Compare commits
1 Commits
main
...
us/US07-07
| Author | SHA1 | Date | |
|---|---|---|---|
| a560086420 |
@@ -1,22 +0,0 @@
|
||||
# Deny-by-default build context (US08-02): the image must contain no secrets, no
|
||||
# photos, no database, no logs, and no .git. An allow list is the only version of this
|
||||
# rule that stays true when a new file appears in the working copy.
|
||||
*
|
||||
|
||||
!pyproject.toml
|
||||
!alembic.ini
|
||||
!README.md
|
||||
!photo_pipeline
|
||||
!migrations
|
||||
!frontend
|
||||
!docker
|
||||
|
||||
# Nothing generated, even under an allowed directory.
|
||||
**/__pycache__
|
||||
**/*.py[cod]
|
||||
**/.DS_Store
|
||||
**/*.env
|
||||
**/*.log
|
||||
**/*.db
|
||||
**/*.db-*
|
||||
**/*.sqlite*
|
||||
9
.gitignore
vendored
9
.gitignore
vendored
@@ -20,12 +20,3 @@ _IGNORE/
|
||||
|
||||
# Test failure evidence (US07-04)
|
||||
.artifacts/
|
||||
|
||||
# Any dotenv, not only the default name.
|
||||
*.env
|
||||
|
||||
# Local virtualenv for running the app.
|
||||
.venv/
|
||||
|
||||
# setuptools editable-install metadata.
|
||||
*.egg-info/
|
||||
|
||||
115
Dockerfile
115
Dockerfile
@@ -1,115 +0,0 @@
|
||||
# One image, two roles (US08-02).
|
||||
#
|
||||
# The application is not self-contained Python: it shells out to `exiftool` for every
|
||||
# EXIF checkpoint and to `immich-go` for every upload, and it serves the static
|
||||
# frontend from `frontend/`. All three are installed here at pinned versions, because
|
||||
# an image whose external tools drift is an image whose metadata checkpoints and
|
||||
# upload reports drift with them (concept §15, "External integration risks").
|
||||
#
|
||||
# Everything is pinned:
|
||||
# * the base image by tag *and* digest, so a moved tag cannot change the runtime;
|
||||
# * exiftool by its Debian package version, verified against `exiftool -ver`;
|
||||
# * immich-go by release version and per-architecture SHA-256 of the release asset.
|
||||
# The verified versions become image labels and /etc/photo-pipeline/versions.json,
|
||||
# which `python -m photo_pipeline diagnostics` reports — so a running container can
|
||||
# prove what it contains instead of being trusted about it.
|
||||
#
|
||||
# The project is installed editable on purpose: `photo_pipeline.db` resolves
|
||||
# `alembic.ini` and `migrations/`, and the API resolves `frontend/`, relative to the
|
||||
# repository root. An editable install keeps that one layout instead of scattering the
|
||||
# same files across site-packages and a source tree.
|
||||
|
||||
ARG PYTHON_IMAGE=python:3.12.14-slim-trixie@sha256:2c941e860699f878900b0edc2403613c234d4b32eda3cc9fa7036991a2a63c4a
|
||||
|
||||
# ── the uploader, fetched and verified outside the final layer ────────────────
|
||||
FROM ${PYTHON_IMAGE} AS uploader
|
||||
|
||||
ARG IMMICH_GO_VERSION=0.32.0
|
||||
ARG IMMICH_GO_SHA256_AMD64=6e2ad86bafdadb9466d6515de7cb882726c0aea1a21d51164dff361d7d480a97
|
||||
ARG IMMICH_GO_SHA256_ARM64=2c35d9284baae407ef9540bdac5f488971b0bdc7be758a4d7c05ab270af09fdb
|
||||
|
||||
COPY docker/fetch-immich-go.py /tmp/fetch-immich-go.py
|
||||
RUN python /tmp/fetch-immich-go.py \
|
||||
--version "${IMMICH_GO_VERSION}" \
|
||||
--sha256-amd64 "${IMMICH_GO_SHA256_AMD64}" \
|
||||
--sha256-arm64 "${IMMICH_GO_SHA256_ARM64}" \
|
||||
--into /usr/local/bin \
|
||||
&& /usr/local/bin/immich-go version
|
||||
|
||||
# ── the application ──────────────────────────────────────────────────────────
|
||||
FROM ${PYTHON_IMAGE} AS runtime
|
||||
|
||||
ARG EXIFTOOL_VERSION=13.25+dfsg-1
|
||||
ARG IMMICH_GO_VERSION=0.32.0
|
||||
# The library is mounted from the host, so the container's identity must match the
|
||||
# ownership that library already has: everything this application renames, writes
|
||||
# EXIF into, or archives has to stay owned by the host user afterwards.
|
||||
ARG UID=1000
|
||||
ARG GID=1000
|
||||
|
||||
LABEL org.opencontainers.image.title="photo_pipeline" \
|
||||
org.opencontainers.image.source="https://github.com/domverse/photoanalyzer" \
|
||||
io.photoanalyzer.exiftool.version="${EXIFTOOL_VERSION}" \
|
||||
io.photoanalyzer.immich-go.version="${IMMICH_GO_VERSION}"
|
||||
|
||||
ENV PYTHONUNBUFFERED=1 \
|
||||
PYTHONDONTWRITEBYTECODE=1 \
|
||||
PATH=/opt/venv/bin:$PATH \
|
||||
PHOTO_PIPELINE_DATA_DIR=/data
|
||||
|
||||
RUN set -eu; \
|
||||
apt-get update; \
|
||||
DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
|
||||
"libimage-exiftool-perl=${EXIFTOOL_VERSION}"; \
|
||||
rm -rf /var/lib/apt/lists/*
|
||||
|
||||
COPY --from=uploader /usr/local/bin/immich-go /usr/local/bin/immich-go
|
||||
|
||||
WORKDIR /app
|
||||
COPY pyproject.toml alembic.ini README.md ./
|
||||
COPY photo_pipeline ./photo_pipeline
|
||||
COPY migrations ./migrations
|
||||
COPY frontend ./frontend
|
||||
COPY docker/entrypoint.sh docker/healthcheck.sh /usr/local/bin/
|
||||
|
||||
# Runtime dependencies only: the `test` extra (pytest, playwright) and the `vision`
|
||||
# extra stay out, and pip's build isolation leaves no build tooling behind.
|
||||
RUN set -eu; \
|
||||
python -m venv /opt/venv; \
|
||||
/opt/venv/bin/pip install --no-cache-dir -e .
|
||||
|
||||
# What is installed must be what was pinned, or the labels and the version record
|
||||
# would be a claim rather than a fact.
|
||||
RUN set -eu; \
|
||||
mkdir -p /etc/photo-pipeline; \
|
||||
exiftool_version="$(exiftool -ver)"; \
|
||||
immich_go_version="$(immich-go version | head -n 1 | tr -d '\r')"; \
|
||||
expected_exiftool="$(printf '%s' "${EXIFTOOL_VERSION}" | cut -d+ -f1 | cut -d- -f1)"; \
|
||||
[ "${exiftool_version}" = "${expected_exiftool}" ] \
|
||||
|| { echo "exiftool ${exiftool_version} is not the pinned ${expected_exiftool}" >&2; exit 1; }; \
|
||||
case "${immich_go_version}" in \
|
||||
*"${IMMICH_GO_VERSION}"*) ;; \
|
||||
*) echo "immich-go '${immich_go_version}' is not pinned ${IMMICH_GO_VERSION}" >&2; exit 1 ;; \
|
||||
esac; \
|
||||
printf '{\n "exiftool": "%s",\n "immich-go": "%s"\n}\n' \
|
||||
"${exiftool_version}" "${IMMICH_GO_VERSION}" > /etc/photo-pipeline/versions.json
|
||||
|
||||
# Non-root, with the host library's ownership. /data is the persistent volume; the
|
||||
# photo library itself is mounted by the deployment (US08-03), never baked in.
|
||||
RUN set -eu; \
|
||||
groupadd --gid "${GID}" --non-unique app; \
|
||||
useradd --uid "${UID}" --gid "${GID}" --non-unique --no-create-home --home-dir /app app; \
|
||||
mkdir -p /data; \
|
||||
chown "${UID}:${GID}" /data
|
||||
USER ${UID}:${GID}
|
||||
|
||||
EXPOSE 8000
|
||||
|
||||
# Readiness, not liveness: an unmigrated or misconfigured database answers
|
||||
# /api/v1/health/ready with 503, and a container that cannot serve must not be
|
||||
# reported healthy. The worker role has no endpoint, so its check is a no-op here.
|
||||
HEALTHCHECK --interval=30s --timeout=10s --start-period=30s --retries=3 \
|
||||
CMD ["/usr/local/bin/healthcheck.sh"]
|
||||
|
||||
ENTRYPOINT ["/usr/local/bin/entrypoint.sh"]
|
||||
CMD ["serve"]
|
||||
116
README.md
116
README.md
@@ -7,49 +7,16 @@ archive workflow. Planning lives in `INTEGRATED_PIPELINE_CONCEPT.md` and
|
||||
## Application (`photo_pipeline`)
|
||||
|
||||
The target application lives in `photo_pipeline/` (FastAPI + SQLAlchemy + Alembic).
|
||||
Install it into a virtualenv once:
|
||||
Run it with:
|
||||
|
||||
```bash
|
||||
python3.12 -m venv .venv
|
||||
.venv/bin/pip install -e ".[vision]" # drop [vision] for a review-only install
|
||||
python -m photo_pipeline migrate # apply database migrations
|
||||
python -m photo_pipeline serve # start the API + static review UI (127.0.0.1:8000)
|
||||
```
|
||||
|
||||
Then run the two processes:
|
||||
|
||||
```bash
|
||||
.venv/bin/python -m photo_pipeline migrate # apply database migrations
|
||||
.venv/bin/python -m photo_pipeline serve # API + review UI at 127.0.0.1:8000/app/
|
||||
.venv/bin/python -m photo_pipeline worker # second terminal: runs the jobs
|
||||
```
|
||||
|
||||
The server enqueues work and serves the UI; nothing actually scans, scores,
|
||||
analyses, uploads, or archives without a worker. `work_item/scripts/python` is the
|
||||
*helper's* launcher — it prefers Conda base and falls back to a bare system
|
||||
interpreter, so it is not how the application is run.
|
||||
|
||||
Configuration comes from `PHOTO_PIPELINE_*` environment variables (see
|
||||
`photo_pipeline/config.py`); secrets are referenced, never logged.
|
||||
|
||||
### Configuration file
|
||||
|
||||
`.env` in the working directory is read at startup, or any path named by
|
||||
`PHOTO_PIPELINE_ENV_FILE`. It is parsed, never executed: `KEY=value` lines,
|
||||
`#` comments, optional quotes — no interpolation and no `export`. **Anything already
|
||||
exported wins**, so the file is the standing configuration and the shell is the
|
||||
override for one run.
|
||||
|
||||
The archived CLI's variable names still work, so an existing `photo_analyzer.env`
|
||||
can be used as-is:
|
||||
|
||||
| in the file | applied as |
|
||||
|---|---|
|
||||
| `LLM_API_KEY` / `GEMINI_API_KEY` | `OPENAI_API_KEY` |
|
||||
| `LLM_BASE_URL` | `OPENAI_BASE_URL` |
|
||||
| `LIBRARY` | `PHOTO_PIPELINE_LIBRARY_ROOTS` |
|
||||
|
||||
`.env` and `*.env` are gitignored and denied by the work-item safety checks: the
|
||||
file holds a real key and must never be committed.
|
||||
|
||||
### API access (US07-02)
|
||||
|
||||
The app listens on loopback, so its attacker is another page in the same browser.
|
||||
@@ -69,83 +36,6 @@ not a loopback name (DNS rebinding), when `Origin` is any other origin, when
|
||||
thumbnail), or when the body exceeds `PHOTO_PIPELINE_MAX_REQUEST_BYTES`. There is no
|
||||
CORS middleware at all, so no other origin can read a response.
|
||||
|
||||
### Reaching it through a hostname or proxy (US08-01)
|
||||
|
||||
| variable | meaning |
|
||||
|---|---|
|
||||
| `PHOTO_PIPELINE_ALLOWED_HOSTS` | comma-separated extra names the app answers to; empty means loopback only |
|
||||
| `PHOTO_PIPELINE_ACCESS_SECRET` | traded for the session cookie at `GET /api/v1/session` via `X-Access-Secret` |
|
||||
| `PHOTO_PIPELINE_TRUSTED_PROXIES` | comma-separated peer addresses whose `X-Forwarded-Proto`/`X-Forwarded-Host` are believed |
|
||||
|
||||
Being reachable *was* the authentication: whoever could open `127.0.0.1:8000` owned
|
||||
the library. So naming any non-loopback host — or binding to one, `0.0.0.0` included
|
||||
— makes the access secret mandatory, and `serve` refuses to start without it rather
|
||||
than publishing the library. Loopback-only deployments need no secret and behave
|
||||
exactly as before.
|
||||
|
||||
```bash
|
||||
curl -sc /tmp/pp.jar -H "X-Access-Secret: $PHOTO_PIPELINE_ACCESS_SECRET" \
|
||||
https://photos.example.com/api/v1/session
|
||||
```
|
||||
|
||||
The browser asks for the secret once per tab and keeps it in `sessionStorage`.
|
||||
Wrong secrets are rate-limited (5 per minute) and logged with the caller's address
|
||||
only. `Host` and `Origin` are judged against the configured names; the *external*
|
||||
scheme and host come from the forwarded headers only when the request arrived from a
|
||||
`PHOTO_PIPELINE_TRUSTED_PROXIES` address, so a client cannot declare its own origin,
|
||||
and the session cookie is marked `Secure` when that external scheme is HTTPS. Health
|
||||
endpoints stay reachable without the secret so an orchestrator can restart the
|
||||
container; nothing else does.
|
||||
|
||||
## Container image (US08-02)
|
||||
|
||||
One image runs either role. It is built from a clean checkout with no arguments:
|
||||
|
||||
```bash
|
||||
docker build -t photo-pipeline:dev .
|
||||
```
|
||||
|
||||
Everything external is pinned, and the build fails rather than drifting: the Python
|
||||
base image by tag *and* digest, `exiftool` by its Debian package version (verified
|
||||
against `exiftool -ver`), and `immich-go` by release version and per-architecture
|
||||
SHA-256 of the release asset. The verified versions become image labels and
|
||||
`/etc/photo-pipeline/versions.json`, which `diagnostics` reports as `tools[].pinned`
|
||||
beside the version actually installed — so a replaced binary shows up as a
|
||||
`tool_version_drift` warning instead of as a misparsed upload report.
|
||||
|
||||
| build argument | default | why change it |
|
||||
|---|---|---|
|
||||
| `UID` / `GID` | `1000` | must match the owner of the mounted photo library |
|
||||
| `PYTHON_IMAGE` | pinned digest | upgrading the base image |
|
||||
| `EXIFTOOL_VERSION` | Debian package version | upgrading exiftool |
|
||||
| `IMMICH_GO_VERSION` + `IMMICH_GO_SHA256_AMD64`/`_ARM64` | pinned release | upgrading the uploader (take the digests from that release's `checksums.txt`) |
|
||||
|
||||
The first argument is the role, and every other management command still works:
|
||||
|
||||
```bash
|
||||
docker run --rm -v /srv/photos:/srv/photos -v pp-data:/data \
|
||||
-e PHOTO_PIPELINE_LIBRARY_ROOTS=/srv/photos photo-pipeline:dev migrate
|
||||
|
||||
docker run -d -p 127.0.0.1:8000:8000 -v /srv/photos:/srv/photos -v pp-data:/data \
|
||||
-e PHOTO_PIPELINE_HOST=0.0.0.0 -e PHOTO_PIPELINE_ACCESS_SECRET=... \
|
||||
-e PHOTO_PIPELINE_LIBRARY_ROOTS=/srv/photos photo-pipeline:dev serve
|
||||
|
||||
docker run -d -v /srv/photos:/srv/photos -v pp-data:/data \
|
||||
-e PHOTO_PIPELINE_LIBRARY_ROOTS=/srv/photos photo-pipeline:dev worker
|
||||
```
|
||||
|
||||
One role per container: `serve` and `worker` each take the library process lock for
|
||||
their role (US07-05), so no supervisor starts both. The container refuses to run as
|
||||
UID 0 — files it renames or writes must keep the ownership the host library expects —
|
||||
and `/data` is the persistent volume holding the database, journals, backups, and
|
||||
thumbnail cache. Binding to `0.0.0.0` makes the access secret mandatory
|
||||
([above](#reaching-it-through-a-hostname-or-proxy-us08-01)); `serve` refuses to start
|
||||
without it. The declared `HEALTHCHECK` polls `/api/v1/health/ready`, so a container
|
||||
whose database is unmigrated or misconfigured is never reported healthy.
|
||||
|
||||
Composing the runtime and mounting the library is US08-03; publishing and deploying
|
||||
the image is US08-04.
|
||||
|
||||
## Testing
|
||||
|
||||
One offline command runs the whole suite (unit, integration, and browser
|
||||
|
||||
@@ -1,22 +0,0 @@
|
||||
#!/bin/sh
|
||||
# One entrypoint, one role per container (US08-02).
|
||||
#
|
||||
# The first argument is the management command the image runs — `serve` and `worker`
|
||||
# are the two roles, and every other `python -m photo_pipeline` command (migrate,
|
||||
# diagnostics, backup, restore, dry-run) is passed through unchanged so operating the
|
||||
# container is operating the same CLI. No supervisor: two roles in one container would
|
||||
# share a process lock they are each meant to hold alone (US07-05).
|
||||
set -eu
|
||||
|
||||
if [ "$(id -u)" = "0" ]; then
|
||||
echo "refusing to run as root: start this image with a non-root UID/GID so files" \
|
||||
"it renames or writes keep the ownership the mounted library expects" >&2
|
||||
exit 1
|
||||
fi
|
||||
|
||||
role="${1:-serve}"
|
||||
# The health check has to know which role it is checking, and only the API has an
|
||||
# endpoint to check. /tmp is writable for the unprivileged user; /run may not be.
|
||||
printf '%s' "${role}" > "${PHOTO_PIPELINE_ROLE_FILE:-/tmp/photo-pipeline-role}" 2>/dev/null || true
|
||||
|
||||
exec python -m photo_pipeline "$@"
|
||||
@@ -1,72 +0,0 @@
|
||||
"""Download one pinned immich-go release and verify it before unpacking (US08-02).
|
||||
|
||||
Run at image build time by the `uploader` stage, with the interpreter that is already
|
||||
in the base image: no curl, no wget, and no download tooling in the layer that ships.
|
||||
The checksum is not advisory — a release asset that does not match the pinned digest
|
||||
is a failed build, not a warning, because the uploader's flags and report format are
|
||||
what the upload parser is written against (concept §15).
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import argparse
|
||||
import hashlib
|
||||
import platform
|
||||
import tarfile
|
||||
import tempfile
|
||||
import urllib.request
|
||||
from pathlib import Path
|
||||
|
||||
RELEASE_URL = "https://github.com/simulot/immich-go/releases/download/v{version}/{asset}"
|
||||
# Debian/BuildKit architecture as the interpreter sees it → release asset name.
|
||||
ASSETS = {
|
||||
"x86_64": ("immich-go_Linux_x86_64.tar.gz", "amd64"),
|
||||
"amd64": ("immich-go_Linux_x86_64.tar.gz", "amd64"),
|
||||
"aarch64": ("immich-go_Linux_arm64.tar.gz", "arm64"),
|
||||
"arm64": ("immich-go_Linux_arm64.tar.gz", "arm64"),
|
||||
}
|
||||
TIMEOUT_SECONDS = 300
|
||||
|
||||
|
||||
def main() -> int:
|
||||
parser = argparse.ArgumentParser(description=__doc__)
|
||||
parser.add_argument("--version", required=True, help="immich-go release, without the v")
|
||||
parser.add_argument("--sha256-amd64", required=True)
|
||||
parser.add_argument("--sha256-arm64", required=True)
|
||||
parser.add_argument("--into", default="/usr/local/bin")
|
||||
args = parser.parse_args()
|
||||
|
||||
machine = platform.machine().lower()
|
||||
if machine not in ASSETS:
|
||||
raise SystemExit(f"unsupported architecture: {machine}")
|
||||
asset, arch = ASSETS[machine]
|
||||
expected = {"amd64": args.sha256_amd64, "arm64": args.sha256_arm64}[arch]
|
||||
url = RELEASE_URL.format(version=args.version, asset=asset)
|
||||
|
||||
with urllib.request.urlopen(url, timeout=TIMEOUT_SECONDS) as response: # noqa: S310
|
||||
payload = response.read()
|
||||
digest = hashlib.sha256(payload).hexdigest()
|
||||
if digest != expected:
|
||||
raise SystemExit(f"checksum mismatch for {url}: {digest} != {expected}")
|
||||
|
||||
target = Path(args.into)
|
||||
target.mkdir(parents=True, exist_ok=True)
|
||||
with tempfile.TemporaryDirectory() as work:
|
||||
archive = Path(work) / asset
|
||||
archive.write_bytes(payload)
|
||||
with tarfile.open(archive) as tar:
|
||||
member = tar.getmember("immich-go")
|
||||
# Extract exactly the one file this pin is about, by name, so nothing
|
||||
# else in the archive can decide where it lands.
|
||||
extracted = tar.extractfile(member)
|
||||
if extracted is None:
|
||||
raise SystemExit("release archive contains no immich-go binary")
|
||||
binary = target / "immich-go"
|
||||
binary.write_bytes(extracted.read())
|
||||
binary.chmod(0o755)
|
||||
print(f"immich-go {args.version} ({arch}) verified {digest}")
|
||||
return 0
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
raise SystemExit(main())
|
||||
@@ -1,34 +0,0 @@
|
||||
#!/bin/sh
|
||||
# Container health for the `serve` role: readiness, not liveness (US08-02).
|
||||
#
|
||||
# /api/v1/health/ready is 503 until the database is reachable, migrated, and in WAL
|
||||
# mode with foreign keys on, so an unmigrated or misconfigured container never reports
|
||||
# healthy. Health endpoints need no session and no access secret, which is what lets an
|
||||
# orchestrator restart a container it holds no credentials for (US08-01).
|
||||
set -eu
|
||||
|
||||
role="$(cat "${PHOTO_PIPELINE_ROLE_FILE:-/tmp/photo-pipeline-role}" 2>/dev/null || echo unknown)"
|
||||
if [ "${role}" != "serve" ]; then
|
||||
# ponytail: the worker has no endpoint to probe; its liveness is its lease and job
|
||||
# heartbeat in the database. Add a `worker --health` command if a restart policy
|
||||
# ever needs to act on it.
|
||||
exit 0
|
||||
fi
|
||||
|
||||
port="${PHOTO_PIPELINE_PORT:-8000}"
|
||||
exec python - "${port}" <<'PY'
|
||||
import sys
|
||||
import urllib.error
|
||||
import urllib.request
|
||||
|
||||
url = f"http://127.0.0.1:{sys.argv[1]}/api/v1/health/ready"
|
||||
try:
|
||||
with urllib.request.urlopen(url, timeout=5) as response: # noqa: S310 — loopback
|
||||
sys.exit(0 if response.status == 200 else 1)
|
||||
except urllib.error.HTTPError as error:
|
||||
print(f"not ready: HTTP {error.code}", file=sys.stderr)
|
||||
sys.exit(1)
|
||||
except OSError as error:
|
||||
print(f"not ready: {error}", file=sys.stderr)
|
||||
sys.exit(1)
|
||||
PY
|
||||
@@ -7,28 +7,9 @@ export const BASE = "/api/v1";
|
||||
// what makes it proof that the caller is this app and not another page.
|
||||
let csrfToken = null;
|
||||
|
||||
// A deployment reachable through a proxy trades an operator secret for that cookie.
|
||||
// Kept per tab: sessionStorage dies with the tab, and the secret never enters a URL.
|
||||
const SECRET_KEY = "pp_access_secret";
|
||||
|
||||
async function bootstrap(secret) {
|
||||
return fetch(BASE + "/session", {
|
||||
credentials: "same-origin",
|
||||
headers: secret ? { "X-Access-Secret": secret } : {},
|
||||
});
|
||||
}
|
||||
|
||||
async function session() {
|
||||
if (csrfToken === null) {
|
||||
let response = await bootstrap(sessionStorage.getItem(SECRET_KEY));
|
||||
if (response.status === 401) {
|
||||
sessionStorage.removeItem(SECRET_KEY);
|
||||
const secret = prompt("Access secret");
|
||||
if (secret) {
|
||||
response = await bootstrap(secret);
|
||||
if (response.ok) sessionStorage.setItem(SECRET_KEY, secret);
|
||||
}
|
||||
}
|
||||
const response = await fetch(BASE + "/session", { credentials: "same-origin" });
|
||||
const body = await response.json().catch(() => null);
|
||||
csrfToken = (body && body.csrf_token) || null;
|
||||
}
|
||||
|
||||
@@ -209,25 +209,15 @@ def main(argv: Sequence[str] | None = None) -> int:
|
||||
lock.release()
|
||||
return 0
|
||||
|
||||
import sys
|
||||
|
||||
import uvicorn
|
||||
|
||||
from photo_pipeline.api.app import ConfigurationRefused, create_app
|
||||
|
||||
# An exposed deployment without an access secret must not reach the port at all,
|
||||
# and the operator needs a sentence, not a traceback (US08-01).
|
||||
try:
|
||||
app = create_app(config)
|
||||
except ConfigurationRefused as error:
|
||||
print(str(error), file=sys.stderr)
|
||||
return 4
|
||||
from photo_pipeline.api.app import create_app
|
||||
|
||||
lock = LibraryLock(config, "api")
|
||||
if (held := _acquire(lock, allow_legacy=args.allow_legacy)) is not None:
|
||||
return held
|
||||
try:
|
||||
uvicorn.run(app, host=config.host, port=config.port)
|
||||
uvicorn.run(create_app(config), host=config.host, port=config.port)
|
||||
finally:
|
||||
lock.release()
|
||||
return 0
|
||||
|
||||
@@ -35,13 +35,7 @@ from photo_pipeline.api.routes import (
|
||||
uploads,
|
||||
workflow,
|
||||
)
|
||||
from photo_pipeline.api.security import (
|
||||
DEFAULT_HEADERS,
|
||||
FailureLimiter,
|
||||
SecurityMiddleware,
|
||||
Session,
|
||||
trust_refusal,
|
||||
)
|
||||
from photo_pipeline.api.security import DEFAULT_HEADERS, SecurityMiddleware, Session
|
||||
|
||||
# Registers the safety_score / analysis job handlers on import.
|
||||
import photo_pipeline.jobs.domain_handlers # noqa: F401
|
||||
@@ -90,16 +84,9 @@ def _install_error_handlers(app: FastAPI) -> None:
|
||||
return _envelope(500, "internal_error", "internal error")
|
||||
|
||||
|
||||
class ConfigurationRefused(RuntimeError):
|
||||
"""The configuration would serve the library to callers it cannot authenticate."""
|
||||
|
||||
|
||||
def create_app(config: Config | None = None) -> FastAPI:
|
||||
config = config or Config.from_env()
|
||||
configure_logging(config.log_level, config.log_format)
|
||||
# Before anything is built, let alone bound to a port (US08-01).
|
||||
if (why := trust_refusal(config)) is not None:
|
||||
raise ConfigurationRefused(why)
|
||||
|
||||
@asynccontextmanager
|
||||
async def lifespan(app: FastAPI):
|
||||
@@ -127,10 +114,6 @@ def create_app(config: Config | None = None) -> FastAPI:
|
||||
# One session per process: the browser exchanges it for a cookie + CSRF token,
|
||||
# and every other origin is refused before a route ever runs (US07-02).
|
||||
app.state.session = Session.create()
|
||||
app.state.access_limiter = FailureLimiter()
|
||||
# Also set in the lifespan, but the bootstrap route reads it, and a caller can
|
||||
# arrive before anything else has touched app.state.
|
||||
app.state.config = config
|
||||
app.add_middleware(SecurityMiddleware, session=app.state.session, config=config)
|
||||
_install_error_handlers(app)
|
||||
app.include_router(session_routes.router, prefix="/api/v1")
|
||||
|
||||
@@ -3,52 +3,21 @@
|
||||
It sets the ``HttpOnly``/``SameSite=Strict`` session cookie and returns the CSRF
|
||||
token in the body. A foreign page can call this — it just cannot read the answer,
|
||||
because the app sends no CORS headers — and the cookie it received is never attached
|
||||
to a request that foreign page initiated.
|
||||
|
||||
When an access secret is configured (mandatory as soon as the app is reachable from
|
||||
another machine, US08-01) this is also the authentication gate: the secret buys the
|
||||
cookie, and every route behind it keeps asking for exactly the session and CSRF token
|
||||
it asked for before. Wrong secrets are counted, and a burst of them stops being
|
||||
answered — otherwise a proxy-exposed deployment could be guessed at indefinitely.
|
||||
to a request that foreign page initiates.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import logging
|
||||
import secrets
|
||||
|
||||
from fastapi import APIRouter, Request
|
||||
from fastapi.responses import JSONResponse
|
||||
|
||||
from photo_pipeline.api.security import ACCESS_SECRET_HEADER, SESSION_COOKIE
|
||||
from photo_pipeline.api.security import SESSION_COOKIE
|
||||
|
||||
router = APIRouter(tags=["session"])
|
||||
|
||||
log = logging.getLogger(__name__)
|
||||
|
||||
|
||||
def _refuse(status: int, code: str, message: str) -> JSONResponse:
|
||||
return JSONResponse(status_code=status, content={"error": {"code": code, "message": message}})
|
||||
|
||||
|
||||
@router.get("/session")
|
||||
def start_session(request: Request) -> JSONResponse:
|
||||
config = request.app.state.config
|
||||
secret = config.access_secret
|
||||
if secret is not None:
|
||||
limiter = request.app.state.access_limiter
|
||||
if limiter.blocked():
|
||||
return _refuse(429, "too_many_attempts", "too many failed attempts; retry later")
|
||||
offered = request.headers.get(ACCESS_SECRET_HEADER, "")
|
||||
if not secrets.compare_digest(offered, secret.get_secret_value()):
|
||||
limiter.record_failure()
|
||||
# The client address is the whole record: the offered secret, the issued
|
||||
# session, and the request body all stay out of the log.
|
||||
log.warning(
|
||||
"access secret rejected", extra={"client": _client(request), "path": "/session"}
|
||||
)
|
||||
return _refuse(401, "access_denied", "a valid access secret is required")
|
||||
|
||||
session = request.app.state.session
|
||||
response = JSONResponse({"csrf_token": session.csrf_token})
|
||||
response.set_cookie(
|
||||
@@ -56,13 +25,6 @@ def start_session(request: Request) -> JSONResponse:
|
||||
session.id,
|
||||
httponly=True,
|
||||
samesite="strict",
|
||||
# HTTPS outside means the cookie must never travel over a plain hop, even one
|
||||
# this process cannot see. Loopback http keeps working unchanged.
|
||||
secure=request.scope.get("state", {}).get("external_scheme") == "https",
|
||||
path="/",
|
||||
)
|
||||
return response
|
||||
|
||||
|
||||
def _client(request: Request) -> str:
|
||||
return request.client.host if request.client else "unknown"
|
||||
|
||||
@@ -19,14 +19,6 @@ The defenses stack, because each one alone has a hole:
|
||||
only in the bootstrap response body, which a foreign page cannot read (no CORS) —
|
||||
so possessing it proves the caller is same-origin.
|
||||
|
||||
Behind a reverse proxy (US08-01) the same stack holds with two substitutions: the
|
||||
allowed host set comes from configuration instead of being the loopback names, and
|
||||
the host/scheme the policy judges is the *external* one, which is only read from
|
||||
``X-Forwarded-*`` when the request actually arrived from a configured proxy. The
|
||||
loopback check was standing in for authentication, so naming a non-loopback host
|
||||
also makes an access secret mandatory — ``trust_refusal`` refuses to start without
|
||||
one, and the secret is what the bootstrap endpoint trades for the session cookie.
|
||||
|
||||
``evaluate`` is a pure function over the request metadata: the whole policy is one
|
||||
table that a unit test can enumerate, and the middleware only applies its verdict.
|
||||
"""
|
||||
@@ -34,7 +26,6 @@ table that a unit test can enumerate, and the middleware only applies its verdic
|
||||
from __future__ import annotations
|
||||
|
||||
import secrets
|
||||
import time
|
||||
from collections.abc import Mapping
|
||||
from dataclasses import dataclass
|
||||
from urllib.parse import urlsplit
|
||||
@@ -44,7 +35,6 @@ from starlette.responses import JSONResponse
|
||||
|
||||
SESSION_COOKIE = "pp_session"
|
||||
CSRF_HEADER = "x-csrf-token"
|
||||
ACCESS_SECRET_HEADER = "x-access-secret"
|
||||
API_PREFIX = "/api/v1"
|
||||
SAFE_METHODS = frozenset({"GET", "HEAD", "OPTIONS"})
|
||||
# Reachable without a session: liveness/readiness (an orchestrator has no cookie)
|
||||
@@ -106,28 +96,6 @@ def split_host(value: str) -> tuple[str, str]:
|
||||
return host, port
|
||||
|
||||
|
||||
def external_view(
|
||||
*,
|
||||
client: str | None,
|
||||
headers: Mapping[str, str],
|
||||
scheme: str,
|
||||
trusted_proxies: frozenset[str],
|
||||
) -> tuple[str, str]:
|
||||
"""The ``(scheme, host)`` the caller used, as opposed to the one this hop saw.
|
||||
|
||||
Forwarded headers are a client-supplied claim. Believing them from anyone lets a
|
||||
request declare its own origin — and origin is half of this module's evidence —
|
||||
so they count only when the connection came from a configured proxy.
|
||||
"""
|
||||
host = headers.get("host", "")
|
||||
if client is None or client not in trusted_proxies:
|
||||
return scheme, host
|
||||
# A chain appends: the first entry is what the original client asked for.
|
||||
forwarded_proto = headers.get("x-forwarded-proto", "").split(",")[0].strip().lower()
|
||||
forwarded_host = headers.get("x-forwarded-host", "").split(",")[0].strip()
|
||||
return forwarded_proto or scheme, forwarded_host or host
|
||||
|
||||
|
||||
def evaluate(
|
||||
*,
|
||||
method: str,
|
||||
@@ -135,26 +103,20 @@ def evaluate(
|
||||
headers: Mapping[str, str],
|
||||
session: Session,
|
||||
allowed_hosts: frozenset[str] = LOOPBACK_HOSTS,
|
||||
scheme: str = "http",
|
||||
max_request_bytes: int,
|
||||
) -> Refusal | None:
|
||||
"""Why this request must be refused, or ``None`` when it may proceed.
|
||||
|
||||
``headers["host"]`` and ``scheme`` are the external ones (see ``external_view``);
|
||||
the allowed origins are the allowed hosts under that scheme and port, so there is
|
||||
no second list that can drift away from the first.
|
||||
"""
|
||||
"""Why this request must be refused, or ``None`` when it may proceed."""
|
||||
host_header = headers.get("host", "")
|
||||
host, port = split_host(host_header)
|
||||
if host.lower() not in allowed_hosts:
|
||||
return Refusal(403, "host_not_allowed", "request host is not an allowed address")
|
||||
return Refusal(403, "host_not_allowed", "request host is not a local address")
|
||||
|
||||
origin = headers.get("origin")
|
||||
if origin is not None and origin != "":
|
||||
parts = urlsplit(origin)
|
||||
origin_host, origin_port = split_host(parts.netloc)
|
||||
if (
|
||||
parts.scheme != scheme
|
||||
parts.scheme not in ("http", "https")
|
||||
or origin_host.lower() not in allowed_hosts
|
||||
or origin_port != port
|
||||
):
|
||||
@@ -180,52 +142,6 @@ def evaluate(
|
||||
return None
|
||||
|
||||
|
||||
def exposed_hosts(config) -> list[str]:
|
||||
"""Configured names by which this application is reachable from another machine."""
|
||||
names = {str(config.host).lower()}
|
||||
names.update(split_host(name)[0].lower() for name in config.allowed_hosts)
|
||||
return sorted(names - LOOPBACK_HOSTS)
|
||||
|
||||
|
||||
def trust_refusal(config) -> str | None:
|
||||
"""Why this configuration must not serve at all, or ``None``.
|
||||
|
||||
Reaching the app used to prove ownership of it. The moment a configuration makes
|
||||
it reachable from elsewhere that stops being true, so serving without a secret
|
||||
would publish the library — refuse at startup rather than at the first request,
|
||||
when the operator is no longer watching (US08-01).
|
||||
"""
|
||||
exposed = exposed_hosts(config)
|
||||
if exposed and config.access_secret is None:
|
||||
return (
|
||||
f"refusing to serve: {', '.join(exposed)} is reachable from outside this "
|
||||
"machine, so PHOTO_PIPELINE_ACCESS_SECRET must be set"
|
||||
)
|
||||
return None
|
||||
|
||||
|
||||
class FailureLimiter:
|
||||
"""Bounded failed access-secret attempts, so the secret cannot be guessed online.
|
||||
|
||||
ponytail: one counter for the whole process rather than per client address —
|
||||
behind a proxy every attempt arrives from the same address anyway. Per-caller
|
||||
buckets if the app is ever exposed without one.
|
||||
"""
|
||||
|
||||
def __init__(self, limit: int = 5, window: float = 60.0) -> None:
|
||||
self.limit = limit
|
||||
self.window = window
|
||||
self._failures: list[float] = []
|
||||
|
||||
def blocked(self) -> bool:
|
||||
now = time.monotonic()
|
||||
self._failures = [at for at in self._failures if now - at < self.window]
|
||||
return len(self._failures) >= self.limit
|
||||
|
||||
def record_failure(self) -> None:
|
||||
self._failures.append(time.monotonic())
|
||||
|
||||
|
||||
class SecurityMiddleware:
|
||||
"""Pure-ASGI so the SSE stream keeps streaming (BaseHTTPMiddleware buffers)."""
|
||||
|
||||
@@ -234,12 +150,7 @@ class SecurityMiddleware:
|
||||
self.session = session
|
||||
self.config = config
|
||||
self.max_request_bytes = config.max_request_bytes
|
||||
self.allowed_hosts = frozenset(
|
||||
LOOPBACK_HOSTS
|
||||
| {str(config.host).lower()}
|
||||
| {split_host(name)[0].lower() for name in config.allowed_hosts}
|
||||
)
|
||||
self.trusted_proxies = frozenset(config.trusted_proxies)
|
||||
self.allowed_hosts = frozenset(LOOPBACK_HOSTS | {str(config.host).lower()})
|
||||
|
||||
async def __call__(self, scope, receive, send) -> None:
|
||||
if scope["type"] != "http":
|
||||
@@ -250,22 +161,12 @@ class SecurityMiddleware:
|
||||
# policy never has to parse a Cookie header.
|
||||
lookup = dict(headers)
|
||||
lookup["cookie-session"] = _cookie(headers.get("cookie", ""), SESSION_COOKIE)
|
||||
client = scope.get("client")
|
||||
scheme, lookup["host"] = external_view(
|
||||
client=client[0] if client else None,
|
||||
headers=headers,
|
||||
scheme=scope.get("scheme", "http"),
|
||||
trusted_proxies=self.trusted_proxies,
|
||||
)
|
||||
# What the session cookie's Secure flag is decided from, one hop later.
|
||||
scope.setdefault("state", {})["external_scheme"] = scheme
|
||||
refusal = evaluate(
|
||||
method=scope.get("method", "GET"),
|
||||
path=scope.get("path", "/"),
|
||||
headers=lookup,
|
||||
session=self.session,
|
||||
allowed_hosts=self.allowed_hosts,
|
||||
scheme=scheme,
|
||||
max_request_bytes=self.max_request_bytes,
|
||||
)
|
||||
if refusal is None:
|
||||
|
||||
@@ -7,9 +7,6 @@ real external call needs them.
|
||||
|
||||
pydantic-settings would do this too, but a prefix-scan over the declared fields
|
||||
is a few lines and one fewer dependency.
|
||||
|
||||
Tuple-valued settings are lists in one variable: library roots are ``os.pathsep``
|
||||
separated because they are paths, everything else is comma separated.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
@@ -21,62 +18,6 @@ from typing import Mapping
|
||||
from pydantic import BaseModel, ConfigDict, SecretStr
|
||||
|
||||
ENV_PREFIX = "PHOTO_PIPELINE_"
|
||||
ENV_FILE_VAR = f"{ENV_PREFIX}ENV_FILE"
|
||||
DEFAULT_ENV_FILE = Path(".env")
|
||||
COMMA_LIST_FIELDS = frozenset({"allowed_hosts", "trusted_proxies"})
|
||||
|
||||
# The archived CLI's variable names, so the configuration file an operator already
|
||||
# has keeps working. The vision provider reads the OpenAI SDK's names, and the
|
||||
# library root is configuration here rather than a bare path (US07-01 donor).
|
||||
LEGACY_ALIASES = {
|
||||
"LLM_API_KEY": "OPENAI_API_KEY",
|
||||
"GEMINI_API_KEY": "OPENAI_API_KEY",
|
||||
"LLM_BASE_URL": "OPENAI_BASE_URL",
|
||||
"LIBRARY": f"{ENV_PREFIX}LIBRARY_ROOTS",
|
||||
}
|
||||
|
||||
|
||||
def parse_env_file(text: str) -> dict[str, str]:
|
||||
"""``KEY=value`` lines into a mapping. Comments, blanks, and quotes handled.
|
||||
|
||||
Deliberately not a shell: no interpolation, no ``export``, no multi-line values.
|
||||
A configuration file that can run code is a configuration file that can be a
|
||||
vulnerability.
|
||||
"""
|
||||
values: dict[str, str] = {}
|
||||
for line in text.splitlines():
|
||||
line = line.strip()
|
||||
if not line or line.startswith("#") or "=" not in line:
|
||||
continue
|
||||
key, _, raw = line.partition("=")
|
||||
key = key.strip()
|
||||
if not key or key.startswith("#"):
|
||||
continue
|
||||
value = raw.strip().strip('"').strip("'")
|
||||
values[key] = value
|
||||
alias = LEGACY_ALIASES.get(key)
|
||||
if alias:
|
||||
values.setdefault(alias, value)
|
||||
return values
|
||||
|
||||
|
||||
def load_env_file(path: Path | str | None = None) -> dict[str, str]:
|
||||
"""Load ``PHOTO_PIPELINE_ENV_FILE`` (or ``./.env``) into the environment.
|
||||
|
||||
Anything already exported wins: a file is the standing configuration, the shell
|
||||
is what you meant *this time*. Returns what it applied, which is what the CLI
|
||||
prints — names only, never values.
|
||||
"""
|
||||
candidate = path or os.environ.get(ENV_FILE_VAR) or DEFAULT_ENV_FILE
|
||||
candidate = Path(candidate)
|
||||
if not candidate.is_file():
|
||||
return {}
|
||||
applied = {}
|
||||
for key, value in parse_env_file(candidate.read_text()).items():
|
||||
if key not in os.environ:
|
||||
os.environ[key] = value
|
||||
applied[key] = value
|
||||
return applied
|
||||
|
||||
|
||||
class Config(BaseModel):
|
||||
@@ -89,19 +30,6 @@ class Config(BaseModel):
|
||||
log_level: str = "INFO"
|
||||
log_format: str = "json" # "json" or "text"
|
||||
|
||||
# Trust boundary (US08-01). Empty means loopback only, which is what the app did
|
||||
# before there was a setting: a request whose Host is not a loopback name is
|
||||
# refused, and no secret is needed because nothing outside this machine can call.
|
||||
# Naming a real hostname here is what makes the app reachable through a reverse
|
||||
# proxy, and it is exactly then that ``access_secret`` becomes mandatory.
|
||||
allowed_hosts: tuple[str, ...] = ()
|
||||
# Addresses whose ``X-Forwarded-Proto``/``X-Forwarded-Host`` may be believed. A
|
||||
# client that is not the proxy can otherwise declare its own origin.
|
||||
trusted_proxies: tuple[str, ...] = ()
|
||||
# Exchanged for the session cookie at the bootstrap endpoint. Once set it is
|
||||
# required even on loopback, so a development setup cannot half-enable it.
|
||||
access_secret: SecretStr | None = None
|
||||
|
||||
# Largest request body the API accepts. Every endpoint takes small JSON commands;
|
||||
# anything larger is a mistake or an attempt to exhaust memory (US07-02).
|
||||
max_request_bytes: int = 1_048_576
|
||||
@@ -140,18 +68,10 @@ class Config(BaseModel):
|
||||
@classmethod
|
||||
def from_env(cls, environ: Mapping[str, str] | None = None) -> "Config":
|
||||
env = os.environ if environ is None else environ
|
||||
if environ is None:
|
||||
load_env_file() # a file never overrides what the shell already set
|
||||
env = os.environ
|
||||
data: dict = {}
|
||||
for name in cls.model_fields:
|
||||
raw = env.get(ENV_PREFIX + name.upper())
|
||||
if not raw:
|
||||
continue
|
||||
if name == "library_roots":
|
||||
data[name] = raw.split(os.pathsep)
|
||||
elif name in COMMA_LIST_FIELDS:
|
||||
data[name] = [part.strip() for part in raw.split(",") if part.strip()]
|
||||
else:
|
||||
data[name] = raw
|
||||
data[name] = raw.split(os.pathsep) if name == "library_roots" else raw
|
||||
return cls(**data)
|
||||
|
||||
@@ -12,10 +12,8 @@ nt-apply-list, pa-nsfw-filter). No dependency on the archived entry points.
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import functools
|
||||
import json
|
||||
import os
|
||||
import shutil
|
||||
import subprocess
|
||||
from collections.abc import Iterable
|
||||
|
||||
@@ -34,27 +32,6 @@ def _timeout() -> float:
|
||||
return DEFAULT_TIMEOUT_SECONDS
|
||||
|
||||
|
||||
def find_binary(binary: str = "exiftool") -> str | None:
|
||||
"""Absolute path of exiftool, or ``None`` when it is not installed."""
|
||||
return shutil.which(binary)
|
||||
|
||||
|
||||
@functools.lru_cache(maxsize=1)
|
||||
def version() -> str | None:
|
||||
"""Reported exiftool version, or ``None`` when it is missing or unusable.
|
||||
|
||||
Cached: it cannot change inside a running process, and diagnostics asks for it
|
||||
on every report (US08-02, where a container image pins this version).
|
||||
"""
|
||||
try:
|
||||
result = subprocess.run(
|
||||
["exiftool", "-ver"], capture_output=True, text=True, timeout=_timeout()
|
||||
)
|
||||
except (OSError, subprocess.SubprocessError):
|
||||
return None
|
||||
return (result.stdout or "").strip() or None
|
||||
|
||||
|
||||
def read_keyword_sets(paths: Iterable[str]) -> dict[str, set[str]]:
|
||||
"""Map each path to its lowercased set of ``Keywords`` + ``Subject`` values.
|
||||
|
||||
|
||||
@@ -15,13 +15,10 @@ application.
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import functools
|
||||
import json
|
||||
import shutil
|
||||
from pathlib import Path
|
||||
|
||||
from photo_pipeline.config import Config
|
||||
from photo_pipeline.integrations import exiftool, immich_go
|
||||
from photo_pipeline.services import app_lock
|
||||
|
||||
# Below this much free space, mutating stages should stop rather than risk a
|
||||
@@ -29,12 +26,6 @@ from photo_pipeline.services import app_lock
|
||||
LOW_DISK_BYTES = 1_000_000_000
|
||||
CRITICAL_DISK_BYTES = 200_000_000
|
||||
|
||||
# Written into the container image at build time (US08-02). The image pins exiftool
|
||||
# and immich-go, and this file is how a running container reports which versions it
|
||||
# was built with — so a drifted or missing binary is visible here rather than in a
|
||||
# failed EXIF checkpoint or a misparsed upload report.
|
||||
IMAGE_VERSIONS_FILE = Path("/etc/photo-pipeline/versions.json")
|
||||
|
||||
|
||||
def _tree_bytes(path: Path) -> int:
|
||||
if not path.exists():
|
||||
@@ -78,47 +69,8 @@ def disk(path: Path) -> dict:
|
||||
}
|
||||
|
||||
|
||||
def _pinned_versions() -> dict[str, str]:
|
||||
"""The versions this image recorded at build time; empty outside a container."""
|
||||
try:
|
||||
recorded = json.loads(IMAGE_VERSIONS_FILE.read_text())
|
||||
except (OSError, ValueError):
|
||||
return {}
|
||||
if not isinstance(recorded, dict):
|
||||
return {}
|
||||
return {str(name): str(value) for name, value in recorded.items()}
|
||||
|
||||
|
||||
@functools.lru_cache(maxsize=4)
|
||||
def _uploader_version(binary: str) -> str | None:
|
||||
"""Cached: the uploader cannot change version inside one process."""
|
||||
return immich_go.version(binary)
|
||||
|
||||
|
||||
def tools(config: Config) -> list[dict]:
|
||||
"""The external executables the pipeline shells out to, and their versions.
|
||||
|
||||
``pinned`` is what the image was built against, ``version`` is what is actually
|
||||
installed. They differ only when the binary was replaced or mounted over.
|
||||
"""
|
||||
return [
|
||||
{
|
||||
"name": "exiftool",
|
||||
"path": exiftool.find_binary(),
|
||||
"version": exiftool.version(),
|
||||
"pinned": _pinned_versions().get("exiftool"),
|
||||
},
|
||||
{
|
||||
"name": "immich-go",
|
||||
"path": immich_go.find_binary(config.immich_go_binary),
|
||||
"version": _uploader_version(config.immich_go_binary),
|
||||
"pinned": _pinned_versions().get("immich-go"),
|
||||
},
|
||||
]
|
||||
|
||||
|
||||
def report(config: Config) -> dict:
|
||||
"""Sizes, disk headroom, tool versions, warnings, and who holds the library lock."""
|
||||
"""Sizes, disk headroom, warnings, and who currently holds the library lock."""
|
||||
database = config.database_path
|
||||
components = [
|
||||
_component("database", database),
|
||||
@@ -180,23 +132,6 @@ def report(config: Config) -> dict:
|
||||
}
|
||||
)
|
||||
|
||||
installed_tools = tools(config)
|
||||
for tool in installed_tools:
|
||||
# A missing tool is reported as ``version: null`` rather than warned about: on a
|
||||
# development machine the uploader is legitimately absent, and the stages that
|
||||
# need it already refuse to run. A *drifted* tool is different — the image pinned
|
||||
# a version and something replaced it.
|
||||
if tool["version"] and tool["pinned"] and tool["pinned"] not in tool["version"]:
|
||||
warnings.append(
|
||||
{
|
||||
"code": "tool_version_drift",
|
||||
"message": (
|
||||
f"{tool['name']} reports {tool['version']} but this image pinned "
|
||||
f"{tool['pinned']}"
|
||||
),
|
||||
}
|
||||
)
|
||||
|
||||
locks = {}
|
||||
for role in ("api", "worker"):
|
||||
holder = app_lock.LibraryLock(config, role).holder()
|
||||
@@ -216,7 +151,6 @@ def report(config: Config) -> dict:
|
||||
"components": components,
|
||||
"total_bytes": sum(component["bytes"] for component in components),
|
||||
"disk": space,
|
||||
"tools": installed_tools,
|
||||
"warnings": warnings,
|
||||
"locks": locks,
|
||||
"legacy_activity": legacy,
|
||||
|
||||
@@ -49,16 +49,8 @@ STAGES: tuple[tuple[str, tuple[str, ...]], ...] = (
|
||||
("browser", ("tests/e2e",)),
|
||||
)
|
||||
|
||||
# Skips the gate accepts, because they describe the machine rather than the code. The
|
||||
# container ones (US08-02) belong here for the same reason exiftool does: the image
|
||||
# build needs a Docker daemon and the network, and its definition is still checked
|
||||
# offline in tests/integration/test_container_image.py.
|
||||
ALLOWED_SKIP_REASONS = (
|
||||
"exiftool not installed",
|
||||
"root ignores directory permissions",
|
||||
"no Docker daemon available",
|
||||
"bind-mount ownership is virtualised",
|
||||
)
|
||||
# Skips the gate accepts, because they describe the machine rather than the code.
|
||||
ALLOWED_SKIP_REASONS = ("exiftool not installed", "root ignores directory permissions")
|
||||
|
||||
|
||||
class ReleaseError(RuntimeError):
|
||||
|
||||
@@ -8,32 +8,18 @@ dependencies = [
|
||||
"sqlalchemy>=2.0",
|
||||
"alembic>=1.13",
|
||||
"pydantic>=2.7",
|
||||
# Imaging is runtime, not test-only: thumbnails decode through Pillow, and the
|
||||
# perceptual hash is a DCT over the decoded pixels (services/hashing.py).
|
||||
"pillow>=10",
|
||||
"numpy>=1.26",
|
||||
"scipy>=1.11",
|
||||
]
|
||||
|
||||
[project.optional-dependencies]
|
||||
# The cloud vision provider. Optional because the analysis stage is the only thing
|
||||
# that needs it, and a local review-only install should not pull an API client.
|
||||
vision = ["openai>=1.30"]
|
||||
test = [
|
||||
"pytest>=8",
|
||||
"httpx>=0.27",
|
||||
"pillow>=10",
|
||||
"numpy>=1.26",
|
||||
"scipy>=1.11",
|
||||
"playwright>=1.40",
|
||||
"pytest-playwright>=0.4",
|
||||
]
|
||||
|
||||
[build-system]
|
||||
requires = ["setuptools>=68"]
|
||||
build-backend = "setuptools.build_meta"
|
||||
|
||||
[tool.setuptools]
|
||||
# The importable application. ``migrations`` and ``work_item`` live beside it but
|
||||
# are not part of the package; without this, an editable install cannot guess.
|
||||
packages = ["photo_pipeline"]
|
||||
# Browser end-to-end tests also require: python -m playwright install chromium
|
||||
|
||||
[tool.ruff]
|
||||
@@ -49,5 +35,4 @@ markers = [
|
||||
"phase_d: Phase D end-to-end acceptance (US04-06) — guarded rename API, fault, and browser journeys",
|
||||
"phase_e: Phase E end-to-end acceptance (US05-06) — upload preflight, uploader, and browser journeys",
|
||||
"phase_f: Phase F end-to-end acceptance (US06-06) — archive destination, transfer, and restore journeys",
|
||||
"container: builds and runs the container image (US08-02) — needs a Docker daemon and network",
|
||||
]
|
||||
|
||||
@@ -1,306 +0,0 @@
|
||||
"""US08-02: the built image, actually built and actually run.
|
||||
|
||||
This is the acceptance test for the image itself, so nothing here is faked: Docker
|
||||
builds from a clean context, the container starts under a chosen UID/GID against a
|
||||
mounted data directory, and the assertions are made over HTTP and against the files
|
||||
the container left on the host.
|
||||
|
||||
It is skipped without a Docker daemon — the build also needs the network for the base
|
||||
image, the pinned exiftool package, and the pinned uploader release. The contract the
|
||||
Dockerfile itself has to keep (pins, non-root, health target, build context) is checked
|
||||
offline in ``tests/integration/test_container_image.py``, so a machine without Docker
|
||||
still fails on a broken image definition; only the running proof needs the daemon. CI
|
||||
builds the image on every change (US08-04), which is where this runs unskipped.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import json
|
||||
import os
|
||||
import platform
|
||||
import re
|
||||
import socket
|
||||
import subprocess
|
||||
import sys
|
||||
import time
|
||||
from pathlib import Path
|
||||
|
||||
import httpx
|
||||
import pytest
|
||||
|
||||
REPO = Path(__file__).resolve().parents[2]
|
||||
IMAGE = "photo-pipeline-test:us08-02"
|
||||
SECRET = "container-acceptance-secret"
|
||||
HOSTNAME = "photos.test"
|
||||
READY_TIMEOUT_SECONDS = 120
|
||||
BUILD_TIMEOUT_SECONDS = 30 * 60
|
||||
|
||||
pytestmark = pytest.mark.container
|
||||
|
||||
|
||||
def docker_available() -> bool:
|
||||
try:
|
||||
return subprocess.run(["docker", "info"], capture_output=True, timeout=60).returncode == 0
|
||||
except (OSError, subprocess.SubprocessError):
|
||||
return False
|
||||
|
||||
|
||||
needs_docker = pytest.mark.skipif(not docker_available(), reason="no Docker daemon available")
|
||||
|
||||
|
||||
def docker(*args: str, check: bool = True, timeout: int = 120) -> subprocess.CompletedProcess:
|
||||
result = subprocess.run(
|
||||
["docker", *args], capture_output=True, text=True, timeout=timeout
|
||||
)
|
||||
if check and result.returncode != 0:
|
||||
raise AssertionError(f"docker {' '.join(args)} failed:\n{result.stdout}\n{result.stderr}")
|
||||
return result
|
||||
|
||||
|
||||
def pins() -> dict[str, str]:
|
||||
"""The pinned versions, read from the Dockerfile that produced the image."""
|
||||
text = (REPO / "Dockerfile").read_text()
|
||||
found = dict(re.findall(r"^ARG\s+([A-Z0-9_]+)=(.+)$", text, re.MULTILINE))
|
||||
return {
|
||||
# The Debian package version carries a packaging suffix; exiftool reports the
|
||||
# upstream version only.
|
||||
"exiftool": found["EXIFTOOL_VERSION"].split("+")[0].split("-")[0],
|
||||
"immich-go": found["IMMICH_GO_VERSION"],
|
||||
}
|
||||
|
||||
|
||||
def free_port() -> int:
|
||||
with socket.socket() as sock:
|
||||
sock.bind(("127.0.0.1", 0))
|
||||
return sock.getsockname()[1]
|
||||
|
||||
|
||||
@pytest.fixture(scope="module")
|
||||
def image() -> str:
|
||||
"""Build from a clean checkout: the build context is the repository, unmodified."""
|
||||
if not docker_available():
|
||||
pytest.skip("no Docker daemon available")
|
||||
docker(
|
||||
"build",
|
||||
"--build-arg",
|
||||
f"UID={os.getuid()}",
|
||||
"--build-arg",
|
||||
f"GID={os.getgid()}",
|
||||
"-t",
|
||||
IMAGE,
|
||||
str(REPO),
|
||||
timeout=BUILD_TIMEOUT_SECONDS,
|
||||
)
|
||||
return IMAGE
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def data_dir(tmp_path) -> Path:
|
||||
data = tmp_path / "data"
|
||||
data.mkdir()
|
||||
return data
|
||||
|
||||
|
||||
def run_detached(image: str, port: int, *args: str, data: Path | None = None) -> str:
|
||||
"""Start a container. ``data`` bind-mounts the host's data directory when the test
|
||||
is about the files themselves; otherwise the image's own /data is used, because a
|
||||
macOS bind mount arrives with an ownership the container did not choose."""
|
||||
result = docker(
|
||||
"run",
|
||||
"--detach",
|
||||
"--rm",
|
||||
"--publish",
|
||||
f"127.0.0.1:{port}:8000",
|
||||
*(("--volume", f"{data}:/data") if data is not None else ()),
|
||||
"--env",
|
||||
# Reachable from outside the container means reachable from another machine as
|
||||
# far as the application is concerned, so the access secret is mandatory
|
||||
# (US08-01) — the image must not weaken that.
|
||||
"PHOTO_PIPELINE_HOST=0.0.0.0",
|
||||
"--env",
|
||||
f"PHOTO_PIPELINE_ACCESS_SECRET={SECRET}",
|
||||
"--env",
|
||||
f"PHOTO_PIPELINE_ALLOWED_HOSTS={HOSTNAME}",
|
||||
image,
|
||||
*args,
|
||||
)
|
||||
return result.stdout.strip()
|
||||
|
||||
|
||||
def wait_until_ready(base: str, container: str) -> None:
|
||||
deadline = time.monotonic() + READY_TIMEOUT_SECONDS
|
||||
while time.monotonic() < deadline:
|
||||
try:
|
||||
if httpx.get(f"{base}/api/v1/health/ready", timeout=5).status_code == 200:
|
||||
return
|
||||
except httpx.HTTPError:
|
||||
pass
|
||||
if docker("inspect", "-f", "{{.State.Running}}", container, check=False).stdout.strip() in (
|
||||
"false",
|
||||
"",
|
||||
):
|
||||
break
|
||||
time.sleep(0.5)
|
||||
logs = docker("logs", container, check=False)
|
||||
raise AssertionError(f"container never became ready:\n{logs.stdout}\n{logs.stderr}")
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def serving(image):
|
||||
port = free_port()
|
||||
container = run_detached(image, port, "serve")
|
||||
try:
|
||||
base = f"http://127.0.0.1:{port}"
|
||||
wait_until_ready(base, container)
|
||||
yield base, container
|
||||
finally:
|
||||
docker("rm", "--force", container, check=False)
|
||||
|
||||
|
||||
def session(base: str) -> httpx.Client:
|
||||
client = httpx.Client(base_url=base, timeout=30)
|
||||
bootstrap = client.get("/api/v1/session", headers={"X-Access-Secret": SECRET})
|
||||
assert bootstrap.status_code == 200, bootstrap.text
|
||||
client.headers["X-CSRF-Token"] = bootstrap.json()["csrf_token"]
|
||||
return client
|
||||
|
||||
|
||||
# ── the image serves, and says what it contains ──────────────────────────────
|
||||
|
||||
|
||||
@needs_docker
|
||||
def test_the_container_serves_the_frontend_and_the_pinned_tool_versions(serving):
|
||||
base, container = serving
|
||||
|
||||
index = httpx.get(f"{base}/app/index.html", timeout=30)
|
||||
assert index.status_code == 200
|
||||
assert "<title" in index.text.lower(), "the application shell, not an API error"
|
||||
|
||||
client = session(base)
|
||||
try:
|
||||
tools = {tool["name"]: tool for tool in client.get("/api/v1/diagnostics").json()["tools"]}
|
||||
finally:
|
||||
client.close()
|
||||
for name, pinned in pins().items():
|
||||
assert tools[name]["pinned"] == pinned, name
|
||||
# Recorded *and* installed: the reported version comes from running the binary.
|
||||
assert pinned in tools[name]["version"], (name, tools[name])
|
||||
assert tools[name]["path"], f"{name} is not on PATH inside the image"
|
||||
|
||||
logs = docker("logs", container, check=False)
|
||||
assert SECRET not in logs.stdout + logs.stderr, "the access secret never reaches the log"
|
||||
|
||||
|
||||
@needs_docker
|
||||
def test_the_declared_health_check_reports_readiness(serving):
|
||||
"""The declared HEALTHCHECK is readiness, so Docker's own verdict is the assertion."""
|
||||
_, container = serving
|
||||
deadline = time.monotonic() + READY_TIMEOUT_SECONDS
|
||||
status = ""
|
||||
while time.monotonic() < deadline:
|
||||
status = docker(
|
||||
"inspect", "-f", "{{.State.Health.Status}}", container, check=False
|
||||
).stdout.strip()
|
||||
if status == "healthy":
|
||||
break
|
||||
time.sleep(1)
|
||||
assert status == "healthy"
|
||||
|
||||
probe = docker("exec", container, "/usr/local/bin/healthcheck.sh", check=False)
|
||||
assert probe.returncode == 0
|
||||
# Point the probe at a port nothing serves: the same script must fail, which is
|
||||
# what makes the healthy verdict above evidence rather than a default.
|
||||
unready = docker(
|
||||
"exec",
|
||||
"--env",
|
||||
"PHOTO_PIPELINE_PORT=1",
|
||||
container,
|
||||
"/usr/local/bin/healthcheck.sh",
|
||||
check=False,
|
||||
)
|
||||
assert unready.returncode != 0
|
||||
|
||||
|
||||
# ── identity: never root, always the configured owner ────────────────────────
|
||||
|
||||
|
||||
@needs_docker
|
||||
def test_the_container_refuses_to_run_as_root(image, data_dir):
|
||||
result = docker(
|
||||
"run",
|
||||
"--rm",
|
||||
"--user",
|
||||
"0:0",
|
||||
"--volume",
|
||||
f"{data_dir}:/data",
|
||||
image,
|
||||
"diagnostics",
|
||||
check=False,
|
||||
)
|
||||
assert result.returncode != 0
|
||||
assert "refusing to run as root" in result.stderr + result.stdout
|
||||
assert not list(data_dir.iterdir()), "a refused container writes nothing"
|
||||
|
||||
|
||||
@needs_docker
|
||||
def test_what_the_container_writes_is_owned_by_the_build_arguments(image):
|
||||
"""The identity the image was built with is the identity on disk afterwards.
|
||||
|
||||
Asserted from inside the container so it holds on every host: a macOS bind mount
|
||||
reports an ownership the container never chose. The host-side proof, which is what
|
||||
the mounted library actually needs, is the Linux test below.
|
||||
"""
|
||||
port = free_port()
|
||||
container = run_detached(image, port, "serve")
|
||||
try:
|
||||
wait_until_ready(f"http://127.0.0.1:{port}", container)
|
||||
owner = docker(
|
||||
"exec", container, "stat", "-c", "%u:%g", "/data/photo_pipeline.db"
|
||||
).stdout.strip()
|
||||
assert owner == f"{os.getuid()}:{os.getgid()}"
|
||||
assert docker("exec", container, "id", "-u").stdout.strip() == str(os.getuid())
|
||||
finally:
|
||||
docker("rm", "--force", container, check=False)
|
||||
|
||||
|
||||
@needs_docker
|
||||
@pytest.mark.skipif(
|
||||
platform.system() != "Linux",
|
||||
reason="bind-mount ownership is virtualised by Docker Desktop on macOS/Windows",
|
||||
)
|
||||
def test_files_the_container_writes_keep_the_configured_ownership(image, data_dir):
|
||||
docker("run", "--rm", "--volume", f"{data_dir}:/data", image, "migrate", timeout=300)
|
||||
|
||||
written = sorted(path for path in data_dir.rglob("*") if path.is_file())
|
||||
assert written, "migrate creates the database in the mounted data directory"
|
||||
for path in written:
|
||||
assert (path.stat().st_uid, path.stat().st_gid) == (os.getuid(), os.getgid()), path
|
||||
|
||||
|
||||
@needs_docker
|
||||
def test_the_worker_role_runs_from_the_same_image(image, data_dir):
|
||||
"""One image, two roles: the worker is the same entrypoint with another argument."""
|
||||
port = free_port()
|
||||
container = run_detached(image, port, "worker", "--id", "container-worker")
|
||||
try:
|
||||
# Taking the worker's library lock is the observable proof that it started,
|
||||
# migrated, and reached its job loop — no sleep required (US07-05).
|
||||
deadline = time.monotonic() + READY_TIMEOUT_SECONDS
|
||||
lock = ""
|
||||
while not lock and time.monotonic() < deadline:
|
||||
assert docker("inspect", "-f", "{{.State.Running}}", container).stdout.strip() == (
|
||||
"true"
|
||||
), docker("logs", container, check=False).stdout
|
||||
lock = docker("exec", container, "cat", "/data/worker.lock.json", check=False).stdout
|
||||
time.sleep(0.5)
|
||||
assert lock, docker("logs", container, check=False).stdout
|
||||
assert json.loads(lock)["role"] == "worker"
|
||||
|
||||
role = docker("exec", container, "cat", "/tmp/photo-pipeline-role").stdout.strip()
|
||||
assert role == "worker", "the health check can tell which role this container is"
|
||||
finally:
|
||||
docker("rm", "--force", container, check=False)
|
||||
|
||||
|
||||
if __name__ == "__main__": # a quick way to run just this file
|
||||
raise SystemExit(pytest.main([__file__, "-v", *sys.argv[1:]]))
|
||||
@@ -1,284 +0,0 @@
|
||||
"""US08-02: the image's build contract, its entrypoint, and its health check.
|
||||
|
||||
Building the image needs a Docker daemon and the network, which is what
|
||||
``tests/e2e/test_container_runtime.py`` does. Everything that can be checked without
|
||||
either is checked here, because the parts most likely to rot silently — a pin that
|
||||
stopped being a pin, a build context that started including the library, a health
|
||||
check pointed at liveness instead of readiness — are all readable from the files.
|
||||
|
||||
The entrypoint and health check are shell, so they are exercised as shell: run with a
|
||||
stubbed ``id`` and ``python`` on ``PATH``, which is enough to prove the refusal, the
|
||||
role marker, and the argument pass-through without a container.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import json
|
||||
import os
|
||||
import re
|
||||
import shutil
|
||||
import stat
|
||||
import subprocess
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
import pytest
|
||||
|
||||
from photo_pipeline.config import Config
|
||||
from photo_pipeline.services import diagnostics
|
||||
|
||||
REPO = Path(__file__).resolve().parents[2]
|
||||
DOCKERFILE = (REPO / "Dockerfile").read_text()
|
||||
DOCKERIGNORE = (REPO / ".dockerignore").read_text()
|
||||
ENTRYPOINT = REPO / "docker" / "entrypoint.sh"
|
||||
HEALTHCHECK = REPO / "docker" / "healthcheck.sh"
|
||||
|
||||
SHA256 = re.compile(r"^[0-9a-f]{64}$")
|
||||
|
||||
|
||||
def instructions(text: str) -> list[str]:
|
||||
"""The lines that do something: comments explain, they do not build."""
|
||||
return [line.strip() for line in text.splitlines() if line.strip() and not line.startswith("#")]
|
||||
|
||||
|
||||
def build_args() -> dict[str, str]:
|
||||
"""Every ``ARG name=default`` in the Dockerfile — the pins, in other words."""
|
||||
found = {}
|
||||
for match in re.finditer(r"^ARG\s+([A-Z0-9_]+)=(.+)$", DOCKERFILE, re.MULTILINE):
|
||||
found[match.group(1)] = match.group(2).strip()
|
||||
return found
|
||||
|
||||
|
||||
# ── pins ─────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_the_base_image_is_pinned_by_version_and_digest():
|
||||
base = build_args()["PYTHON_IMAGE"]
|
||||
assert base.startswith("python:3.12.")
|
||||
assert "@sha256:" in base, "a tag can be moved; a digest cannot"
|
||||
assert ":latest" not in DOCKERFILE
|
||||
# Both stages build from the same pinned base, so the tool that was verified in one
|
||||
# is the tool that ships in the other.
|
||||
assert DOCKERFILE.count("FROM ${PYTHON_IMAGE}") == 2
|
||||
|
||||
|
||||
def test_exiftool_and_the_uploader_are_pinned_and_verified():
|
||||
args = build_args()
|
||||
assert re.match(r"^\d+\.\d+", args["EXIFTOOL_VERSION"])
|
||||
assert re.match(r"^\d+\.\d+\.\d+$", args["IMMICH_GO_VERSION"])
|
||||
for arch in ("AMD64", "ARM64"):
|
||||
assert SHA256.match(args[f"IMMICH_GO_SHA256_{arch}"]), arch
|
||||
# The pinned exiftool package is installed by version, not by name alone.
|
||||
assert 'libimage-exiftool-perl=${EXIFTOOL_VERSION}"' in DOCKERFILE
|
||||
# And the build fails if what got installed is not what was pinned.
|
||||
assert "is not the pinned" in DOCKERFILE and "is not pinned" in DOCKERFILE
|
||||
|
||||
|
||||
def test_the_uploader_download_refuses_a_mismatching_checksum(tmp_path):
|
||||
"""The verification is the point of pinning a URL, so it is run, not read."""
|
||||
script = REPO / "docker" / "fetch-immich-go.py"
|
||||
result = subprocess.run(
|
||||
[
|
||||
sys.executable,
|
||||
str(script),
|
||||
"--version",
|
||||
"0.0.0-does-not-exist",
|
||||
"--sha256-amd64",
|
||||
"0" * 64,
|
||||
"--sha256-arm64",
|
||||
"0" * 64,
|
||||
"--into",
|
||||
str(tmp_path),
|
||||
],
|
||||
capture_output=True,
|
||||
text=True,
|
||||
)
|
||||
assert result.returncode != 0
|
||||
assert not list(tmp_path.iterdir()), "nothing is written before it is verified"
|
||||
|
||||
|
||||
def test_the_recorded_versions_are_reported_by_diagnostics(tmp_path, monkeypatch):
|
||||
"""What the image records is what `diagnostics` answers with (acceptance criterion 2)."""
|
||||
recorded = tmp_path / "versions.json"
|
||||
recorded.write_text(json.dumps({"exiftool": "13.25", "immich-go": "0.32.0"}))
|
||||
monkeypatch.setattr(diagnostics, "IMAGE_VERSIONS_FILE", recorded)
|
||||
monkeypatch.setattr(diagnostics.exiftool, "version", lambda: "13.25")
|
||||
monkeypatch.setattr(diagnostics, "_uploader_version", lambda _binary: "immich-go 0.32.0")
|
||||
|
||||
config = Config(data_dir=tmp_path / "data")
|
||||
reported = {tool["name"]: tool for tool in diagnostics.tools(config)}
|
||||
|
||||
assert reported["exiftool"]["pinned"] == "13.25"
|
||||
assert reported["immich-go"]["pinned"] == "0.32.0"
|
||||
assert "0.32.0" in reported["immich-go"]["version"]
|
||||
assert diagnostics.report(config)["tools"] == list(reported.values())
|
||||
assert "tool_version_drift" not in {w["code"] for w in diagnostics.report(config)["warnings"]}
|
||||
|
||||
|
||||
def test_a_replaced_tool_is_reported_as_drift(tmp_path, monkeypatch):
|
||||
recorded = tmp_path / "versions.json"
|
||||
recorded.write_text(json.dumps({"exiftool": "13.25"}))
|
||||
monkeypatch.setattr(diagnostics, "IMAGE_VERSIONS_FILE", recorded)
|
||||
monkeypatch.setattr(diagnostics.exiftool, "version", lambda: "12.57")
|
||||
|
||||
report = diagnostics.report(Config(data_dir=tmp_path / "data"))
|
||||
|
||||
drift = [w for w in report["warnings"] if w["code"] == "tool_version_drift"]
|
||||
assert drift and "13.25" in drift[0]["message"] and "12.57" in drift[0]["message"]
|
||||
|
||||
|
||||
def test_versions_are_absent_rather_than_invented_outside_a_container(tmp_path, monkeypatch):
|
||||
monkeypatch.setattr(diagnostics, "IMAGE_VERSIONS_FILE", tmp_path / "nothing.json")
|
||||
for tool in diagnostics.tools(Config(data_dir=tmp_path / "data")):
|
||||
assert tool["pinned"] is None
|
||||
|
||||
|
||||
# ── the final layer ──────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
def test_no_test_or_build_tooling_is_installed_in_the_image():
|
||||
runtime = "\n".join(instructions(DOCKERFILE.split("AS runtime", 1)[1]))
|
||||
for unwanted in ("[test]", "pytest", "playwright", "build-essential", "gcc"):
|
||||
assert unwanted not in runtime, unwanted
|
||||
assert "pip install --no-cache-dir -e ." in runtime
|
||||
|
||||
|
||||
def test_neither_secrets_nor_library_data_can_enter_the_build_context():
|
||||
lines = instructions(DOCKERIGNORE)
|
||||
assert lines[0] == "*", "the context is deny-by-default"
|
||||
allowed = {line[1:] for line in lines if line.startswith("!")}
|
||||
# Everything the Dockerfile copies has to be allowed, and nothing else is.
|
||||
copied = {
|
||||
source
|
||||
for match in re.finditer(r"^COPY (?!--from)(.+)$", DOCKERFILE, re.MULTILINE)
|
||||
for source in match.group(1).split()[:-1]
|
||||
}
|
||||
assert {Path(source).parts[0] for source in copied} <= allowed
|
||||
assert not {"data", ".git", ".env", "tests", ".venv"} & allowed
|
||||
for generated in ("**/*.env", "**/*.db", "**/*.log", "**/__pycache__"):
|
||||
assert generated in lines, generated
|
||||
|
||||
|
||||
def test_the_image_runs_as_a_non_root_user_whose_ids_are_build_arguments():
|
||||
args = build_args()
|
||||
assert args["UID"] == "1000" and args["GID"] == "1000"
|
||||
assert "USER ${UID}:${GID}" in DOCKERFILE
|
||||
assert re.search(r"^USER (root|0)", DOCKERFILE, re.MULTILINE) is None
|
||||
assert 'useradd --uid "${UID}" --gid "${GID}"' in DOCKERFILE
|
||||
|
||||
|
||||
def test_the_health_check_is_readiness_and_the_default_role_is_serve():
|
||||
assert "HEALTHCHECK" in DOCKERFILE
|
||||
assert "/usr/local/bin/healthcheck.sh" in DOCKERFILE
|
||||
assert 'CMD ["serve"]' in DOCKERFILE
|
||||
assert 'ENTRYPOINT ["/usr/local/bin/entrypoint.sh"]' in DOCKERFILE
|
||||
assert "/api/v1/health/ready" in HEALTHCHECK.read_text()
|
||||
assert "/api/v1/health/live" not in HEALTHCHECK.read_text()
|
||||
# No supervisor: one role per container (acceptance criterion 4).
|
||||
for supervisor in ("supervisord", "s6-overlay", "runit"):
|
||||
assert supervisor not in DOCKERFILE
|
||||
|
||||
|
||||
@pytest.mark.parametrize("script", [ENTRYPOINT, HEALTHCHECK])
|
||||
def test_the_scripts_are_executable(script):
|
||||
assert script.stat().st_mode & stat.S_IXUSR, f"{script.name} must be executable in git"
|
||||
|
||||
|
||||
# ── the entrypoint, run as shell ─────────────────────────────────────────────
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def stubs(tmp_path):
|
||||
"""A PATH where ``python`` records its arguments and ``id`` can be told a UID."""
|
||||
bin_dir = tmp_path / "bin"
|
||||
bin_dir.mkdir()
|
||||
recorded = tmp_path / "argv"
|
||||
python = bin_dir / "python"
|
||||
python.write_text(f'#!/bin/sh\nprintf "%s\\n" "$@" > {recorded}\nexit 0\n')
|
||||
python.chmod(0o755)
|
||||
(bin_dir / "id").write_text('#!/bin/sh\nprintf "%s" "${STUB_UID:-1000}"\n')
|
||||
(bin_dir / "id").chmod(0o755)
|
||||
return bin_dir, recorded, tmp_path / "role"
|
||||
|
||||
|
||||
def run_script(script: Path, *args, stubs, env=None):
|
||||
bin_dir, recorded, role_file = stubs
|
||||
result = subprocess.run(
|
||||
["/bin/sh", str(script), *args],
|
||||
capture_output=True,
|
||||
text=True,
|
||||
env={
|
||||
"PATH": f"{bin_dir}:{os.environ['PATH']}",
|
||||
"PHOTO_PIPELINE_ROLE_FILE": str(role_file),
|
||||
**(env or {}),
|
||||
},
|
||||
)
|
||||
argv = recorded.read_text().splitlines() if recorded.exists() else []
|
||||
return result, argv
|
||||
|
||||
|
||||
def test_the_container_refuses_to_run_as_root(stubs):
|
||||
result, argv = run_script(ENTRYPOINT, "serve", stubs=stubs, env={"STUB_UID": "0"})
|
||||
assert result.returncode == 1
|
||||
assert "refusing to run as root" in result.stderr
|
||||
assert argv == [], "the application is never started as root"
|
||||
assert not stubs[2].exists(), "not even the role marker is written"
|
||||
|
||||
|
||||
@pytest.mark.parametrize(
|
||||
"given,expected",
|
||||
[
|
||||
(["serve"], ["-m", "photo_pipeline", "serve"]),
|
||||
(["worker", "--id", "worker-2"], ["-m", "photo_pipeline", "worker", "--id", "worker-2"]),
|
||||
# Every other management command stays reachable: operating the container is
|
||||
# operating the same CLI.
|
||||
(["diagnostics"], ["-m", "photo_pipeline", "diagnostics"]),
|
||||
([], ["-m", "photo_pipeline"]),
|
||||
],
|
||||
)
|
||||
def test_the_role_selects_the_command_and_arguments_pass_through(given, expected, stubs):
|
||||
result, argv = run_script(ENTRYPOINT, *given, stubs=stubs)
|
||||
assert result.returncode == 0, result.stderr
|
||||
assert argv == expected
|
||||
|
||||
|
||||
def test_the_role_is_recorded_for_the_health_check(stubs):
|
||||
run_script(ENTRYPOINT, "worker", stubs=stubs)
|
||||
assert stubs[2].read_text() == "worker"
|
||||
|
||||
|
||||
def test_the_health_check_only_probes_the_serving_role(stubs):
|
||||
stubs[2].write_text("worker")
|
||||
result, argv = run_script(HEALTHCHECK, stubs=stubs)
|
||||
assert result.returncode == 0 and argv == [], "a worker has no endpoint to probe"
|
||||
|
||||
stubs[2].write_text("serve")
|
||||
result, argv = run_script(HEALTHCHECK, stubs=stubs, env={"PHOTO_PIPELINE_PORT": "9123"})
|
||||
assert result.returncode == 0, result.stderr
|
||||
assert argv == ["-", "9123"], "the configured port is the one probed"
|
||||
|
||||
|
||||
def test_the_health_check_fails_while_the_api_is_not_ready(stubs):
|
||||
"""No python stub: the real interpreter probes a port nothing is listening on."""
|
||||
stubs[2].write_text("serve")
|
||||
result = subprocess.run(
|
||||
["/bin/sh", str(HEALTHCHECK)],
|
||||
capture_output=True,
|
||||
text=True,
|
||||
env={
|
||||
"PATH": os.path.dirname(sys.executable) + os.pathsep + os.environ["PATH"],
|
||||
"PHOTO_PIPELINE_ROLE_FILE": str(stubs[2]),
|
||||
"PHOTO_PIPELINE_PORT": "1",
|
||||
},
|
||||
)
|
||||
assert result.returncode == 1
|
||||
assert "not ready" in result.stderr
|
||||
|
||||
|
||||
def test_the_scripts_are_posix_shell():
|
||||
"""They run in the image's /bin/sh, which is dash — not bash."""
|
||||
shells = ["/bin/sh"] + ([dash] if (dash := shutil.which("dash")) else [])
|
||||
for shell in shells:
|
||||
for script in (ENTRYPOINT, HEALTHCHECK):
|
||||
checked = subprocess.run([shell, "-n", str(script)], capture_output=True, text=True)
|
||||
assert checked.returncode == 0, f"{shell} {script.name}: {checked.stderr}"
|
||||
@@ -1,253 +0,0 @@
|
||||
"""US08-01: the configurable trust boundary and its authentication gate.
|
||||
|
||||
Until now, reaching the app proved ownership of it: it answered only to loopback
|
||||
names. A container behind a reverse proxy answers to a real hostname, so these tests
|
||||
pin the two halves that replace that proof — the app refuses to start exposed without
|
||||
an access secret, and the secret is the only way to obtain the session every other
|
||||
route already required (US07-02, unchanged and re-asserted here).
|
||||
|
||||
The suite's ``conftest`` bootstraps a session for any ``TestClient`` automatically,
|
||||
which is precisely what an unauthenticated caller does not get; ``raw_client``
|
||||
pre-seeds a placeholder CSRF header to opt out of that convenience.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import pytest
|
||||
from starlette.testclient import TestClient
|
||||
|
||||
from photo_pipeline.api.app import ConfigurationRefused, create_app
|
||||
from photo_pipeline.api.security import ACCESS_SECRET_HEADER, CSRF_HEADER, SESSION_COOKIE
|
||||
from photo_pipeline.config import Config
|
||||
|
||||
SECRET = "operator-secret-value"
|
||||
HOSTNAME = "photos.example.com"
|
||||
# What Starlette reports as the peer address of an in-process request.
|
||||
TESTCLIENT_ADDRESS = "testclient"
|
||||
|
||||
# One of each route class: a read, a mutation, and a media endpoint.
|
||||
PROTECTED = [
|
||||
("GET", "/api/v1/workflow", None),
|
||||
("POST", "/api/v1/albums/proposals", {}),
|
||||
("GET", "/api/v1/assets/unknown-asset/thumbnail?size=256", None),
|
||||
]
|
||||
|
||||
|
||||
def config(tmp_path, **overrides) -> Config:
|
||||
return Config(data_dir=tmp_path / "data", **overrides)
|
||||
|
||||
|
||||
def raw_client(app, base_url="http://127.0.0.1") -> TestClient:
|
||||
client = TestClient(app, base_url=base_url)
|
||||
client.headers[CSRF_HEADER] = "placeholder"
|
||||
return client
|
||||
|
||||
|
||||
def exchange(client, secret=SECRET, headers=None):
|
||||
return client.get("/api/v1/session", headers={ACCESS_SECRET_HEADER: secret, **(headers or {})})
|
||||
|
||||
|
||||
# ── startup: exposure without a secret is refused, loopback is unchanged ──────
|
||||
|
||||
|
||||
@pytest.mark.parametrize(
|
||||
"exposure,exposed",
|
||||
[({"allowed_hosts": (HOSTNAME,)}, HOSTNAME), ({"host": "0.0.0.0"}, "0.0.0.0")],
|
||||
)
|
||||
def test_an_exposed_configuration_refuses_to_serve_without_a_secret(tmp_path, exposure, exposed):
|
||||
with pytest.raises(ConfigurationRefused) as refused:
|
||||
create_app(config(tmp_path, **exposure))
|
||||
assert "PHOTO_PIPELINE_ACCESS_SECRET" in str(refused.value)
|
||||
# The message names what is exposed, so the operator knows which setting did it.
|
||||
assert exposed in str(refused.value)
|
||||
|
||||
|
||||
def test_the_serve_command_reports_the_refusal_instead_of_binding(tmp_path, monkeypatch, capsys):
|
||||
"""Exit before the port, the lock, and the database, with a sentence not a trace."""
|
||||
from photo_pipeline.__main__ import main
|
||||
|
||||
monkeypatch.setenv("PHOTO_PIPELINE_DATA_DIR", str(tmp_path / "data"))
|
||||
monkeypatch.setenv("PHOTO_PIPELINE_ALLOWED_HOSTS", HOSTNAME)
|
||||
monkeypatch.delenv("PHOTO_PIPELINE_ACCESS_SECRET", raising=False)
|
||||
|
||||
assert main(["serve"]) == 4
|
||||
assert "PHOTO_PIPELINE_ACCESS_SECRET" in capsys.readouterr().err
|
||||
|
||||
|
||||
def test_an_exposed_configuration_with_a_secret_starts(tmp_path):
|
||||
app = create_app(config(tmp_path, allowed_hosts=(HOSTNAME,), access_secret=SECRET))
|
||||
with raw_client(app, base_url=f"http://{HOSTNAME}") as client:
|
||||
assert exchange(client).status_code == 200
|
||||
|
||||
|
||||
def test_a_loopback_configuration_still_needs_no_secret(tmp_path):
|
||||
"""An unset trust boundary must behave exactly as it did before this story."""
|
||||
with raw_client(create_app(config(tmp_path))) as client:
|
||||
response = client.get("/api/v1/session")
|
||||
assert response.status_code == 200
|
||||
assert response.json()["csrf_token"]
|
||||
assert "secure" not in response.headers["set-cookie"].lower()
|
||||
|
||||
|
||||
# ── the exchange: secret in, session out ─────────────────────────────────────
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def gated(tmp_path):
|
||||
app = create_app(
|
||||
config(
|
||||
tmp_path,
|
||||
allowed_hosts=(HOSTNAME,),
|
||||
access_secret=SECRET,
|
||||
trusted_proxies=(TESTCLIENT_ADDRESS,),
|
||||
)
|
||||
)
|
||||
with raw_client(app, base_url=f"http://{HOSTNAME}") as client:
|
||||
yield client
|
||||
|
||||
|
||||
def test_the_secret_buys_the_session_and_the_session_buys_the_routes(gated):
|
||||
response = exchange(gated)
|
||||
assert response.status_code == 200
|
||||
cookie = response.headers["set-cookie"].lower()
|
||||
assert "httponly" in cookie and "samesite=strict" in cookie
|
||||
gated.headers[CSRF_HEADER] = response.json()["csrf_token"]
|
||||
|
||||
# The session and CSRF requirements behind the gate are the ones US07-02 set.
|
||||
assert gated.get("/api/v1/workflow").status_code == 200
|
||||
assert gated.post("/api/v1/albums/proposals", json={}).status_code == 200
|
||||
refused = gated.post("/api/v1/albums/proposals", json={}, headers={CSRF_HEADER: "guessed"})
|
||||
assert refused.json()["error"]["code"] == "csrf_failed"
|
||||
|
||||
|
||||
@pytest.mark.parametrize("offered", ["", "wrong-secret", SECRET + "x", SECRET.upper()])
|
||||
def test_a_wrong_secret_buys_nothing(gated, offered):
|
||||
response = exchange(gated, secret=offered)
|
||||
assert response.status_code == 401
|
||||
assert response.json()["error"]["code"] == "access_denied"
|
||||
assert "set-cookie" not in response.headers
|
||||
|
||||
|
||||
def test_a_refusal_never_echoes_the_secret_or_the_session(gated, caplog):
|
||||
with caplog.at_level("WARNING"):
|
||||
response = exchange(gated, secret="wrong-secret")
|
||||
assert SECRET not in response.text and "wrong-secret" not in response.text
|
||||
assert SECRET not in caplog.text
|
||||
# Logged as an event with its caller, without the session it did not get.
|
||||
assert "access secret rejected" in caplog.text
|
||||
|
||||
|
||||
def test_guessing_is_rate_limited(gated):
|
||||
codes = [exchange(gated, secret=f"guess-{n}").status_code for n in range(6)]
|
||||
assert codes.count(401) == 5 and codes[-1] == 429
|
||||
assert gated.get("/api/v1/session").status_code == 429
|
||||
# The right secret is refused too while the limiter holds: that is the point.
|
||||
blocked = exchange(gated)
|
||||
assert blocked.status_code == 429
|
||||
assert SECRET not in blocked.text
|
||||
|
||||
|
||||
def test_every_route_class_is_unreachable_without_the_secret(gated):
|
||||
for method, path, body in PROTECTED:
|
||||
response = gated.request(method, path, json=body)
|
||||
assert response.status_code == 401, path
|
||||
assert response.json()["error"]["code"] == "unauthenticated", path
|
||||
# Health stays open: an orchestrator restarting the container holds no secret.
|
||||
assert gated.get("/api/v1/health/live").status_code == 200
|
||||
assert gated.get("/api/v1/health/ready").status_code == 200
|
||||
|
||||
|
||||
def test_a_session_from_another_process_is_not_replayable(tmp_path):
|
||||
"""Sessions live in the process, so a cookie captured from a previous one — a
|
||||
restarted container, or a second deployment — must not open this one."""
|
||||
settings = dict(allowed_hosts=(HOSTNAME,), access_secret=SECRET)
|
||||
first, second = (create_app(config(tmp_path / str(n), **settings)) for n in (1, 2))
|
||||
with raw_client(first, base_url=f"http://{HOSTNAME}") as client:
|
||||
exchange(client)
|
||||
stolen = client.cookies[SESSION_COOKIE]
|
||||
|
||||
with raw_client(second, base_url=f"http://{HOSTNAME}") as client:
|
||||
client.cookies.set(SESSION_COOKIE, stolen, domain=HOSTNAME)
|
||||
response = client.get("/api/v1/workflow")
|
||||
assert response.status_code == 401
|
||||
assert response.json()["error"]["code"] == "unauthenticated"
|
||||
|
||||
|
||||
def test_a_cross_site_request_is_still_refused_behind_the_gate(gated):
|
||||
gated.headers[CSRF_HEADER] = exchange(gated).json()["csrf_token"]
|
||||
refused = gated.post(
|
||||
"/api/v1/albums/proposals", json={}, headers={"Origin": "https://evil.example"}
|
||||
)
|
||||
assert refused.json()["error"]["code"] == "origin_not_allowed"
|
||||
embedded = gated.get(
|
||||
"/api/v1/assets/unknown-asset/thumbnail?size=256", headers={"Sec-Fetch-Site": "cross-site"}
|
||||
)
|
||||
assert embedded.json()["error"]["code"] == "cross_site_blocked"
|
||||
|
||||
|
||||
def test_an_unconfigured_host_is_refused_even_with_a_valid_session(gated):
|
||||
gated.headers[CSRF_HEADER] = exchange(gated).json()["csrf_token"]
|
||||
for host in ("other.example.com", "192.168.1.10"):
|
||||
response = gated.get("/api/v1/workflow", headers={"Host": host})
|
||||
assert response.status_code == 403, host
|
||||
assert response.json()["error"]["code"] == "host_not_allowed", host
|
||||
|
||||
|
||||
# ── forwarded headers: believed from the proxy, ignored from anyone else ──────
|
||||
|
||||
|
||||
def test_a_trusted_proxys_https_makes_the_cookie_secure(gated):
|
||||
"""The proxy speaks HTTPS outward and HTTP to this app, so only the header knows."""
|
||||
assert "secure" in exchange(gated, headers={"X-Forwarded-Proto": "https"}).headers[
|
||||
"set-cookie"
|
||||
].lower()
|
||||
assert "secure" not in exchange(gated).headers["set-cookie"].lower()
|
||||
|
||||
|
||||
def test_the_external_scheme_is_part_of_the_accepted_origin(gated):
|
||||
gated.headers[CSRF_HEADER] = exchange(gated).json()["csrf_token"]
|
||||
allowed = gated.post(
|
||||
"/api/v1/albums/proposals",
|
||||
json={},
|
||||
headers={"X-Forwarded-Proto": "https", "Origin": f"https://{HOSTNAME}"},
|
||||
)
|
||||
assert allowed.status_code == 200
|
||||
# The scheme is part of the origin: the same name over plain HTTP is not it.
|
||||
refused = gated.post(
|
||||
"/api/v1/albums/proposals",
|
||||
json={},
|
||||
headers={"X-Forwarded-Proto": "https", "Origin": f"http://{HOSTNAME}"},
|
||||
)
|
||||
assert refused.json()["error"]["code"] == "origin_not_allowed"
|
||||
|
||||
|
||||
def test_a_trusted_proxys_forwarded_host_is_the_host_that_is_judged(tmp_path):
|
||||
"""The proxy terminates the operator's hostname and dials this app by address."""
|
||||
app = create_app(
|
||||
config(
|
||||
tmp_path,
|
||||
allowed_hosts=(HOSTNAME,),
|
||||
access_secret=SECRET,
|
||||
trusted_proxies=(TESTCLIENT_ADDRESS,),
|
||||
)
|
||||
)
|
||||
with raw_client(app, base_url="http://10.0.0.5") as client:
|
||||
forwarded = {"X-Forwarded-Host": HOSTNAME}
|
||||
assert exchange(client, headers=forwarded).status_code == 200
|
||||
# Without the header the address it was dialled by is not an allowed name.
|
||||
assert exchange(client).json()["error"]["code"] == "host_not_allowed"
|
||||
|
||||
|
||||
def test_forwarded_headers_from_an_untrusted_client_are_ignored(tmp_path):
|
||||
"""Otherwise any caller could declare the hostname and scheme of its choosing."""
|
||||
app = create_app(config(tmp_path, allowed_hosts=(HOSTNAME,), access_secret=SECRET))
|
||||
with raw_client(app, base_url="http://evil.example") as client:
|
||||
forged = exchange(client, headers={"X-Forwarded-Host": HOSTNAME})
|
||||
assert forged.json()["error"]["code"] == "host_not_allowed"
|
||||
|
||||
with raw_client(app, base_url=f"http://{HOSTNAME}") as client:
|
||||
# A forged scheme would flip the cookie's Secure flag on a plain connection,
|
||||
# which is how a cookie gets set and then never sent again.
|
||||
response = exchange(client, headers={"X-Forwarded-Proto": "https"})
|
||||
assert response.status_code == 200
|
||||
assert "secure" not in response.headers["set-cookie"].lower()
|
||||
@@ -15,10 +15,9 @@
|
||||
"tests/characterization/test_webapp_query.py"
|
||||
],
|
||||
"US01-02": [
|
||||
"tests/integration/test_app_lifecycle.py",
|
||||
"tests/integration/test_migrations.py",
|
||||
"tests/unit/test_config.py",
|
||||
"tests/unit/test_env_file.py"
|
||||
"tests/integration/test_migrations.py",
|
||||
"tests/integration/test_app_lifecycle.py"
|
||||
],
|
||||
"US01-03": [
|
||||
"tests/unit/test_path_policy.py",
|
||||
@@ -174,17 +173,11 @@
|
||||
"US07-07": [
|
||||
"tests/e2e/test_release_gate.py",
|
||||
"tests/e2e/test_release_journey.py"
|
||||
],
|
||||
"US08-01": [
|
||||
"tests/unit/test_security_policy.py",
|
||||
"tests/integration/test_trusted_hosts.py"
|
||||
],
|
||||
"US08-02": [
|
||||
"tests/integration/test_container_image.py",
|
||||
"tests/e2e/test_container_runtime.py"
|
||||
]
|
||||
},
|
||||
"planned": [
|
||||
"US08-01",
|
||||
"US08-02",
|
||||
"US08-03",
|
||||
"US08-04",
|
||||
"US08-05"
|
||||
|
||||
@@ -1,69 +0,0 @@
|
||||
"""Configuration from a dotenv file, including the archived CLI's variable names.
|
||||
|
||||
An operator who already has a ``photo_analyzer.env`` should not have to rewrite it
|
||||
to run the application it was replaced by. The file is standing configuration; the
|
||||
shell is what you meant this time, so the shell always wins.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import os
|
||||
|
||||
from photo_pipeline.config import Config, load_env_file, parse_env_file
|
||||
|
||||
SAMPLE = """
|
||||
# The archived CLI's shape, comments and all.
|
||||
LLM_API_KEY=not-real
|
||||
LLM_BASE_URL="https://example.invalid/v1beta/openai/"
|
||||
LLM_MODEL='gemini-2.5-flash'
|
||||
LIBRARY=/tmp/pictures
|
||||
|
||||
MAX_WORKERS=4
|
||||
# commented=ignored
|
||||
malformed line without an equals sign
|
||||
"""
|
||||
|
||||
|
||||
def test_the_file_is_parsed_and_never_executed():
|
||||
values = parse_env_file(SAMPLE)
|
||||
|
||||
assert values["LLM_BASE_URL"] == "https://example.invalid/v1beta/openai/" # quotes stripped
|
||||
assert values["LLM_MODEL"] == "gemini-2.5-flash"
|
||||
assert values["MAX_WORKERS"] == "4"
|
||||
assert "commented" not in values and "malformed line without an equals sign" not in values
|
||||
|
||||
|
||||
def test_the_archived_cli_names_still_configure_the_application():
|
||||
values = parse_env_file(SAMPLE)
|
||||
|
||||
assert values["OPENAI_API_KEY"] == "not-real"
|
||||
assert values["OPENAI_BASE_URL"] == "https://example.invalid/v1beta/openai/"
|
||||
assert values["PHOTO_PIPELINE_LIBRARY_ROOTS"] == "/tmp/pictures"
|
||||
|
||||
|
||||
def test_an_explicit_shell_variable_beats_the_file(tmp_path, monkeypatch):
|
||||
path = tmp_path / "photo_analyzer.env"
|
||||
path.write_text(SAMPLE)
|
||||
monkeypatch.setenv("OPENAI_API_KEY", "from-the-shell")
|
||||
monkeypatch.delenv("PHOTO_PIPELINE_LIBRARY_ROOTS", raising=False)
|
||||
|
||||
applied = load_env_file(path)
|
||||
|
||||
assert "OPENAI_API_KEY" not in applied, "the file overrode an exported value"
|
||||
assert os.environ["OPENAI_API_KEY"] == "from-the-shell"
|
||||
assert os.environ["PHOTO_PIPELINE_LIBRARY_ROOTS"] == "/tmp/pictures"
|
||||
assert Config.from_env().library_roots[0].name == "pictures"
|
||||
|
||||
|
||||
def test_the_file_is_found_through_its_variable(tmp_path, monkeypatch):
|
||||
path = tmp_path / "custom.env"
|
||||
path.write_text("PHOTO_PIPELINE_PORT=9123\n")
|
||||
monkeypatch.delenv("PHOTO_PIPELINE_PORT", raising=False)
|
||||
monkeypatch.setenv("PHOTO_PIPELINE_ENV_FILE", str(path))
|
||||
monkeypatch.chdir(tmp_path) # no ./.env here, so only the variable can find it
|
||||
|
||||
assert Config.from_env().port == 9123
|
||||
|
||||
|
||||
def test_a_missing_file_is_not_an_error(tmp_path):
|
||||
assert load_env_file(tmp_path / "nothing-here.env") == {}
|
||||
@@ -13,31 +13,18 @@ import pytest
|
||||
|
||||
from photo_pipeline.api.security import (
|
||||
CSRF_HEADER,
|
||||
LOOPBACK_HOSTS,
|
||||
PUBLIC_PATHS,
|
||||
FailureLimiter,
|
||||
Session,
|
||||
evaluate,
|
||||
exposed_hosts,
|
||||
external_view,
|
||||
split_host,
|
||||
trust_refusal,
|
||||
)
|
||||
from photo_pipeline.config import Config
|
||||
|
||||
SESSION = Session(id="session-id", csrf_token="csrf-token")
|
||||
HOST = "127.0.0.1:8000"
|
||||
LIMIT = 1024
|
||||
HOSTNAME = "photos.example.com"
|
||||
|
||||
|
||||
def check(
|
||||
method="GET",
|
||||
path="/api/v1/workflow",
|
||||
allowed_hosts=LOOPBACK_HOSTS,
|
||||
scheme="http",
|
||||
**headers,
|
||||
):
|
||||
def check(method="GET", path="/api/v1/workflow", **headers):
|
||||
"""Evaluate a request that is authenticated and same-origin unless overridden."""
|
||||
sent = {
|
||||
"host": HOST,
|
||||
@@ -51,8 +38,6 @@ def check(
|
||||
path=path,
|
||||
headers=sent,
|
||||
session=SESSION,
|
||||
allowed_hosts=allowed_hosts,
|
||||
scheme=scheme,
|
||||
max_request_bytes=LIMIT,
|
||||
)
|
||||
|
||||
@@ -170,115 +155,6 @@ def test_refusals_name_no_path_secret_or_internal():
|
||||
assert "/" not in refusal.message
|
||||
|
||||
|
||||
# ── US08-01: the same table with a configured trust boundary ─────────────────
|
||||
|
||||
CONFIGURED = frozenset(LOOPBACK_HOSTS | {HOSTNAME})
|
||||
|
||||
|
||||
def test_a_configured_host_is_accepted_and_its_neighbours_are_not():
|
||||
assert check(host=HOSTNAME, allowed_hosts=CONFIGURED) is None
|
||||
for host in ("other.example.com", f"evil-{HOSTNAME}", "192.168.1.10"):
|
||||
refusal = check(host=host, allowed_hosts=CONFIGURED)
|
||||
assert (refusal.status, refusal.code) == (403, "host_not_allowed"), host
|
||||
|
||||
|
||||
def test_the_loopback_default_refuses_a_host_nobody_configured():
|
||||
"""The default set is what the app enforced before there was a setting."""
|
||||
refusal = check(host=HOSTNAME)
|
||||
assert (refusal.status, refusal.code) == (403, "host_not_allowed")
|
||||
|
||||
|
||||
def test_the_origin_must_match_the_external_scheme():
|
||||
for scheme in ("http", "https"):
|
||||
assert (
|
||||
check(
|
||||
method="POST",
|
||||
host=HOSTNAME,
|
||||
origin=f"{scheme}://{HOSTNAME}",
|
||||
allowed_hosts=CONFIGURED,
|
||||
scheme=scheme,
|
||||
)
|
||||
is None
|
||||
)
|
||||
# An HTTPS deployment whose caller claims plain HTTP is a different origin.
|
||||
refusal = check(
|
||||
method="POST",
|
||||
host=HOSTNAME,
|
||||
origin=f"http://{HOSTNAME}",
|
||||
allowed_hosts=CONFIGURED,
|
||||
scheme="https",
|
||||
)
|
||||
assert (refusal.status, refusal.code) == (403, "origin_not_allowed")
|
||||
|
||||
|
||||
def view(client, *, trusted=(), **headers):
|
||||
sent = {name.replace("_", "-"): value for name, value in headers.items()}
|
||||
return external_view(
|
||||
client=client,
|
||||
headers={"host": HOST, **sent},
|
||||
scheme="http",
|
||||
trusted_proxies=frozenset(trusted),
|
||||
)
|
||||
|
||||
|
||||
def test_forwarded_headers_are_ignored_without_a_trusted_proxy():
|
||||
forged = {"x_forwarded_proto": "https", "x_forwarded_host": HOSTNAME}
|
||||
assert view("10.0.0.9", **forged) == ("http", HOST)
|
||||
assert view(None, **forged) == ("http", HOST)
|
||||
# Configuring *a* proxy does not trust a caller that is not it.
|
||||
assert view("10.0.0.9", trusted=("10.0.0.1",), **forged) == ("http", HOST)
|
||||
|
||||
|
||||
def test_a_trusted_proxy_defines_the_external_scheme_and_host():
|
||||
assert view(
|
||||
"10.0.0.1", trusted=("10.0.0.1",), x_forwarded_proto="https", x_forwarded_host=HOSTNAME
|
||||
) == ("https", HOSTNAME)
|
||||
# A chain: the first entry is what the original client asked for.
|
||||
assert view(
|
||||
"10.0.0.1",
|
||||
trusted=("10.0.0.1",),
|
||||
x_forwarded_proto="https, http",
|
||||
x_forwarded_host=f"{HOSTNAME}, inner.internal",
|
||||
) == ("https", HOSTNAME)
|
||||
# Trusted but silent: this hop's own view stands.
|
||||
assert view("10.0.0.1", trusted=("10.0.0.1",)) == ("http", HOST)
|
||||
|
||||
|
||||
@pytest.mark.parametrize(
|
||||
"settings,exposed",
|
||||
[
|
||||
({}, []),
|
||||
({"host": "127.0.0.1"}, []),
|
||||
({"allowed_hosts": ("localhost", "127.0.0.1")}, []),
|
||||
({"allowed_hosts": (f"{HOSTNAME}:8443",)}, [HOSTNAME]),
|
||||
({"host": "0.0.0.0", "allowed_hosts": (HOSTNAME,)}, ["0.0.0.0", HOSTNAME]),
|
||||
],
|
||||
)
|
||||
def test_exposed_hosts_names_only_what_another_machine_can_reach(settings, exposed):
|
||||
assert exposed_hosts(Config(**settings)) == exposed
|
||||
|
||||
|
||||
def test_an_exposed_configuration_without_a_secret_must_not_serve():
|
||||
refusal = trust_refusal(Config(allowed_hosts=(HOSTNAME,)))
|
||||
assert HOSTNAME in refusal and "PHOTO_PIPELINE_ACCESS_SECRET" in refusal
|
||||
assert trust_refusal(Config(allowed_hosts=(HOSTNAME,), access_secret="s")) is None
|
||||
# Loopback-only, with and without a secret, is unchanged.
|
||||
assert trust_refusal(Config()) is None
|
||||
assert trust_refusal(Config(access_secret="s")) is None
|
||||
|
||||
|
||||
def test_failed_attempts_are_bounded_per_window():
|
||||
limiter = FailureLimiter(limit=2, window=60.0)
|
||||
assert not limiter.blocked()
|
||||
limiter.record_failure()
|
||||
assert not limiter.blocked()
|
||||
limiter.record_failure()
|
||||
assert limiter.blocked()
|
||||
# Attempts age out, so a locked-out operator is not locked out forever.
|
||||
limiter._failures = [-120.0, -120.0]
|
||||
assert not limiter.blocked()
|
||||
|
||||
|
||||
@pytest.mark.parametrize(
|
||||
"value,expected",
|
||||
[
|
||||
|
||||
Reference in New Issue
Block a user