POSA_Copyrighter/docs/operations/offline-install.md

# Offline (air-gapped / 폐쇄망) install

The deployment target has **no internet**. All Python dependencies must be
carried across the air gap as pre-downloaded wheels. The build/staging machine
(which has internet) does the downloading; the target only installs.

## Prerequisites

- Same OS + CPU architecture + **Python 3.13** on build and target machines
  (wheels are platform- and version-specific).
- The repository itself transferred to the target.

## 1. On the build machine (has internet)

```bash
# Runtime only (--only-binary keeps sdists out so nothing has to build on the
# offline target; also carry pip/setuptools/wheel themselves):
pip download --only-binary=:all: -r requirements.txt -d wheelhouse/
pip download --only-binary=:all: pip setuptools wheel -d wheelhouse/

# Or runtime + test tooling:
pip download --only-binary=:all: -r requirements-dev.txt -d wheelhouse/
```

This fills `wheelhouse/` with every wheel (including transitive deps).

### Hash-verified lock (recommended)

Pin the full transitive tree with hashes so a tampered/substituted wheel is
rejected at install time:

```bash
# On the build machine (pip-tools):
pip-compile --generate-hashes -o requirements.lock requirements.txt
```

Then install on the target with `--require-hashes` (step 3).

### Model weights (`models/`)

Some analysis paths use pre-trained weights that are **not** pip packages and are
**not** committed to the repository (`models/` is gitignored, same treatment as
`wheelhouse/`). Fetch them on the build machine:

```bash
mkdir -p models
# YuNet face detector (OpenCV Zoo, MIT). Used by face/person presence detection;
# cv2.FaceDetectorYN itself ships inside opencv-python-headless, only the weights
# are external.
curl -sSL -o models/face_detection_yunet_2023mar.onnx \
  https://github.com/opencv/opencv_zoo/raw/main/models/face_detection_yunet/face_detection_yunet_2023mar.onnx
```

Verify before transferring — a truncated or substituted model must not cross the
air gap unnoticed:

| file | bytes | sha256 |
|---|---:|---|
| `face_detection_yunet_2023mar.onnx` | 232589 | `8f2383e4dd3cfbb4553ea8718107fc0423210dc964f9f4280604804ed2552fa4` |

```bash
python - <<'PY'
import hashlib, pathlib
p = pathlib.Path("models/face_detection_yunet_2023mar.onnx")
print(len(p.read_bytes()), hashlib.sha256(p.read_bytes()).hexdigest())
PY
```

If `models/` is absent on the target the server still starts: face detection
falls back to the bundled Haar cascades and logs the downgrade once. It does not
fail closed, because a missing optional weight must not stop an operator from
reviewing.

## 2. Transfer

Copy the repository, the `wheelhouse/` directory, and the `models/` directory to
the target via the approved air-gap transfer process (e.g. removable media).

## 3. On the target machine (no internet)

```bash
python -m venv .venv
. .venv/bin/activate          # Windows: .venv\Scripts\activate
pip install --no-index --find-links wheelhouse/ -r requirements.txt
```

`--no-index` guarantees pip never reaches out to the internet; `--find-links`
points it at the local wheelhouse.

## 4. Verify offline

```bash
# Confirm no network access is attempted and imports resolve:
python -c "import numpy, cv2, PIL; print('deps ok')"
python -m pytest -q          # if test deps were installed
```

## Optional: Ollama LLM model (evidence narration)

The LLM assistant is optional. If you want it on the air-gapped target, the
model weights must be carried across too — they are **not** part of the Python
wheelhouse:

Carry the **`ollama` binary itself** as well as the weights — neither is in the
Python wheelhouse.

Use `qwen2.5:7b` (about 4.7 GB), the code default. Do **not** substitute
`qwen2.5:0.5b-instruct`: it was measured against this prompt at 55.7 s and it
broke the output contract, reprinting the prompt's format example as if it were
a finding. A 3B-class model is the floor for reading evidence and citing it.

```bash
# Build machine (has internet):
ollama pull qwen2.5:7b                   # or your pinned OLLAMA_MODEL
# Copy the model store across the air gap:
#   default: ~/.ollama/models  (or set OLLAMA_MODELS to a directory you transfer)

# Target machine (no internet):
export OLLAMA_MODELS=/path/to/transferred/models
ollama serve &
# Smoke test:
ollama run qwen2.5:7b "hello"
```

Set `OLLAMA_BASE_URL` / `OLLAMA_MODEL` in `.env`. Leaving them unset does **not**
turn the LLM off — the client falls back to `http://127.0.0.1:11434` and the code
default model, so a target without Ollama spends about 2 s per case before
writing a readable failure row (scores are unaffected either way). To stop the
attempts, disable the `llm` provider in the operator console.

Hardware: the 7B default was timed on a GPU host (about 2 s warm, 7 s cold). On a
CPU-only target, raise `OLLAMA_TIMEOUT` (default 120 s) and expect narration to
dominate seed time.

## Optional: Playwright Chromium (E2E tests)

The operator-GUI E2E suite needs a Chromium binary, which Playwright normally
downloads. For the air gap, fetch it on the build machine and carry it:

```bash
# Build machine:
PLAYWRIGHT_BROWSERS_PATH=./pw-browsers playwright install chromium
# Transfer the pw-browsers/ directory to the target, then:

# Target machine:
export PLAYWRIGHT_BROWSERS_PATH=/path/to/pw-browsers
python -m pytest tests/operator_gui -q
```

## Notes

- **Which network mode?** Read `operating-modes.md` before deploying: what
  scores mean with full egress, API hosts only, or no egress at all (and why a
  fully closed site must remove the API keys rather than just block the network).
- The operator GUI (`web/operator-gui/`) is fully self-contained: all
  scripts/styles are local and the Pretendard font is bundled
  (`assets/fonts/PretendardVariable.woff2`). No CDN fetch at runtime.
- External **search** APIs (Google / Naver / Cloud Vision) are an allowed
  runtime exception and require outbound access to those providers. They
  are optional; missing credentials disable the provider (see the startup
  provider-readiness log).
- **Broader egress:** beyond the API hosts, the image-comparison feature also
  fetches arbitrary public result-page / image / stylesheet URLs returned by
  those searches (SSRF-guarded: private/loopback/CGNAT/link-local ranges are
  blocked). Restricting egress to the API hosts only disables image comparison
  of external results; it does not break core review, but it weakens the web-copy
  signal silently (no failure row) - see `operating-modes.md` mode B.
- `COPYRIGHTER_AUTH_TOKEN` should be set in `.env` for any multi-operator or
  networked deployment so the data routes require a shared token.