octet-stream"),
+            headers={"Content-Disposition": f"attachment; filename={os.path.basename(path)}"},
+        )
+
+    path_parts = path.split("/")
+    if len(path_parts) >= 2:
+        repo_path = f"{path_parts[0]}/{path_parts[1]}"
+        file_path = "/".join(path_parts[2:])
+    else:
+        repo_path = path
+        file_path = path
+
+    mutable_patterns = config.get_mutable_patterns(remote_name)
+    if not cache.is_mutable_file(file_path, mutable_patterns) and not cache.is_mutable_file(path, mutable_patterns):
+        patterns = config.get_immutable_patterns(remote_name, repo_path)
+        if patterns and not any(re.search(p, file_path) or re.search(p, path) for p in patterns):
+            logger.info(f"PATTERN BLOCKED: {remote_name}/{path} - not matching include patterns")
+            raise HTTPException(status_code=403, detail="Artifact not allowed by configuration patterns")
+
+    remote_url = construct_url(remote_config, path)
+    if not remote_config.get("base_url"):
+        raise HTTPException(status_code=500, detail=f"No base_url configured for remote '{remote_name}'")
+
+    cached_key = storage.get_object_key(remote_name, path)
+    if not storage.exists(cached_key):
+        cached_key = None
+
+    filename = os.path.basename(path)
+    is_mutable = cache.is_mutable_file(path, mutable_patterns)
+
+    if cached_key and is_mutable:
+        if not cache.is_index_valid(remote_name, path):
+            if not await handle_expired_mutable(remote_name, path, remote_url, config, cache, storage):
+                cached_key = None
+
+    if cached_key:
+        if not is_mutable:
+            published = cache.get_artifact_published(remote_name, path)
+            if not published:
+                published = await _fetch_last_modified(remote_url, remote_config)
+                if published:
+                    cache.store_artifact_published(remote_name, path, published)
+            _check_quarantine(remote_name, published, config)
+
+        try:
+            artifact_data = storage.download_object(cached_key)
+            artifact_data, content_type = _resolve_content(artifact_data, path, filename, remote_config, request, remote_name)
+            logger.info(f"Cache HIT: {remote_name}/{path} (size: {len(artifact_data)} bytes, key: {cached_key})")
+            metrics.record_cache_hit(remote_name, len(artifact_data))
+            database.record_artifact_mapping(cached_key, remote_name, path, len(artifact_data))
+            return Response(
+                content=artifact_data,
+                media_type=content_type,
+                headers={
+                    "Content-Disposition": f"attachment; filename={filename}",
+                    "X-Artifact-Source": "cache",
+                    "X-Artifact-Size": str(len(artifact_data)),
+                },
+            )
+        except HTTPException:
+            raise
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=f"Error retrieving cached artifact: {str(e)}")
+
+    logger.info(f"Cache MISS: {remote_name}/{path} - fetching from remote: {remote_url}")
+    result = await cache_single_artifact(remote_url, remote_name, path, storage, remote_config)
+
+    if result["status"] == "error":
+        logger.error(f"Cache ADD FAILED: {remote_name}/{path} - {result['error']}")
+        raise HTTPException(status_code=502, detail=f"Failed to fetch artifact: {result['error']}")
+
+    if result["status"] == "cached" and is_mutable:
+        cache_config = config.get_cache_config(remote_name)
+        mutable_ttl = cache_config.get("mutable_ttl", 3600)
+        cache.mark_index_cached(remote_name, path, mutable_ttl)
+        logger.info(f"Mutable file cached with TTL: {remote_name}/{path} (ttl: {mutable_ttl}s)")
+        if result.get("etag") or result.get("last_modified"):
+            cache.store_mutable_meta(remote_name, path, result.get("etag"), result.get("last_modified"))
+
+    if not is_mutable:
+        published = result.get("last_modified")
+        if published:
+            cache.store_artifact_published(remote_name, path, published)
+        _check_quarantine(remote_name, published, config)
+
+    try:
+        cache_key = storage.get_object_key(remote_name, path)
+        artifact_data = storage.download_object(cache_key)
+        artifact_data, content_type = _resolve_content(artifact_data, path, filename, remote_config, request, remote_name)
+        metrics.record_cache_miss(remote_name, len(artifact_data))
+        database.record_artifact_mapping(cache_key, remote_name, path, len(artifact_data))
+        return Response(
+            content=artifact_data,
+            media_type=content_type,
+            headers={
+                "Content-Disposition": f"attachment; filename={filename}",
+                "X-Artifact-Source": "remote",
+                "X-Artifact-Size": str(len(artifact_data)),
+            },
+        )
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Error serving artifact: {str(e)}")

src/artifactapi/artifact/virtual.py

@@ -0,0 +1,227 @@
+import asyncio
+import base64
+import logging
+import time
+from datetime import UTC, date, datetime
+from typing import Protocol, runtime_checkable
+
+import httpx
+import yaml
+from fastapi import HTTPException, Request, Response
+
+from ..remote import helm as _helm
+
+logger = logging.getLogger(__name__)
+
+
+class _HelmDumper(yaml.Dumper):
+    """YAML dumper that serializes datetime/date objects back to ISO 8601 strings.
+
+    yaml.safe_load converts timestamp-shaped YAML scalars (e.g. chart `created`
+    fields) to Python datetime objects. Without a custom representer, yaml.dump
+    would render them as "2022-12-16 11:08:49+00:00" (space, not T), which
+    Go's YAML parser cannot unmarshal into time.Time.
+    """
+
+
+def _repr_datetime(dumper: yaml.Dumper, data: datetime) -> yaml.ScalarNode:
+    s = data.strftime("%Y-%m-%dT%H:%M:%S.%f") + ("Z" if data.tzinfo else "")
+    return dumper.represent_scalar("tag:yaml.org,2002:str", s)
+
+
+def _repr_date(dumper: yaml.Dumper, data: date) -> yaml.ScalarNode:
+    return dumper.represent_scalar("tag:yaml.org,2002:str", data.isoformat())
+
+
+_HelmDumper.add_representer(datetime, _repr_datetime)
+_HelmDumper.add_representer(date, _repr_date)
+
+
+async def _get_member_index(
+    member_name: str,
+    member_cfg: dict,
+    path: str,
+    storage,
+    cache,
+) -> tuple[str, dict, int, bytes | None]:
+    """Fetch or retrieve cached index.yaml for one member remote.
+
+    Returns (member_name, member_cfg, ttl, raw_bytes).
+    raw_bytes is None if the member is unreachable and not in S3.
+    """
+    member_ttl = member_cfg.get("cache", {}).get("mutable_ttl", 3600)
+    s3_key = storage.get_object_key(member_name, path)
+    raw_data: bytes | None = None
+
+    if storage.exists(s3_key) and cache.is_index_valid(member_name, path):
+        try:
+            raw_data = storage.download_object(s3_key)
+            logger.info(f"Virtual: cache hit for member '{member_name}'")
+        except Exception:
+            raw_data = None
+
+    if raw_data is None:
+        base_url = member_cfg.get("base_url", "").rstrip("/")
+        upstream_url = f"{base_url}/index.yaml"
+        headers = {}
+        username = member_cfg.get("username")
+        password = member_cfg.get("password")
+        if username and password:
+            token = base64.b64encode(f"{username}:{password}".encode()).decode()
+            headers["Authorization"] = f"Basic {token}"
+        try:
+            async with httpx.AsyncClient(follow_redirects=True) as client:
+                response = await client.get(upstream_url, headers=headers, timeout=30.0)
+                response.raise_for_status()
+                raw_data = response.content
+        except Exception as e:
+            logger.warning(f"Virtual: failed to fetch index.yaml from member '{member_name}': {e}")
+            return member_name, member_cfg, member_ttl, None
+        try:
+            storage.upload(s3_key, raw_data)
+            cache.mark_index_cached(member_name, path, member_ttl)
+        except Exception as e:
+            logger.warning(f"Virtual: failed to cache index.yaml for member '{member_name}': {e}")
+
+    return member_name, member_cfg, member_ttl, raw_data
+
+
+def _merge_helm_indexes(raw_indexes: list[bytes], member_names: list[str], member_configs: list[dict], proxy_base: str) -> bytes:
+    """Merge helm index.yaml files with per-member URL rewriting.
+
+    Priority is determined by position in member_names: earlier members win
+    when the same chart name + version appears in multiple remotes.
+    """
+    merged_entries: dict[str, list] = {}
+
+    for raw_data, member_name, member_cfg in zip(raw_indexes, member_names, member_configs):
+        base_url = member_cfg.get("base_url", "").rstrip("/")
+        rewritten, _ = _helm.resolve_content(raw_data, "index.yaml", "index.yaml", base_url, proxy_base, member_name)
+
+        try:
+            index = yaml.safe_load(rewritten)
+        except Exception as e:
+            logger.warning(f"Virtual: failed to parse index.yaml from member '{member_name}': {e}")
+            continue
+
+        for chart_name, versions in (index.get("entries") or {}).items():
+            if chart_name not in merged_entries:
+                merged_entries[chart_name] = list(versions)
+            else:
+                existing = {(v.get("name"), v.get("version")) for v in merged_entries[chart_name]}
+                for version_entry in versions:
+                    key = (version_entry.get("name"), version_entry.get("version"))
+                    if key not in existing:
+                        merged_entries[chart_name].append(version_entry)
+                        existing.add(key)
+
+    merged = {
+        "apiVersion": "v1",
+        "entries": merged_entries,
+        "generated": datetime.now(UTC).strftime("%Y-%m-%dT%H:%M:%S.000Z"),
+    }
+    return yaml.dump(merged, Dumper=_HelmDumper, default_flow_style=False, allow_unicode=True).encode()
+
+
+@runtime_checkable
+class _VirtualHandler(Protocol):
+    def accepts_path(self, path: str) -> bool: ...
+    def merge(self, raw_indexes: list[bytes], member_names: list[str], member_configs: list[dict], proxy_base: str) -> bytes: ...
+    def path_error(self) -> str: ...
+
+
+class _HelmHandler:
+    def accepts_path(self, path: str) -> bool:
+        return path == "index.yaml"
+
+    def merge(self, raw_indexes: list[bytes], member_names: list[str], member_configs: list[dict], proxy_base: str) -> bytes:
+        return _merge_helm_indexes(raw_indexes, member_names, member_configs, proxy_base)
+
+    def path_error(self) -> str:
+        return "Virtual helm repositories only serve index.yaml; chart tarballs are served directly by member remotes"
+
+
+_HANDLERS: dict[str, _VirtualHandler] = {
+    "helm": _HelmHandler(),
+}
+
+
+async def handle(request: Request, virtual_name: str, path: str, storage, cache, config) -> Response:
+    virtual_cfg = config.get_remote_config(virtual_name)
+    if not virtual_cfg:
+        raise HTTPException(status_code=404, detail=f"Virtual repository '{virtual_name}' not configured")
+    if virtual_cfg.get("type") != "virtual":
+        raise HTTPException(status_code=400, detail=f"'{virtual_name}' is not a virtual repository")
+
+    package = virtual_cfg.get("package")
+    handler = _HANDLERS.get(package)
+    if handler is None:
+        raise HTTPException(status_code=400, detail=f"Virtual repositories with package '{package}' are not yet supported")
+
+    if not handler.accepts_path(path):
+        raise HTTPException(status_code=404, detail=handler.path_error())
+
+    members = virtual_cfg.get("members", [])
+    if not members:
+        raise HTTPException(status_code=500, detail=f"Virtual repository '{virtual_name}' has no members configured")
+
+    virtual_key = storage.get_object_key(virtual_name, path)
+
+    if cache.is_index_valid(virtual_name, path) and storage.exists(virtual_key):
+        data = storage.download_object(virtual_key)
+        logger.info(f"Virtual HIT: {virtual_name}/{path}")
+        return Response(content=data, media_type="text/yaml")
+
+    # Resolve configs first (config reads are sync/cheap)
+    member_entries = []
+    for member_name in members:
+        member_cfg = config.get_remote_config(member_name)
+        if not member_cfg:
+            logger.warning(f"Virtual '{virtual_name}': member '{member_name}' not found in config, skipping")
+            continue
+        member_entries.append((member_name, member_cfg))
+
+    # Fetch all member indexes in parallel; asyncio.gather preserves input order
+    proxy_base = str(request.base_url).rstrip("/")
+    t_fetch = time.perf_counter()
+    results = await asyncio.gather(*[_get_member_index(name, cfg, path, storage, cache) for name, cfg in member_entries])
+    fetch_ms = int((time.perf_counter() - t_fetch) * 1000)
+
+    raw_indexes: list[bytes] = []
+    used_members: list[str] = []
+    used_configs: list[dict] = []
+    min_ttl: int | None = None
+
+    for member_name, member_cfg, member_ttl, raw_data in results:
+        if min_ttl is None or member_ttl < min_ttl:
+            min_ttl = member_ttl
+        if raw_data is None:
+            logger.warning(f"Virtual '{virtual_name}': skipping unreachable member '{member_name}'")
+            continue
+        raw_indexes.append(raw_data)
+        used_members.append(member_name)
+        used_configs.append(member_cfg)
+
+    if not raw_indexes:
+        raise HTTPException(status_code=502, detail=f"Virtual repository '{virtual_name}': no member indices could be fetched")
+
+    if min_ttl is None:
+        min_ttl = 3600
+
+    t_merge = time.perf_counter()
+    merged = handler.merge(raw_indexes, used_members, used_configs, proxy_base)
+    merge_ms = int((time.perf_counter() - t_merge) * 1000)
+
+    try:
+        t_store = time.perf_counter()
+        storage.upload(virtual_key, merged)
+        cache.mark_index_cached(virtual_name, path, min_ttl)
+        store_ms = int((time.perf_counter() - t_store) * 1000)
+        logger.info(
+            f"Virtual MISS: {virtual_name}/{path} rebuilt from {used_members} "
+            f"(fetch={fetch_ms}ms merge={merge_ms}ms store={store_ms}ms ttl={min_ttl}s)"
+        )
+    except Exception as e:
+        logger.warning(f"Virtual: failed to store merged index for '{virtual_name}': {e}")
+
+    return Response(content=merged, media_type="text/yaml")

src/artifactapi/auth/__init__.py

@@ -0,0 +1,3 @@
+from .docker import fetch_token, get_docker_token_for_response, parse_www_authenticate
+
+__all__ = ["fetch_token", "get_docker_token_for_response", "parse_www_authenticate"]

src/artifactapi/auth/docker.py

@@ -0,0 +1,96 @@
+import logging
+import re
+import time
+
+import httpx
+
+logger = logging.getLogger(__name__)
+
+# In-memory token cache: key -> (token, expires_at)
+_token_cache: dict[str, tuple[str, float]] = {}
+
+_WWW_AUTH_RE = re.compile(
+    r'Bearer\s+realm="(?P<realm>[^"]+)"'
+    r'(?:,service="(?P<service>[^"]*)")?'
+    r'(?:,scope="(?P<scope>[^"]*)")?',
+    re.IGNORECASE,
+)
+
+
+def _cache_key(realm: str, service: str, scope: str, username: str | None) -> str:
+    return f"{realm}|{service}|{scope}|{username or ''}"
+
+
+def _get_cached_token(key: str) -> str | None:
+    entry = _token_cache.get(key)
+    if entry and entry[1] > time.time():
+        return entry[0]
+    _token_cache.pop(key, None)
+    return None
+
+
+def _store_token(key: str, token: str, expires_in: int) -> None:
+    # Expire 30s early to avoid using a token right as it expires
+    _token_cache[key] = (token, time.time() + max(expires_in - 30, 10))
+
+
+async def fetch_token(
+    realm: str,
+    service: str,
+    scope: str,
+    username: str | None = None,
+    password: str | None = None,
+) -> str | None:
+    """Fetch a Bearer token from a Docker registry auth server."""
+    key = _cache_key(realm, service, scope, username)
+    cached = _get_cached_token(key)
+    if cached:
+        return cached
+
+    params: dict[str, str] = {}
+    if service:
+        params["service"] = service
+    if scope:
+        params["scope"] = scope
+
+    auth = (username, password) if username and password else None
+
+    try:
+        async with httpx.AsyncClient(follow_redirects=True) as client:
+            response = await client.get(realm, params=params, auth=auth)
+            response.raise_for_status()
+            data = response.json()
+    except Exception as e:
+        logger.warning(f"Docker token fetch failed ({realm}): {e}")
+        return None
+
+    token = data.get("token") or data.get("access_token")
+    if not token:
+        logger.warning(f"Docker token response missing token field: {data}")
+        return None
+
+    expires_in = int(data.get("expires_in", 300))
+    _store_token(key, token, expires_in)
+    logger.debug(f"Docker token obtained (realm={realm}, service={service}, scope={scope}, expires_in={expires_in}s)")
+    return token
+
+
+def parse_www_authenticate(header: str) -> tuple[str, str, str] | None:
+    """Parse WWW-Authenticate: Bearer header. Returns (realm, service, scope) or None."""
+    m = _WWW_AUTH_RE.search(header)
+    if not m:
+        return None
+    return m.group("realm"), m.group("service") or "", m.group("scope") or ""
+
+
+async def get_docker_token_for_response(
+    www_authenticate: str,
+    username: str | None = None,
+    password: str | None = None,
+) -> str | None:
+    """Given a WWW-Authenticate header value, fetch and return a Bearer token."""
+    parsed = parse_www_authenticate(www_authenticate)
+    if not parsed:
+        return None
+    realm, service, scope = parsed
+    return await fetch_token(realm, service, scope, username, password)

src/artifactapi/cache.py

@@ -1,69 +0,0 @@
-import time
-import hashlib
-import redis
-
-
-class RedisCache:
-    def __init__(self, redis_url: str):
-        self.redis_url = redis_url
-
-        try:
-            self.client = redis.from_url(self.redis_url, decode_responses=True)
-            # Test connection
-            self.client.ping()
-            self.available = True
-        except Exception as e:
-            print(f"Redis not available: {e}")
-            self.client = None
-            self.available = False
-
-    def is_index_file(self, file_path: str) -> bool:
-        """Check if the file is an index file that should have TTL"""
-        return (
-            file_path.endswith("APKINDEX.tar.gz")
-            or file_path.endswith("Packages.gz")
-            or file_path.endswith("repomd.xml")
-            or "repodata/" in file_path
-            and file_path.endswith((".xml", ".xml.gz", ".xml.bz2", ".xml.xz"))
-        )
-
-    def get_index_cache_key(self, remote_name: str, path: str) -> str:
-        """Generate cache key for index files"""
-        return f"index:{remote_name}:{hashlib.sha256(path.encode()).hexdigest()[:16]}"
-
-    def is_index_valid(
-        self, remote_name: str, path: str, ttl_override: int = None
-    ) -> bool:
-        """Check if index file is still valid (not expired)"""
-        if not self.available:
-            return False
-
-        try:
-            key = self.get_index_cache_key(remote_name, path)
-            return self.client.exists(key) > 0
-        except Exception:
-            return False
-
-    def mark_index_cached(self, remote_name: str, path: str, ttl: int = 300) -> None:
-        """Mark index file as cached with TTL"""
-        if not self.available:
-            return
-
-        try:
-            key = self.get_index_cache_key(remote_name, path)
-            self.client.setex(key, ttl, str(int(time.time())))
-        except Exception:
-            pass
-
-    def cleanup_expired_index(self, storage, remote_name: str, path: str) -> None:
-        """Remove expired index from S3 storage"""
-        if not self.available:
-            return
-
-        try:
-            # Get the S3 key and remove it
-            s3_key = storage.get_object_key_from_path(remote_name, path)
-            if storage.exists(s3_key):
-                storage.client.delete_object(Bucket=storage.bucket, Key=s3_key)
-        except Exception:
-            pass

src/artifactapi/cache/__init__.py

@@ -0,0 +1,3 @@
+from .redis import RedisCache
+
+__all__ = ["RedisCache"]

src/artifactapi/cache/redis.py

@@ -0,0 +1,124 @@
+import hashlib
+import re
+import time
+
+import redis
+
+
+class RedisCache:
+    def __init__(self, redis_url: str):
+        self.redis_url = redis_url
+
+        try:
+            self.client = redis.from_url(self.redis_url, decode_responses=True)
+            self.client.ping()
+            self.available = True
+        except Exception as e:
+            print(f"Redis not available: {e}")
+            self.client = None
+            self.available = False
+
+    def is_mutable_file(self, file_path: str, patterns: list[str] | None = None) -> bool:
+        if patterns is None:
+            patterns = []
+        return any(re.search(p, file_path) for p in patterns)
+
+    def get_index_cache_key(self, remote_name: str, path: str) -> str:
+        return f"index:{remote_name}:{hashlib.sha256(path.encode()).hexdigest()[:16]}"
+
+    def get_mutable_meta_key(self, remote_name: str, path: str) -> str:
+        return f"mutable:meta:{remote_name}:{hashlib.sha256(path.encode()).hexdigest()[:16]}"
+
+    def is_index_valid(self, remote_name: str, path: str) -> bool:
+        if not self.available:
+            return False
+        try:
+            key = self.get_index_cache_key(remote_name, path)
+            return self.client.exists(key) > 0
+        except Exception:
+            return False
+
+    def mark_index_cached(self, remote_name: str, path: str, ttl: int = 300) -> None:
+        if not self.available:
+            return
+        try:
+            key = self.get_index_cache_key(remote_name, path)
+            self.client.setex(key, ttl, str(int(time.time())))
+        except Exception:
+            pass
+
+    def store_mutable_meta(self, remote_name: str, path: str, etag: str | None, last_modified: str | None) -> None:
+        if not self.available:
+            return
+        data = {}
+        if etag:
+            data["etag"] = etag
+        if last_modified:
+            data["last_modified"] = last_modified
+        if not data:
+            return
+        try:
+            self.client.hset(self.get_mutable_meta_key(remote_name, path), mapping=data)
+        except Exception:
+            pass
+
+    def get_mutable_meta(self, remote_name: str, path: str) -> dict:
+        if not self.available:
+            return {}
+        try:
+            return self.client.hgetall(self.get_mutable_meta_key(remote_name, path)) or {}
+        except Exception:
+            return {}
+
+    def delete_mutable_meta(self, remote_name: str, path: str) -> None:
+        if not self.available:
+            return
+        try:
+            self.client.delete(self.get_mutable_meta_key(remote_name, path))
+        except Exception:
+            pass
+
+    def get_artifact_published_key(self, remote_name: str, path: str) -> str:
+        return f"pkg:published:{remote_name}:{hashlib.sha256(path.encode()).hexdigest()[:16]}"
+
+    def store_artifact_published(self, remote_name: str, path: str, last_modified: str) -> None:
+        """Persist the upstream Last-Modified header for a (typically immutable) artifact."""
+        if not self.available:
+            return
+        try:
+            self.client.set(self.get_artifact_published_key(remote_name, path), last_modified)
+        except Exception:
+            pass
+
+    def get_artifact_published(self, remote_name: str, path: str) -> str | None:
+        """Return the stored Last-Modified string for an artifact, or None."""
+        if not self.available:
+            return None
+        try:
+            return self.client.get(self.get_artifact_published_key(remote_name, path))
+        except Exception:
+            return None
+
+    def cleanup_expired_index(self, storage, remote_name: str, path: str) -> None:
+        if not self.available:
+            return
+
+        try:
+            import os
+
+            from ..config import ConfigManager
+
+            config_path = os.environ.get("CONFIG_PATH")
+            if config_path:
+                config = ConfigManager(config_path)
+                remote_config = config.get_remote_config(remote_name)
+                if remote_config:
+                    base_url = remote_config.get("base_url")
+                    if base_url:
+                        s3_key = storage.get_object_key(remote_name, path)
+                        if storage.exists(s3_key):
+                            storage.client.delete_object(Bucket=storage.bucket, Key=s3_key)
+        except Exception:
+            pass
+
+        self.delete_mutable_meta(remote_name, path)

src/artifactapi/config.py

@@ -1,64 +1,156 @@
-import os
+import glob
 import json
+import os
+
 import yaml
-from typing import Optional
+
+_TYPE_KEYS = ("remote", "virtual", "local")
+
+
+def _normalize_loaded(raw: dict) -> dict:
+    """Convert {remote: {...}, virtual: {...}, local: {...}} into {remotes: {name: {type: ..., ...}}}."""
+    remotes = {}
+    for type_key in _TYPE_KEYS:
+        for name, cfg in (raw.get(type_key) or {}).items():
+            remotes[name] = {"type": type_key, **cfg}
+    result = {k: v for k, v in raw.items() if k not in _TYPE_KEYS}
+    if remotes:
+        result["remotes"] = remotes
+    return result
+
+
+_PACKAGE_MUTABLE_PATTERNS: dict[str, list[str]] = {
+    "alpine": [
+        r"APKINDEX\.tar\.gz$",
+    ],
+    "rpm": [
+        r"repomd\.xml$",
+        r"repodata/.*\.(xml|xml\.gz|xml\.bz2|xml\.xz|xml\.zck|xml\.zst"
+        r"|sqlite|sqlite\.gz|sqlite\.bz2|sqlite\.xz|sqlite\.zck|sqlite\.zst"
+        r"|yaml\.xz|yaml\.gz|yaml\.bz2|yaml\.zst|asc|txt)$",
+        r"Packages\.gz$",
+    ],
+    "docker": [
+        r"/manifests/(?!sha256:)[^/]+$",
+        r"/tags/list$",
+    ],
+    "pypi": [
+        r"simple/",  # Per-package and top-level simple index pages
+    ],
+    "npm": [],
+    "helm": [
+        r"index\.yaml$",
+    ],
+    "generic": [],
+}
 
 
 class ConfigManager:
-    def __init__(self, config_file: str = "remotes.yaml"):
-        self.config_file = config_file
-        self._last_modified = 0
+    def __init__(self, config_path: str = "remotes.yaml"):
+        self.config_path = config_path
+        self._config_dir: str | None = None
+        self._last_modified: float = 0.0
         self.config = self._load_config()
 
-    def _load_config(self) -> dict:
+    def _load_single_file(self, path: str) -> dict:
         try:
-            with open(self.config_file, "r") as f:
-                if self.config_file.endswith(".yaml") or self.config_file.endswith(
-                    ".yml"
-                ):
-                    return yaml.safe_load(f)
+            with open(path) as f:
+                if path.endswith((".yaml", ".yml")):
+                    raw = yaml.safe_load(f) or {}
                 else:
-                    return json.load(f)
+                    raw = json.load(f)
+            return _normalize_loaded(raw)
         except FileNotFoundError:
+            return {}
+
+    @staticmethod
+    def _merge(base: dict, overlay: dict) -> dict:
+        result = {**base}
+        for key, value in overlay.items():
+            if key == "remotes" and isinstance(base.get("remotes"), dict) and isinstance(value, dict):
+                result["remotes"] = {**base.get("remotes", {}), **value}
+            else:
+                result[key] = value
+        return result
+
+    def _load_from_dir(self, dir_path: str) -> dict:
+        merged: dict = {}
+        files = sorted(glob.glob(os.path.join(dir_path, "*.yaml")) + glob.glob(os.path.join(dir_path, "*.yml")))
+        for path in files:
+            merged = self._merge(merged, self._load_single_file(path))
+        return merged
+
+    def _load_config(self) -> dict:
+        self._config_dir = None
+
+        if os.path.isdir(self.config_path):
+            return self._load_from_dir(self.config_path) or {"remotes": {}}
+
+        config = self._load_single_file(self.config_path)
+        if not config:
             return {"remotes": {}}
 
-    def _check_reload(self) -> None:
-        """Check if config file has been modified and reload if needed"""
-        try:
-            import os
+        config_dir = config.pop("config_dir", None)
+        if config_dir:
+            if not os.path.isabs(config_dir):
+                config_dir = os.path.join(os.path.dirname(os.path.abspath(self.config_path)), config_dir)
+            self._config_dir = config_dir
+            config = self._merge(config, self._load_from_dir(config_dir))
 
-            current_modified = os.path.getmtime(self.config_file)
+        return config
+
+    def _file_mtimes(self) -> list[float]:
+        mtimes: list[float] = []
+        if os.path.isdir(self.config_path):
+            for f in glob.glob(os.path.join(self.config_path, "*.yaml")) + glob.glob(os.path.join(self.config_path, "*.yml")):
+                try:
+                    mtimes.append(os.path.getmtime(f))
+                except OSError:
+                    pass
+        else:
+            try:
+                mtimes.append(os.path.getmtime(self.config_path))
+            except OSError:
+                pass
+
+        if self._config_dir and os.path.isdir(self._config_dir):
+            for f in glob.glob(os.path.join(self._config_dir, "*.yaml")) + glob.glob(os.path.join(self._config_dir, "*.yml")):
+                try:
+                    mtimes.append(os.path.getmtime(f))
+                except OSError:
+                    pass
+
+        return mtimes
+
+    def _check_reload(self) -> None:
+        try:
+            current_modified = max(self._file_mtimes(), default=0.0)
             if current_modified > self._last_modified:
                 self._last_modified = current_modified
                 self.config = self._load_config()
-                print(f"Config reloaded from {self.config_file}")
+                print(f"Config reloaded from {self.config_path}")
         except OSError:
             pass
 
-    def get_remote_config(self, remote_name: str) -> Optional[dict]:
+    def get_remote_config(self, remote_name: str) -> dict | None:
         self._check_reload()
         return self.config.get("remotes", {}).get(remote_name)
 
-    def get_repository_patterns(self, remote_name: str, repo_path: str) -> list:
+    def get_immutable_patterns(self, remote_name: str, repo_path: str = "") -> list[str]:
         remote_config = self.get_remote_config(remote_name)
         if not remote_config:
             return []
 
         repositories = remote_config.get("repositories", {})
 
-        # Handle both dict (GitHub style) and list (Alpine style) repositories
         if isinstance(repositories, dict):
             repo_config = repositories.get(repo_path)
             if repo_config:
-                patterns = repo_config.get("include_patterns", [])
+                patterns = repo_config.get("immutable_patterns", [])
             else:
-                patterns = remote_config.get("include_patterns", [])
-        elif isinstance(repositories, list):
-            # For Alpine, repositories is just a list of allowed repo names
-            # Pattern matching is handled by the main include_patterns
-            patterns = remote_config.get("include_patterns", [])
+                patterns = remote_config.get("immutable_patterns", [])
         else:
-            patterns = remote_config.get("include_patterns", [])
+            patterns = remote_config.get("immutable_patterns", [])
 
         return patterns
 
@@ -92,9 +184,7 @@ class ConfigManager:
         if not redis_url:
             raise ValueError("REDIS_URL environment variable is required")
 
-        return {
-            "url": redis_url
-        }
+        return {"url": redis_url}
 
     def get_database_config(self) -> dict:
         """Get database configuration from environment variables"""
@@ -105,12 +195,37 @@ class ConfigManager:
         db_name = os.getenv("DBNAME")
 
         if not all([db_host, db_port, db_user, db_pass, db_name]):
-            missing = [var for var, val in [("DBHOST", db_host), ("DBPORT", db_port), ("DBUSER", db_user), ("DBPASS", db_pass), ("DBNAME", db_name)] if not val]
+            missing = [
+                var
+                for var, val in [("DBHOST", db_host), ("DBPORT", db_port), ("DBUSER", db_user), ("DBPASS", db_pass), ("DBNAME", db_name)]
+                if not val
+            ]
             raise ValueError(f"All database environment variables are required: {', '.join(missing)}")
 
         db_url = f"postgresql://{db_user}:{db_pass}@{db_host}:{db_port}/{db_name}"
         return {"url": db_url}
 
+    def get_user_mutable_patterns(self, remote_name: str) -> list[str]:
+        """Return only user-configured mutable_patterns, excluding package-type defaults."""
+        remote_config = self.get_remote_config(remote_name)
+        if not remote_config:
+            return []
+        return remote_config.get("mutable_patterns", [])
+
+    def get_mutable_patterns(self, remote_name: str) -> list[str]:
+        """Return mutable-file patterns for a remote (TTL is configured per-remote in cache.index_ttl).
+
+        Merges the package-level defaults with any extra patterns listed under
+        ``mutable_patterns`` in the remote's config.
+        """
+        remote_config = self.get_remote_config(remote_name)
+        if not remote_config:
+            return []
+        package = remote_config.get("package", "generic")
+        defaults = _PACKAGE_MUTABLE_PATTERNS.get(package, [])
+        extra = remote_config.get("mutable_patterns", [])
+        return defaults + [p for p in extra if p not in defaults]
+
     def get_cache_config(self, remote_name: str) -> dict:
         """Get cache configuration for a specific remote"""
         remote_config = self.get_remote_config(remote_name)
@@ -118,3 +233,16 @@ class ConfigManager:
             return {}
 
         return remote_config.get("cache", {})
+
+    def get_quarantine_config(self, remote_name: str) -> tuple[bool, int]:
+        """Return (enabled, quarantine_days) for a remote.
+
+        When enabled=True and quarantine_days>0, immutable artifacts published
+        within the last quarantine_days days are blocked with a 404.
+        """
+        remote_config = self.get_remote_config(remote_name)
+        if not remote_config:
+            return False, 0
+        enabled = bool(remote_config.get("quarantine_new", False))
+        days = int(remote_config.get("quarantine_days", 0))
+        return enabled, days

src/artifactapi/database/__init__.py

@@ -0,0 +1,3 @@
+from .postgres import DatabaseManager
+
+__all__ = ["DatabaseManager"]

src/artifactapi/database/postgres.py

@@ -1,5 +1,3 @@
-import os
-from typing import Optional
 import psycopg2
 from psycopg2.extras import RealDictCursor
 
@@ -11,7 +9,6 @@ class DatabaseManager:
         self._init_database()
 
     def _init_database(self):
-        """Initialize database connection and create schema if needed"""
         try:
             self.connection = psycopg2.connect(self.db_url)
             self.connection.autocommit = True
@@ -23,10 +20,8 @@ class DatabaseManager:
             self.available = False
 
     def _create_schema(self):
-        """Create tables if they don't exist"""
         try:
             with self.connection.cursor() as cursor:
-                # Create table to map S3 keys to remote names
                 cursor.execute("""
                     CREATE TABLE IF NOT EXISTS artifact_mappings (
                         id SERIAL PRIMARY KEY,
@@ -53,27 +48,15 @@ class DatabaseManager:
                     )
                 """)
 
-                # Create indexes separately
-                cursor.execute(
-                    "CREATE INDEX IF NOT EXISTS idx_s3_key ON artifact_mappings (s3_key)"
-                )
-                cursor.execute(
-                    "CREATE INDEX IF NOT EXISTS idx_remote_name ON artifact_mappings (remote_name)"
-                )
-                cursor.execute(
-                    "CREATE INDEX IF NOT EXISTS idx_local_repo_path ON local_files (repository_name, file_path)"
-                )
-                cursor.execute(
-                    "CREATE INDEX IF NOT EXISTS idx_local_s3_key ON local_files (s3_key)"
-                )
+                cursor.execute("CREATE INDEX IF NOT EXISTS idx_s3_key ON artifact_mappings (s3_key)")
+                cursor.execute("CREATE INDEX IF NOT EXISTS idx_remote_name ON artifact_mappings (remote_name)")
+                cursor.execute("CREATE INDEX IF NOT EXISTS idx_local_repo_path ON local_files (repository_name, file_path)")
+                cursor.execute("CREATE INDEX IF NOT EXISTS idx_local_s3_key ON local_files (s3_key)")
                 print("Database schema initialized")
         except Exception as e:
             print(f"Error creating schema: {e}")
 
-    def record_artifact_mapping(
-        self, s3_key: str, remote_name: str, file_path: str, size_bytes: int
-    ):
-        """Record mapping between S3 key and remote"""
+    def record_artifact_mapping(self, s3_key: str, remote_name: str, file_path: str, size_bytes: int):
         if not self.available:
             return
 
@@ -95,7 +78,6 @@ class DatabaseManager:
             print(f"Error recording artifact mapping: {e}")
 
     def get_storage_by_remote(self) -> dict[str, int]:
-        """Get storage size breakdown by remote from database"""
         if not self.available:
             return {}
 
@@ -112,8 +94,7 @@ class DatabaseManager:
             print(f"Error getting storage by remote: {e}")
             return {}
 
-    def get_remote_for_s3_key(self, s3_key: str) -> Optional[str]:
-        """Get remote name for given S3 key"""
+    def get_remote_for_s3_key(self, s3_key: str) -> str | None:
         if not self.available:
             return None
 
@@ -138,7 +119,6 @@ class DatabaseManager:
         sha256_sum: str,
         content_type: str = None,
     ):
-        """Add a file to local repository"""
         if not self.available:
             return False
 
@@ -165,7 +145,6 @@ class DatabaseManager:
             return False
 
     def get_local_file_metadata(self, repository_name: str, file_path: str):
-        """Get metadata for a local file"""
         if not self.available:
             return None
 
@@ -197,7 +176,6 @@ class DatabaseManager:
             return None
 
     def list_local_files(self, repository_name: str, prefix: str = ""):
-        """List files in local repository with optional path prefix"""
         if not self.available:
             return []
 
@@ -241,7 +219,6 @@ class DatabaseManager:
             return []
 
     def delete_local_file(self, repository_name: str, file_path: str):
-        """Delete a file from local repository"""
         if not self.available:
             return False
 
@@ -263,7 +240,6 @@ class DatabaseManager:
             return None
 
     def file_exists(self, repository_name: str, file_path: str):
-        """Check if file exists in local repository"""
         if not self.available:
             return False
 

src/artifactapi/docker_auth.py

@@ -0,0 +1,19 @@
+from .auth.docker import (
+    _cache_key,
+    _get_cached_token,
+    _store_token,
+    _token_cache,
+    fetch_token,
+    get_docker_token_for_response,
+    parse_www_authenticate,
+)
+
+__all__ = [
+    "_cache_key",
+    "_get_cached_token",
+    "_store_token",
+    "_token_cache",
+    "fetch_token",
+    "get_docker_token_for_response",
+    "parse_www_authenticate",
+]

src/artifactapi/main.py

@@ -1,18 +1,44 @@
+import logging
 import os
-import re
-import hashlib
-from typing import Dict, Any, Optional
-import httpx
-from fastapi import FastAPI, HTTPException, Response, Query, File, UploadFile
-from fastapi.responses import PlainTextResponse, JSONResponse
-from pydantic import BaseModel
-from prometheus_client import generate_latest, CONTENT_TYPE_LATEST
 
+from fastapi import FastAPI, File, Query, Request, UploadFile
+from fastapi.responses import PlainTextResponse
+from prometheus_client import CONTENT_TYPE_LATEST, generate_latest
+from pydantic import BaseModel
+
+try:
+    from importlib.metadata import version
+
+    __version__ = version("artifactapi")
+except ImportError:
+    __version__ = "dev"
+
+from .artifact import discovery, flush, local, proxy, virtual
+from .artifact import docker as docker_handler
+from .cache import RedisCache
 from .config import ConfigManager
 from .database import DatabaseManager
-from .storage import S3Storage
-from .cache import RedisCache
 from .metrics import MetricsManager
+from .storage import S3Storage
+
+logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s")
+logger = logging.getLogger(__name__)
+
+app = FastAPI(title="Artifact Storage API", version=__version__)
+
+config_path = os.environ.get("CONFIG_PATH")
+if not config_path:
+    raise ValueError("CONFIG_PATH environment variable is required")
+config = ConfigManager(config_path)
+
+s3_config = config.get_s3_config()
+redis_config = config.get_redis_config()
+db_config = config.get_database_config()
+
+storage = S3Storage(**s3_config)
+cache = RedisCache(redis_config["url"])
+database = DatabaseManager(db_config["url"])
+metrics = MetricsManager(cache, database)
 
 
 class ArtifactRequest(BaseModel):
@@ -20,34 +46,10 @@ class ArtifactRequest(BaseModel):
     include_pattern: str
 
 
-app = FastAPI(title="Artifact Storage API", version="2.0.0")
-
-# Initialize components using config
-config_path = os.environ.get("CONFIG_PATH")
-if not config_path:
-    raise ValueError("CONFIG_PATH environment variable is required")
-config = ConfigManager(config_path)
-
-# Get configurations
-s3_config = config.get_s3_config()
-redis_config = config.get_redis_config()
-db_config = config.get_database_config()
-
-# Initialize services
-storage = S3Storage(**s3_config)
-cache = RedisCache(redis_config["url"])
-database = DatabaseManager(db_config["url"])
-metrics = MetricsManager(cache, database)
-
-
 @app.get("/")
 def read_root():
     config._check_reload()
-    return {
-        "message": "Artifact Storage API",
-        "version": "2.0.0",
-        "remotes": list(config.config.get("remotes", {}).keys()),
-    }
+    return {"message": "Artifact Storage API", "version": app.version, "remotes": list(config.config.get("remotes", {}).keys())}
 
 
 @app.get("/health")
@@ -55,522 +57,71 @@ def health_check():
     return {"status": "healthy"}
 
 
-async def construct_remote_url(remote_name: str, path: str) -> str:
-    remote_config = config.get_remote_config(remote_name)
-    if not remote_config:
-        raise HTTPException(
-            status_code=404, detail=f"Remote '{remote_name}' not configured"
-        )
-
-    base_url = remote_config.get("base_url")
-    if not base_url:
-        raise HTTPException(
-            status_code=500, detail=f"No base_url configured for remote '{remote_name}'"
-        )
-
-    return f"{base_url}/{path}"
+@app.get("/config")
+def get_config():
+    return config.config
 
 
-async def check_artifact_patterns(
-    remote_name: str, repo_path: str, file_path: str, full_path: str
-) -> bool:
-    # First check if this is an index file - always allow index files
-    if cache.is_index_file(file_path) or cache.is_index_file(full_path):
-        return True
-
-    # Then check basic include patterns
-    patterns = config.get_repository_patterns(remote_name, repo_path)
-    if not patterns:
-        return True  # Allow all if no patterns configured
-
-    pattern_matched = False
-    for pattern in patterns:
-        # Check both file_path and full_path to handle different pattern types
-        if re.search(pattern, file_path) or re.search(pattern, full_path):
-            pattern_matched = True
-            break
-
-    if not pattern_matched:
-        return False
-
-    # All remotes now use pattern-based filtering only - no additional checks needed
-    return True
+@app.get("/metrics")
+def get_metrics(json: bool | None = Query(False, description="Return JSON format instead of Prometheus")):
+    config._check_reload()
+    if json:
+        return metrics.get_metrics(storage, config)
+    metrics.get_metrics(storage, config)
+    return PlainTextResponse(generate_latest().decode("utf-8"), media_type=CONTENT_TYPE_LATEST)
 
 
-async def cache_single_artifact(url: str, remote_name: str, path: str) -> dict:
-    # Check if using URL-based key or path-based key
-    if url.startswith("http"):
-        key = storage.get_object_key(url)
-    else:
-        key = storage.get_object_key_from_path(remote_name, path)
+@app.put("/cache/flush")
+def flush_cache(
+    remote: str = Query(default=None, description="Specific remote to flush (optional)"),
+    cache_type: str = Query(default="all", description="Type to flush: 'all', 'index', 'files', 'metrics'"),
+):
+    return flush.handle(remote, cache_type, cache, storage)
 
-    if storage.exists(key):
-        return {
-            "url": url,
-            "cached_url": storage.get_url(key),
-            "status": "already_cached",
-        }
 
-    try:
-        async with httpx.AsyncClient(follow_redirects=True) as client:
-            response = await client.get(url)
-            response.raise_for_status()
+@app.get("/v2/")
+async def docker_v2_ping():
+    return docker_handler.ping()
 
-            storage_path = storage.upload(key, response.content)
 
-            return {
-                "url": url,
-                "cached_url": storage.get_url(key),
-                "storage_path": storage_path,
-                "size": len(response.content),
-                "status": "cached",
-            }
+@app.api_route("/v2/{remote_name}/{path:path}", methods=["GET", "HEAD"])
+async def docker_v2_proxy(request: Request, remote_name: str, path: str):
+    return await docker_handler.proxy(request, remote_name, path, storage, cache, config, metrics)
 
-    except Exception as e:
-        return {"url": url, "status": "error", "error": str(e)}
+
+@app.get("/api/v1/virtual/{virtual_name}/{path:path}")
+async def get_virtual_artifact(request: Request, virtual_name: str, path: str):
+    return await virtual.handle(request, virtual_name, path, storage, cache, config)
 
 
 @app.get("/api/v1/remote/{remote_name}/{path:path}")
-async def get_artifact(remote_name: str, path: str):
-    # Check if remote is configured
-    remote_config = config.get_remote_config(remote_name)
-    if not remote_config:
-        raise HTTPException(
-            status_code=404, detail=f"Remote '{remote_name}' not configured"
-        )
-
-    # Check if this is a local repository
-    if remote_config.get("type") == "local":
-        # Handle local repository download
-        metadata = database.get_local_file_metadata(remote_name, path)
-        if not metadata:
-            raise HTTPException(status_code=404, detail="File not found")
-
-        # Get file from S3
-        content = storage.download_object(metadata["s3_key"])
-        if content is None:
-            raise HTTPException(status_code=500, detail="File not accessible")
-
-        # Determine content type
-        content_type = metadata.get("content_type", "application/octet-stream")
-
-        return Response(
-            content=content,
-            media_type=content_type,
-            headers={
-                "Content-Disposition": f"attachment; filename={os.path.basename(path)}"
-            },
-        )
-
-    # Extract repository path for pattern checking
-    path_parts = path.split("/")
-    if len(path_parts) >= 2:
-        repo_path = f"{path_parts[0]}/{path_parts[1]}"
-        file_path = "/".join(path_parts[2:])
-    else:
-        repo_path = path
-        file_path = path
-
-    # Check if artifact matches configured patterns
-    if not await check_artifact_patterns(remote_name, repo_path, file_path, path):
-        raise HTTPException(
-            status_code=403, detail="Artifact not allowed by configuration patterns"
-        )
-
-    # Construct the remote URL
-    remote_url = await construct_remote_url(remote_name, path)
-
-    # Check if artifact is already cached (try both URL and path-based keys)
-    url_key = storage.get_object_key(remote_url)
-    path_key = storage.get_object_key_from_path(remote_name, path)
-
-    cached_key = None
-    if storage.exists(url_key):
-        cached_key = url_key
-    elif storage.exists(path_key):
-        cached_key = path_key
-
-    # For index files, check Redis TTL validity
-    filename = os.path.basename(path)
-    is_index = cache.is_index_file(filename)
-
-    if cached_key and is_index:
-        # Index file exists, but check if it's still valid
-        if not cache.is_index_valid(remote_name, path):
-            # Index has expired, remove it from S3
-            cache.cleanup_expired_index(storage, remote_name, path)
-            cached_key = None  # Force re-download
-
-    if cached_key:
-        # Return cached artifact
-        try:
-            artifact_data = storage.download_object(cached_key)
-            filename = os.path.basename(path)
-
-            # Determine content type based on file extension
-            content_type = "application/octet-stream"
-            if filename.endswith(".tar.gz"):
-                content_type = "application/gzip"
-            elif filename.endswith(".zip"):
-                content_type = "application/zip"
-            elif filename.endswith(".exe"):
-                content_type = "application/x-msdownload"
-            elif filename.endswith(".rpm"):
-                content_type = "application/x-rpm"
-            elif filename.endswith(".xml"):
-                content_type = "application/xml"
-            elif filename.endswith((".xml.gz", ".xml.bz2", ".xml.xz")):
-                content_type = "application/gzip"
-
-            # Record cache hit metrics
-            metrics.record_cache_hit(remote_name, len(artifact_data))
-
-            # Record artifact mapping in database if not already recorded
-            database.record_artifact_mapping(
-                cached_key, remote_name, path, len(artifact_data)
-            )
-
-            return Response(
-                content=artifact_data,
-                media_type=content_type,
-                headers={
-                    "Content-Disposition": f"attachment; filename={filename}",
-                    "X-Artifact-Source": "cache",
-                    "X-Artifact-Size": str(len(artifact_data)),
-                },
-            )
-        except Exception as e:
-            raise HTTPException(
-                status_code=500, detail=f"Error retrieving cached artifact: {str(e)}"
-            )
-
-    # Artifact not cached, cache it first
-    result = await cache_single_artifact(remote_url, remote_name, path)
-
-    if result["status"] == "error":
-        raise HTTPException(
-            status_code=502, detail=f"Failed to fetch artifact: {result['error']}"
-        )
-
-    # Mark index files as cached in Redis if this was a new download
-    if result["status"] == "cached" and is_index:
-        # Get TTL from remote config
-        cache_config = config.get_cache_config(remote_name)
-        index_ttl = cache_config.get("index_ttl", 300)  # Default 5 minutes
-        cache.mark_index_cached(remote_name, path, index_ttl)
-
-    # Now return the cached artifact
-    try:
-        cache_key = storage.get_object_key(remote_url)
-        artifact_data = storage.download_object(cache_key)
-        filename = os.path.basename(path)
-
-        content_type = "application/octet-stream"
-        if filename.endswith(".tar.gz"):
-            content_type = "application/gzip"
-        elif filename.endswith(".zip"):
-            content_type = "application/zip"
-        elif filename.endswith(".exe"):
-            content_type = "application/x-msdownload"
-        elif filename.endswith(".rpm"):
-            content_type = "application/x-rpm"
-        elif filename.endswith(".xml"):
-            content_type = "application/xml"
-        elif filename.endswith((".xml.gz", ".xml.bz2", ".xml.xz")):
-            content_type = "application/gzip"
-
-        # Record cache miss metrics
-        metrics.record_cache_miss(remote_name, len(artifact_data))
-
-        # Record artifact mapping in database
-        cache_key = storage.get_object_key(remote_url)
-        database.record_artifact_mapping(
-            cache_key, remote_name, path, len(artifact_data)
-        )
-
-        return Response(
-            content=artifact_data,
-            media_type=content_type,
-            headers={
-                "Content-Disposition": f"attachment; filename={filename}",
-                "X-Artifact-Source": "remote",
-                "X-Artifact-Size": str(len(artifact_data)),
-            },
-        )
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=f"Error serving artifact: {str(e)}")
-
-
-async def discover_artifacts(remote: str, include_pattern: str) -> list[str]:
-    if "github.com" in remote:
-        return await discover_github_releases(remote, include_pattern)
-    else:
-        raise HTTPException(status_code=400, detail=f"Unsupported remote: {remote}")
-
-
-async def discover_github_releases(remote: str, include_pattern: str) -> list[str]:
-    match = re.match(r"github\.com/([^/]+)/([^/]+)", remote)
-    if not match:
-        raise HTTPException(status_code=400, detail="Invalid GitHub remote format")
-
-    owner, repo = match.groups()
-
-    async with httpx.AsyncClient(follow_redirects=True) as client:
-        response = await client.get(
-            f"https://api.github.com/repos/{owner}/{repo}/releases"
-        )
-
-        if response.status_code != 200:
-            raise HTTPException(
-                status_code=response.status_code,
-                detail=f"Failed to fetch releases: {response.text}",
-            )
-
-        releases = response.json()
-
-        matching_urls = []
-        pattern = include_pattern.replace("*", ".*")
-        regex = re.compile(pattern)
-
-        for release in releases:
-            for asset in release.get("assets", []):
-                download_url = asset["browser_download_url"]
-                if regex.search(download_url):
-                    matching_urls.append(download_url)
-
-        return matching_urls
+async def get_artifact(request: Request, remote_name: str, path: str):
+    return await proxy.handle(request, remote_name, path, storage, cache, config, database, metrics)
 
 
 @app.put("/api/v1/remote/{remote_name}/{path:path}")
 async def upload_file(remote_name: str, path: str, file: UploadFile = File(...)):
-    """Upload a file to local repository"""
-    # Check if remote is configured and is local
-    remote_config = config.get_remote_config(remote_name)
-    if not remote_config:
-        raise HTTPException(
-            status_code=404, detail=f"Remote '{remote_name}' not configured"
-        )
-
-    if remote_config.get("type") != "local":
-        raise HTTPException(
-            status_code=400, detail="Upload only supported for local repositories"
-        )
-
-    try:
-        # Read file content
-        content = await file.read()
-
-        # Calculate SHA256
-        sha256_sum = hashlib.sha256(content).hexdigest()
-
-        # Check if file already exists (prevent overwrite)
-        if database.file_exists(remote_name, path):
-            raise HTTPException(status_code=409, detail="File already exists")
-
-        # Generate S3 key
-        s3_key = f"local/{remote_name}/{path}"
-
-        # Determine content type
-        content_type = file.content_type or "application/octet-stream"
-
-        # Upload to S3
-        try:
-            storage.upload(s3_key, content)
-        except Exception as e:
-            raise HTTPException(status_code=500, detail=f"Upload failed: {e}")
-
-        # Add to database
-        success = database.add_local_file(
-            repository_name=remote_name,
-            file_path=path,
-            s3_key=s3_key,
-            size_bytes=len(content),
-            sha256_sum=sha256_sum,
-            content_type=content_type,
-        )
-
-        if not success:
-            # Clean up S3 if database insert failed
-            storage.delete_object(s3_key)
-            raise HTTPException(status_code=500, detail="Failed to save file metadata")
-
-        return JSONResponse(
-            {
-                "message": "File uploaded successfully",
-                "file_path": path,
-                "size_bytes": len(content),
-                "sha256_sum": sha256_sum,
-                "content_type": content_type,
-            }
-        )
-
-    except HTTPException:
-        raise
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=f"Upload failed: {str(e)}")
+    return await local.upload(remote_name, path, file, storage, database, config)
 
 
 @app.head("/api/v1/remote/{remote_name}/{path:path}")
 def check_file_exists(remote_name: str, path: str):
-    """Check if file exists (for CI jobs) - supports local repositories only"""
-    # Check if remote is configured
-    remote_config = config.get_remote_config(remote_name)
-    if not remote_config:
-        raise HTTPException(
-            status_code=404, detail=f"Remote '{remote_name}' not configured"
-        )
-
-    # Handle local repository
-    if remote_config.get("type") == "local":
-        try:
-            metadata = database.get_local_file_metadata(remote_name, path)
-            if not metadata:
-                raise HTTPException(status_code=404, detail="File not found")
-
-            return Response(
-                headers={
-                    "Content-Length": str(metadata["size_bytes"]),
-                    "Content-Type": metadata.get(
-                        "content_type", "application/octet-stream"
-                    ),
-                    "X-SHA256": metadata["sha256_sum"],
-                    "X-Created-At": metadata["created_at"].isoformat()
-                    if metadata["created_at"]
-                    else "",
-                    "X-Uploaded-At": metadata["uploaded_at"].isoformat()
-                    if metadata["uploaded_at"]
-                    else "",
-                }
-            )
-        except HTTPException:
-            raise
-        except Exception as e:
-            raise HTTPException(status_code=500, detail=f"Check failed: {str(e)}")
-    else:
-        # For remote repositories, just return 405 Method Not Allowed
-        raise HTTPException(
-            status_code=405, detail="HEAD method only supported for local repositories"
-        )
+    return local.check_exists(remote_name, path, database, config)
 
 
 @app.delete("/api/v1/remote/{remote_name}/{path:path}")
 def delete_file(remote_name: str, path: str):
-    """Delete a file from local repository"""
-    # Check if remote is configured and is local
-    remote_config = config.get_remote_config(remote_name)
-    if not remote_config:
-        raise HTTPException(
-            status_code=404, detail=f"Remote '{remote_name}' not configured"
-        )
-
-    if remote_config.get("type") != "local":
-        raise HTTPException(
-            status_code=400, detail="Delete only supported for local repositories"
-        )
-
-    try:
-        # Get S3 key before deleting from database
-        s3_key = database.delete_local_file(remote_name, path)
-        if not s3_key:
-            raise HTTPException(status_code=404, detail="File not found")
-
-        # Delete from S3
-        if not storage.delete_object(s3_key):
-            # File was deleted from database but not from S3 - log warning but continue
-            print(f"Warning: Failed to delete S3 object {s3_key}")
-
-        return JSONResponse({"message": "File deleted successfully"})
-    except HTTPException:
-        raise
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=f"Delete failed: {str(e)}")
+    return local.delete(remote_name, path, storage, database, config)
 
 
 @app.post("/api/v1/artifacts/cache")
-async def cache_artifact(request: ArtifactRequest) -> Dict[str, Any]:
-    try:
-        matching_urls = await discover_artifacts(
-            request.remote, request.include_pattern
-        )
-
-        if not matching_urls:
-            return {
-                "message": "No matching artifacts found",
-                "cached_count": 0,
-                "artifacts": [],
-            }
-
-        cached_artifacts = []
-
-        for url in matching_urls:
-            result = await cache_single_artifact(url, "", "")
-            cached_artifacts.append(result)
-
-        cached_count = sum(
-            1
-            for artifact in cached_artifacts
-            if artifact["status"] in ["cached", "already_cached"]
-        )
-
-        return {
-            "message": f"Processed {len(matching_urls)} artifacts, {cached_count} successfully cached",
-            "cached_count": cached_count,
-            "artifacts": cached_artifacts,
-        }
-
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
+async def cache_artifact(request: ArtifactRequest):
+    return await discovery.cache_artifacts(request.remote, request.include_pattern, storage)
 
 
 @app.get("/api/v1/artifacts/{remote:path}")
-async def list_cached_artifacts(
-    remote: str, include_pattern: str = ".*"
-) -> Dict[str, Any]:
-    try:
-        matching_urls = await discover_artifacts(remote, include_pattern)
-
-        cached_artifacts = []
-        for url in matching_urls:
-            key = storage.get_object_key(url)
-            if storage.exists(key):
-                cached_artifacts.append(
-                    {"url": url, "cached_url": storage.get_url(key), "key": key}
-                )
-
-        return {
-            "remote": remote,
-            "pattern": include_pattern,
-            "total_found": len(matching_urls),
-            "cached_count": len(cached_artifacts),
-            "artifacts": cached_artifacts,
-        }
-
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.get("/metrics")
-def get_metrics(
-    json: Optional[bool] = Query(
-        False, description="Return JSON format instead of Prometheus"
-    ),
-):
-    """Get comprehensive metrics about the artifact storage system"""
-    config._check_reload()
-
-    if json:
-        # Return JSON format
-        return metrics.get_metrics(storage, config)
-    else:
-        # Return Prometheus format
-        metrics.get_metrics(storage, config)  # Update gauges
-        prometheus_data = generate_latest().decode("utf-8")
-        return PlainTextResponse(prometheus_data, media_type=CONTENT_TYPE_LATEST)
-
-
-@app.get("/config")
-def get_config():
-    return config.config
+async def list_cached_artifacts(remote: str, include_pattern: str = ".*"):
+    return await discovery.list_artifacts(remote, include_pattern, storage)
 
 
 def main():

src/artifactapi/metrics.py

@@ -1,22 +1,14 @@
 from datetime import datetime
-from typing import Dict, Any
+from typing import Any
+
 from prometheus_client import Counter, Gauge
 
-
 # Prometheus metrics
-request_counter = Counter(
-    "artifact_requests_total", "Total artifact requests", ["remote", "status"]
-)
+request_counter = Counter("artifact_requests_total", "Total artifact requests", ["remote", "status"])
 cache_hit_counter = Counter("artifact_cache_hits_total", "Total cache hits", ["remote"])
-cache_miss_counter = Counter(
-    "artifact_cache_misses_total", "Total cache misses", ["remote"]
-)
-bandwidth_saved_counter = Counter(
-    "artifact_bandwidth_saved_bytes_total", "Total bandwidth saved", ["remote"]
-)
-storage_size_gauge = Gauge(
-    "artifact_storage_size_bytes", "Storage size by remote", ["remote"]
-)
+cache_miss_counter = Counter("artifact_cache_misses_total", "Total cache misses", ["remote"])
+bandwidth_saved_counter = Counter("artifact_bandwidth_saved_bytes_total", "Total bandwidth saved", ["remote"])
+storage_size_gauge = Gauge("artifact_storage_size_bytes", "Storage size by remote", ["remote"])
 redis_keys_gauge = Gauge("artifact_redis_keys_total", "Total Redis keys")
 
 
@@ -44,9 +36,7 @@ class MetricsManager:
                 # Increment per-remote counters
                 self.redis_client.client.incr(f"metrics:cache_hits:{remote_name}")
                 self.redis_client.client.incr(f"metrics:total_requests:{remote_name}")
-                self.redis_client.client.incrby(
-                    f"metrics:bandwidth_saved:{remote_name}", size_bytes
-                )
+                self.redis_client.client.incrby(f"metrics:bandwidth_saved:{remote_name}", size_bytes)
             except Exception:
                 pass
 
@@ -91,7 +81,7 @@ class MetricsManager:
         except Exception:
             return 0
 
-    def get_s3_size_by_remote(self, storage, config_manager) -> Dict[str, int]:
+    def get_s3_size_by_remote(self, storage, config_manager) -> dict[str, int]:
         """Get size of stored data per remote using database mappings"""
         if self.database_manager and self.database_manager.available:
             # Get from database if available
@@ -146,7 +136,7 @@ class MetricsManager:
         except Exception:
             return {}
 
-    def get_metrics(self, storage, config_manager) -> Dict[str, Any]:
+    def get_metrics(self, storage, config_manager) -> dict[str, Any]:
         """Get comprehensive metrics"""
         # Update Redis keys gauge
         redis_key_count = self.get_redis_key_count()
@@ -173,54 +163,31 @@ class MetricsManager:
         if self.redis_client and self.redis_client.available:
             try:
                 # Get global metrics
-                cache_hits = int(
-                    self.redis_client.client.get("metrics:cache_hits") or 0
-                )
-                cache_misses = int(
-                    self.redis_client.client.get("metrics:cache_misses") or 0
-                )
+                cache_hits = int(self.redis_client.client.get("metrics:cache_hits") or 0)
+                cache_misses = int(self.redis_client.client.get("metrics:cache_misses") or 0)
                 total_requests = cache_hits + cache_misses
-                bandwidth_saved = int(
-                    self.redis_client.client.get("metrics:bandwidth_saved") or 0
-                )
+                bandwidth_saved = int(self.redis_client.client.get("metrics:bandwidth_saved") or 0)
 
                 metrics["requests"]["cache_hits"] = cache_hits
                 metrics["requests"]["cache_misses"] = cache_misses
                 metrics["requests"]["total_requests"] = total_requests
-                metrics["requests"]["cache_hit_ratio"] = (
-                    cache_hits / total_requests if total_requests > 0 else 0.0
-                )
+                metrics["requests"]["cache_hit_ratio"] = cache_hits / total_requests if total_requests > 0 else 0.0
                 metrics["bandwidth"]["saved_bytes"] = bandwidth_saved
 
                 # Get per-remote metrics
                 for remote in config_manager.config.get("remotes", {}).keys():
-                    remote_cache_hits = int(
-                        self.redis_client.client.get(f"metrics:cache_hits:{remote}")
-                        or 0
-                    )
-                    remote_cache_misses = int(
-                        self.redis_client.client.get(f"metrics:cache_misses:{remote}")
-                        or 0
-                    )
+                    remote_cache_hits = int(self.redis_client.client.get(f"metrics:cache_hits:{remote}") or 0)
+                    remote_cache_misses = int(self.redis_client.client.get(f"metrics:cache_misses:{remote}") or 0)
                     remote_total = remote_cache_hits + remote_cache_misses
-                    remote_bandwidth_saved = int(
-                        self.redis_client.client.get(
-                            f"metrics:bandwidth_saved:{remote}"
-                        )
-                        or 0
-                    )
+                    remote_bandwidth_saved = int(self.redis_client.client.get(f"metrics:bandwidth_saved:{remote}") or 0)
 
                     metrics["per_remote"][remote] = {
                         "cache_hits": remote_cache_hits,
                         "cache_misses": remote_cache_misses,
                         "total_requests": remote_total,
-                        "cache_hit_ratio": remote_cache_hits / remote_total
-                        if remote_total > 0
-                        else 0.0,
+                        "cache_hit_ratio": remote_cache_hits / remote_total if remote_total > 0 else 0.0,
                         "bandwidth_saved_bytes": remote_bandwidth_saved,
-                        "storage_size_bytes": metrics["storage"]["size_by_remote"].get(
-                            remote, 0
-                        ),
+                        "storage_size_bytes": metrics["storage"]["size_by_remote"].get(remote, 0),
                     }
 
             except Exception:

src/artifactapi/remote/__init__.py

@@ -0,0 +1,4 @@
+from . import generic, helm, npm, python, rpm
+from .base import get_content_type
+
+__all__ = ["generic", "helm", "npm", "python", "rpm", "get_content_type"]

src/artifactapi/remote/base.py

@@ -0,0 +1,16 @@
+def get_content_type(filename: str) -> str:
+    if filename.endswith((".tar.gz", ".tgz")):
+        return "application/gzip"
+    if filename.endswith(".zip") or filename.endswith(".whl"):
+        return "application/zip"
+    if filename.endswith(".exe"):
+        return "application/x-msdownload"
+    if filename.endswith(".rpm"):
+        return "application/x-rpm"
+    if filename.endswith(".xml"):
+        return "application/xml"
+    if filename.endswith((".xml.gz", ".xml.bz2", ".xml.xz")):
+        return "application/gzip"
+    if filename.endswith((".yaml", ".yml")):
+        return "text/yaml"
+    return "application/octet-stream"

src/artifactapi/remote/generic.py

@@ -0,0 +1,3 @@
+from .base import get_content_type
+
+__all__ = ["get_content_type"]

src/artifactapi/remote/helm.py

@@ -0,0 +1,18 @@
+from .base import get_content_type
+
+
+def resolve_content(
+    data: bytes,
+    path: str,
+    filename: str,
+    base_url: str,
+    proxy_url: str,
+    remote_name: str,
+) -> tuple[bytes, str]:
+    if filename == "index.yaml":
+        data = data.replace(
+            base_url.encode(),
+            f"{proxy_url}/api/v1/remote/{remote_name}".encode(),
+        )
+        return data, "text/yaml"
+    return data, get_content_type(filename)

src/artifactapi/remote/npm.py

@@ -0,0 +1,21 @@
+import re
+
+from .base import get_content_type
+
+
+def resolve_content(
+    data: bytes,
+    path: str,
+    filename: str,
+    immutable_patterns: list[str],
+    base_url: str,
+    proxy_url: str,
+    remote_name: str,
+) -> tuple[bytes, str]:
+    if not any(re.search(p, path) for p in immutable_patterns):
+        data = data.replace(
+            base_url.encode(),
+            f"{proxy_url}/api/v1/remote/{remote_name}".encode(),
+        )
+        return data, "application/json"
+    return data, get_content_type(filename)

src/artifactapi/remote/python.py

@@ -0,0 +1,32 @@
+import re
+
+from .base import get_content_type
+
+
+def construct_url(base_url: str, path: str) -> str:
+    """Build the upstream URL for a PyPI request.
+
+    PyPI splits simple/ index pages (pypi.org) from file downloads
+    (files.pythonhosted.org), so simple/ requests are redirected to pypi.org.
+    """
+    if base_url.rstrip("/") == "https://files.pythonhosted.org" and "simple/" in path:
+        return f"https://pypi.org/{path}"
+    return f"{base_url}/{path}"
+
+
+def resolve_content(
+    data: bytes,
+    path: str,
+    filename: str,
+    immutable_patterns: list[str],
+    base_url: str,
+    proxy_url: str,
+    remote_name: str,
+) -> tuple[bytes, str]:
+    if not any(re.search(p, path) for p in immutable_patterns):
+        data = data.replace(
+            base_url.encode(),
+            f"{proxy_url}/api/v1/remote/{remote_name}".encode(),
+        )
+        return data, "text/html; charset=utf-8"
+    return data, get_content_type(filename)

src/artifactapi/remote/rpm.py

@@ -0,0 +1,3 @@
+from .base import get_content_type
+
+__all__ = ["get_content_type"]

src/artifactapi/storage/__init__.py

@@ -0,0 +1,3 @@
+from .s3 import S3Storage
+
+__all__ = ["S3Storage"]

src/artifactapi/storage/s3.py

@@ -1,6 +1,6 @@
-import os
 import hashlib
-from urllib.parse import urlparse
+import os
+
 import boto3
 from botocore.config import Config
 from botocore.exceptions import ClientError
@@ -22,27 +22,25 @@ class S3Storage:
         self.bucket = bucket
         self.secure = secure
 
-        ca_bundle = os.environ.get('REQUESTS_CA_BUNDLE') or os.environ.get('SSL_CERT_FILE')
-        config_kwargs = {
-            "request_checksum_calculation": "when_required",
-            "response_checksum_validation": "when_required"
-        }
+        ca_bundle = os.environ.get("REQUESTS_CA_BUNDLE") or os.environ.get("SSL_CERT_FILE")
+        config_kwargs = {"request_checksum_calculation": "when_required", "response_checksum_validation": "when_required"}
         client_kwargs = {
             "endpoint_url": f"http{'s' if self.secure else ''}://{self.endpoint}",
             "aws_access_key_id": self.access_key,
             "aws_secret_access_key": self.secret_key,
-            "config": Config(**config_kwargs)
+            "config": Config(**config_kwargs),
         }
 
         if ca_bundle and os.path.exists(ca_bundle):
             client_kwargs["verify"] = ca_bundle
             print(f"Debug: Using CA bundle: {ca_bundle}")
         else:
-            print(f"Debug: No CA bundle found. REQUESTS_CA_BUNDLE={os.environ.get('REQUESTS_CA_BUNDLE')}, SSL_CERT_FILE={os.environ.get('SSL_CERT_FILE')}")
+            print(
+                f"Debug: No CA bundle found. REQUESTS_CA_BUNDLE={os.environ.get('REQUESTS_CA_BUNDLE')}, SSL_CERT_FILE={os.environ.get('SSL_CERT_FILE')}"
+            )
 
         self.client = boto3.client("s3", **client_kwargs)
 
-        # Try to ensure bucket exists, but don't fail if MinIO isn't ready yet
         try:
             self._ensure_bucket_exists()
         except Exception as e:
@@ -55,17 +53,23 @@ class S3Storage:
         except ClientError:
             self.client.create_bucket(Bucket=self.bucket)
 
-    def get_object_key(self, url: str) -> str:
-        url_hash = hashlib.sha256(url.encode()).hexdigest()[:16]
-        parsed = urlparse(url)
-        filename = os.path.basename(parsed.path)
-        return f"{parsed.netloc}/{url_hash}/{filename}"
+    def get_object_key(self, remote_name: str, path: str) -> str:
+        clean_path = path.lstrip("/")
+        filename = os.path.basename(clean_path)
+        directory_path = os.path.dirname(clean_path)
 
-    def get_object_key_from_path(self, remote_name: str, path: str) -> str:
-        # Create a key based on the API path for direct access
-        path_hash = hashlib.sha256(path.encode()).hexdigest()[:16]
-        filename = os.path.basename(path)
-        return f"{remote_name}/{path_hash}/{filename}"
+        # Docker blobs are keyed by digest for deduplication across images
+        if "/blobs/sha256:" in clean_path:
+            parts = clean_path.split("/blobs/sha256:")
+            if len(parts) == 2:
+                digest = parts[1]
+                return f"{remote_name}/blobs/sha256/{digest}"
+
+        if directory_path:
+            path_hash = hashlib.sha256(directory_path.encode()).hexdigest()[:16]
+            return f"{remote_name}/{path_hash}/{filename}"
+        else:
+            return f"{remote_name}/{filename}"
 
     def exists(self, key: str) -> bool:
         try:

tests/conftest.py

@@ -0,0 +1,205 @@
+"""
+Pytest configuration and shared fixtures.
+
+Module-level setup (env vars + connection patches) runs before any test
+module is imported, so the FastAPI app initialises against mocks rather
+than real S3 / Redis / PostgreSQL services.
+"""
+
+import os
+import tempfile
+from unittest.mock import MagicMock, patch
+
+import yaml
+
+# ---------------------------------------------------------------------------
+# Test remote configuration
+# ---------------------------------------------------------------------------
+
+TEST_REMOTES = {
+    "remotes": {
+        "alpine-test": {
+            "base_url": "https://dl-cdn.alpinelinux.org",
+            "type": "remote",
+            "package": "alpine",
+            "immutable_patterns": [".*/x86_64/.*\\.apk$"],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 3600},
+        },
+        "rpm-test": {
+            "base_url": "https://example.com/rpm",
+            "type": "remote",
+            "package": "rpm",
+            "immutable_patterns": [".*/x86_64/.*\\.rpm$", ".*/repodata/.*$"],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 3600},
+        },
+        "docker-test": {
+            "base_url": "https://registry.example.com",
+            "type": "remote",
+            "package": "docker",
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 300},
+        },
+        "docker-restricted": {
+            "base_url": "https://registry.example.com",
+            "type": "remote",
+            "package": "docker",
+            "immutable_patterns": ["^library/nginx"],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 300},
+        },
+        "generic-test": {
+            "base_url": "https://releases.example.com",
+            "type": "remote",
+            "package": "generic",
+            "immutable_patterns": [".*\\.tar\\.gz$"],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 0},
+        },
+        "custom-index-test": {
+            "base_url": "https://example.com",
+            "type": "remote",
+            "package": "generic",
+            "mutable_patterns": ["metadata\\.json$"],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 600},
+        },
+        "check-mutable-test": {
+            "base_url": "https://example.com",
+            "type": "remote",
+            "package": "generic",
+            "mutable_patterns": ["metadata\\.json$"],
+            "check_mutable_updates": True,
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 600},
+        },
+        "local-test": {
+            "type": "local",
+            "package": "generic",
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 0},
+        },
+        "pypi-test": {
+            "base_url": "https://files.pythonhosted.org",
+            "type": "remote",
+            "package": "pypi",
+            "immutable_patterns": [
+                r"packages/.*\.whl$",
+                r"packages/.*\.whl\.metadata$",
+                r"packages/.*\.tar\.gz$",
+            ],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 600},
+        },
+        "npm-test": {
+            "base_url": "https://registry.npmjs.org",
+            "type": "remote",
+            "package": "npm",
+            "immutable_patterns": [r"\.tgz$"],
+            "mutable_patterns": [r"^(?!.*\.tgz$).*"],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 600},
+        },
+        "helm-test": {
+            "base_url": "https://helm.releases.hashicorp.com",
+            "type": "remote",
+            "package": "helm",
+            "immutable_patterns": [r"\.tgz$"],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 3600},
+        },
+        "quarantine-test": {
+            "base_url": "https://releases.example.com",
+            "type": "remote",
+            "package": "generic",
+            "immutable_patterns": [r".*\.tar\.gz$"],
+            "quarantine_new": True,
+            "quarantine_days": 3,
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 0},
+        },
+        "quarantine-disabled": {
+            "base_url": "https://releases.example.com",
+            "type": "remote",
+            "package": "generic",
+            "immutable_patterns": [r".*\.tar\.gz$"],
+            "quarantine_new": False,
+            "quarantine_days": 3,
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 0},
+        },
+        "helm-member-2": {
+            "base_url": "https://charts.example.com",
+            "type": "remote",
+            "package": "helm",
+            "immutable_patterns": [r"\.tgz$"],
+            "cache": {"immutable_ttl": 0, "mutable_ttl": 1800},
+        },
+        "helm-virtual-test": {
+            "type": "virtual",
+            "package": "helm",
+            "members": ["helm-test", "helm-member-2"],
+        },
+        "unsupported-virtual-test": {
+            "type": "virtual",
+            "package": "rpm",
+            "members": ["rpm-test"],
+        },
+        "empty-virtual-test": {
+            "type": "virtual",
+            "package": "helm",
+            "members": [],
+        },
+    }
+}
+
+# ---------------------------------------------------------------------------
+# Write temp config and set env vars BEFORE importing the package
+# ---------------------------------------------------------------------------
+
+_tmpdir = tempfile.mkdtemp()
+_config_path = os.path.join(_tmpdir, "remotes.yaml")
+with open(_config_path, "w") as _f:
+    yaml.dump(TEST_REMOTES, _f)
+
+os.environ.update(
+    {
+        "CONFIG_PATH": _config_path,
+        "MINIO_ENDPOINT": "localhost:9000",
+        "MINIO_ACCESS_KEY": "testkey",
+        "MINIO_SECRET_KEY": "testsecret",
+        "MINIO_BUCKET": "testbucket",
+        "REDIS_URL": "redis://localhost:6379/0",
+        "DBHOST": "localhost",
+        "DBPORT": "5432",
+        "DBUSER": "test",
+        "DBPASS": "test",
+        "DBNAME": "test",
+    }
+)
+
+# Patch external service connections before the package is imported.
+# These stay active for the whole session (process exits after tests finish).
+_boto3_patch = patch("boto3.client", return_value=MagicMock())
+_redis_patch = patch("redis.from_url", return_value=MagicMock())
+_psycopg2_patch = patch("psycopg2.connect", return_value=MagicMock())
+_boto3_patch.start()
+_redis_patch.start()
+_psycopg2_patch.start()
+
+# ---------------------------------------------------------------------------
+# Shared fixtures
+# ---------------------------------------------------------------------------
+
+import pytest  # noqa: E402
+from fastapi.testclient import TestClient  # noqa: E402
+
+
+@pytest.fixture(scope="session")
+def app():
+    from artifactapi.main import app as fastapi_app
+
+    return fastapi_app
+
+
+@pytest.fixture(scope="session")
+def client(app):
+    return TestClient(app)
+
+
+@pytest.fixture
+def config_path():
+    return _config_path
+
+
+@pytest.fixture
+def test_remotes():
+    return TEST_REMOTES

tests/test_cache.py

@@ -0,0 +1,329 @@
+"""Tests for RedisCache, focusing on is_mutable_file with configurable patterns."""
+
+import hashlib
+from unittest.mock import ANY, MagicMock, patch
+
+import pytest
+
+from artifactapi.cache import RedisCache
+from artifactapi.config import _PACKAGE_MUTABLE_PATTERNS
+
+
+@pytest.fixture
+def bare_cache():
+    """RedisCache instance bypassing __init__ (no Redis needed for pure-logic tests)."""
+    return RedisCache.__new__(RedisCache)
+
+
+@pytest.fixture
+def unavailable_cache():
+    """RedisCache where Redis is not reachable."""
+    with patch("redis.from_url", side_effect=Exception("connection refused")):
+        return RedisCache("redis://localhost:6379/0")
+
+
+@pytest.fixture
+def mock_redis_client():
+    return MagicMock()
+
+
+@pytest.fixture
+def cache_with_redis(mock_redis_client):
+    """RedisCache backed by a MagicMock Redis client."""
+    with patch("redis.from_url", return_value=mock_redis_client):
+        c = RedisCache("redis://localhost:6379/0")
+    c.client = mock_redis_client
+    c.available = True
+    return c
+
+
+# ---------------------------------------------------------------------------
+# is_mutable_file — alpine patterns
+# ---------------------------------------------------------------------------
+
+
+class TestIsMutableFileAlpine:
+    def test_apkindex_tarball_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["alpine"]
+        assert bare_cache.is_mutable_file("alpine/v3.18/x86_64/APKINDEX.tar.gz", patterns)
+
+    def test_nested_apkindex_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["alpine"]
+        assert bare_cache.is_mutable_file("mirrors/dl-cdn/alpine/v3.19/community/x86_64/APKINDEX.tar.gz", patterns)
+
+    def test_apk_package_is_not_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["alpine"]
+        assert not bare_cache.is_mutable_file("alpine/v3.18/x86_64/musl-1.2.4-r2.apk", patterns)
+
+    def test_random_tarball_is_not_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["alpine"]
+        assert not bare_cache.is_mutable_file("some/path/archive.tar.gz", patterns)
+
+    def test_apkindex_signature_file_is_not_index(self, bare_cache):
+        # Signature file adjacent to the index should not be treated as an index
+        patterns = _PACKAGE_MUTABLE_PATTERNS["alpine"]
+        assert not bare_cache.is_mutable_file("alpine/v3.18/x86_64/APKINDEX.tar.gz.sig", patterns)
+
+    def test_apkindex_tmp_file_is_not_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["alpine"]
+        assert not bare_cache.is_mutable_file("alpine/v3.18/x86_64/APKINDEX.tar.gz.tmp", patterns)
+
+
+# ---------------------------------------------------------------------------
+# is_mutable_file — rpm patterns
+# ---------------------------------------------------------------------------
+
+
+class TestIsMutableFileRpm:
+    def test_repomd_xml_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["rpm"]
+        assert bare_cache.is_mutable_file("almalinux/9/x86_64/repomd.xml", patterns)
+
+    def test_repodata_primary_xml_gz_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["rpm"]
+        assert bare_cache.is_mutable_file("repo/repodata/primary.xml.gz", patterns)
+
+    def test_repodata_sqlite_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["rpm"]
+        assert bare_cache.is_mutable_file("repo/repodata/primary.sqlite", patterns)
+
+    def test_repodata_sqlite_bz2_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["rpm"]
+        assert bare_cache.is_mutable_file("repo/repodata/other.sqlite.bz2", patterns)
+
+    def test_repodata_yaml_xz_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["rpm"]
+        assert bare_cache.is_mutable_file("repo/repodata/comps.yaml.xz", patterns)
+
+    def test_packages_gz_pattern_matches_any_path(self, bare_cache):
+        # The Packages.gz$ regex is a carryover from the original hardcoded logic and
+        # deliberately matches any path ending in Packages.gz — including Debian-style paths.
+        # This test documents that intentional behaviour.
+        patterns = _PACKAGE_MUTABLE_PATTERNS["rpm"]
+        assert bare_cache.is_mutable_file("debian/dists/stable/main/binary-amd64/Packages.gz", patterns)
+
+    def test_rpm_package_is_not_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["rpm"]
+        assert not bare_cache.is_mutable_file("almalinux/9/x86_64/Packages/bash-5.1.8.x86_64.rpm", patterns)
+
+    def test_arbitrary_xml_outside_repodata_is_not_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["rpm"]
+        assert not bare_cache.is_mutable_file("some/path/config.xml", patterns)
+
+
+# ---------------------------------------------------------------------------
+# is_mutable_file — docker patterns
+# ---------------------------------------------------------------------------
+
+
+class TestIsMutableFileDocker:
+    def test_tag_manifest_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["docker"]
+        assert bare_cache.is_mutable_file("library/nginx/manifests/latest", patterns)
+
+    def test_version_tag_manifest_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["docker"]
+        assert bare_cache.is_mutable_file("library/nginx/manifests/1.25.3", patterns)
+
+    def test_hyphenated_tag_manifest_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["docker"]
+        assert bare_cache.is_mutable_file("library/nginx/manifests/latest-rc", patterns)
+
+    def test_numeric_date_tag_manifest_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["docker"]
+        assert bare_cache.is_mutable_file("library/nginx/manifests/20240101", patterns)
+
+    def test_digest_manifest_is_not_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["docker"]
+        digest = "sha256:" + "a" * 64
+        assert not bare_cache.is_mutable_file(f"library/nginx/manifests/{digest}", patterns)
+
+    def test_tags_list_is_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["docker"]
+        assert bare_cache.is_mutable_file("library/nginx/tags/list", patterns)
+
+    def test_blob_is_not_index(self, bare_cache):
+        patterns = _PACKAGE_MUTABLE_PATTERNS["docker"]
+        assert not bare_cache.is_mutable_file("library/nginx/blobs/sha256:abc123", patterns)
+
+
+# ---------------------------------------------------------------------------
+# is_mutable_file — edge cases
+# ---------------------------------------------------------------------------
+
+
+class TestIsMutableFileEdgeCases:
+    def test_empty_patterns_nothing_is_index(self, bare_cache):
+        assert not bare_cache.is_mutable_file("APKINDEX.tar.gz", [])
+        assert not bare_cache.is_mutable_file("repomd.xml", [])
+        assert not bare_cache.is_mutable_file("library/nginx/manifests/latest", [])
+
+    def test_none_patterns_nothing_is_index(self, bare_cache):
+        assert not bare_cache.is_mutable_file("APKINDEX.tar.gz", None)
+        assert not bare_cache.is_mutable_file("repomd.xml", None)
+
+    def test_custom_patterns_match(self, bare_cache):
+        patterns = [r"metadata\.json$", r"index\.yaml$"]
+        assert bare_cache.is_mutable_file("repo/metadata.json", patterns)
+        assert bare_cache.is_mutable_file("repo/subdir/index.yaml", patterns)
+        assert not bare_cache.is_mutable_file("repo/data.tar.gz", patterns)
+
+    def test_custom_pattern_does_not_match_standard_index(self, bare_cache):
+        patterns = [r"metadata\.json$"]
+        assert not bare_cache.is_mutable_file("APKINDEX.tar.gz", patterns)
+
+
+# ---------------------------------------------------------------------------
+# get_index_cache_key
+# ---------------------------------------------------------------------------
+
+
+class TestGetIndexCacheKey:
+    def test_key_format_is_deterministic(self, bare_cache):
+        # Assert against a pre-computed value to pin the hash algorithm,
+        # truncation length, and format string in one assertion.
+        path = "alpine/v3.18/x86_64/APKINDEX.tar.gz"
+        expected_hash = hashlib.sha256(path.encode()).hexdigest()[:16]
+        key = bare_cache.get_index_cache_key("alpine-test", path)
+        assert key == f"index:alpine-test:{expected_hash}"
+
+    def test_different_paths_produce_different_keys(self, bare_cache):
+        k1 = bare_cache.get_index_cache_key("alpine-test", "alpine/v3.18/x86_64/APKINDEX.tar.gz")
+        k2 = bare_cache.get_index_cache_key("alpine-test", "alpine/v3.19/x86_64/APKINDEX.tar.gz")
+        assert k1 != k2
+
+    def test_different_remotes_produce_different_keys(self, bare_cache):
+        k1 = bare_cache.get_index_cache_key("remote-a", "path/to/APKINDEX.tar.gz")
+        k2 = bare_cache.get_index_cache_key("remote-b", "path/to/APKINDEX.tar.gz")
+        assert k1 != k2
+
+    def test_key_starts_with_index_prefix_and_remote(self, bare_cache):
+        key = bare_cache.get_index_cache_key("myremote", "some/path")
+        assert key.startswith("index:myremote:")
+
+    def test_key_hash_segment_is_16_chars(self, bare_cache):
+        key = bare_cache.get_index_cache_key("myremote", "some/path/file.xml")
+        # Format: index:<remote>:<16-char hash> — the fixed length matters for key-space hygiene
+        parts = key.split(":")
+        assert len(parts) == 3
+        assert len(parts[2]) == 16
+
+
+# ---------------------------------------------------------------------------
+# mark_index_cached / is_index_valid
+# ---------------------------------------------------------------------------
+
+
+class TestIndexValidity:
+    def test_mark_index_cached_calls_setex_with_correct_ttl(self, cache_with_redis, mock_redis_client):
+        cache_with_redis.mark_index_cached("remote", "path/APKINDEX.tar.gz", 300)
+        expected_key = cache_with_redis.get_index_cache_key("remote", "path/APKINDEX.tar.gz")
+        mock_redis_client.setex.assert_called_once_with(expected_key, 300, ANY)
+
+    def test_present_key_is_valid(self, cache_with_redis, mock_redis_client):
+        mock_redis_client.exists.return_value = 1
+        assert cache_with_redis.is_index_valid("remote", "path/APKINDEX.tar.gz")
+
+    def test_missing_key_is_not_valid(self, cache_with_redis, mock_redis_client):
+        mock_redis_client.exists.return_value = 0
+        assert not cache_with_redis.is_index_valid("remote", "path/APKINDEX.tar.gz")
+
+    def test_unavailable_redis_is_not_valid(self, unavailable_cache):
+        assert not unavailable_cache.is_index_valid("remote", "some/path")
+
+    def test_mark_cached_no_op_when_unavailable(self, unavailable_cache):
+        # client is None when Redis is unavailable — setex cannot be called
+        assert unavailable_cache.client is None
+        unavailable_cache.mark_index_cached("remote", "some/path", 300)  # must not raise
+
+
+# ---------------------------------------------------------------------------
+# mutable meta (ETag / Last-Modified storage)
+# ---------------------------------------------------------------------------
+
+
+class TestMutableMeta:
+    def test_meta_key_format(self, bare_cache):
+        path = "repo/metadata.json"
+        expected_hash = hashlib.sha256(path.encode()).hexdigest()[:16]
+        assert bare_cache.get_mutable_meta_key("myremote", path) == f"mutable:meta:myremote:{expected_hash}"
+
+    def test_meta_key_hash_is_16_chars(self, bare_cache):
+        key = bare_cache.get_mutable_meta_key("remote", "some/path/file.json")
+        assert len(key.split(":")[-1]) == 16
+
+    def test_store_and_retrieve_etag(self, cache_with_redis, mock_redis_client):
+        mock_redis_client.hgetall.return_value = {"etag": '"abc123"'}
+        cache_with_redis.store_mutable_meta("remote", "path/meta.json", '"abc123"', None)
+        mock_redis_client.hset.assert_called_once()
+        meta = cache_with_redis.get_mutable_meta("remote", "path/meta.json")
+        assert meta["etag"] == '"abc123"'
+
+    def test_store_and_retrieve_last_modified(self, cache_with_redis, mock_redis_client):
+        lm = "Mon, 01 Jan 2024 00:00:00 GMT"
+        mock_redis_client.hgetall.return_value = {"last_modified": lm}
+        cache_with_redis.store_mutable_meta("remote", "path/meta.json", None, lm)
+        meta = cache_with_redis.get_mutable_meta("remote", "path/meta.json")
+        assert meta["last_modified"] == lm
+
+    def test_store_no_op_when_both_none(self, cache_with_redis, mock_redis_client):
+        cache_with_redis.store_mutable_meta("remote", "path/meta.json", None, None)
+        mock_redis_client.hset.assert_not_called()
+
+    def test_store_no_op_when_unavailable(self, unavailable_cache):
+        unavailable_cache.store_mutable_meta("remote", "path", "etag", None)  # must not raise
+
+    def test_get_returns_empty_when_unavailable(self, unavailable_cache):
+        assert unavailable_cache.get_mutable_meta("remote", "path") == {}
+
+    def test_delete_removes_meta_key(self, cache_with_redis, mock_redis_client):
+        expected_key = cache_with_redis.get_mutable_meta_key("remote", "path/meta.json")
+        cache_with_redis.delete_mutable_meta("remote", "path/meta.json")
+        mock_redis_client.delete.assert_called_once_with(expected_key)
+
+    def test_delete_no_op_when_unavailable(self, unavailable_cache):
+        unavailable_cache.delete_mutable_meta("remote", "path")  # must not raise
+
+
+# ---------------------------------------------------------------------------
+# artifact published date (quarantine support)
+# ---------------------------------------------------------------------------
+
+
+class TestArtifactPublished:
+    def test_key_format_is_deterministic(self, bare_cache):
+        path = "some/path/package-1.0.tar.gz"
+        expected_hash = hashlib.sha256(path.encode()).hexdigest()[:16]
+        assert bare_cache.get_artifact_published_key("myremote", path) == f"pkg:published:myremote:{expected_hash}"
+
+    def test_key_hash_is_16_chars(self, bare_cache):
+        key = bare_cache.get_artifact_published_key("remote", "path/to/file.whl")
+        assert len(key.split(":")[-1]) == 16
+
+    def test_different_paths_produce_different_keys(self, bare_cache):
+        k1 = bare_cache.get_artifact_published_key("remote", "pkg-1.0.tar.gz")
+        k2 = bare_cache.get_artifact_published_key("remote", "pkg-2.0.tar.gz")
+        assert k1 != k2
+
+    def test_store_calls_set_with_correct_value(self, cache_with_redis, mock_redis_client):
+        lm = "Mon, 01 Jan 2024 00:00:00 GMT"
+        cache_with_redis.store_artifact_published("remote", "path/pkg.tar.gz", lm)
+        expected_key = cache_with_redis.get_artifact_published_key("remote", "path/pkg.tar.gz")
+        mock_redis_client.set.assert_called_once_with(expected_key, lm)
+
+    def test_get_returns_stored_value(self, cache_with_redis, mock_redis_client):
+        lm = "Tue, 15 Mar 2022 12:00:00 GMT"
+        mock_redis_client.get.return_value = lm
+        result = cache_with_redis.get_artifact_published("remote", "path/pkg.tar.gz")
+        assert result == lm
+
+    def test_get_returns_none_when_not_stored(self, cache_with_redis, mock_redis_client):
+        mock_redis_client.get.return_value = None
+        result = cache_with_redis.get_artifact_published("remote", "path/pkg.tar.gz")
+        assert result is None
+
+    def test_store_no_op_when_unavailable(self, unavailable_cache):
+        unavailable_cache.store_artifact_published("remote", "path", "Mon, 01 Jan 2024 00:00:00 GMT")
+
+    def test_get_returns_none_when_unavailable(self, unavailable_cache):
+        assert unavailable_cache.get_artifact_published("remote", "path") is None

tests/test_config.py

@@ -0,0 +1,598 @@
+"""Tests for ConfigManager, focusing on get_mutable_patterns and get_immutable_patterns."""
+
+import os
+
+import pytest
+import yaml
+
+from artifactapi.config import ConfigManager
+
+
+@pytest.fixture
+def make_config(tmp_path):
+    """Factory: write a remote dict to a temp YAML and return a ConfigManager."""
+
+    def _make(remotes_dict):
+        cfg_file = tmp_path / "remotes.yaml"
+        cfg_file.write_text(yaml.dump({"remote": remotes_dict}))
+        return ConfigManager(str(cfg_file))
+
+    return _make
+
+
+# ---------------------------------------------------------------------------
+# get_mutable_patterns
+# ---------------------------------------------------------------------------
+
+
+class TestGetMutablePatterns:
+    def test_alpine_returns_package_defaults(self, make_config):
+        cfg = make_config({"r": {"package": "alpine", "base_url": "https://x.com"}})
+        patterns = cfg.get_mutable_patterns("r")
+        assert r"APKINDEX\.tar\.gz$" in patterns
+
+    def test_rpm_returns_package_defaults(self, make_config):
+        cfg = make_config({"r": {"package": "rpm", "base_url": "https://x.com"}})
+        patterns = cfg.get_mutable_patterns("r")
+        assert r"repomd\.xml$" in patterns
+        assert any("repodata" in p for p in patterns)
+
+    def test_docker_returns_package_defaults(self, make_config):
+        cfg = make_config({"r": {"package": "docker", "base_url": "https://x.com"}})
+        patterns = cfg.get_mutable_patterns("r")
+        assert any("manifests" in p for p in patterns)
+        assert any("tags/list" in p for p in patterns)
+
+    def test_generic_returns_empty_list(self, make_config):
+        cfg = make_config({"r": {"package": "generic", "base_url": "https://x.com"}})
+        assert cfg.get_mutable_patterns("r") == []
+
+    def test_unknown_remote_returns_empty_list(self, make_config):
+        cfg = make_config({})
+        assert cfg.get_mutable_patterns("nonexistent") == []
+
+    def test_missing_package_field_defaults_to_generic(self, make_config):
+        cfg = make_config({"r": {"base_url": "https://x.com"}})
+        assert cfg.get_mutable_patterns("r") == []
+
+    def test_unknown_package_type_returns_empty_list(self, make_config):
+        # A mis-spelled package type silently returns [] — this is a known footgun
+        cfg = make_config({"r": {"package": "deb", "base_url": "https://x.com"}})
+        assert cfg.get_mutable_patterns("r") == []
+
+    def test_extra_patterns_appended_after_defaults(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "alpine",
+                    "base_url": "https://x.com",
+                    "mutable_patterns": [r"custom\.json$"],
+                }
+            }
+        )
+        patterns = cfg.get_mutable_patterns("r")
+        assert r"APKINDEX\.tar\.gz$" in patterns
+        assert r"custom\.json$" in patterns
+        # Defaults come first
+        assert patterns.index(r"APKINDEX\.tar\.gz$") < patterns.index(r"custom\.json$")
+
+    def test_explicit_empty_extra_patterns_returns_defaults(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "alpine",
+                    "base_url": "https://x.com",
+                    "mutable_patterns": [],
+                }
+            }
+        )
+        assert r"APKINDEX\.tar\.gz$" in cfg.get_mutable_patterns("r")
+
+    def test_duplicate_extra_pattern_not_added_twice(self, make_config):
+        existing = r"APKINDEX\.tar\.gz$"
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "alpine",
+                    "base_url": "https://x.com",
+                    "mutable_patterns": [existing],
+                }
+            }
+        )
+        patterns = cfg.get_mutable_patterns("r")
+        assert patterns.count(existing) == 1
+
+    def test_generic_with_only_extra_patterns(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "generic",
+                    "base_url": "https://x.com",
+                    "mutable_patterns": [r"meta\.json$", r"index\.yaml$"],
+                }
+            }
+        )
+        assert cfg.get_mutable_patterns("r") == [r"meta\.json$", r"index\.yaml$"]
+
+    def test_rpm_extra_patterns_merged(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "rpm",
+                    "base_url": "https://x.com",
+                    "mutable_patterns": [r"custom-meta\.xml$"],
+                }
+            }
+        )
+        patterns = cfg.get_mutable_patterns("r")
+        assert r"repomd\.xml$" in patterns
+        assert r"custom-meta\.xml$" in patterns
+
+    def test_npm_has_no_package_defaults(self, make_config):
+        cfg = make_config({"r": {"package": "npm", "base_url": "https://x.com"}})
+        assert cfg.get_mutable_patterns("r") == []
+
+    def test_npm_explicit_mutable_pattern_matches_metadata(self, make_config):
+        import re
+
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "npm",
+                    "base_url": "https://x.com",
+                    "mutable_patterns": [r"^(?!.*\.tgz$).*"],
+                }
+            }
+        )
+        patterns = cfg.get_mutable_patterns("r")
+        assert any(re.search(p, "express") for p in patterns)
+        assert any(re.search(p, "@babel/core") for p in patterns)
+
+    def test_helm_returns_index_yaml_as_mutable(self, make_config):
+        cfg = make_config({"r": {"package": "helm", "base_url": "https://helm.example.com"}})
+        patterns = cfg.get_mutable_patterns("r")
+        assert r"index\.yaml$" in patterns
+
+    def test_helm_chart_tarballs_not_mutable_by_default(self, make_config):
+        import re
+
+        cfg = make_config({"r": {"package": "helm", "base_url": "https://helm.example.com"}})
+        patterns = cfg.get_mutable_patterns("r")
+        # Only index.yaml is mutable; .tgz chart tarballs are not
+        assert not any(re.search(p, "vault-0.29.1.tgz") for p in patterns)
+        assert not any(re.search(p, "consul-1.5.0.tgz") for p in patterns)
+
+    def test_npm_explicit_mutable_pattern_excludes_tarballs(self, make_config):
+        import re
+
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "npm",
+                    "base_url": "https://x.com",
+                    "mutable_patterns": [r"^(?!.*\.tgz$).*"],
+                }
+            }
+        )
+        patterns = cfg.get_mutable_patterns("r")
+        assert not any(re.search(p, "express-4.18.2.tgz") for p in patterns)
+        assert not any(re.search(p, "express/-/express-4.18.2.tgz") for p in patterns)
+
+
+# ---------------------------------------------------------------------------
+# get_immutable_patterns
+# ---------------------------------------------------------------------------
+
+
+class TestGetImmutablePatterns:
+    def test_returns_immutable_patterns(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "generic",
+                    "base_url": "https://x.com",
+                    "immutable_patterns": [r".*\.tar\.gz$"],
+                }
+            }
+        )
+        assert cfg.get_immutable_patterns("r") == [r".*\.tar\.gz$"]
+
+    def test_returns_empty_for_missing_remote(self, make_config):
+        cfg = make_config({})
+        assert cfg.get_immutable_patterns("nonexistent") == []
+
+    def test_returns_empty_when_no_patterns_configured(self, make_config):
+        cfg = make_config({"r": {"package": "generic", "base_url": "https://x.com"}})
+        assert cfg.get_immutable_patterns("r") == []
+
+    def test_multiple_patterns_returned(self, make_config):
+        patterns = [r".*\.rpm$", r".*/repodata/.*$"]
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "rpm",
+                    "base_url": "https://x.com",
+                    "immutable_patterns": patterns,
+                }
+            }
+        )
+        assert cfg.get_immutable_patterns("r") == patterns
+
+    def test_dict_keyed_repositories_returns_per_repo_patterns(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "generic",
+                    "base_url": "https://x.com",
+                    "immutable_patterns": [r".*\.tar\.gz$"],
+                    "repositories": {
+                        "/path/to/repo": {"immutable_patterns": [r".*\.rpm$"]},
+                    },
+                }
+            }
+        )
+        assert cfg.get_immutable_patterns("r", "/path/to/repo") == [r".*\.rpm$"]
+
+    def test_dict_keyed_repositories_falls_back_to_remote_patterns(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "generic",
+                    "base_url": "https://x.com",
+                    "immutable_patterns": [r".*\.tar\.gz$"],
+                    "repositories": {
+                        "/path/to/repo": {"immutable_patterns": [r".*\.rpm$"]},
+                    },
+                }
+            }
+        )
+        assert cfg.get_immutable_patterns("r", "/unknown/path") == [r".*\.tar\.gz$"]
+
+
+# ---------------------------------------------------------------------------
+# get_user_mutable_patterns
+# ---------------------------------------------------------------------------
+
+
+class TestGetUserMutablePatterns:
+    def test_returns_only_user_patterns(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "alpine",
+                    "base_url": "https://x.com",
+                    "mutable_patterns": [r"custom\.json$"],
+                }
+            }
+        )
+        assert cfg.get_user_mutable_patterns("r") == [r"custom\.json$"]
+
+    def test_excludes_package_defaults(self, make_config):
+        # Package defaults (APKINDEX etc.) must NOT appear here
+        cfg = make_config({"r": {"package": "alpine", "base_url": "https://x.com"}})
+        assert cfg.get_user_mutable_patterns("r") == []
+
+    def test_returns_empty_for_missing_remote(self, make_config):
+        cfg = make_config({})
+        assert cfg.get_user_mutable_patterns("nonexistent") == []
+
+    def test_returns_empty_when_key_absent(self, make_config):
+        cfg = make_config({"r": {"package": "generic", "base_url": "https://x.com"}})
+        assert cfg.get_user_mutable_patterns("r") == []
+
+
+# ---------------------------------------------------------------------------
+# get_cache_config
+# ---------------------------------------------------------------------------
+
+
+class TestGetCacheConfig:
+    def test_returns_cache_section(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "generic",
+                    "base_url": "https://x.com",
+                    "cache": {"immutable_ttl": 0, "mutable_ttl": 7200},
+                }
+            }
+        )
+        assert cfg.get_cache_config("r") == {"immutable_ttl": 0, "mutable_ttl": 7200}
+
+    def test_returns_empty_dict_for_missing_remote(self, make_config):
+        cfg = make_config({})
+        assert cfg.get_cache_config("nonexistent") == {}
+
+    def test_returns_empty_dict_when_no_cache_key(self, make_config):
+        cfg = make_config({"r": {"package": "generic", "base_url": "https://x.com"}})
+        assert cfg.get_cache_config("r") == {}
+
+
+# ---------------------------------------------------------------------------
+# Config file reload
+# ---------------------------------------------------------------------------
+
+
+class TestConfigReload:
+    def test_reloads_when_file_mtime_advances(self, tmp_path):
+        cfg_file = tmp_path / "remotes.yaml"
+        cfg_file.write_text(yaml.dump({"remote": {"repo-a": {"package": "generic", "base_url": "https://x.com"}}}))
+        cfg = ConfigManager(str(cfg_file))
+        assert "repo-a" in cfg.config["remotes"]
+
+        cfg_file.write_text(yaml.dump({"remote": {"repo-b": {"package": "generic", "base_url": "https://y.com"}}}))
+        future_mtime = cfg._last_modified + 1
+        os.utime(str(cfg_file), (future_mtime, future_mtime))
+
+        cfg._check_reload()
+
+        assert "repo-b" in cfg.config["remotes"]
+        assert "repo-a" not in cfg.config["remotes"]
+
+    def test_no_reload_when_file_unchanged(self, tmp_path):
+        cfg_file = tmp_path / "remotes.yaml"
+        cfg_file.write_text(yaml.dump({"remote": {"repo-a": {"package": "generic", "base_url": "https://x.com"}}}))
+        cfg = ConfigManager(str(cfg_file))
+
+        # Call check_reload without touching the file — should not reload
+        cfg._check_reload()
+
+        assert "repo-a" in cfg.config["remotes"]
+
+
+# ---------------------------------------------------------------------------
+# get_quarantine_config
+# ---------------------------------------------------------------------------
+
+
+class TestGetQuarantineConfig:
+    def test_returns_false_zero_when_not_configured(self, make_config):
+        cfg = make_config({"r": {"package": "generic", "base_url": "https://x.com"}})
+        enabled, days = cfg.get_quarantine_config("r")
+        assert enabled is False
+        assert days == 0
+
+    def test_returns_false_zero_for_missing_remote(self, make_config):
+        cfg = make_config({})
+        enabled, days = cfg.get_quarantine_config("nonexistent")
+        assert enabled is False
+        assert days == 0
+
+    def test_enabled_true_and_days_returned(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "generic",
+                    "base_url": "https://x.com",
+                    "quarantine_new": True,
+                    "quarantine_days": 7,
+                }
+            }
+        )
+        enabled, days = cfg.get_quarantine_config("r")
+        assert enabled is True
+        assert days == 7
+
+    def test_quarantine_new_false_returns_disabled(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "generic",
+                    "base_url": "https://x.com",
+                    "quarantine_new": False,
+                    "quarantine_days": 7,
+                }
+            }
+        )
+        enabled, days = cfg.get_quarantine_config("r")
+        assert enabled is False
+        assert days == 7
+
+    def test_enabled_with_zero_days_returns_zero(self, make_config):
+        cfg = make_config(
+            {
+                "r": {
+                    "package": "generic",
+                    "base_url": "https://x.com",
+                    "quarantine_new": True,
+                    "quarantine_days": 0,
+                }
+            }
+        )
+        enabled, days = cfg.get_quarantine_config("r")
+        assert enabled is True
+        assert days == 0
+
+
+# ---------------------------------------------------------------------------
+# Directory mode (CONFIG_PATH points to a directory)
+# ---------------------------------------------------------------------------
+
+
+def _remote(base_url: str = "https://x.com") -> dict:
+    return {"package": "generic", "base_url": base_url}
+
+
+class TestConfigDirMode:
+    def test_loads_all_yaml_files(self, tmp_path):
+        (tmp_path / "a.yaml").write_text(yaml.dump({"remote": {"repo-a": _remote()}}))
+        (tmp_path / "b.yaml").write_text(yaml.dump({"remote": {"repo-b": _remote("https://y.com")}}))
+        cfg = ConfigManager(str(tmp_path))
+        assert "repo-a" in cfg.config["remotes"]
+        assert "repo-b" in cfg.config["remotes"]
+
+    def test_later_file_overrides_earlier_on_same_key(self, tmp_path):
+        (tmp_path / "a.yaml").write_text(yaml.dump({"remote": {"r": _remote("https://first.com")}}))
+        (tmp_path / "b.yaml").write_text(yaml.dump({"remote": {"r": _remote("https://second.com")}}))
+        cfg = ConfigManager(str(tmp_path))
+        assert cfg.config["remotes"]["r"]["base_url"] == "https://second.com"
+
+    def test_empty_directory_returns_empty_remotes(self, tmp_path):
+        cfg = ConfigManager(str(tmp_path))
+        assert cfg.config == {"remotes": {}}
+
+    def test_ignores_non_yaml_files(self, tmp_path):
+        (tmp_path / "notes.txt").write_text("not yaml")
+        (tmp_path / "a.yaml").write_text(yaml.dump({"remote": {"repo-a": _remote()}}))
+        cfg = ConfigManager(str(tmp_path))
+        assert list(cfg.config["remotes"].keys()) == ["repo-a"]
+
+    def test_reload_picks_up_new_file(self, tmp_path):
+        (tmp_path / "a.yaml").write_text(yaml.dump({"remote": {"repo-a": _remote()}}))
+        cfg = ConfigManager(str(tmp_path))
+        assert "repo-a" in cfg.config["remotes"]
+        assert "repo-b" not in cfg.config["remotes"]
+
+        new_file = tmp_path / "b.yaml"
+        new_file.write_text(yaml.dump({"remote": {"repo-b": _remote("https://y.com")}}))
+        future_mtime = cfg._last_modified + 1
+        os.utime(str(new_file), (future_mtime, future_mtime))
+
+        cfg._check_reload()
+
+        assert "repo-a" in cfg.config["remotes"]
+        assert "repo-b" in cfg.config["remotes"]
+
+
+# ---------------------------------------------------------------------------
+# config_dir key (main file contains a config_dir pointer)
+# ---------------------------------------------------------------------------
+
+
+class TestConfigDirKey:
+    def test_merges_remotes_from_config_dir(self, tmp_path):
+        conf_d = tmp_path / "conf.d"
+        conf_d.mkdir()
+        (conf_d / "remotes.yaml").write_text(yaml.dump({"remote": {"repo-extra": _remote("https://extra.com")}}))
+        main = tmp_path / "config.yaml"
+        main.write_text(yaml.dump({"config_dir": str(conf_d), "remote": {"repo-main": _remote()}}))
+        cfg = ConfigManager(str(main))
+        assert "repo-main" in cfg.config["remotes"]
+        assert "repo-extra" in cfg.config["remotes"]
+
+    def test_relative_config_dir_resolved_from_main_file(self, tmp_path):
+        conf_d = tmp_path / "conf.d"
+        conf_d.mkdir()
+        (conf_d / "r.yaml").write_text(yaml.dump({"remote": {"repo-a": _remote()}}))
+        main = tmp_path / "config.yaml"
+        main.write_text(yaml.dump({"config_dir": "conf.d"}))
+        cfg = ConfigManager(str(main))
+        assert "repo-a" in cfg.config["remotes"]
+
+    def test_config_dir_key_not_present_in_loaded_config(self, tmp_path):
+        conf_d = tmp_path / "conf.d"
+        conf_d.mkdir()
+        main = tmp_path / "config.yaml"
+        main.write_text(yaml.dump({"config_dir": str(conf_d), "remote": {}}))
+        cfg = ConfigManager(str(main))
+        assert "config_dir" not in cfg.config
+
+    def test_dir_remote_overrides_main_file_remote(self, tmp_path):
+        conf_d = tmp_path / "conf.d"
+        conf_d.mkdir()
+        (conf_d / "override.yaml").write_text(yaml.dump({"remote": {"r": _remote("https://new.com")}}))
+        main = tmp_path / "config.yaml"
+        main.write_text(yaml.dump({"config_dir": str(conf_d), "remote": {"r": _remote("https://old.com")}}))
+        cfg = ConfigManager(str(main))
+        assert cfg.config["remotes"]["r"]["base_url"] == "https://new.com"
+
+    def test_empty_config_dir_uses_main_file_only(self, tmp_path):
+        conf_d = tmp_path / "conf.d"
+        conf_d.mkdir()
+        main = tmp_path / "config.yaml"
+        main.write_text(yaml.dump({"config_dir": str(conf_d), "remote": {"repo-main": _remote()}}))
+        cfg = ConfigManager(str(main))
+        assert list(cfg.config["remotes"].keys()) == ["repo-main"]
+
+    def test_reload_picks_up_changed_dir_file(self, tmp_path):
+        conf_d = tmp_path / "conf.d"
+        conf_d.mkdir()
+        dir_file = conf_d / "r.yaml"
+        dir_file.write_text(yaml.dump({"remote": {"repo-v1": _remote()}}))
+        main = tmp_path / "config.yaml"
+        main.write_text(yaml.dump({"config_dir": str(conf_d), "remote": {}}))
+        cfg = ConfigManager(str(main))
+        assert "repo-v1" in cfg.config["remotes"]
+
+        dir_file.write_text(yaml.dump({"remote": {"repo-v2": _remote("https://v2.com")}}))
+        future_mtime = cfg._last_modified + 1
+        os.utime(str(dir_file), (future_mtime, future_mtime))
+
+        cfg._check_reload()
+
+        assert "repo-v2" in cfg.config["remotes"]
+        assert "repo-v1" not in cfg.config["remotes"]
+
+
+# ---------------------------------------------------------------------------
+# YAML format normalisation — top-level type keys
+# ---------------------------------------------------------------------------
+
+
+class TestYamlTypeKeys:
+    def test_remote_key_injects_type_remote(self, tmp_path):
+        f = tmp_path / "r.yaml"
+        f.write_text(yaml.dump({"remote": {"my-remote": {"package": "generic", "base_url": "https://x.com"}}}))
+        cfg = ConfigManager(str(f))
+        assert cfg.config["remotes"]["my-remote"]["type"] == "remote"
+
+    def test_virtual_key_injects_type_virtual(self, tmp_path):
+        f = tmp_path / "r.yaml"
+        f.write_text(yaml.dump({"virtual": {"my-virtual": {"package": "helm", "members": ["a", "b"]}}}))
+        cfg = ConfigManager(str(f))
+        assert cfg.config["remotes"]["my-virtual"]["type"] == "virtual"
+        assert cfg.config["remotes"]["my-virtual"]["members"] == ["a", "b"]
+
+    def test_local_key_injects_type_local(self, tmp_path):
+        f = tmp_path / "r.yaml"
+        f.write_text(yaml.dump({"local": {"my-local": {"package": "generic"}}}))
+        cfg = ConfigManager(str(f))
+        assert cfg.config["remotes"]["my-local"]["type"] == "local"
+
+    def test_mixed_file_all_three_types(self, tmp_path):
+        f = tmp_path / "r.yaml"
+        f.write_text(
+            yaml.dump(
+                {
+                    "remote": {"r": {"package": "helm", "base_url": "https://helm.example.com"}},
+                    "virtual": {"v": {"package": "helm", "members": ["r"]}},
+                    "local": {"l": {"package": "generic"}},
+                }
+            )
+        )
+        cfg = ConfigManager(str(f))
+        assert cfg.config["remotes"]["r"]["type"] == "remote"
+        assert cfg.config["remotes"]["v"]["type"] == "virtual"
+        assert cfg.config["remotes"]["l"]["type"] == "local"
+
+    def test_type_field_not_required_in_yaml(self, tmp_path):
+        f = tmp_path / "r.yaml"
+        f.write_text(yaml.dump({"remote": {"r": {"package": "alpine", "base_url": "https://x.com"}}}))
+        cfg = ConfigManager(str(f))
+        raw = cfg.config["remotes"]["r"]
+        # type is injected by the loader; the original dict had no type key
+        assert "type" in raw
+        assert raw["type"] == "remote"
+
+    def test_other_fields_preserved_after_normalisation(self, tmp_path):
+        f = tmp_path / "r.yaml"
+        f.write_text(
+            yaml.dump(
+                {
+                    "remote": {
+                        "r": {
+                            "package": "helm",
+                            "base_url": "https://helm.example.com",
+                            "immutable_patterns": [r"\.tgz$"],
+                            "cache": {"immutable_ttl": 0, "mutable_ttl": 1800},
+                        }
+                    }
+                }
+            )
+        )
+        cfg = ConfigManager(str(f))
+        remote = cfg.config["remotes"]["r"]
+        assert remote["package"] == "helm"
+        assert remote["base_url"] == "https://helm.example.com"
+        assert remote["cache"] == {"immutable_ttl": 0, "mutable_ttl": 1800}
+        assert r"\.tgz$" in remote["immutable_patterns"]

tests/test_docker_auth.py

@@ -0,0 +1,273 @@
+"""Tests for docker_auth: WWW-Authenticate parsing and token caching."""
+
+import time
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import httpx
+import pytest
+
+from artifactapi import docker_auth
+from artifactapi.docker_auth import (
+    _cache_key,
+    _get_cached_token,
+    _store_token,
+    fetch_token,
+    get_docker_token_for_response,
+    parse_www_authenticate,
+)
+
+
+@pytest.fixture(autouse=True)
+def clear_token_cache():
+    """Isolate tests: wipe the module-level token cache before and after each test."""
+    docker_auth._token_cache.clear()
+    yield
+    docker_auth._token_cache.clear()
+
+
+# ---------------------------------------------------------------------------
+# parse_www_authenticate
+# ---------------------------------------------------------------------------
+
+
+class TestParseWwwAuthenticate:
+    def test_full_bearer_header(self):
+        header = 'Bearer realm="https://auth.docker.io/token",service="registry.docker.io",scope="repository:library/nginx:pull"'
+        result = parse_www_authenticate(header)
+        assert result is not None
+        realm, service, scope = result
+        assert realm == "https://auth.docker.io/token"
+        assert service == "registry.docker.io"
+        assert scope == "repository:library/nginx:pull"
+
+    def test_realm_only(self):
+        header = 'Bearer realm="https://auth.example.com/token"'
+        result = parse_www_authenticate(header)
+        assert result is not None
+        realm, service, scope = result
+        assert realm == "https://auth.example.com/token"
+        assert service == ""
+        assert scope == ""
+
+    def test_realm_and_service_only(self):
+        header = 'Bearer realm="https://auth.example.com",service="registry.example.com"'
+        result = parse_www_authenticate(header)
+        assert result is not None
+        _, service, scope = result
+        assert service == "registry.example.com"
+        assert scope == ""
+
+    def test_invalid_scheme_returns_none(self):
+        assert parse_www_authenticate('Basic realm="example"') is None
+
+    def test_empty_header_returns_none(self):
+        assert parse_www_authenticate("") is None
+
+    def test_case_insensitive_bearer_parses_realm(self):
+        header = 'bearer realm="https://auth.example.com/token"'
+        result = parse_www_authenticate(header)
+        assert result is not None
+        realm, _, _ = result
+        assert realm == "https://auth.example.com/token"
+
+    def test_field_order_scope_before_service_drops_service(self):
+        # The regex requires realm,service,scope order; scope before service
+        # results in service being silently dropped. This test documents the known limitation.
+        header = 'Bearer realm="https://auth.example.com",scope="repo:pull",service="svc"'
+        result = parse_www_authenticate(header)
+        assert result is not None
+        realm, service, scope = result
+        assert realm == "https://auth.example.com"
+        assert scope == "repo:pull"
+        assert service == ""  # silently dropped when out of order
+
+
+# ---------------------------------------------------------------------------
+# _cache_key
+# ---------------------------------------------------------------------------
+
+
+class TestCacheKey:
+    def test_key_contains_all_components(self):
+        key = _cache_key("https://realm.com", "svc", "scope", "user")
+        assert "https://realm.com" in key
+        assert "svc" in key
+        assert "scope" in key
+        assert "user" in key
+
+    def test_none_username_uses_empty_string(self):
+        key = _cache_key("https://realm.com", "svc", "scope", None)
+        assert key.endswith("|")
+
+    def test_different_services_give_different_keys(self):
+        k1 = _cache_key("realm", "svc1", "scope", None)
+        k2 = _cache_key("realm", "svc2", "scope", None)
+        assert k1 != k2
+
+    def test_different_scopes_give_different_keys(self):
+        k1 = _cache_key("realm", "svc", "scope:read", None)
+        k2 = _cache_key("realm", "svc", "scope:write", None)
+        assert k1 != k2
+
+    def test_pipe_in_field_value_can_collide_with_adjacent_fields(self):
+        # The "|" separator is not escaped, so a pipe embedded in one field
+        # produces the same key as the same pipe appearing as a separator boundary.
+        # This is a known limitation: _cache_key("a|b","c","d",None) ==
+        # _cache_key("a","b|c","d",None). Documents the behaviour, not a claim it's correct.
+        k1 = _cache_key("a|b", "c", "d", None)
+        k2 = _cache_key("a", "b|c", "d", None)
+        assert k1 == k2
+
+
+# ---------------------------------------------------------------------------
+# _get_cached_token / _store_token
+# ---------------------------------------------------------------------------
+
+
+class TestTokenCaching:
+    def test_get_returns_none_when_not_cached(self):
+        assert _get_cached_token("no-such-key") is None
+
+    def test_get_returns_token_when_valid(self):
+        _store_token("mykey", "tok-abc", 300)
+        assert _get_cached_token("mykey") == "tok-abc"
+
+    def test_get_returns_none_when_expired(self):
+        docker_auth._token_cache["mykey"] = ("old-token", time.time() - 1)
+        assert _get_cached_token("mykey") is None
+
+    def test_expired_entry_is_removed_from_cache(self):
+        docker_auth._token_cache["mykey"] = ("old-token", time.time() - 1)
+        _get_cached_token("mykey")
+        assert "mykey" not in docker_auth._token_cache
+
+    def test_store_expires_30s_before_stated_time(self):
+        before = time.time()
+        _store_token("mykey", "tok", 100)
+        _, expires_at = docker_auth._token_cache["mykey"]
+        # expires_in - 30 = 70; allow ±2 s clock wiggle
+        assert before + 68 <= expires_at <= before + 72
+
+    def test_store_enforces_minimum_10s_expiry(self):
+        before = time.time()
+        _store_token("mykey", "tok", 5)  # expires_in - 30 would be negative
+        _, expires_at = docker_auth._token_cache["mykey"]
+        assert expires_at >= before + 10
+
+
+# ---------------------------------------------------------------------------
+# fetch_token (async, mocks httpx)
+# ---------------------------------------------------------------------------
+
+
+def _make_mock_http_client(token_payload: dict):
+    mock_response = MagicMock()
+    mock_response.raise_for_status = MagicMock()
+    mock_response.json.return_value = token_payload
+
+    mock_client = AsyncMock()
+    mock_client.get = AsyncMock(return_value=mock_response)
+
+    ctx = MagicMock()
+    ctx.__aenter__ = AsyncMock(return_value=mock_client)
+    ctx.__aexit__ = AsyncMock(return_value=False)
+    return ctx, mock_client
+
+
+class TestFetchToken:
+    async def test_returns_token_field(self):
+        ctx, _ = _make_mock_http_client({"token": "bearer-tok", "expires_in": 300})
+        with patch("httpx.AsyncClient", return_value=ctx):
+            token = await fetch_token("https://auth.example.com", "svc", "scope")
+        assert token == "bearer-tok"
+
+    async def test_falls_back_to_access_token_field(self):
+        ctx, _ = _make_mock_http_client({"access_token": "access-tok", "expires_in": 300})
+        with patch("httpx.AsyncClient", return_value=ctx):
+            token = await fetch_token("https://auth.example.com", "svc", "scope")
+        assert token == "access-tok"
+
+    async def test_returns_none_when_response_missing_token_field(self):
+        ctx, _ = _make_mock_http_client({"not_token": "value", "expires_in": 300})
+        with patch("httpx.AsyncClient", return_value=ctx):
+            token = await fetch_token("https://auth.example.com", "svc", "scope")
+        assert token is None
+
+    async def test_defaults_expires_in_to_300_when_missing(self):
+        ctx, _ = _make_mock_http_client({"token": "tok"})  # no expires_in key
+        before = time.time()
+        with patch("httpx.AsyncClient", return_value=ctx):
+            token = await fetch_token("https://auth.example.com", "svc", "scope")
+        assert token == "tok"
+        key = _cache_key("https://auth.example.com", "svc", "scope", None)
+        _, expires_at = docker_auth._token_cache[key]
+        # Default expires_in=300, stored as time.time() + max(300-30, 10) = 270
+        assert before + 268 <= expires_at <= before + 272
+
+    async def test_uses_cache_on_second_call_without_http(self):
+        ctx, mock_client = _make_mock_http_client({"token": "cached-tok", "expires_in": 300})
+        with patch("httpx.AsyncClient", return_value=ctx):
+            await fetch_token("https://auth.example.com", "svc", "scope")
+            mock_client.get.reset_mock()
+            token = await fetch_token("https://auth.example.com", "svc", "scope")
+            mock_client.get.assert_not_called()
+        assert token == "cached-tok"
+
+    async def test_returns_none_on_network_error(self):
+        mock_client = AsyncMock()
+        mock_client.get = AsyncMock(side_effect=Exception("connection refused"))
+        ctx = MagicMock()
+        ctx.__aenter__ = AsyncMock(return_value=mock_client)
+        ctx.__aexit__ = AsyncMock(return_value=False)
+        with patch("httpx.AsyncClient", return_value=ctx):
+            token = await fetch_token("https://auth.example.com", "svc", "scope")
+        assert token is None
+
+    async def test_returns_none_on_http_status_error(self):
+        mock_response = MagicMock()
+        mock_response.raise_for_status.side_effect = httpx.HTTPStatusError("401 Unauthorized", request=MagicMock(), response=MagicMock())
+        mock_client = AsyncMock()
+        mock_client.get = AsyncMock(return_value=mock_response)
+        ctx = MagicMock()
+        ctx.__aenter__ = AsyncMock(return_value=mock_client)
+        ctx.__aexit__ = AsyncMock(return_value=False)
+        with patch("httpx.AsyncClient", return_value=ctx):
+            token = await fetch_token("https://auth.example.com", "svc", "scope")
+        assert token is None
+
+    async def test_passes_credentials_as_auth_tuple(self):
+        ctx, mock_client = _make_mock_http_client({"token": "authed-tok", "expires_in": 300})
+        with patch("httpx.AsyncClient", return_value=ctx):
+            await fetch_token("https://auth.example.com", "svc", "scope", "user", "pass")
+        call_kwargs = mock_client.get.call_args.kwargs
+        assert call_kwargs.get("auth") == ("user", "pass")
+
+    async def test_no_auth_when_no_credentials(self):
+        ctx, mock_client = _make_mock_http_client({"token": "anon-tok", "expires_in": 300})
+        with patch("httpx.AsyncClient", return_value=ctx):
+            await fetch_token("https://auth.example.com", "svc", "scope")
+        call_kwargs = mock_client.get.call_args.kwargs
+        assert call_kwargs.get("auth") is None
+
+
+# ---------------------------------------------------------------------------
+# get_docker_token_for_response
+# ---------------------------------------------------------------------------
+
+
+class TestGetDockerTokenForResponse:
+    async def test_returns_none_for_non_bearer_header(self):
+        token = await get_docker_token_for_response('Basic realm="example"')
+        assert token is None
+
+    async def test_end_to_end_parse_and_fetch(self):
+        """parse_www_authenticate → fetch_token wired together end-to-end."""
+        header = 'Bearer realm="https://auth.example.com",service="svc",scope="repo:pull"'
+        ctx, mock_client = _make_mock_http_client({"token": "e2e-tok", "expires_in": 300})
+        with patch("httpx.AsyncClient", return_value=ctx):
+            token = await get_docker_token_for_response(header, "user", "pass")
+        assert token == "e2e-tok"
+        call_kwargs = mock_client.get.call_args.kwargs
+        assert call_kwargs["params"]["service"] == "svc"
+        assert call_kwargs["params"]["scope"] == "repo:pull"
+        assert call_kwargs["auth"] == ("user", "pass")

tests/test_storage.py

@@ -0,0 +1,132 @@
+"""Tests for S3Storage: get_object_key (pure logic) and I/O methods."""
+
+import hashlib
+from unittest.mock import MagicMock, patch
+
+import pytest
+from botocore.exceptions import ClientError
+from fastapi import HTTPException
+
+from artifactapi.storage import S3Storage
+
+
+@pytest.fixture
+def storage():
+    """S3Storage with a mocked boto3 client."""
+    with patch("boto3.client", return_value=MagicMock()):
+        s = S3Storage(
+            endpoint="localhost:9000",
+            access_key="testkey",
+            secret_key="testsecret",
+            bucket="testbucket",
+            secure=False,
+        )
+    s.client = MagicMock()
+    return s
+
+
+# ---------------------------------------------------------------------------
+# get_object_key
+# ---------------------------------------------------------------------------
+
+
+class TestGetObjectKey:
+    def test_key_has_three_part_structure(self, storage):
+        # remote / hash-segment / filename
+        key = storage.get_object_key("myremote", "some/path/to/file.rpm")
+        parts = key.split("/")
+        assert len(parts) == 3
+        assert parts[0] == "myremote"
+        assert parts[2] == "file.rpm"
+        assert len(parts[1]) == 16  # SHA-256 hex truncated to 16 chars
+
+    def test_key_uses_sha256_of_directory_path(self, storage):
+        # Pin the hash algorithm, truncation length, and format in one assertion
+        key = storage.get_object_key("myremote", "some/path/to/file.rpm")
+        expected_hash = hashlib.sha256(b"some/path/to").hexdigest()[:16]
+        assert key == f"myremote/{expected_hash}/file.rpm"
+
+    def test_different_remotes_give_different_keys(self, storage):
+        k1 = storage.get_object_key("remote-a", "path/to/file.rpm")
+        k2 = storage.get_object_key("remote-b", "path/to/file.rpm")
+        assert k1 != k2
+
+    def test_different_directories_give_different_keys(self, storage):
+        k1 = storage.get_object_key("myremote", "path/version-1/file.rpm")
+        k2 = storage.get_object_key("myremote", "path/version-2/file.rpm")
+        assert k1 != k2
+        assert k1.split("/")[-1] == k2.split("/")[-1] == "file.rpm"
+
+    def test_leading_slash_stripped(self, storage):
+        k1 = storage.get_object_key("myremote", "/path/to/file.rpm")
+        k2 = storage.get_object_key("myremote", "path/to/file.rpm")
+        assert k1 == k2
+
+    def test_file_with_no_directory(self, storage):
+        key = storage.get_object_key("myremote", "file.rpm")
+        assert key == "myremote/file.rpm"
+
+    def test_docker_blob_uses_digest_path(self, storage):
+        digest = "a" * 64  # realistic 64-char SHA-256 hex string
+        path = f"library/nginx/blobs/sha256:{digest}"
+        key = storage.get_object_key("dockerhub", path)
+        assert key == f"dockerhub/blobs/sha256/{digest}"
+
+    def test_docker_blob_deduplication_across_images(self, storage):
+        """Same blob digest pulled from different images maps to the same S3 key."""
+        digest = "deadbeef" * 8  # 64-char hex
+        k1 = storage.get_object_key("dockerhub", f"library/nginx/blobs/sha256:{digest}")
+        k2 = storage.get_object_key("dockerhub", f"library/ubuntu/blobs/sha256:{digest}")
+        assert k1 == k2
+
+    def test_docker_blob_different_digests_different_keys(self, storage):
+        k1 = storage.get_object_key("dockerhub", "library/nginx/blobs/sha256:" + "a" * 64)
+        k2 = storage.get_object_key("dockerhub", "library/nginx/blobs/sha256:" + "b" * 64)
+        assert k1 != k2
+
+    def test_docker_blob_different_remotes_different_keys(self, storage):
+        digest = "abc" * 21 + "d"  # 64-char hex
+        k1 = storage.get_object_key("remote-a", f"library/nginx/blobs/sha256:{digest}")
+        k2 = storage.get_object_key("remote-b", f"library/nginx/blobs/sha256:{digest}")
+        assert k1 != k2
+
+
+# ---------------------------------------------------------------------------
+# get_url
+# ---------------------------------------------------------------------------
+
+
+class TestGetUrl:
+    def test_returns_http_url_for_insecure_endpoint(self, storage):
+        url = storage.get_url("myremote/abc123/file.rpm")
+        assert url == "http://localhost:9000/testbucket/myremote/abc123/file.rpm"
+
+    def test_returns_http_url_for_secure_storage(self):
+        with patch("boto3.client", return_value=MagicMock()):
+            s = S3Storage(endpoint="s3.example.com", access_key="k", secret_key="s", bucket="b", secure=True)
+        s.client = MagicMock()
+        # get_url uses http:// always (direct internal access address, not the S3 protocol)
+        assert s.get_url("path/to/file.rpm") == "http://s3.example.com/b/path/to/file.rpm"
+
+
+# ---------------------------------------------------------------------------
+# upload / download_object
+# ---------------------------------------------------------------------------
+
+
+class TestUpload:
+    def test_upload_returns_s3_uri(self, storage):
+        storage.client.put_object.return_value = {}
+        result = storage.upload("myremote/abc123/file.rpm", b"content")
+        assert result == "s3://testbucket/myremote/abc123/file.rpm"
+
+
+class TestDownloadObject:
+    def test_download_object_raises_404_on_client_error(self, storage):
+        storage.client.get_object.side_effect = ClientError(
+            {"Error": {"Code": "NoSuchKey", "Message": "The specified key does not exist"}},
+            "GetObject",
+        )
+        with pytest.raises(HTTPException) as exc_info:
+            storage.download_object("nonexistent/key")
+        assert exc_info.value.status_code == 404

tests/test_virtual.py

@@ -0,0 +1,596 @@
+"""Unit tests for the virtual repository handler (artifact/virtual.py)."""
+
+from datetime import UTC, date, datetime
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+import yaml
+
+from artifactapi.artifact.virtual import (
+    _HANDLERS,
+    _get_member_index,
+    _HelmDumper,
+    _HelmHandler,
+    _merge_helm_indexes,
+    _VirtualHandler,
+)
+
+# ---------------------------------------------------------------------------
+# Shared sample data
+# ---------------------------------------------------------------------------
+
+_INDEX_A = b"""\
+apiVersion: v1
+entries:
+  vault:
+    - name: vault
+      version: "0.27.0"
+      urls:
+        - https://helm.releases.hashicorp.com/vault-0.27.0.tgz
+  consul:
+    - name: consul
+      version: "1.2.0"
+      urls:
+        - https://helm.releases.hashicorp.com/consul-1.2.0.tgz
+generated: "2023-01-01T00:00:00.000Z"
+"""
+
+_INDEX_B = b"""\
+apiVersion: v1
+entries:
+  nginx:
+    - name: nginx
+      version: "15.0.0"
+      urls:
+        - https://charts.example.com/nginx-15.0.0.tgz
+  vault:
+    - name: vault
+      version: "0.27.0"
+      urls:
+        - https://charts.example.com/vault-0.27.0.tgz
+    - name: vault
+      version: "0.26.0"
+      urls:
+        - https://charts.example.com/vault-0.26.0.tgz
+generated: "2023-01-01T00:00:00.000Z"
+"""
+
+_INDEX_SIMPLE = b"""\
+apiVersion: v1
+entries:
+  mychart:
+    - name: mychart
+      version: "1.0.0"
+      urls:
+        - https://helm.releases.hashicorp.com/mychart-1.0.0.tgz
+generated: "2023-01-01T00:00:00.000Z"
+"""
+
+_CFG_A = {"base_url": "https://helm.releases.hashicorp.com", "cache": {"mutable_ttl": 3600}}
+_CFG_B = {"base_url": "https://charts.example.com", "cache": {"mutable_ttl": 1800}}
+
+
+def _identity_resolve(data, *args, **kwargs):
+    return data, None
+
+
+# ---------------------------------------------------------------------------
+# _HelmDumper — datetime/date YAML serialization
+# ---------------------------------------------------------------------------
+
+
+class TestHelmDumper:
+    def _dump(self, value):
+        return yaml.dump({"v": value}, Dumper=_HelmDumper)
+
+    def test_datetime_with_tz_includes_Z_suffix(self):
+        dt = datetime(2023, 6, 15, 12, 0, 0, tzinfo=UTC)
+        assert "Z" in self._dump(dt)
+
+    def test_datetime_without_tz_has_no_Z_suffix(self):
+        dt = datetime(2023, 6, 15, 12, 0, 0)
+        assert "Z" not in self._dump(dt)
+
+    def test_datetime_uses_T_separator_not_space(self):
+        dt = datetime(2023, 6, 15, 12, 30, 0, tzinfo=UTC)
+        assert "T12:30:00" in self._dump(dt)
+
+    def test_date_serialized_as_iso_string(self):
+        assert "2023-01-15" in self._dump(date(2023, 1, 15))
+
+    def test_datetime_round_trips_as_string_not_python_datetime(self):
+        dt = datetime(2023, 6, 15, 12, 0, 0, tzinfo=UTC)
+        parsed = yaml.safe_load(self._dump(dt))
+        # yaml.safe_load must not re-parse this as a datetime object
+        assert isinstance(parsed["v"], str)
+
+    def test_date_round_trips_as_string_not_python_date(self):
+        parsed = yaml.safe_load(self._dump(date(2023, 1, 15)))
+        assert isinstance(parsed["v"], str)
+
+
+# ---------------------------------------------------------------------------
+# _HelmHandler
+# ---------------------------------------------------------------------------
+
+
+class TestHelmHandler:
+    def setup_method(self):
+        self.handler = _HelmHandler()
+
+    def test_accepts_index_yaml(self):
+        assert self.handler.accepts_path("index.yaml") is True
+
+    def test_rejects_tgz_path(self):
+        assert self.handler.accepts_path("vault-0.27.0.tgz") is False
+
+    def test_rejects_subdirectory_index(self):
+        assert self.handler.accepts_path("charts/index.yaml") is False
+
+    def test_rejects_empty_path(self):
+        assert self.handler.accepts_path("") is False
+
+    def test_path_error_is_non_empty_string(self):
+        msg = self.handler.path_error()
+        assert isinstance(msg, str) and len(msg) > 0
+
+    def test_merge_returns_bytes(self):
+        with patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve):
+            result = self.handler.merge([_INDEX_A], ["member-a"], [_CFG_A], "http://proxy.example.com")
+        assert isinstance(result, bytes)
+
+    def test_merge_delegates_to_merge_helm_indexes(self):
+        with patch("artifactapi.artifact.virtual._merge_helm_indexes", return_value=b"merged") as mock_fn:
+            result = self.handler.merge([b"data"], ["m"], [{}], "http://proxy")
+        mock_fn.assert_called_once_with([b"data"], ["m"], [{}], "http://proxy")
+        assert result == b"merged"
+
+
+# ---------------------------------------------------------------------------
+# _HANDLERS registry
+# ---------------------------------------------------------------------------
+
+
+class TestHandlersRegistry:
+    def test_helm_handler_is_registered(self):
+        assert "helm" in _HANDLERS
+        assert isinstance(_HANDLERS["helm"], _HelmHandler)
+
+    def test_helm_handler_satisfies_protocol(self):
+        assert isinstance(_HANDLERS["helm"], _VirtualHandler)
+
+
+# ---------------------------------------------------------------------------
+# _merge_helm_indexes
+# ---------------------------------------------------------------------------
+
+
+class TestMergeHelmIndexes:
+    def _merge(self, raw_indexes, member_names, member_configs, proxy_base="http://proxy.example.com"):
+        with patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve):
+            return _merge_helm_indexes(raw_indexes, member_names, member_configs, proxy_base)
+
+    def _parse(self, raw):
+        return yaml.safe_load(raw)
+
+    def test_single_member_all_charts_present(self):
+        index = self._parse(self._merge([_INDEX_A], ["member-a"], [_CFG_A]))
+        assert "vault" in index["entries"]
+        assert "consul" in index["entries"]
+
+    def test_two_members_non_overlapping_charts_all_present(self):
+        index = self._parse(self._merge([_INDEX_A, _INDEX_B], ["member-a", "member-b"], [_CFG_A, _CFG_B]))
+        assert "vault" in index["entries"]
+        assert "consul" in index["entries"]
+        assert "nginx" in index["entries"]
+
+    def test_first_member_wins_on_duplicate_name_and_version(self):
+        index = self._parse(self._merge([_INDEX_A, _INDEX_B], ["member-a", "member-b"], [_CFG_A, _CFG_B]))
+        v027 = next(e for e in index["entries"]["vault"] if e["version"] == "0.27.0")
+        assert "helm.releases.hashicorp.com" in v027["urls"][0]
+
+    def test_different_versions_of_same_chart_both_included(self):
+        index = self._parse(self._merge([_INDEX_A, _INDEX_B], ["member-a", "member-b"], [_CFG_A, _CFG_B]))
+        versions = {e["version"] for e in index["entries"]["vault"]}
+        assert "0.27.0" in versions
+        assert "0.26.0" in versions
+
+    def test_malformed_yaml_from_member_is_skipped(self):
+        index = self._parse(self._merge([_INDEX_A, b"{bad yaml"], ["member-a", "bad"], [_CFG_A, _CFG_B]))
+        assert "vault" in index["entries"]
+        assert "consul" in index["entries"]
+
+    def test_output_has_apiVersion_v1(self):
+        index = self._parse(self._merge([_INDEX_A], ["member-a"], [_CFG_A]))
+        assert index["apiVersion"] == "v1"
+
+    def test_output_has_generated_field(self):
+        index = self._parse(self._merge([_INDEX_A], ["member-a"], [_CFG_A]))
+        assert "generated" in index
+
+    def test_output_is_valid_yaml(self):
+        raw = self._merge([_INDEX_A, _INDEX_B], ["member-a", "member-b"], [_CFG_A, _CFG_B])
+        assert isinstance(yaml.safe_load(raw), dict)
+
+    def test_empty_index_from_member_produces_no_entries(self):
+        empty = b"apiVersion: v1\nentries: {}\ngenerated: '2023-01-01T00:00:00.000Z'\n"
+        index = self._parse(self._merge([empty], ["member-a"], [_CFG_A]))
+        assert index["entries"] == {}
+
+
+# ---------------------------------------------------------------------------
+# _get_member_index  (async)
+# ---------------------------------------------------------------------------
+
+
+class TestGetMemberIndex:
+    @pytest.fixture
+    def storage(self):
+        m = MagicMock()
+        m.get_object_key.return_value = "member/key/index.yaml"
+        m.exists.return_value = False
+        m.download_object.return_value = b"cached bytes"
+        return m
+
+    @pytest.fixture
+    def cache(self):
+        m = MagicMock()
+        m.is_index_valid.return_value = False
+        return m
+
+    @pytest.fixture
+    def member_cfg(self):
+        return {"base_url": "https://helm.releases.hashicorp.com", "cache": {"mutable_ttl": 3600}}
+
+    def _fake_response(self, content=b"upstream bytes"):
+        r = MagicMock()
+        r.content = content
+        r.raise_for_status = MagicMock()
+        return r
+
+    def _patch_httpx(self, response):
+        mock_client_cls = patch("artifactapi.artifact.virtual.httpx.AsyncClient")
+        p = mock_client_cls.start()
+        mock_client = AsyncMock()
+        p.return_value.__aenter__.return_value = mock_client
+        mock_client.get.return_value = response
+        return mock_client_cls, mock_client
+
+    async def test_cache_hit_returns_stored_bytes(self, storage, cache, member_cfg):
+        storage.exists.return_value = True
+        cache.is_index_valid.return_value = True
+
+        _, _, _, raw_data = await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        assert raw_data == b"cached bytes"
+
+    async def test_cache_hit_does_not_fetch_upstream(self, storage, cache, member_cfg):
+        storage.exists.return_value = True
+        cache.is_index_valid.return_value = True
+
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        mock_cls.assert_not_called()
+
+    async def test_cache_hit_storage_error_falls_through_to_upstream(self, storage, cache, member_cfg):
+        storage.exists.return_value = True
+        cache.is_index_valid.return_value = True
+        storage.download_object.side_effect = Exception("S3 read error")
+
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response(b"fresh bytes")
+
+            _, _, _, raw_data = await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        assert raw_data == b"fresh bytes"
+
+    async def test_cache_miss_fetches_from_upstream(self, storage, cache, member_cfg):
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response()
+
+            _, _, _, raw_data = await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        assert raw_data == b"upstream bytes"
+
+    async def test_cache_miss_stores_result_in_s3(self, storage, cache, member_cfg):
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response()
+
+            await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        storage.upload.assert_called_once()
+
+    async def test_cache_miss_marks_cache_with_configured_ttl(self, storage, cache, member_cfg):
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response()
+
+            await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        cache.mark_index_cached.assert_called_once_with("m", "index.yaml", 3600)
+
+    async def test_cache_miss_with_auth_sends_basic_auth_header(self, storage, cache):
+        cfg = {
+            "base_url": "https://private.example.com",
+            "username": "user",
+            "password": "pass",
+            "cache": {"mutable_ttl": 3600},
+        }
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response()
+
+            await _get_member_index("m", cfg, "index.yaml", storage, cache)
+
+        headers = mock_client.get.call_args.kwargs["headers"]
+        assert "Authorization" in headers
+        assert headers["Authorization"].startswith("Basic ")
+
+    async def test_no_credentials_sends_no_auth_header(self, storage, cache, member_cfg):
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response()
+
+            await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        headers = mock_client.get.call_args.kwargs["headers"]
+        assert "Authorization" not in headers
+
+    async def test_upstream_fetch_failure_returns_none(self, storage, cache, member_cfg):
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.side_effect = Exception("connection refused")
+
+            _, _, _, raw_data = await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        assert raw_data is None
+
+    async def test_s3_upload_failure_still_returns_data(self, storage, cache, member_cfg):
+        storage.upload.side_effect = Exception("S3 write error")
+
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response()
+
+            _, _, _, raw_data = await _get_member_index("m", member_cfg, "index.yaml", storage, cache)
+
+        assert raw_data == b"upstream bytes"
+
+    async def test_returns_ttl_from_config(self, storage, cache):
+        cfg = {"base_url": "https://example.com", "cache": {"mutable_ttl": 900}}
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response()
+
+            _, _, ttl, _ = await _get_member_index("m", cfg, "index.yaml", storage, cache)
+
+        assert ttl == 900
+
+    async def test_defaults_ttl_to_3600_when_not_configured(self, storage, cache):
+        cfg = {"base_url": "https://example.com"}
+        with patch("artifactapi.artifact.virtual.httpx.AsyncClient") as mock_cls:
+            mock_client = AsyncMock()
+            mock_cls.return_value.__aenter__.return_value = mock_client
+            mock_client.get.return_value = self._fake_response()
+
+            _, _, ttl, _ = await _get_member_index("m", cfg, "index.yaml", storage, cache)
+
+        assert ttl == 3600
+
+
+# ---------------------------------------------------------------------------
+# Virtual route  GET /api/v1/virtual/{name}/{path}
+# ---------------------------------------------------------------------------
+
+
+@pytest.fixture
+def mock_storage_v():
+    m = MagicMock()
+    m.get_object_key.return_value = "virtual/helm-virtual-test/index.yaml"
+    m.exists.return_value = False
+    m.download_object.return_value = b"apiVersion: v1\nentries: {}\n"
+    return m
+
+
+@pytest.fixture
+def mock_cache_v():
+    m = MagicMock()
+    m.is_index_valid.return_value = False
+    m.available = False
+    m.client = None
+    return m
+
+
+@pytest.fixture
+def patched_virtual_deps(mock_storage_v, mock_cache_v):
+    import artifactapi.main as main_mod
+
+    with (
+        patch.object(main_mod, "storage", mock_storage_v),
+        patch.object(main_mod, "cache", mock_cache_v),
+    ):
+        yield {"storage": mock_storage_v, "cache": mock_cache_v}
+
+
+class TestVirtualRoute:
+    def test_unknown_virtual_name_returns_404(self, client, patched_virtual_deps):
+        response = client.get("/api/v1/virtual/no-such-virtual/index.yaml")
+        assert response.status_code == 404
+
+    def test_non_virtual_type_returns_400(self, client, patched_virtual_deps):
+        # helm-test is type "remote", not "virtual"
+        response = client.get("/api/v1/virtual/helm-test/index.yaml")
+        assert response.status_code == 400
+
+    def test_unsupported_package_returns_400(self, client, patched_virtual_deps):
+        # unsupported-virtual-test has package "rpm"
+        response = client.get("/api/v1/virtual/unsupported-virtual-test/index.yaml")
+        assert response.status_code == 400
+
+    def test_non_index_path_returns_404(self, client, patched_virtual_deps):
+        response = client.get("/api/v1/virtual/helm-virtual-test/vault-0.27.0.tgz")
+        assert response.status_code == 404
+
+    def test_no_members_returns_500(self, client, patched_virtual_deps):
+        response = client.get("/api/v1/virtual/empty-virtual-test/index.yaml")
+        assert response.status_code == 500
+
+    def test_virtual_cache_hit_returns_200(self, client, patched_virtual_deps):
+        deps = patched_virtual_deps
+        deps["storage"].exists.return_value = True
+        deps["cache"].is_index_valid.return_value = True
+
+        response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+        assert response.status_code == 200
+
+    def test_virtual_cache_hit_content_type_is_yaml(self, client, patched_virtual_deps):
+        deps = patched_virtual_deps
+        deps["storage"].exists.return_value = True
+        deps["cache"].is_index_valid.return_value = True
+
+        response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+        assert "text/yaml" in response.headers["content-type"]
+
+    def test_virtual_cache_hit_returns_stored_content(self, client, patched_virtual_deps):
+        deps = patched_virtual_deps
+        deps["storage"].exists.return_value = True
+        deps["cache"].is_index_valid.return_value = True
+        deps["storage"].download_object.return_value = b"apiVersion: v1\nentries: {}\n"
+
+        response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+        assert response.content == b"apiVersion: v1\nentries: {}\n"
+
+    def test_virtual_cache_hit_skips_member_fetch(self, client, patched_virtual_deps):
+        deps = patched_virtual_deps
+        deps["storage"].exists.return_value = True
+        deps["cache"].is_index_valid.return_value = True
+
+        with patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get:
+            client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        mock_get.assert_not_called()
+
+    def test_cache_miss_returns_200_with_yaml_content_type(self, client, patched_virtual_deps):
+        with (
+            patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get,
+            patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve),
+        ):
+            mock_get.return_value = ("helm-test", _CFG_A, 3600, _INDEX_SIMPLE)
+            response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        assert response.status_code == 200
+        assert "text/yaml" in response.headers["content-type"]
+
+    def test_cache_miss_response_contains_merged_entries(self, client, patched_virtual_deps):
+        with (
+            patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get,
+            patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve),
+        ):
+            mock_get.return_value = ("helm-test", _CFG_A, 3600, _INDEX_SIMPLE)
+            response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        index = yaml.safe_load(response.content)
+        assert "mychart" in index["entries"]
+
+    def test_cache_miss_stores_result_in_s3(self, client, patched_virtual_deps):
+        deps = patched_virtual_deps
+        with (
+            patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get,
+            patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve),
+        ):
+            mock_get.return_value = ("helm-test", _CFG_A, 3600, _INDEX_SIMPLE)
+            client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        deps["storage"].upload.assert_called_once()
+
+    def test_cache_miss_marks_index_cached(self, client, patched_virtual_deps):
+        deps = patched_virtual_deps
+        with (
+            patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get,
+            patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve),
+        ):
+            mock_get.return_value = ("helm-test", _CFG_A, 3600, _INDEX_SIMPLE)
+            client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        deps["cache"].mark_index_cached.assert_called_once()
+
+    def test_cache_miss_uses_min_ttl_across_members(self, client, patched_virtual_deps):
+        deps = patched_virtual_deps
+        with (
+            patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get,
+            patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve),
+        ):
+            mock_get.side_effect = [
+                ("helm-test", _CFG_A, 3600, _INDEX_SIMPLE),
+                ("helm-member-2", _CFG_B, 1800, _INDEX_SIMPLE),
+            ]
+            client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        _, _, ttl = deps["cache"].mark_index_cached.call_args[0]
+        assert ttl == 1800
+
+    def test_all_members_unreachable_returns_502(self, client, patched_virtual_deps):
+        with patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get:
+            mock_get.return_value = ("helm-test", _CFG_A, 3600, None)
+            response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        assert response.status_code == 502
+
+    def test_one_member_unreachable_still_returns_200(self, client, patched_virtual_deps):
+        with (
+            patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get,
+            patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve),
+        ):
+            mock_get.side_effect = [
+                ("helm-test", _CFG_A, 3600, _INDEX_SIMPLE),
+                ("helm-member-2", _CFG_B, 1800, None),
+            ]
+            response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        assert response.status_code == 200
+
+    def test_member_not_in_config_is_skipped(self, client, patched_virtual_deps):
+        import artifactapi.main as main_mod
+
+        real_get = main_mod.config.get_remote_config
+
+        def patched_get(name):
+            return None if name == "helm-member-2" else real_get(name)
+
+        with (
+            patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get,
+            patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve),
+            patch.object(main_mod.config, "get_remote_config", side_effect=patched_get),
+        ):
+            mock_get.return_value = ("helm-test", _CFG_A, 3600, _INDEX_SIMPLE)
+            response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        # only helm-test was available — should succeed
+        assert response.status_code == 200
+        mock_get.assert_called_once()
+
+    def test_s3_store_failure_still_returns_200(self, client, patched_virtual_deps):
+        deps = patched_virtual_deps
+        deps["storage"].upload.side_effect = Exception("S3 write error")
+
+        with (
+            patch("artifactapi.artifact.virtual._get_member_index", new_callable=AsyncMock) as mock_get,
+            patch("artifactapi.artifact.virtual._helm.resolve_content", side_effect=_identity_resolve),
+        ):
+            mock_get.return_value = ("helm-test", _CFG_A, 3600, _INDEX_SIMPLE)
+            response = client.get("/api/v1/virtual/helm-virtual-test/index.yaml")
+
+        assert response.status_code == 200

tox.ini

@@ -0,0 +1,8 @@
+[tox]
+envlist = py311
+isolated_build = true
+
+[testenv]
+extras = dev
+commands =
+    pytest {posargs:tests}