.dockerignore

@@ -0,0 +1,28 @@
+.git
+.direnv
+.mypy_cache
+.pytest_cache
+.ruff_cache
+.venv
+**/.venv
+.env
+.cache
+.claude
+.coverage
+.vscode
+.stfolder
+.literotica_data
+esphome
+htmlcov
+data
+ebooks
+__pycache__
+**/__pycache__
+*.pyc
+*.pyo
+.ebook_search_bm25
+result
+result-*
+*.egg-info
+dist
+build

.github/workflows/test_ebook_search.yml

@@ -0,0 +1,26 @@
+name: test ebook search
+
+on:
+  workflow_dispatch:
+  push:
+    branches:
+      - main
+  pull_request:
+    branches:
+      - main
+
+env:
+  UV_PYTHON_DOWNLOADS: never
+  UV_CACHE_DIR: /var/cache/uv
+  UV_LINK_MODE: copy
+
+jobs:
+  test-ebook-search:
+    runs-on: self-hosted
+
+    steps:
+      - uses: actions/checkout@v4
+      - name: Install dependencies
+        run: nix develop .#devShells.x86_64-linux.ebook-search -c uv sync --locked --project python/ebook_search/docker
+      - name: Run ebook search tests
+        run: nix develop .#devShells.x86_64-linux.ebook-search -c uv run --project python/ebook_search/docker --no-sync pytest tests/ebook_search --override-ini addopts="-n auto -ra"

overlays/default.nix

@@ -17,45 +17,17 @@
 
   python-env = final: _prev: {
     my_python = final.python314.withPackages (
-      ps:
-      let
-        bm25s = ps.buildPythonPackage rec {
-          pname = "bm25s";
-          version = "0.3.9";
-          pyproject = true;
-
-          src = final.fetchPypi {
-            inherit pname version;
-            hash = "sha256-iVxnnZUrfeg1XttfPhpiCh4vKU0dQrkZvwghzOLi9Zc=";
-          };
-
-          build-system = [ ps.setuptools ];
-          dependencies = with ps; [
-            numpy
-            scipy
-          ];
-
-          pythonImportsCheck = [ "bm25s" ];
-        };
-      in
-      with ps;
-      [
+      ps: with ps; [
         alembic
         apprise
         apscheduler
-        beautifulsoup4
-        ebooklib
         fastapi
         fastapi-cli
         httpx
         mypy
-        numpy
-        orjson
         pgvector
-        polars
         psycopg
         pydantic
-        pydantic-settings
         pyfakefs
         pytest
         pytest-cov
@@ -63,12 +35,8 @@
         pytest-xdist
         python-multipart
         ruff
-        scalene
         sqlalchemy
-        bm25s
         tenacity
-        textual
-        tiktoken
         tinytuya
         typer
         websockets

pyproject.toml

@@ -12,9 +12,14 @@ dependencies = [
     "alembic",
     "apprise",
     "apscheduler",
+    "beautifulsoup4",
+    "bm25s",
+    "ebooklib",
     "fastapi",
     "fastapi-cli",
     "httpx",
+    "jinja2",
+    "pgvector",
     "polars",
     "psycopg[binary]",
     "pydantic",
@@ -22,9 +27,12 @@ dependencies = [
     "python-multipart",
     "sqlalchemy",
     "tenacity",
+    "tiktoken",
     "tinytuya",
     "typer",
+    "uvicorn",
     "websockets",
+    "yake",
 ]
 
 [project.scripts]
@@ -109,6 +117,6 @@ exclude_lines = [
 ]
 
 [tool.pytest.ini_options]
-addopts = "-n auto -ra"
+addopts = "-n auto -ra --ignore=tests/ebook_search"
 testpaths = ["tests"]
 # --cov=system_tools --cov-report=term-missing --cov-report=xml --cov-report=html --cov-branch

python/ebook_search/docker/Dockerfile

@@ -0,0 +1,54 @@
+FROM python:3.14-slim AS base
+
+COPY --from=ghcr.io/astral-sh/uv:0.11.26 /uv /uvx /bin/
+
+ENV PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1 \
+    APP_DIR=/home/richie/dotfiles \
+    UV_PROJECT_ENVIRONMENT=/opt/venv \
+    UV_PYTHON_DOWNLOADS=never \
+    UV_NO_CACHE=1
+
+# Separate ENV instruction so ${APP_DIR} and ${PATH} from above resolve.
+ENV PYTHONPATH=${APP_DIR} \
+    PATH=/opt/venv/bin:${PATH}
+
+WORKDIR ${APP_DIR}
+
+RUN apt-get update \
+    && apt-get install -y --no-install-recommends build-essential curl \
+    && rm -rf /var/lib/apt/lists/*
+
+COPY python/ebook_search/docker/pyproject.toml python/ebook_search/docker/uv.lock ./
+
+RUN uv sync --locked --no-dev
+
+
+FROM base AS test
+
+RUN uv sync --locked
+
+COPY python ./python
+COPY tests/__init__.py ./tests/__init__.py
+COPY tests/ebook_search ./tests/ebook_search
+
+CMD ["pytest"]
+
+
+FROM base AS runtime
+
+ENV EBOOK_SEARCH_HOST=0.0.0.0 \
+    EBOOK_SEARCH_PORT=8070 \
+    EBOOK_SEARCH_BM25_INDEX_DIR=/data/bm25
+
+COPY python ./python
+
+RUN useradd --create-home --uid 10001 app \
+    && mkdir -p /data \
+    && chown -R app:app /home/richie /data
+
+USER app
+
+EXPOSE 8070
+
+CMD ["sh", "-c", "exec python -m python.ebook_search.api.main --host \"${EBOOK_SEARCH_HOST}\" --port \"${EBOOK_SEARCH_PORT}\" --log-level \"${EBOOK_SEARCH_LOG_LEVEL:-INFO}\""]

python/ebook_search/docker/README.md

@@ -0,0 +1,77 @@
+# Ebook Search Docker
+
+Run the EPUB search app against the existing Postgres database on `jeeves`:
+
+```sh
+python -m python.ebook_search.docker.containers start --library-path /path/to/epubs --build
+```
+
+All ebook-search Docker files live in this directory:
+
+- `Dockerfile` — multi-stage: `test` (runs pytest) and `runtime` (default target, the app image)
+- `docker-compose.yml`
+- `containers.py` — Typer lifecycle CLI
+- `pyproject.toml` / `uv.lock` — the container's uv-locked dependencies
+
+The app listens on `http://localhost:8070`.
+
+Useful lifecycle commands:
+
+```sh
+python -m python.ebook_search.docker.containers build
+python -m python.ebook_search.docker.containers start --library-path /path/to/epubs
+python -m python.ebook_search.docker.containers test
+python -m python.ebook_search.docker.containers logs
+python -m python.ebook_search.docker.containers ps
+python -m python.ebook_search.docker.containers stop
+```
+
+Direct compose usage from the repo root:
+
+```sh
+docker compose -f python/ebook_search/docker/docker-compose.yml ps
+```
+
+## Dependencies
+
+The image builds its environment with uv from `pyproject.toml` + `uv.lock` in this
+directory — this is the source of truth for the container's dependencies. To add or
+update a dependency, edit `pyproject.toml` here and regenerate the lock (uv is
+available in the `ebook-search` dev shell):
+
+```sh
+nix develop .#ebook-search -c uv lock --project python/ebook_search/docker
+```
+
+## Tests
+
+The main pytest suite excludes `tests/ebook_search` (its dependencies are no longer
+in the nix dev shell). The `test ebook search` CI workflow runs them in a uv env
+built from the lockfile in this directory — same commands work locally from the
+repo root (the `--override-ini` drops the main suite's ignore):
+
+```sh
+uv sync --locked --project python/ebook_search/docker
+uv run --project python/ebook_search/docker --no-sync pytest tests/ebook_search --override-ini addopts="-n auto -ra"
+```
+
+They can also run inside the Docker `test` image, which validates the image itself:
+
+```sh
+python -m python.ebook_search.docker.containers test
+```
+
+or the raw docker equivalent:
+
+```sh
+docker build --file python/ebook_search/docker/Dockerfile --target test --tag ebook-search:test .
+docker run --rm ebook-search:test
+```
+
+## Configuration
+
+The compose service loads the repo root `.env` into the container via `env_file`.
+
+Mount your EPUB directory by setting `EBOOK_LIBRARY_HOST_PATH` in an env file or on the command line. The container sees it as `/library`, and `EBOOK_SEARCH_LIBRARY_PATHS` is set to `/library` inside the container.
+
+Database connection settings are controlled by `RICHIE_DB`, `RICHIE_HOST`, `RICHIE_PORT`, `RICHIE_USER`, and `RICHIE_PASSWORD`. The default host is `jeeves`.

python/ebook_search/docker/__init__.py

@@ -0,0 +1 @@
+"""Docker packaging and lifecycle tooling for ebook search."""

python/ebook_search/docker/containers.py

@@ -0,0 +1,259 @@
+"""Docker container lifecycle management for ebook search."""
+
+from __future__ import annotations
+
+import logging
+import os
+import subprocess
+from pathlib import Path
+from typing import Annotated
+
+import typer
+
+from python.common import configure_logger, get_repo_dir
+
+logger = logging.getLogger(__name__)
+
+
+def get_compose_file() -> Path:
+    """Return the path to the docker-compose.yml file."""
+    return Path(__file__).resolve().with_name("docker-compose.yml")
+
+
+def compose_base_args() -> list[str]:
+    """Return the common docker compose arguments for the ebook search stack."""
+    return ["compose", "-f", str(get_compose_file())]
+
+
+def docker_run(
+    arguments: list[str],
+    *,
+    env: dict[str, str] | None = None,
+    capture_output: bool = False,
+) -> subprocess.CompletedProcess[str]:
+    """Run docker with repo-root cwd and consistent error handling."""
+    logger.info("docker %s", " ".join(arguments))
+    return subprocess.run(
+        ["docker", *arguments],
+        cwd=get_repo_dir(),
+        env=env,
+        text=True,
+        check=False,
+        capture_output=capture_output,
+    )
+
+
+def compose_env(*, library_path: Path | None = None, port: int | None = None) -> dict[str, str]:
+    """Return environment variables passed to docker compose."""
+    env = os.environ.copy()
+    if library_path is not None:
+        resolved_library = library_path.expanduser().resolve()
+        if not resolved_library.exists():
+            msg = f"EPUB library path does not exist: {resolved_library}"
+            raise FileNotFoundError(msg)
+        env["EBOOK_LIBRARY_HOST_PATH"] = str(resolved_library)
+    if port is not None:
+        env["EBOOK_SEARCH_PORT"] = str(port)
+    return env
+
+
+def ensure_compose_file() -> None:
+    """Raise if the ebook search compose file is missing."""
+    if not get_compose_file().is_file():
+        msg = f"Compose file not found: {get_compose_file()}"
+        raise FileNotFoundError(msg)
+
+
+def build_image() -> None:
+    """Build the ebook search app image."""
+    ensure_compose_file()
+    result = docker_run([*compose_base_args(), "build"])
+    if result.returncode != 0:
+        msg = "Failed to build ebook search image"
+        raise RuntimeError(msg)
+
+
+def build_test_image() -> None:
+    """Build the ebook search test Docker image."""
+    dockerfile = Path(__file__).resolve().with_name("Dockerfile")
+    result = docker_run(["build", "--file", str(dockerfile), "--target", "test", "--tag", "ebook-search:test", "."])
+    if result.returncode != 0:
+        msg = "Failed to build ebook search test image"
+        raise RuntimeError(msg)
+
+
+def run_test_image() -> None:
+    """Run the ebook search test suite inside Docker."""
+    result = docker_run(["run", "--rm", "ebook-search:test"])
+    if result.returncode != 0:
+        msg = f"Ebook search tests failed with code {result.returncode}"
+        raise RuntimeError(msg)
+
+
+def start_stack(
+    *,
+    library_path: Path | None = None,
+    port: int | None = None,
+    build: bool = False,
+) -> None:
+    """Start the ebook search Docker compose stack."""
+    ensure_compose_file()
+    env = compose_env(library_path=library_path, port=port)
+    if build:
+        build_image()
+    result = docker_run(
+        [*compose_base_args(), "up", "-d"],
+        env=env,
+    )
+    if result.returncode != 0:
+        msg = f"Ebook search stack failed to start with code {result.returncode}"
+        raise RuntimeError(msg)
+    logger.info("Ebook search started.")
+
+
+def stop_stack(
+    *,
+    volumes: bool = False,
+) -> None:
+    """Stop and remove ebook search containers."""
+    ensure_compose_file()
+    command = [*compose_base_args(), "down"]
+    if volumes:
+        command.append("-v")
+    result = docker_run(command)
+    if result.returncode != 0:
+        msg = f"Ebook search stack failed to stop with code {result.returncode}"
+        raise RuntimeError(msg)
+
+
+def logs_stack(
+    *,
+    service: str | None = None,
+    tail: int = 100,
+    follow: bool = False,
+) -> str | None:
+    """Return recent logs from the ebook search stack."""
+    ensure_compose_file()
+    command = [*compose_base_args(), "logs", "--tail", str(tail)]
+    if follow:
+        command.append("--follow")
+    if service:
+        command.append(service)
+    result = docker_run(command, capture_output=not follow)
+    if result.returncode != 0:
+        return None
+    if follow:
+        return ""
+    return result.stdout + result.stderr
+
+
+def ps_stack() -> str | None:
+    """Return docker compose ps output for the ebook search stack."""
+    ensure_compose_file()
+    result = docker_run([*compose_base_args(), "ps"], capture_output=True)
+    if result.returncode != 0:
+        return None
+    return result.stdout + result.stderr
+
+
+app = typer.Typer(help="Ebook search Docker container management.", no_args_is_help=True)
+
+
+@app.command()
+def build() -> None:
+    """Build the ebook search Docker image."""
+    build_image()
+
+
+@app.command()
+def start(
+    library_path: Annotated[Path | None, typer.Option(help="Override host path containing EPUB files.")] = None,
+    port: Annotated[int | None, typer.Option(help="Override host port for the web UI.")] = None,
+    *,
+    build: Annotated[bool, typer.Option("--build", help="Build the image before starting.")] = False,
+    log_level: Annotated[str, typer.Option(help="Log level.")] = "INFO",
+) -> None:
+    """Start the ebook search container."""
+    configure_logger(log_level)
+    start_stack(
+        library_path=library_path,
+        port=port,
+        build=build,
+    )
+
+
+@app.command()
+def stop(
+    *,
+    volumes: Annotated[bool, typer.Option("--volumes", help="Also remove ebook search data volumes.")] = False,
+    log_level: Annotated[str, typer.Option(help="Log level.")] = "INFO",
+) -> None:
+    """Stop and remove ebook search containers."""
+    configure_logger(log_level)
+    stop_stack(volumes=volumes)
+
+
+@app.command()
+def restart(
+    library_path: Annotated[Path | None, typer.Option(help="Override host path containing EPUB files.")] = None,
+    port: Annotated[int | None, typer.Option(help="Override host port for the web UI.")] = None,
+    *,
+    build: Annotated[bool, typer.Option("--build", help="Build the image before starting.")] = False,
+    log_level: Annotated[str, typer.Option(help="Log level.")] = "INFO",
+) -> None:
+    """Restart the ebook search stack."""
+    configure_logger(log_level)
+    stop_stack()
+    start_stack(
+        library_path=library_path,
+        port=port,
+        build=build,
+    )
+
+
+@app.command()
+def logs(
+    service: Annotated[str | None, typer.Option(help="Service name, or omit for all services.")] = None,
+    tail: Annotated[int, typer.Option(help="Number of recent log lines.")] = 100,
+    *,
+    follow: Annotated[bool, typer.Option("--follow", "-f", help="Follow logs.")] = False,
+) -> None:
+    """Show recent ebook search container logs."""
+    output = logs_stack(service=service, tail=tail, follow=follow)
+    if output is None:
+        typer.echo("No ebook search containers found.")
+        raise typer.Exit(code=1)
+    if output:
+        typer.echo(output)
+
+
+@app.command("test")
+def run_tests(
+    *,
+    build: Annotated[bool, typer.Option("--build/--no-build", help="Build the test image before running.")] = True,
+    log_level: Annotated[str, typer.Option(help="Log level.")] = "INFO",
+) -> None:
+    """Run ebook search tests inside the Docker test image."""
+    configure_logger(log_level)
+    if build:
+        build_test_image()
+    run_test_image()
+
+
+@app.command("ps")
+def ps() -> None:
+    """Show ebook search container status."""
+    output = ps_stack()
+    if output is None:
+        typer.echo("No ebook search containers found.")
+        raise typer.Exit(code=1)
+    typer.echo(output)
+
+
+def cli() -> None:
+    """Typer entry point."""
+    app()
+
+
+if __name__ == "__main__":
+    cli()

python/ebook_search/docker/docker-compose.yml

@@ -0,0 +1,36 @@
+name: ebook-search
+
+services:
+  ebook-search:
+    build:
+      context: ../../..
+      dockerfile: python/ebook_search/docker/Dockerfile
+    image: ebook-search:latest
+    restart: unless-stopped
+    ports:
+      - "${EBOOK_SEARCH_PORT:-8070}:8070"
+    extra_hosts:
+      - "jeeves:192.168.90.40"
+    env_file:
+      - ../../../.env
+    environment:
+      EBOOK_SEARCH_HOST: "0.0.0.0"
+      EBOOK_SEARCH_PORT: "8070"
+      EBOOK_SEARCH_LIBRARY_PATHS: "/library"
+      EBOOK_SEARCH_BM25_INDEX_DIR: "/data/bm25"
+    volumes:
+      - "${EBOOK_LIBRARY_HOST_PATH:-/home/richie/ebooks}:/library:ro"
+      - ebook-search-data:/data
+    healthcheck:
+      test:
+        [
+          "CMD-SHELL",
+          "curl -fsS http://127.0.0.1:8070/health >/dev/null || exit 1",
+        ]
+      interval: 30s
+      timeout: 5s
+      retries: 5
+      start_period: 30s
+
+volumes:
+  ebook-search-data:

python/ebook_search/docker/pyproject.toml

@@ -0,0 +1,38 @@
+[project]
+name = "ebook-search"
+version = "0.1.0"
+description = "Locked runtime environment for the ebook search container."
+requires-python = "~=3.14.0"
+dependencies = [
+    "alembic",
+    "beautifulsoup4",
+    "bm25s",
+    "ebooklib",
+    "fastapi",
+    "httpx",
+    "jinja2",
+    "pgvector",
+    "psycopg[binary]",
+    "pydantic",
+    "pydantic-settings",
+    "python-multipart",
+    "sqlalchemy",
+    "tiktoken",
+    "typer",
+    "uvicorn[standard]",
+    "yake",
+]
+
+[dependency-groups]
+dev = [
+    "pytest",
+    "pytest-mock",
+    "pytest-xdist",
+]
+
+[tool.uv]
+package = false
+
+[tool.pytest.ini_options]
+addopts = "-n auto -ra"
+testpaths = ["tests/ebook_search"]

shell.nix

@@ -3,6 +3,20 @@
   ...
 }:
 {
+  # For running the uv-managed ebook-search venv outside the container:
+  # PyPI manylinux wheels (numpy via bm25s) expect libstdc++.so.6 on the
+  # loader path, which NixOS does not provide globally.
+  ebook-search = pkgs.mkShell {
+    nativeBuildInputs = with pkgs; [
+      my_python
+      uv
+    ];
+    LD_LIBRARY_PATH = pkgs.lib.makeLibraryPath [
+      pkgs.stdenv.cc.cc.lib
+      pkgs.zlib
+    ];
+  };
+
   default = pkgs.mkShell {
     NIX_CONFIG = "extra-experimental-features = nix-command flakes ca-derivations";
     nativeBuildInputs = with pkgs; [

systems/jeeves/runners/default.nix

@@ -1,6 +1,9 @@
 { ... }:
 {
-  imports = [ ./nix_builder.nix ];
+  imports = [
+    ./nix_builder.nix
+    ./uv_cache_prune.nix
+  ];
 
   services.nix_builder.containers = {
     nix-builder-00.enable = true;

systems/jeeves/runners/nix_builder.nix

@@ -62,6 +62,11 @@ in
             mountPoint = "/run/secrets/gitea-runners";
             isReadOnly = true;
           };
+          uv-cache = {
+            hostPath = vars.uv_cache;
+            mountPoint = "/var/cache/uv";
+            isReadOnly = false;
+          };
         };
         config =
           {
@@ -153,7 +158,8 @@ in
       }
     ) cfg.containers;
 
-    systemd.services = builtins.listToAttrs (
+    systemd = {
+      services = builtins.listToAttrs (
         map (name: {
           name = "container@${name}";
           value = {
@@ -162,5 +168,10 @@ in
           };
         }) (builtins.attrNames (filterAttrs (_: c: c.enable) cfg.containers))
       );
+
+      tmpfiles.rules = [
+        "d ${vars.uv_cache} 0755 ${runnerUsername} ${runnerUsername} - -"
+      ];
+    };
   };
 }

systems/jeeves/runners/uv_cache_prune.nix

@@ -0,0 +1,28 @@
+{ pkgs, ... }:
+let
+  vars = import ../vars.nix;
+  runnerUsername = "gitea-runner";
+in
+{
+  systemd = {
+    services.uv-cache-prune = {
+      description = "Prune the shared gitea runner uv cache";
+      environment.UV_CACHE_DIR = vars.uv_cache;
+      serviceConfig = {
+        Type = "oneshot";
+        User = runnerUsername;
+        Group = runnerUsername;
+        ExecStart = "${pkgs.uv}/bin/uv cache prune";
+      };
+    };
+
+    timers.uv-cache-prune = {
+      description = "Monthly prune of the shared gitea runner uv cache";
+      wantedBy = [ "timers.target" ];
+      timerConfig = {
+        OnCalendar = "*-*-01 04:00:00";
+        Persistent = true;
+      };
+    };
+  };
+}

systems/jeeves/scripts/zfs.sh

@@ -30,6 +30,7 @@ sudo zfs create media/secure/share -o mountpoint=/zfs/media/share -o exec=off
 # scratch datasets
 sudo zfs create scratch/kafka -o mountpoint=/zfs/scratch/kafka -o recordsize=1M
 sudo zfs create scratch/transmission -o mountpoint=/zfs/scratch/transmission -o recordsize=16k -o sync=disabled -o redundant_metadata=none
+sudo zfs create scratch/uv_cache -o mountpoint=/zfs/scratch/uv_cache
 
 # storage datasets
 sudo zfs create storage/ollama -o recordsize=1M -o compression=zstd-19 -o sync=disabled

systems/jeeves/snapshot_config.toml

@@ -121,3 +121,9 @@ monthly = 2
 hourly = 0
 daily = 0
 monthly = 0
+
+["scratch/uv_cache"]
+15_min = 2
+hourly = 0
+daily = 0
+monthly = 0

systems/jeeves/vars.nix

@@ -17,5 +17,6 @@ in
   transmission = "${zfs_storage}/transmission";
   ollama = "${zfs_storage}/ollama";
   transmission_scratch = "${zfs_scratch}/transmission";
+  uv_cache = "${zfs_scratch}/uv_cache";
   kafka = "${zfs_scratch}/kafka";
 }