experimental feature for one-click deployment

.github/workflows/tests-deploy.yaml

@@ -0,0 +1,29 @@
+name: Deploy Tests
+
+on:
+  push:
+    branches: [ main ]
+    paths:
+      - deploy/**
+  pull_request:
+    paths:
+      - deploy/**
+
+jobs:
+  tests:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Set up Python 3.12
+        uses: actions/setup-python@v6
+        with:
+          python-version: "3.12"
+
+      - name: Install dependencies
+        run: pip install pytest fastapi httpx python-multipart pyyaml
+
+      - name: Run Deploy Tests
+        working-directory: deploy
+        run: python -m pytest tests/ -v

README.md

@@ -22,7 +22,40 @@ Auto Archiver is a Python tool to automatically archive content on the web in a
 Read the [article about Auto Archiver on bellingcat.com](https://www.bellingcat.com/resources/2022/09/22/preserve-vital-online-content-with-bellingcats-auto-archiver-tool/).
 
 
-## Installation
+## One-Click Cloud Deploy
+
+Deploy your own Auto Archiver instance to the cloud — no coding required:
+
+[![Deploy on Railway](https://railway.app/button.svg)](https://railway.app/new/template?template=https://github.com/bellingcat/auto-archiver&envs=AUTH_PASSWORD,GSHEET_URL,GOOGLE_SERVICE_ACCOUNT_JSON,POLL_INTERVAL,S3_BUCKET,S3_KEY,S3_SECRET,S3_REGION,TELEGRAM_API_ID,TELEGRAM_API_HASH,TELEGRAM_BOT_TOKEN,ENABLE_SCREENSHOTS,LOG_LEVEL&optionalEnvs=GSHEET_URL,GOOGLE_SERVICE_ACCOUNT_JSON,POLL_INTERVAL,S3_BUCKET,S3_KEY,S3_SECRET,S3_REGION,TELEGRAM_API_ID,TELEGRAM_API_HASH,TELEGRAM_BOT_TOKEN,ENABLE_SCREENSHOTS,LOG_LEVEL&AUTH_PASSWORDDesc=Password+to+access+your+archiver+web+interface&GSHEET_URLDesc=Google+Sheet+URL+to+monitor+for+new+URLs+(leave+empty+to+disable)&POLL_INTERVALDesc=Seconds+between+Google+Sheet+checks+(min+60)&POLL_INTERVALDefault=300&S3_BUCKETDesc=S3+bucket+name+for+storage+(leave+empty+for+local+only)&S3_REGIONDefault=us-east-1&LOG_LEVELDefault=INFO)
+
+**What you get:** A web interface where you can paste URLs and archive them instantly. Optionally connect a Google Sheet for automated monitoring, S3 for cloud storage, and Telegram for archiving channels.
+
+**Only required setting:** `AUTH_PASSWORD` — everything else is optional and can be configured later via the Railway dashboard.
+
+<details>
+<summary>📋 Environment variables reference</summary>
+
+| Variable | Required | Description |
+|----------|----------|-------------|
+| `AUTH_PASSWORD` | **Yes** | Password to access the web interface |
+| `GSHEET_URL` | No | Google Sheet URL to monitor for new URLs [use this template](https://docs.google.com/spreadsheets/d/1NJZo_XZUBKTI1Ghlgi4nTPVvCfb0HXAs6j5tNGas72k/edit?gid=0#gid=0) |
+| `GOOGLE_SERVICE_ACCOUNT_JSON` | No | Google service account JSON (required with Sheets) [follow these instructions](https://auto-archiver.readthedocs.io/en/v1.0.1/how_to/gsheets_setup.html) |
+| `POLL_INTERVAL` | No | Seconds between Sheet checks (default: 300) |
+| `S3_BUCKET` | No | S3 bucket name for archived content, ideal for cloud hosting your archives but not mandatory, any S3-compatible storage works |
+| `S3_KEY` / `S3_SECRET` | No | S3 credentials |
+| `S3_REGION` | No | S3 region (default: us-east-1) |
+| `S3_ENDPOINT` | No | S3 endpoint URL |
+| `TELEGRAM_API_ID` / `TELEGRAM_API_HASH` | No | Telegram API credentials |
+| `TELEGRAM_BOT_TOKEN` | No | Telegram bot token |
+| `ENABLE_SCREENSHOTS` | No | Set to `true` for full-page screenshots |
+| `ENABLE_THUMBNAILS` | No | Set to `true` for video thumbnails |
+| `ENABLE_CSV_DB` | No | Set to `true` for CSV logging |
+| `LOG_LEVEL` | No | DEBUG, INFO, WARNING, ERROR (default: INFO) |
+
+</details>
+
+
+## Traditional Installation
 
 View the [Installation Guide](https://auto-archiver.readthedocs.io/en/latest/installation/installation.html) for full instructions
 

deploy/Dockerfile

@@ -0,0 +1,34 @@
+# ── Cloud Deploy ──────────────────────────────────────────────────────
+# Thin web UI + config generator layer on top of the published
+# auto-archiver Docker image. Used by the Railway one-click deploy.
+#
+# Build:
+#   docker build -f deploy/Dockerfile -t auto-archiver-deploy .
+#
+# Run:
+#   docker run -p 8080:8080 -e PORT=8080 -e AUTH_PASSWORD=secret auto-archiver-deploy
+# ──────────────────────────────────────────────────────────────────────
+
+FROM bellingcat/auto-archiver:latest
+
+USER root
+
+# Install the lightweight web layer dependencies
+RUN pip install --no-cache-dir fastapi uvicorn[standard] python-multipart pyyaml
+
+# Copy deploy scripts into the image
+COPY deploy/ /app/deploy/
+
+# Ensure writable dirs exist
+RUN mkdir -p /app/local_archive /app/secrets && \
+    chown -R 1000:1000 /app/local_archive /app/secrets /app/deploy
+
+USER 1000
+
+# Railway sets PORT; default to 8080
+ENV PORT=8080
+
+EXPOSE ${PORT}
+
+# Override the CLI entrypoint with the web server
+ENTRYPOINT ["python3", "-m", "deploy.start"]

deploy/__init__.py

@@ -0,0 +1 @@
+# Cloud deployment layer for auto-archiver

deploy/generate_config.py

@@ -0,0 +1,163 @@
+#!/usr/bin/env python3
+"""
+Generates orchestration.yaml from environment variables.
+
+This script bridges Railway's env-var-based configuration with
+auto-archiver's YAML-based configuration system. It runs at container
+startup before the web UI server starts.
+"""
+
+import os
+from pathlib import Path
+
+import yaml
+
+
+CONFIG_PATH = Path("/app/secrets/orchestration.yaml")
+SECRETS_DIR = Path("/app/secrets")
+
+
+def build_config() -> dict:
+    """Build an orchestration config dict from environment variables."""
+
+    # -- Base config: always present ------------------------------------
+    config = {
+        "steps": {
+            "feeders": ["cli_feeder"],
+            "extractors": ["generic_extractor"],
+            "enrichers": ["hash_enricher"],
+            "databases": ["console_db"],
+            "storages": ["local_storage"],
+            "formatters": ["html_formatter"],
+        },
+        "logging": {
+            "level": os.environ.get("LOG_LEVEL", "INFO"),
+        },
+        "local_storage": {
+            "save_to": "/app/local_archive",
+            "path_generator": "flat",
+            "filename_generator": "static",
+        },
+        "generic_extractor": {
+            "subtitles": os.environ.get("SUBTITLES", "false").lower() == "true",
+            "comments": False,
+            "livestreams": False,
+            "live_from_start": False,
+            "end_means_success": True,
+            "allow_playlist": False,
+        },
+        "hash_enricher": {
+            "algorithm": "SHA-256",
+        },
+        "html_formatter": {
+            "detect_thumbnails": True,
+        },
+        "authentication": {},
+    }
+
+    # -- Google Sheets feeder (optional) --------------------------------
+    gsheet_url = os.environ.get("GSHEET_URL", "")
+    if gsheet_url:
+        config["steps"]["feeders"].append("gsheet_feeder")
+        config["steps"]["databases"].append("gsheet_db")
+        config["gsheet_feeder"] = {
+            "sheet": gsheet_url,
+            "header": 1,
+            "service_account": str(SECRETS_DIR / "service_account.json"),
+            "use_sheet_names_in_stored_paths": False,
+            "columns": {
+                "url": "link",
+                "status": "archive status",
+                "folder": "destination folder",
+                "archive": "archive location",
+                "date": "archive date",
+                "thumbnail": "thumbnail",
+                "timestamp": "upload timestamp",
+                "title": "upload title",
+                "text": "textual content",
+                "screenshot": "screenshot",
+                "hash": "hash",
+                "pdq_hash": "perceptual hashes",
+            },
+        }
+
+    # -- Google service account JSON (optional) -------------------------
+    sa_json = os.environ.get("GOOGLE_SERVICE_ACCOUNT_JSON", "")
+    if sa_json:
+        SECRETS_DIR.mkdir(parents=True, exist_ok=True)
+        sa_path = SECRETS_DIR / "service_account.json"
+        sa_path.write_text(sa_json)
+        print(f"[deploy] Wrote Google service account to {sa_path}")
+
+    # -- S3 storage (optional) ------------------------------------------
+    s3_bucket = os.environ.get("S3_BUCKET", "")
+    if s3_bucket:
+        config["steps"]["storages"].append("s3_storage")
+        config["s3_storage"] = {
+            "bucket": s3_bucket,
+            "region": os.environ.get("S3_REGION", "us-east-1"),
+            "key": os.environ.get("S3_KEY", ""),
+            "secret": os.environ.get("S3_SECRET", ""),
+            "endpoint_url": os.environ.get("S3_ENDPOINT", "https://s3.{region}.amazonaws.com"),
+            "cdn_url": os.environ.get(
+                "S3_CDN_URL",
+                "https://{bucket}.s3.{region}.amazonaws.com/{key}",
+            ),
+            "private": os.environ.get("S3_PRIVATE", "false").lower() == "true",
+            "random_no_duplicate": True,
+            "key_path": "random",
+        }
+
+    # -- Telegram extractor (optional) ----------------------------------
+    tg_api_id = os.environ.get("TELEGRAM_API_ID", "")
+    tg_api_hash = os.environ.get("TELEGRAM_API_HASH", "")
+    if tg_api_id and tg_api_hash:
+        config["steps"]["extractors"].append("telegram_extractor")
+        config["telegram_extractor"] = {
+            "api_id": tg_api_id,
+            "api_hash": tg_api_hash,
+        }
+        bot_token = os.environ.get("TELEGRAM_BOT_TOKEN", "")
+        if bot_token:
+            config["telegram_extractor"]["bot_token"] = bot_token
+
+    # -- Screenshot enricher (optional) ---------------------------------
+    if os.environ.get("ENABLE_SCREENSHOTS", "").lower() == "true":
+        config["steps"]["enrichers"].append("screenshot_enricher")
+        config["screenshot_enricher"] = {
+            "width": 1280,
+            "height": 7200,
+            "save_to_pdf": True,
+        }
+
+    # -- Thumbnail enricher (optional) ----------------------------------
+    if os.environ.get("ENABLE_THUMBNAILS", "").lower() == "true":
+        config["steps"]["enrichers"].append("thumbnail_enricher")
+        config["thumbnail_enricher"] = {
+            "thumbnails_per_minute": 60,
+            "max_thumbnails": 16,
+        }
+
+    # -- CSV database (optional) ----------------------------------------
+    if os.environ.get("ENABLE_CSV_DB", "").lower() == "true":
+        config["steps"]["databases"].append("csv_db")
+        config["csv_db"] = {
+            "csv_file": "/app/local_archive/db.csv",
+        }
+
+    return config
+
+
+def main():
+    config = build_config()
+
+    CONFIG_PATH.parent.mkdir(parents=True, exist_ok=True)
+    with open(CONFIG_PATH, "w") as f:
+        yaml.dump(config, f, default_flow_style=False, sort_keys=False)
+
+    print(f"[deploy] Generated config at {CONFIG_PATH}")
+    print(f"[deploy] Active steps: {config['steps']}")
+
+
+if __name__ == "__main__":
+    main()

deploy/gsheet_poller.py

@@ -0,0 +1,71 @@
+#!/usr/bin/env python3
+"""
+Background Google Sheets poller for auto-archiver cloud deployments.
+
+When GSHEET_URL is set, periodically runs auto-archiver with gsheet_feeder
+to check for new URLs in the configured spreadsheet. Runs as a daemon thread
+alongside the web UI.
+"""
+
+import logging
+import os
+import subprocess
+import threading
+import time
+
+logger = logging.getLogger("gsheet_poller")
+
+CONFIG_PATH = "/app/secrets/orchestration.yaml"
+
+
+def _poll_once():
+    """Run auto-archiver once to process any new rows in the Google Sheet."""
+    logger.info("Polling Google Sheet for new URLs...")
+    try:
+        result = subprocess.run(
+            ["python3", "-m", "auto_archiver", "--config", CONFIG_PATH],
+            capture_output=True,
+            text=True,
+            cwd="/app",
+            timeout=600,  # 10 minute timeout per poll
+        )
+        if result.returncode == 0:
+            logger.info("Sheet poll completed successfully.")
+        else:
+            logger.warning("Sheet poll exited with code %d: %s", result.returncode, result.stderr[-500:])
+    except subprocess.TimeoutExpired:
+        logger.error("Sheet poll timed out after 600s")
+    except Exception:
+        logger.exception("Sheet poll failed")
+
+
+def _poll_loop(interval: int):
+    """Run the poll loop at the given interval (seconds)."""
+    logger.info("Google Sheets poller started (interval=%ds)", interval)
+    while True:
+        _poll_once()
+        time.sleep(interval)
+
+
+def start_poller():
+    """
+    Start the Google Sheets poller as a daemon thread if GSHEET_URL is set.
+    Call this once at application startup.
+    """
+    gsheet_url = os.environ.get("GSHEET_URL", "")
+    if not gsheet_url:
+        logger.info("GSHEET_URL not set – Sheet poller disabled.")
+        return
+
+    interval = int(os.environ.get("POLL_INTERVAL", "300"))
+    if interval < 60:
+        interval = 60  # minimum 1 minute
+
+    thread = threading.Thread(
+        target=_poll_loop,
+        args=(interval,),
+        daemon=True,
+        name="gsheet-poller",
+    )
+    thread.start()
+    logger.info("Google Sheets poller thread started.")

deploy/pytest.ini

@@ -0,0 +1,2 @@
+[pytest]
+testpaths = tests

deploy/start.py

@@ -0,0 +1,37 @@
+#!/usr/bin/env python3
+"""
+Startup entrypoint for cloud deployments.
+
+1. Generates orchestration.yaml from environment variables
+2. Starts the Google Sheets poller (if GSHEET_URL is set)
+3. Starts the FastAPI web UI
+"""
+
+import os
+import logging
+
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s [%(name)s] %(levelname)s: %(message)s",
+)
+
+# Generate config from env vars
+from deploy.generate_config import main as generate_config  # noqa: E402
+
+generate_config()
+
+# Start gsheet poller (no-op if GSHEET_URL not set)
+from deploy.gsheet_poller import start_poller  # noqa: E402
+
+start_poller()
+
+# Start web server
+import uvicorn  # noqa: E402
+
+port = int(os.environ.get("PORT", "8080"))
+uvicorn.run(
+    "deploy.web_ui:app",
+    host="0.0.0.0",
+    port=port,
+    log_level="info",
+)

deploy/tests/test_generate_config.py

@@ -0,0 +1,354 @@
+"""Tests for deploy/generate_config.py – config generation from env vars."""
+
+import json
+import os
+from unittest.mock import patch
+
+import yaml
+
+from deploy.generate_config import build_config, main
+
+
+# ── Helpers ───────────────────────────────────────────────────────────
+
+
+def _env(**overrides):
+    """Return a clean env dict with only the given overrides (no leak from host)."""
+    # Clear all deploy-relevant env vars, then apply overrides
+    deploy_vars = [
+        "LOG_LEVEL",
+        "SUBTITLES",
+        "GSHEET_URL",
+        "GOOGLE_SERVICE_ACCOUNT_JSON",
+        "S3_BUCKET",
+        "S3_KEY",
+        "S3_SECRET",
+        "S3_REGION",
+        "S3_ENDPOINT",
+        "S3_CDN_URL",
+        "S3_PRIVATE",
+        "TELEGRAM_API_ID",
+        "TELEGRAM_API_HASH",
+        "TELEGRAM_BOT_TOKEN",
+        "ENABLE_SCREENSHOTS",
+        "ENABLE_THUMBNAILS",
+        "ENABLE_CSV_DB",
+    ]
+    clean = {k: v for k, v in os.environ.items() if k not in deploy_vars}
+    clean.update(overrides)
+    return clean
+
+
+# ── Base config (no optional env vars) ────────────────────────────────
+
+
+class TestBaseConfig:
+    """When no optional env vars are set, build_config returns a minimal working config."""
+
+    def test_base_steps(self):
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        steps = cfg["steps"]
+        assert steps["feeders"] == ["cli_feeder"]
+        assert steps["extractors"] == ["generic_extractor"]
+        assert steps["enrichers"] == ["hash_enricher"]
+        assert steps["databases"] == ["console_db"]
+        assert steps["storages"] == ["local_storage"]
+        assert steps["formatters"] == ["html_formatter"]
+
+    def test_base_has_required_module_configs(self):
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        assert "local_storage" in cfg
+        assert "generic_extractor" in cfg
+        assert "hash_enricher" in cfg
+        assert "html_formatter" in cfg
+
+    def test_default_log_level_is_info(self):
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        assert cfg["logging"]["level"] == "INFO"
+
+    def test_custom_log_level(self):
+        with patch.dict(os.environ, _env(LOG_LEVEL="DEBUG"), clear=True):
+            cfg = build_config()
+        assert cfg["logging"]["level"] == "DEBUG"
+
+    def test_authentication_present_and_empty(self):
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        assert cfg["authentication"] == {}
+
+    def test_local_storage_defaults(self):
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        ls = cfg["local_storage"]
+        assert ls["save_to"] == "/app/local_archive"
+        assert ls["path_generator"] == "flat"
+        assert ls["filename_generator"] == "static"
+
+    def test_subtitles_default_false(self):
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        assert cfg["generic_extractor"]["subtitles"] is False
+
+    def test_subtitles_enabled(self):
+        with patch.dict(os.environ, _env(SUBTITLES="true"), clear=True):
+            cfg = build_config()
+        assert cfg["generic_extractor"]["subtitles"] is True
+
+    def test_subtitles_case_insensitive(self):
+        with patch.dict(os.environ, _env(SUBTITLES="True"), clear=True):
+            cfg = build_config()
+        assert cfg["generic_extractor"]["subtitles"] is True
+
+    def test_no_optional_modules_present(self):
+        """Ensure optional modules don't appear when their env vars are absent."""
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        assert "gsheet_feeder" not in cfg
+        assert "s3_storage" not in cfg
+        assert "telegram_extractor" not in cfg
+        assert "screenshot_enricher" not in cfg
+        assert "thumbnail_enricher" not in cfg
+        assert "csv_db" not in cfg
+
+    def test_config_is_valid_yaml(self):
+        """The output dict should round-trip through YAML cleanly."""
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        dumped = yaml.dump(cfg)
+        reloaded = yaml.safe_load(dumped)
+        assert reloaded == cfg
+
+
+# ── Google Sheets ─────────────────────────────────────────────────────
+
+
+class TestGSheetConfig:
+    def test_gsheet_adds_feeder_and_db(self):
+        with patch.dict(os.environ, _env(GSHEET_URL="https://docs.google.com/spreadsheets/d/abc"), clear=True):
+            cfg = build_config()
+        assert "gsheet_feeder" in cfg["steps"]["feeders"]
+        assert "gsheet_db" in cfg["steps"]["databases"]
+
+    def test_gsheet_feeder_config(self):
+        url = "https://docs.google.com/spreadsheets/d/abc123"
+        with patch.dict(os.environ, _env(GSHEET_URL=url), clear=True):
+            cfg = build_config()
+        gf = cfg["gsheet_feeder"]
+        assert gf["sheet"] == url
+        assert gf["header"] == 1
+        assert "service_account" in gf
+        assert gf["columns"]["url"] == "link"
+        assert gf["columns"]["status"] == "archive status"
+
+    def test_gsheet_preserves_cli_feeder(self):
+        """cli_feeder should still be present even when gsheet is added."""
+        with patch.dict(os.environ, _env(GSHEET_URL="https://example.com/sheet"), clear=True):
+            cfg = build_config()
+        assert "cli_feeder" in cfg["steps"]["feeders"]
+
+    def test_service_account_json_written(self, tmp_path):
+        """When GOOGLE_SERVICE_ACCOUNT_JSON is set, it writes the file."""
+        sa_data = json.dumps({"type": "service_account", "project_id": "test"})
+        secrets_dir = tmp_path / "secrets"
+        with (
+            patch.dict(os.environ, _env(GOOGLE_SERVICE_ACCOUNT_JSON=sa_data), clear=True),
+            patch("deploy.generate_config.SECRETS_DIR", secrets_dir),
+        ):
+            build_config()
+        sa_path = secrets_dir / "service_account.json"
+        assert sa_path.exists()
+        assert json.loads(sa_path.read_text())["project_id"] == "test"
+
+
+# ── S3 storage ────────────────────────────────────────────────────────
+
+
+class TestS3Config:
+    def test_s3_adds_storage(self):
+        with patch.dict(os.environ, _env(S3_BUCKET="my-bucket"), clear=True):
+            cfg = build_config()
+        assert "s3_storage" in cfg["steps"]["storages"]
+        assert "local_storage" in cfg["steps"]["storages"]  # local still there
+
+    def test_s3_config_values(self):
+        env = _env(
+            S3_BUCKET="my-bucket",
+            S3_KEY="AKID",
+            S3_SECRET="shhh",
+            S3_REGION="eu-west-1",
+        )
+        with patch.dict(os.environ, env, clear=True):
+            cfg = build_config()
+        s3 = cfg["s3_storage"]
+        assert s3["bucket"] == "my-bucket"
+        assert s3["key"] == "AKID"
+        assert s3["secret"] == "shhh"
+        assert s3["region"] == "eu-west-1"
+        assert s3["private"] is False
+        assert s3["random_no_duplicate"] is True
+
+    def test_s3_defaults(self):
+        with patch.dict(os.environ, _env(S3_BUCKET="b"), clear=True):
+            cfg = build_config()
+        s3 = cfg["s3_storage"]
+        assert s3["region"] == "us-east-1"
+        assert "{region}" in s3["endpoint_url"]
+
+    def test_s3_private_flag(self):
+        with patch.dict(os.environ, _env(S3_BUCKET="b", S3_PRIVATE="true"), clear=True):
+            cfg = build_config()
+        assert cfg["s3_storage"]["private"] is True
+
+    def test_s3_custom_endpoint(self):
+        endpoint = "https://nyc3.digitaloceanspaces.com"
+        with patch.dict(os.environ, _env(S3_BUCKET="b", S3_ENDPOINT=endpoint), clear=True):
+            cfg = build_config()
+        assert cfg["s3_storage"]["endpoint_url"] == endpoint
+
+
+# ── Telegram ──────────────────────────────────────────────────────────
+
+
+class TestTelegramConfig:
+    def test_telegram_added_when_both_set(self):
+        env = _env(TELEGRAM_API_ID="12345", TELEGRAM_API_HASH="abc")
+        with patch.dict(os.environ, env, clear=True):
+            cfg = build_config()
+        assert "telegram_extractor" in cfg["steps"]["extractors"]
+        assert cfg["telegram_extractor"]["api_id"] == "12345"
+        assert cfg["telegram_extractor"]["api_hash"] == "abc"
+
+    def test_telegram_not_added_if_only_id(self):
+        with patch.dict(os.environ, _env(TELEGRAM_API_ID="12345"), clear=True):
+            cfg = build_config()
+        assert "telegram_extractor" not in cfg["steps"]["extractors"]
+
+    def test_telegram_not_added_if_only_hash(self):
+        with patch.dict(os.environ, _env(TELEGRAM_API_HASH="abc"), clear=True):
+            cfg = build_config()
+        assert "telegram_extractor" not in cfg["steps"]["extractors"]
+
+    def test_telegram_bot_token_optional(self):
+        env = _env(TELEGRAM_API_ID="12345", TELEGRAM_API_HASH="abc", TELEGRAM_BOT_TOKEN="bot:tok")
+        with patch.dict(os.environ, env, clear=True):
+            cfg = build_config()
+        assert cfg["telegram_extractor"]["bot_token"] == "bot:tok"
+
+    def test_telegram_no_bot_token(self):
+        env = _env(TELEGRAM_API_ID="12345", TELEGRAM_API_HASH="abc")
+        with patch.dict(os.environ, env, clear=True):
+            cfg = build_config()
+        assert "bot_token" not in cfg["telegram_extractor"]
+
+
+# ── Optional enrichers / databases ────────────────────────────────────
+
+
+class TestOptionalModules:
+    def test_screenshots_disabled_by_default(self):
+        with patch.dict(os.environ, _env(), clear=True):
+            cfg = build_config()
+        assert "screenshot_enricher" not in cfg["steps"]["enrichers"]
+
+    def test_screenshots_enabled(self):
+        with patch.dict(os.environ, _env(ENABLE_SCREENSHOTS="true"), clear=True):
+            cfg = build_config()
+        assert "screenshot_enricher" in cfg["steps"]["enrichers"]
+        assert cfg["screenshot_enricher"]["width"] == 1280
+
+    def test_thumbnails_enabled(self):
+        with patch.dict(os.environ, _env(ENABLE_THUMBNAILS="true"), clear=True):
+            cfg = build_config()
+        assert "thumbnail_enricher" in cfg["steps"]["enrichers"]
+        assert cfg["thumbnail_enricher"]["max_thumbnails"] == 16
+
+    def test_csv_db_enabled(self):
+        with patch.dict(os.environ, _env(ENABLE_CSV_DB="true"), clear=True):
+            cfg = build_config()
+        assert "csv_db" in cfg["steps"]["databases"]
+        assert cfg["csv_db"]["csv_file"] == "/app/local_archive/db.csv"
+
+    def test_case_insensitive_boolean(self):
+        with patch.dict(os.environ, _env(ENABLE_SCREENSHOTS="TRUE"), clear=True):
+            cfg = build_config()
+        assert "screenshot_enricher" in cfg["steps"]["enrichers"]
+
+
+# ── Combined / full config ────────────────────────────────────────────
+
+
+class TestCombinedConfig:
+    def test_all_optional_modules_together(self):
+        """Enable everything at once and verify no conflicts."""
+        env = _env(
+            GSHEET_URL="https://example.com/sheet",
+            S3_BUCKET="bucket",
+            S3_KEY="key",
+            S3_SECRET="secret",
+            TELEGRAM_API_ID="123",
+            TELEGRAM_API_HASH="abc",
+            TELEGRAM_BOT_TOKEN="tok",
+            ENABLE_SCREENSHOTS="true",
+            ENABLE_THUMBNAILS="true",
+            ENABLE_CSV_DB="true",
+        )
+        with patch.dict(os.environ, env, clear=True):
+            cfg = build_config()
+
+        steps = cfg["steps"]
+        assert "gsheet_feeder" in steps["feeders"]
+        assert "telegram_extractor" in steps["extractors"]
+        assert "screenshot_enricher" in steps["enrichers"]
+        assert "thumbnail_enricher" in steps["enrichers"]
+        assert "csv_db" in steps["databases"]
+        assert "gsheet_db" in steps["databases"]
+        assert "s3_storage" in steps["storages"]
+        assert "local_storage" in steps["storages"]
+
+        # All module configs present
+        for key in [
+            "gsheet_feeder",
+            "s3_storage",
+            "telegram_extractor",
+            "screenshot_enricher",
+            "thumbnail_enricher",
+            "csv_db",
+        ]:
+            assert key in cfg, f"{key} config missing"
+
+    def test_full_config_valid_yaml(self):
+        env = _env(
+            GSHEET_URL="https://example.com/sheet",
+            S3_BUCKET="bucket",
+            TELEGRAM_API_ID="123",
+            TELEGRAM_API_HASH="abc",
+            ENABLE_SCREENSHOTS="true",
+            ENABLE_CSV_DB="true",
+        )
+        with patch.dict(os.environ, env, clear=True):
+            cfg = build_config()
+        dumped = yaml.dump(cfg)
+        reloaded = yaml.safe_load(dumped)
+        assert reloaded == cfg
+
+
+# ── main() writes file ───────────────────────────────────────────────
+
+
+class TestMainFunction:
+    def test_main_writes_config_file(self, tmp_path):
+        config_path = tmp_path / "orchestration.yaml"
+        with patch.dict(os.environ, _env(), clear=True), patch("deploy.generate_config.CONFIG_PATH", config_path):
+            main()
+        assert config_path.exists()
+        cfg = yaml.safe_load(config_path.read_text())
+        assert cfg["steps"]["feeders"] == ["cli_feeder"]
+
+    def test_main_creates_parent_dirs(self, tmp_path):
+        config_path = tmp_path / "nested" / "dir" / "orchestration.yaml"
+        with patch.dict(os.environ, _env(), clear=True), patch("deploy.generate_config.CONFIG_PATH", config_path):
+            main()
+        assert config_path.exists()

deploy/tests/test_gsheet_poller.py

@@ -0,0 +1,124 @@
+"""Tests for deploy/gsheet_poller.py – background Google Sheets polling."""
+
+import os
+from unittest.mock import patch, MagicMock
+
+
+from deploy.gsheet_poller import start_poller, _poll_once
+
+
+# ── start_poller ──────────────────────────────────────────────────────
+
+
+class TestStartPoller:
+    def test_disabled_when_no_gsheet_url(self):
+        """No thread should be started when GSHEET_URL is empty."""
+        with (
+            patch.dict(os.environ, {"GSHEET_URL": ""}, clear=False),
+            patch("deploy.gsheet_poller.threading.Thread") as mock_thread,
+        ):
+            start_poller()
+        mock_thread.assert_not_called()
+
+    def test_disabled_when_gsheet_url_absent(self):
+        env = {k: v for k, v in os.environ.items() if k != "GSHEET_URL"}
+        with patch.dict(os.environ, env, clear=True), patch("deploy.gsheet_poller.threading.Thread") as mock_thread:
+            start_poller()
+        mock_thread.assert_not_called()
+
+    def test_starts_thread_when_gsheet_url_set(self):
+        with (
+            patch.dict(os.environ, {"GSHEET_URL": "https://example.com/sheet"}, clear=False),
+            patch("deploy.gsheet_poller.threading.Thread") as mock_thread,
+        ):
+            mock_instance = MagicMock()
+            mock_thread.return_value = mock_instance
+            start_poller()
+        mock_thread.assert_called_once()
+        assert mock_thread.call_args.kwargs["daemon"] is True
+        assert mock_thread.call_args.kwargs["name"] == "gsheet-poller"
+        mock_instance.start.assert_called_once()
+
+    def test_default_interval_300(self):
+        env = {"GSHEET_URL": "https://example.com/sheet"}
+        # Remove POLL_INTERVAL if present
+        clean_env = {k: v for k, v in os.environ.items() if k != "POLL_INTERVAL"}
+        clean_env.update(env)
+        with (
+            patch.dict(os.environ, clean_env, clear=True),
+            patch("deploy.gsheet_poller.threading.Thread") as mock_thread,
+        ):
+            mock_thread.return_value = MagicMock()
+            start_poller()
+        # interval should be passed as arg to _poll_loop
+        args = mock_thread.call_args.kwargs.get("args") or mock_thread.call_args[1].get("args")
+        assert args == (300,)
+
+    def test_custom_interval(self):
+        with (
+            patch.dict(os.environ, {"GSHEET_URL": "x", "POLL_INTERVAL": "600"}, clear=False),
+            patch("deploy.gsheet_poller.threading.Thread") as mock_thread,
+        ):
+            mock_thread.return_value = MagicMock()
+            start_poller()
+        args = mock_thread.call_args.kwargs.get("args") or mock_thread.call_args[1].get("args")
+        assert args == (600,)
+
+    def test_interval_minimum_enforced(self):
+        """Intervals below 60 should be clamped to 60."""
+        with (
+            patch.dict(os.environ, {"GSHEET_URL": "x", "POLL_INTERVAL": "10"}, clear=False),
+            patch("deploy.gsheet_poller.threading.Thread") as mock_thread,
+        ):
+            mock_thread.return_value = MagicMock()
+            start_poller()
+        args = mock_thread.call_args.kwargs.get("args") or mock_thread.call_args[1].get("args")
+        assert args == (60,)
+
+
+# ── _poll_once ────────────────────────────────────────────────────────
+
+
+class TestPollOnce:
+    def test_calls_subprocess_with_config(self):
+        with patch("deploy.gsheet_poller.subprocess.run") as mock_run:
+            mock_run.return_value = MagicMock(returncode=0, stderr="")
+            _poll_once()
+        mock_run.assert_called_once()
+        cmd = mock_run.call_args[0][0]
+        assert "auto_archiver" in " ".join(cmd)
+        assert "--config" in cmd
+
+    def test_handles_nonzero_exit(self):
+        """Should not raise on non-zero exit, just log a warning."""
+        with patch("deploy.gsheet_poller.subprocess.run") as mock_run:
+            mock_run.return_value = MagicMock(returncode=1, stderr="some error")
+            _poll_once()  # should not raise
+
+    def test_handles_timeout(self):
+        """Should not raise on timeout, just log."""
+        import subprocess
+
+        with patch("deploy.gsheet_poller.subprocess.run") as mock_run:
+            mock_run.side_effect = subprocess.TimeoutExpired(cmd="test", timeout=600)
+            _poll_once()  # should not raise
+
+    def test_handles_exception(self):
+        """Should not raise on arbitrary exceptions."""
+        with patch("deploy.gsheet_poller.subprocess.run") as mock_run:
+            mock_run.side_effect = OSError("broken")
+            _poll_once()  # should not raise
+
+    def test_uses_correct_config_path(self):
+        with patch("deploy.gsheet_poller.subprocess.run") as mock_run:
+            mock_run.return_value = MagicMock(returncode=0, stderr="")
+            _poll_once()
+        cmd = mock_run.call_args[0][0]
+        config_idx = cmd.index("--config")
+        assert cmd[config_idx + 1] == "/app/secrets/orchestration.yaml"
+
+    def test_timeout_set(self):
+        with patch("deploy.gsheet_poller.subprocess.run") as mock_run:
+            mock_run.return_value = MagicMock(returncode=0, stderr="")
+            _poll_once()
+        assert mock_run.call_args[1]["timeout"] == 600

deploy/tests/test_web_ui.py

@@ -0,0 +1,310 @@
+"""Tests for deploy/web_ui.py – FastAPI web interface."""
+
+from unittest.mock import patch, AsyncMock
+
+import pytest
+from fastapi.testclient import TestClient
+
+
+# ── Fixtures ──────────────────────────────────────────────────────────
+
+
+@pytest.fixture(autouse=True)
+def _reset_state():
+    """Reset in-memory state between tests."""
+    import deploy.web_ui as mod
+
+    mod._valid_sessions.clear()
+    mod._jobs.clear()
+    yield
+    mod._valid_sessions.clear()
+    mod._jobs.clear()
+
+
+@pytest.fixture
+def client_no_auth():
+    """Test client with auth disabled (no AUTH_PASSWORD)."""
+    with patch.object(__import__("deploy.web_ui", fromlist=["web_ui"]), "AUTH_PASSWORD", ""):
+        from deploy.web_ui import app
+
+        yield TestClient(app, raise_server_exceptions=False)
+
+
+@pytest.fixture
+def client_with_auth():
+    """Test client with auth enabled."""
+    with patch.object(__import__("deploy.web_ui", fromlist=["web_ui"]), "AUTH_PASSWORD", "secret123"):
+        from deploy.web_ui import app
+
+        yield TestClient(app, raise_server_exceptions=False)
+
+
+def _login(client, password="secret123"):
+    """Helper: log in and return the session cookie."""
+    resp = client.post("/login", data={"password": password}, follow_redirects=False)
+    return resp.cookies.get("aa_session")
+
+
+# ── Health check ──────────────────────────────────────────────────────
+
+
+class TestHealthCheck:
+    def test_status_returns_ok(self, client_no_auth):
+        resp = client_no_auth.get("/status")
+        assert resp.status_code == 200
+        assert resp.json() == {"status": "ok"}
+
+    def test_status_no_auth_required(self, client_with_auth):
+        resp = client_with_auth.get("/status")
+        assert resp.status_code == 200
+        assert resp.json() == {"status": "ok"}
+
+
+# ── Auth disabled ─────────────────────────────────────────────────────
+
+
+class TestNoAuth:
+    def test_index_accessible(self, client_no_auth):
+        resp = client_no_auth.get("/")
+        assert resp.status_code == 200
+        assert "Auto Archiver" in resp.text
+
+    def test_login_page_redirects_to_index(self, client_no_auth):
+        resp = client_no_auth.get("/login", follow_redirects=False)
+        assert resp.status_code == 302
+        assert resp.headers["location"] == "/"
+
+    def test_login_post_redirects_to_index(self, client_no_auth):
+        resp = client_no_auth.post("/login", data={"password": "anything"}, follow_redirects=False)
+        assert resp.status_code == 302
+
+    def test_no_logout_link_shown(self, client_no_auth):
+        resp = client_no_auth.get("/")
+        assert "Logout" not in resp.text
+
+
+# ── Auth enabled ──────────────────────────────────────────────────────
+
+
+class TestAuth:
+    def test_index_redirects_to_login(self, client_with_auth):
+        resp = client_with_auth.get("/", follow_redirects=False)
+        assert resp.status_code == 307
+        assert resp.headers["location"] == "/login"
+
+    def test_login_page_renders(self, client_with_auth):
+        resp = client_with_auth.get("/login")
+        assert resp.status_code == 200
+        assert "Password" in resp.text
+
+    def test_wrong_password_returns_401(self, client_with_auth):
+        resp = client_with_auth.post("/login", data={"password": "wrong"})
+        assert resp.status_code == 401
+        assert "Wrong password" in resp.text
+
+    def test_correct_password_sets_cookie(self, client_with_auth):
+        resp = client_with_auth.post("/login", data={"password": "secret123"}, follow_redirects=False)
+        assert resp.status_code == 302
+        assert "aa_session" in resp.cookies
+
+    def test_authenticated_access(self, client_with_auth):
+        cookie = _login(client_with_auth)
+        client_with_auth.cookies.set("aa_session", cookie)
+        resp = client_with_auth.get("/")
+        assert resp.status_code == 200
+        assert "Auto Archiver" in resp.text
+
+    def test_logout_clears_session(self, client_with_auth):
+        cookie = _login(client_with_auth)
+        client_with_auth.cookies.set("aa_session", cookie)
+        resp = client_with_auth.get("/logout", follow_redirects=False)
+        assert resp.status_code == 302
+        # After logout, index should redirect to login again
+        client_with_auth.cookies.clear()
+        resp = client_with_auth.get("/", follow_redirects=False)
+        assert resp.status_code == 307
+
+    def test_logout_link_shown_when_auth_enabled(self, client_with_auth):
+        cookie = _login(client_with_auth)
+        client_with_auth.cookies.set("aa_session", cookie)
+        resp = client_with_auth.get("/")
+        assert "Logout" in resp.text
+
+    def test_results_requires_auth(self, client_with_auth):
+        resp = client_with_auth.get("/results", follow_redirects=False)
+        assert resp.status_code == 307
+
+    def test_invalid_session_rejected(self, client_with_auth):
+        client_with_auth.cookies.set("aa_session", "bogus-token")
+        resp = client_with_auth.get("/", follow_redirects=False)
+        assert resp.status_code == 307
+
+
+# ── Archive submission ────────────────────────────────────────────────
+
+
+class TestArchive:
+    def test_archive_creates_job(self, client_no_auth):
+        with patch("deploy.web_ui._run_archive", new_callable=AsyncMock):
+            resp = client_no_auth.post(
+                "/archive",
+                data={"urls": "https://example.com\nhttps://example.org"},
+                follow_redirects=False,
+            )
+        assert resp.status_code == 303
+        assert resp.headers["location"] == "/"
+
+        from deploy.web_ui import _jobs
+
+        assert len(_jobs) == 1
+        assert _jobs[0]["urls"] == ["https://example.com", "https://example.org"]
+        assert _jobs[0]["status"] == "running"
+
+    def test_archive_empty_urls_returns_400(self, client_no_auth):
+        resp = client_no_auth.post("/archive", data={"urls": "   \n  \n"})
+        assert resp.status_code == 400
+
+    def test_archive_strips_whitespace(self, client_no_auth):
+        with patch("deploy.web_ui._run_archive", new_callable=AsyncMock):
+            client_no_auth.post(
+                "/archive",
+                data={"urls": "  https://example.com  \n\n  https://example.org  \n"},
+                follow_redirects=False,
+            )
+        from deploy.web_ui import _jobs
+
+        assert _jobs[0]["urls"] == ["https://example.com", "https://example.org"]
+
+    def test_archive_requires_auth(self, client_with_auth):
+        resp = client_with_auth.post(
+            "/archive",
+            data={"urls": "https://example.com"},
+            follow_redirects=False,
+        )
+        assert resp.status_code == 307
+
+
+# ── Results page ──────────────────────────────────────────────────────
+
+
+class TestResults:
+    def test_results_empty(self, client_no_auth, tmp_path):
+        with patch("deploy.web_ui.ARCHIVE_DIR", tmp_path):
+            resp = client_no_auth.get("/results")
+        assert resp.status_code == 200
+        assert "No archived files yet" in resp.text
+
+    def test_results_lists_files(self, client_no_auth, tmp_path):
+        (tmp_path / "test.html").write_text("<html>archived</html>")
+        (tmp_path / "video.mp4").write_bytes(b"\x00" * 10)
+        with patch("deploy.web_ui.ARCHIVE_DIR", tmp_path):
+            resp = client_no_auth.get("/results")
+        assert resp.status_code == 200
+        assert "test.html" in resp.text
+        assert "video.mp4" in resp.text
+
+    def test_results_nonexistent_dir(self, client_no_auth, tmp_path):
+        with patch("deploy.web_ui.ARCHIVE_DIR", tmp_path / "nonexistent"):
+            resp = client_no_auth.get("/results")
+        assert resp.status_code == 200
+        assert "No archived files yet" in resp.text
+
+
+# ── File serving ──────────────────────────────────────────────────────
+
+
+class TestFileServing:
+    def test_serve_existing_file(self, client_no_auth, tmp_path):
+        (tmp_path / "report.html").write_text("<html>done</html>")
+        with patch("deploy.web_ui.ARCHIVE_DIR", tmp_path):
+            resp = client_no_auth.get("/files/report.html")
+        assert resp.status_code == 200
+
+    def test_serve_nonexistent_file(self, client_no_auth, tmp_path):
+        with patch("deploy.web_ui.ARCHIVE_DIR", tmp_path):
+            resp = client_no_auth.get("/files/nope.txt")
+        assert resp.status_code == 404
+
+    def test_path_traversal_blocked(self, client_no_auth, tmp_path):
+        # Create a file outside the archive dir
+        outside = tmp_path / "outside"
+        outside.mkdir()
+        (outside / "secret.txt").write_text("secret")
+        archive = tmp_path / "archive"
+        archive.mkdir()
+        # Symlink into archive pointing outside
+        (archive / "escape").symlink_to(outside / "secret.txt")
+        with patch("deploy.web_ui.ARCHIVE_DIR", archive):
+            resp = client_no_auth.get("/files/escape")
+        assert resp.status_code == 403
+
+
+# ── Job rendering ─────────────────────────────────────────────────────
+
+
+class TestJobRendering:
+    def test_no_jobs_shows_message(self, client_no_auth):
+        resp = client_no_auth.get("/")
+        assert "No archiving jobs yet" in resp.text
+
+    def test_jobs_shown_in_table(self, client_no_auth):
+        from deploy.web_ui import _jobs
+
+        _jobs.append(
+            {
+                "id": 1,
+                "urls": ["https://example.com"],
+                "status": "done",
+                "started": "2026-01-01 00:00 UTC",
+                "output": "",
+            }
+        )
+        resp = client_no_auth.get("/")
+        assert "example.com" in resp.text
+        assert "done" in resp.text
+
+    def test_many_urls_truncated(self, client_no_auth):
+        from deploy.web_ui import _jobs
+
+        _jobs.append(
+            {
+                "id": 1,
+                "urls": [f"https://example.com/{i}" for i in range(10)],
+                "status": "running",
+                "started": "2026-01-01 00:00 UTC",
+                "output": "",
+            }
+        )
+        resp = client_no_auth.get("/")
+        assert "+7 more" in resp.text
+
+
+# ── HTML template rendering ──────────────────────────────────────────
+
+
+class TestTemplates:
+    """Verify HTML templates can be .format()-ed without KeyError."""
+
+    def test_login_html_renders(self):
+        from deploy.web_ui import LOGIN_HTML
+
+        result = LOGIN_HTML.format(error="")
+        assert "Auto Archiver" in result
+
+    def test_login_html_renders_with_error(self):
+        from deploy.web_ui import LOGIN_HTML
+
+        result = LOGIN_HTML.format(error='<p class="err">Nope</p>')
+        assert "Nope" in result
+
+    def test_main_html_renders(self):
+        from deploy.web_ui import MAIN_HTML
+
+        result = MAIN_HTML.format(logout="", jobs_html="")
+        assert "Auto Archiver" in result
+
+    def test_results_html_renders(self):
+        from deploy.web_ui import RESULTS_HTML
+
+        result = RESULTS_HTML.format(file_list="<p>empty</p>")
+        assert "Archived Files" in result

deploy/web_ui.py

@@ -0,0 +1,269 @@
+#!/usr/bin/env python3
+"""
+Minimal web UI for auto-archiver cloud deployments.
+
+Provides:
+  - GET  /          → HTML form to submit URLs for archiving
+  - POST /archive   → Runs auto-archiver on submitted URLs
+  - GET  /results   → Lists archived files available for download
+  - GET  /files/{path} → Serves archived files
+  - GET  /status    → Health check
+"""
+
+import asyncio
+import html
+import os
+import secrets
+from datetime import datetime, timezone
+from pathlib import Path
+
+from fastapi import Depends, FastAPI, Form, HTTPException, Request, status
+from fastapi.responses import FileResponse, HTMLResponse, RedirectResponse
+
+AUTH_PASSWORD = os.environ.get("AUTH_PASSWORD", "")
+ARCHIVE_DIR = Path("/app/local_archive")
+CONFIG_PATH = Path("/app/secrets/orchestration.yaml")
+COOKIE_NAME = "aa_session"
+
+# In-memory session tokens (reset on restart, which is fine for this use case)
+_valid_sessions: set[str] = set()
+# In-memory job log
+_jobs: list[dict] = []
+
+app = FastAPI(title="Auto Archiver", docs_url=None, redoc_url=None)
+
+
+# ── Auth helpers ──────────────────────────────────────────────────────
+
+
+def _check_auth(request: Request):
+    """Dependency: redirect to /login if auth is enabled and session is missing."""
+    if not AUTH_PASSWORD:
+        return  # auth disabled
+    token = request.cookies.get(COOKIE_NAME, "")
+    if token not in _valid_sessions:
+        raise HTTPException(
+            status_code=status.HTTP_307_TEMPORARY_REDIRECT,
+            headers={"Location": "/login"},
+        )
+
+
+# ── Pages ─────────────────────────────────────────────────────────────
+
+LOGIN_HTML = """<!DOCTYPE html>
+<html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1">
+<title>Auto Archiver – Login</title>
+<style>
+  body {{ font-family: system-ui, sans-serif; max-width: 420px; margin: 80px auto; padding: 0 1rem; }}
+  h1 {{ font-size: 1.4rem; }}
+  input[type=password], button {{ font-size: 1rem; padding: .5rem .8rem; }}
+  input[type=password] {{ width: 100%; box-sizing: border-box; margin: .5rem 0; }}
+  button {{ cursor: pointer; background: #2563eb; color: #fff; border: none; border-radius: 4px; }}
+  .err {{ color: #dc2626; }}
+</style></head><body>
+<h1>🔐 Auto Archiver</h1>
+<form method="POST" action="/login">
+  <label>Password<br><input type="password" name="password" autofocus required></label><br>
+  <button type="submit">Log in</button>
+  {error}
+</form></body></html>"""
+
+
+MAIN_HTML = """<!DOCTYPE html>
+<html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1">
+<title>Auto Archiver</title>
+<style>
+  body {{ font-family: system-ui, sans-serif; max-width: 700px; margin: 2rem auto; padding: 0 1rem; line-height: 1.6; }}
+  h1 {{ font-size: 1.5rem; }}
+  textarea {{ width: 100%; box-sizing: border-box; font-size: .95rem; font-family: monospace; }}
+  button {{ font-size: 1rem; padding: .5rem 1.2rem; cursor: pointer; background: #2563eb; color: #fff; border: none; border-radius: 4px; margin-top: .5rem; }}
+  table {{ border-collapse: collapse; width: 100%; margin-top: 1rem; }}
+  th, td {{ border: 1px solid #e5e7eb; padding: .4rem .6rem; text-align: left; font-size: .9rem; }}
+  th {{ background: #f9fafb; }}
+  .status {{ padding: 2px 8px; border-radius: 4px; font-size: .85rem; }}
+  .running {{ background: #fef3c7; color: #92400e; }}
+  .done {{ background: #d1fae5; color: #065f46; }}
+  .failed {{ background: #fee2e2; color: #991b1b; }}
+  a {{ color: #2563eb; }}
+  .info {{ color: #6b7280; font-size: .9rem; }}
+  nav {{ display: flex; gap: 1rem; align-items: center; }}
+  nav a {{ text-decoration: none; }}
+</style></head><body>
+<nav>
+  <h1>📦 Auto Archiver</h1>
+  <a href="/results">Browse files</a>
+  {logout}
+</nav>
+<form method="POST" action="/archive">
+  <label for="urls"><strong>URLs to archive</strong> (one per line)</label><br>
+  <textarea id="urls" name="urls" rows="5" placeholder="https://example.com/post&#10;https://youtube.com/watch?v=..." required></textarea><br>
+  <button type="submit">Archive</button>
+</form>
+{jobs_html}
+</body></html>"""
+
+
+RESULTS_HTML = """<!DOCTYPE html>
+<html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1">
+<title>Auto Archiver – Files</title>
+<style>
+  body {{ font-family: system-ui, sans-serif; max-width: 700px; margin: 2rem auto; padding: 0 1rem; }}
+  h1 {{ font-size: 1.4rem; }}
+  a {{ color: #2563eb; }}
+  li {{ margin: .3rem 0; font-family: monospace; font-size: .9rem; }}
+</style></head><body>
+<h1>📁 Archived Files</h1>
+<p><a href="/">← Back</a></p>
+{file_list}
+</body></html>"""
+
+
+# ── Routes ────────────────────────────────────────────────────────────
+
+
+@app.get("/login", response_class=HTMLResponse)
+async def login_page():
+    if not AUTH_PASSWORD:
+        return RedirectResponse("/", status_code=302)
+    return LOGIN_HTML.format(error="")
+
+
+@app.post("/login")
+async def login_submit(password: str = Form(...)):
+    if not AUTH_PASSWORD:
+        return RedirectResponse("/", status_code=302)
+    if password != AUTH_PASSWORD:
+        return HTMLResponse(
+            LOGIN_HTML.format(error='<p class="err">Wrong password.</p>'),
+            status_code=401,
+        )
+    token = secrets.token_urlsafe(32)
+    _valid_sessions.add(token)
+    resp = RedirectResponse("/", status_code=302)
+    resp.set_cookie(COOKIE_NAME, token, httponly=True, samesite="lax", max_age=86400 * 30)
+    return resp
+
+
+@app.get("/", response_class=HTMLResponse)
+async def index(request: Request, _=Depends(_check_auth)):
+    logout = '<a href="/logout">Logout</a>' if AUTH_PASSWORD else ""
+    jobs_html = _render_jobs()
+    return MAIN_HTML.format(logout=logout, jobs_html=jobs_html)
+
+
+@app.post("/archive")
+async def archive(request: Request, urls: str = Form(...), _=Depends(_check_auth)):
+    url_list = [u.strip() for u in urls.strip().splitlines() if u.strip()]
+    if not url_list:
+        raise HTTPException(400, "No URLs provided")
+
+    job = {
+        "id": len(_jobs) + 1,
+        "urls": url_list,
+        "status": "running",
+        "started": datetime.now(timezone.utc).strftime("%Y-%m-%d %H:%M UTC"),
+        "output": "",
+    }
+    _jobs.insert(0, job)
+
+    # Run in background so the user sees the page immediately
+    asyncio.create_task(_run_archive(job))
+    return RedirectResponse("/", status_code=303)
+
+
+@app.get("/results", response_class=HTMLResponse)
+async def results(request: Request, _=Depends(_check_auth)):
+    if not ARCHIVE_DIR.exists():
+        return RESULTS_HTML.format(file_list="<p>No archived files yet.</p>")
+
+    files = sorted(ARCHIVE_DIR.rglob("*"), key=lambda p: p.stat().st_mtime, reverse=True)
+    files = [f for f in files if f.is_file()]
+
+    if not files:
+        return RESULTS_HTML.format(file_list="<p>No archived files yet.</p>")
+
+    items = []
+    for f in files[:200]:  # cap listing
+        rel = f.relative_to(ARCHIVE_DIR)
+        items.append(f'<li><a href="/files/{rel}">{html.escape(str(rel))}</a></li>')
+
+    return RESULTS_HTML.format(file_list="<ul>" + "\n".join(items) + "</ul>")
+
+
+@app.get("/files/{path:path}")
+async def serve_file(path: str, request: Request, _=Depends(_check_auth)):
+    full = ARCHIVE_DIR / path
+    if not full.exists() or not full.is_file():
+        raise HTTPException(404, "File not found")
+    # Security: ensure the resolved path is within ARCHIVE_DIR
+    try:
+        full.resolve().relative_to(ARCHIVE_DIR.resolve())
+    except ValueError:
+        raise HTTPException(403, "Forbidden")
+    return FileResponse(full)
+
+
+@app.get("/status")
+async def health():
+    return {"status": "ok"}
+
+
+@app.get("/logout")
+async def logout(request: Request):
+    token = request.cookies.get(COOKIE_NAME, "")
+    _valid_sessions.discard(token)
+    resp = RedirectResponse("/login", status_code=302)
+    resp.delete_cookie(COOKIE_NAME)
+    return resp
+
+
+# ── Helpers ───────────────────────────────────────────────────────────
+
+
+async def _run_archive(job: dict):
+    """Run auto-archiver as a subprocess for the given URLs."""
+    cmd = [
+        "python3",
+        "-m",
+        "auto_archiver",
+        "--config",
+        str(CONFIG_PATH),
+    ] + job["urls"]
+
+    try:
+        proc = await asyncio.create_subprocess_exec(
+            *cmd,
+            stdout=asyncio.subprocess.PIPE,
+            stderr=asyncio.subprocess.STDOUT,
+            cwd="/app",
+        )
+        stdout, _ = await proc.communicate()
+        job["output"] = stdout.decode(errors="replace")[-5000:]  # keep last 5k chars
+        job["status"] = "done" if proc.returncode == 0 else "failed"
+    except Exception as e:
+        job["output"] = str(e)
+        job["status"] = "failed"
+
+
+def _render_jobs() -> str:
+    if not _jobs:
+        return '<p class="info">No archiving jobs yet. Submit URLs above to get started.</p>'
+
+    rows = []
+    for j in _jobs[:50]:
+        urls_str = html.escape(", ".join(j["urls"][:3]))
+        if len(j["urls"]) > 3:
+            urls_str += f" (+{len(j['urls']) - 3} more)"
+        status_cls = j["status"]
+        rows.append(
+            f"<tr><td>{j['id']}</td>"
+            f"<td>{urls_str}</td>"
+            f'<td><span class="status {status_cls}">{j["status"]}</span></td>'
+            f"<td>{j['started']}</td></tr>"
+        )
+
+    return (
+        "<h2>Recent Jobs</h2>"
+        "<table><thead><tr><th>#</th><th>URLs</th><th>Status</th><th>Started</th></tr></thead>"
+        "<tbody>" + "\n".join(rows) + "</tbody></table>"
+    )

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [project]
 name = "auto-archiver"
-version = "1.2.3"
+version = "1.2.5"
 description = "Automatically archive links to videos, images, and social media content from Google Sheets (and more)."
 
 requires-python = ">=3.10,<3.13"

railway.json

@@ -0,0 +1,99 @@
+{
+  "$schema": "https://railway.app/railway.schema.json",
+  "build": {
+    "dockerfilePath": "deploy/Dockerfile"
+  },
+  "deploy": {
+    "startCommand": "python3 -m deploy.start",
+    "healthcheckPath": "/status",
+    "healthcheckTimeout": 30,
+    "restartPolicyType": "ON_FAILURE",
+    "restartPolicyMaxRetries": 5
+  },
+  "variables": {
+    "AUTH_PASSWORD": {
+      "description": "Password to access your archiver web interface",
+      "required": true
+    },
+    "GSHEET_URL": {
+      "description": "Google Sheet URL to monitor for new URLs (leave empty to disable)",
+      "required": false,
+      "default": ""
+    },
+    "GOOGLE_SERVICE_ACCOUNT_JSON": {
+      "description": "Full JSON contents of your Google service account key (required for Sheets)",
+      "required": false,
+      "default": ""
+    },
+    "POLL_INTERVAL": {
+      "description": "Seconds between Google Sheet checks (min 60)",
+      "required": false,
+      "default": "300"
+    },
+    "S3_BUCKET": {
+      "description": "S3 bucket name for storage (leave empty for local-only)",
+      "required": false,
+      "default": ""
+    },
+    "S3_KEY": {
+      "description": "S3 access key ID",
+      "required": false,
+      "default": ""
+    },
+    "S3_SECRET": {
+      "description": "S3 secret access key",
+      "required": false,
+      "default": ""
+    },
+    "S3_REGION": {
+      "description": "S3 region (e.g. us-east-1, nyc3 for DO Spaces)",
+      "required": false,
+      "default": "us-east-1"
+    },
+    "S3_ENDPOINT": {
+      "description": "S3 endpoint URL template",
+      "required": false,
+      "default": "https://s3.{region}.amazonaws.com"
+    },
+    "S3_CDN_URL": {
+      "description": "Public CDN URL template for archived files",
+      "required": false,
+      "default": "https://{bucket}.s3.{region}.amazonaws.com/{key}"
+    },
+    "TELEGRAM_API_ID": {
+      "description": "Telegram API ID from https://my.telegram.org",
+      "required": false,
+      "default": ""
+    },
+    "TELEGRAM_API_HASH": {
+      "description": "Telegram API hash from https://my.telegram.org",
+      "required": false,
+      "default": ""
+    },
+    "TELEGRAM_BOT_TOKEN": {
+      "description": "Telegram bot token from @BotFather",
+      "required": false,
+      "default": ""
+    },
+    "ENABLE_SCREENSHOTS": {
+      "description": "Set to true to capture full-page screenshots",
+      "required": false,
+      "default": "false"
+    },
+    "ENABLE_THUMBNAILS": {
+      "description": "Set to true to generate video thumbnails",
+      "required": false,
+      "default": "false"
+    },
+    "ENABLE_CSV_DB": {
+      "description": "Set to true to save a CSV log of archived items",
+      "required": false,
+      "default": "false"
+    },
+    "LOG_LEVEL": {
+      "description": "Logging level: DEBUG, INFO, WARNING, ERROR",
+      "required": false,
+      "default": "INFO"
+    }
+  }
+}

src/auto_archiver/core/metadata.py

@@ -181,6 +181,9 @@ class Metadata:
         media_hashes = set()
         new_media = []
         for m in self.media:
+            if not m.filename:
+                new_media.append(m)
+                continue
             h = m.get("hash")
             if not h:
                 h = calculate_hash_in_chunks(hashlib.sha256(), int(1.6e7), m.filename)

src/auto_archiver/modules/antibot_extractor_enricher/antibot_extractor_enricher.py

@@ -73,6 +73,7 @@ class AntibotExtractorEnricher(Extractor, Enricher):
         if self.enrich(result):
             result.status = "antibot"
             return result
+        return False
 
     def _prepare_user_data_dir(self):
         if self.user_data_dir:

src/auto_archiver/modules/generic_extractor/bluesky.py

@@ -39,12 +39,18 @@ class Bluesky(GenericDropin):
         media_url = "https://bsky.social/xrpc/com.atproto.sync.getBlob?cid={}&did={}"
         for image_media in image_medias:
             url = media_url.format(image_media["image"]["ref"]["$link"], post["author"]["did"])
-            image_media = archiver.download_from_url(url)
-            media.append(Media(image_media))
+            filename = archiver.download_from_url(url)
+            if filename:
+                media.append(Media(filename))
+            else:
+                logger.warning(f"Failed to download Bluesky image from {url}")
         for video_media in video_medias:
             url = media_url.format(video_media["ref"]["$link"], post["author"]["did"])
-            video_media = archiver.download_from_url(url)
-            media.append(Media(video_media))
+            filename = archiver.download_from_url(url)
+            if filename:
+                media.append(Media(filename))
+            else:
+                logger.warning(f"Failed to download Bluesky video from {url}")
         return media
 
     def _get_post_data(self, post: dict) -> dict:

src/auto_archiver/modules/generic_extractor/generic_extractor.py

@@ -204,8 +204,11 @@ class GenericExtractor(Extractor):
         if thumbnail_url:
             try:
                 cover_image_path = self.download_from_url(thumbnail_url)
-                media = Media(cover_image_path)
-                metadata.add_media(media, id="cover")
+                if cover_image_path:
+                    media = Media(cover_image_path)
+                    metadata.add_media(media, id="cover")
+                else:
+                    logger.warning(f"Failed to download cover image from {thumbnail_url}")
             except Exception as e:
                 logger.error(f"Could not download cover image {thumbnail_url}: {e}")
 

src/auto_archiver/modules/generic_extractor/truth.py

@@ -1,6 +1,7 @@
 from typing import Type
 
 from auto_archiver.utils import traverse_obj
+from auto_archiver.utils.custom_logger import logger
 from auto_archiver.core.metadata import Metadata, Media
 from auto_archiver.core.extractor import Extractor
 from yt_dlp.extractor.common import InfoExtractor
@@ -58,6 +59,9 @@ class Truth(GenericDropin):
         # add the media
         for media in post.get("media_attachments", []):
             filename = archiver.download_from_url(media["url"])
+            if not filename:
+                logger.warning(f"Failed to download media from {media['url']}")
+                continue
             result.add_media(Media(filename), id=media.get("id"))
 
         return result

src/auto_archiver/modules/generic_extractor/twitter.py

@@ -157,5 +157,8 @@ class Twitter(GenericDropin):
                 mimetype = variant["content_type"]
             ext = mimetypes.guess_extension(mimetype)
             media.filename = archiver.download_from_url(media.get("src"), f"{slugify(url)}_{i}{ext}")
+            if not media.filename:
+                logger.warning(f"Failed to download media from {media.get('src')}")
+                continue
             result.add_media(media)
         return result

src/auto_archiver/modules/hash_enricher/hash_enricher.py

@@ -25,6 +25,9 @@ class HashEnricher(Enricher):
         logger.debug(f"Calculating media hashes with algo={self.algorithm}")
 
         for i, m in enumerate(to_enrich.media):
+            if not m.filename:
+                logger.warning(f"Skipping hash for media without filename: {m}")
+                continue
             if len(hd := self.calculate_hash(m.filename)):
                 to_enrich.media[i].set("hash", f"{self.algorithm}:{hd}")
 

src/auto_archiver/modules/instagram_api_extractor/instagram_api_extractor.py

@@ -99,7 +99,10 @@ class InstagramAPIExtractor(Extractor):
         result.set_title(user.get("full_name", username)).set("data", user)
         if pic_url := user.get("profile_pic_url_hd", user.get("profile_pic_url")):
             filename = self.download_from_url(pic_url)
-            result.add_media(Media(filename=filename), id="profile_picture")
+            if filename:
+                result.add_media(Media(filename=filename), id="profile_picture")
+            else:
+                logger.warning(f"Failed to download profile picture from {pic_url}")
 
         count_posts = 0
         if self.full_profile:
@@ -202,7 +205,10 @@ class InstagramAPIExtractor(Extractor):
 
         if cover_media := h_info.get("cover_media", {}).get("cropped_image_version", {}).get("url"):
             filename = self.download_from_url(cover_media)
-            result.add_media(Media(filename=filename), id=f"cover_media highlight {id}")
+            if filename:
+                result.add_media(Media(filename=filename), id=f"cover_media highlight {id}")
+            else:
+                logger.warning(f"Failed to download cover media from {cover_media}")
 
         items = h_info.get("items", [])[::-1]  # newest to oldest
         items = items[: min(max_to_download, len(items))]
@@ -345,7 +351,10 @@ class InstagramAPIExtractor(Extractor):
         image_media = None
         if image_url := item.get("thumbnail_url"):
             filename = self.download_from_url(image_url, verbose=False)
-            image_media = Media(filename=filename)
+            if filename:
+                image_media = Media(filename=filename)
+            else:
+                logger.warning(f"Failed to download thumbnail from {image_url}")
 
         # retrieve video info
         best_id = item.get("id", item.get("pk"))
@@ -357,16 +366,19 @@ class InstagramAPIExtractor(Extractor):
 
         if video_url := item.get("video_url"):
             filename = self.download_from_url(video_url, verbose=False)
-            video_media = Media(filename=filename)
-            if taken_at:
-                video_media.set("date", taken_at)
-            if code:
-                video_media.set("url", f"https://www.instagram.com/p/{code}")
-            if caption_text:
-                video_media.set("text", caption_text)
-            video_media.set("preview", [image_media])
-            video_media.set("data", [item])
-            return item, video_media, f"{context or 'video'} {best_id}"
+            if filename:
+                video_media = Media(filename=filename)
+                if taken_at:
+                    video_media.set("date", taken_at)
+                if code:
+                    video_media.set("url", f"https://www.instagram.com/p/{code}")
+                if caption_text:
+                    video_media.set("text", caption_text)
+                video_media.set("preview", [image_media])
+                video_media.set("data", [item])
+                return item, video_media, f"{context or 'video'} {best_id}"
+            else:
+                logger.warning(f"Failed to download video from {video_url}")
         elif image_media:
             if taken_at:
                 image_media.set("date", taken_at)

src/auto_archiver/modules/meta_enricher/meta_enricher.py

@@ -25,6 +25,9 @@ class MetaEnricher(Enricher):
         logger.debug(f"Calculating archive file sizes for {len(to_enrich.media)} media files")
         total_size = 0
         for media in to_enrich.get_all_media():
+            if not media.filename:
+                logger.warning(f"Skipping file size for media without filename: {media}")
+                continue
             file_stats = os.stat(media.filename)
             media.set("bytes", file_stats.st_size)
             media.set("size", self.human_readable_bytes(file_stats.st_size))

src/auto_archiver/modules/telegram_extractor/telegram_extractor.py

@@ -49,10 +49,18 @@ class TelegramExtractor(Extractor):
             if not len(image_urls):
                 return False
             for img_url in image_urls:
-                result.add_media(Media(self.download_from_url(img_url)))
+                filename = self.download_from_url(img_url)
+                if not filename:
+                    logger.warning(f"Failed to download image from {img_url}")
+                    continue
+                result.add_media(Media(filename))
         else:
             video_url = video.get("src")
-            m_video = Media(self.download_from_url(video_url))
+            video_filename = self.download_from_url(video_url)
+            if not video_filename:
+                logger.warning(f"Failed to download video from {video_url}")
+                return False
+            m_video = Media(video_filename)
             # extract duration from HTML
             try:
                 duration = s.find_all("time")[0].contents[0]

src/auto_archiver/modules/telethon_extractor/telethon_extractor.py

@@ -1,3 +1,4 @@
+import asyncio
 import os
 import shutil
 import re
@@ -53,6 +54,16 @@ class TelethonExtractor(Extractor):
         logger.debug(f"Making a copy of the session file {base_session_filepath} to {self.session_file}.session")
         shutil.copy(base_session_filepath, f"{self.session_file}.session")
 
+        # ensure a running event loop exists (Needed when used by Celery workers which may close the default one)
+        try:
+            loop = asyncio.get_event_loop()
+            if loop.is_closed():
+                loop = asyncio.new_event_loop()
+                asyncio.set_event_loop(loop)
+        except RuntimeError:
+            loop = asyncio.new_event_loop()
+            asyncio.set_event_loop(loop)
+
         # initiate the client
         self.client = TelegramClient(self.session_file, self.api_id, self.api_hash)
 
@@ -190,6 +201,9 @@ class TelethonExtractor(Extractor):
                             )
                         for i, om_url in enumerate(other_media_urls):
                             filename = self.download_from_url(om_url, f"{chat}_{group_id}_{i}")
+                            if not filename:
+                                logger.warning(f"Failed to download media from {om_url}")
+                                continue
                             result.add_media(Media(filename=filename), id=f"{group_id}_{i}")
 
                     filename_dest = os.path.join(self.tmp_dir, f"{chat}_{group_id}", str(mp.id))

src/auto_archiver/modules/twitter_api_extractor/twitter_api_extractor.py

@@ -114,6 +114,9 @@ class TwitterApiExtractor(Extractor):
                 logger.info(f"Found media {media}")
                 ext = mimetypes.guess_extension(mimetype)
                 media.filename = self.download_from_url(media.get("src"), f"{slugify(url)}_{i}{ext}")
+                if not media.filename:
+                    logger.warning(f"Failed to download media from {media.get('src')}")
+                    continue
                 result.add_media(media)
 
         result.set_content(

tests/extractors/test_antibot_extractor_enricher.py

@@ -53,6 +53,7 @@ class TestAntibotExtractorEnricher(TestExtractorBase):
     }
 
     @pytest.mark.download
+    @pytest.mark.flaky(reruns=2, reruns_delay=5)
     @pytest.mark.parametrize(
         "url,in_title,in_text,image_count,video_count,skip_ci",
         [
@@ -128,6 +129,7 @@ class TestAntibotExtractorEnricher(TestExtractorBase):
         item = make_item(url)
         result = self.extractor.download(item)
 
+        assert result, f"download() returned {result!r} — Selenium may have failed (e.g., window close timeout)"
         assert result.status == "antibot", "Expected status to be 'antibot'"
 
         # Check title contains all required words (case-insensitive)

tests/extractors/test_telethon_extractor.py

@@ -1,3 +1,4 @@
+import asyncio
 import os
 from datetime import date
 
@@ -60,3 +61,53 @@ def test_valid_url_regex(url, expected, get_lazy_module):
 def test_invite_pattern_regex(invite, expected, get_lazy_module):
     match = TelethonExtractor.invite_pattern.search(invite)
     assert bool(match) == expected
+
+
+def test_setup_with_closed_event_loop(get_lazy_module, tmp_path, mocker):
+    """
+    Simulate the Celery worker scenario where the asyncio event loop is closed
+    before setup() runs. The fix should create a new event loop so that
+    TelegramClient.start() does not raise 'Event loop is closed'.
+    """
+    # create a session file so setup doesn't fail on missing file
+    session_file = tmp_path / "test.session"
+    session_file.touch()
+
+    # close the current event loop to simulate a Celery worker environment
+    loop = asyncio.new_event_loop()
+    asyncio.set_event_loop(loop)
+    loop.close()
+
+    lazy_module = get_lazy_module("telethon_extractor")
+    module = lazy_module.load(
+        {"telethon_extractor": {"session_file": str(session_file), "api_id": 123, "api_hash": "ABC"}}
+    )
+
+    # setup should have succeeded and a new open event loop should exist
+    new_loop = asyncio.get_event_loop()
+    assert not new_loop.is_closed()
+    assert module.client is not None
+
+
+def test_setup_with_no_event_loop(get_lazy_module, tmp_path, mocker):
+    """
+    Simulate the scenario where there is no current event loop at all
+    (e.g. running in a non-main thread). The fix should create one.
+    """
+    session_file = tmp_path / "test.session"
+    session_file.touch()
+
+    # Remove the current event loop entirely
+    # In Python 3.12+, get_event_loop() in a non-main thread raises RuntimeError
+    mocker.patch("asyncio.get_event_loop", side_effect=RuntimeError("no current event loop"))
+    new_loop_mock = mocker.MagicMock()
+    new_loop_mock.is_closed.return_value = False
+    mocker.patch("asyncio.new_event_loop", return_value=new_loop_mock)
+    set_loop = mocker.patch("asyncio.set_event_loop")
+
+    lazy_module = get_lazy_module("telethon_extractor")
+    lazy_module.load({"telethon_extractor": {"session_file": str(session_file), "api_id": 123, "api_hash": "ABC"}})
+
+    # a new event loop should have been created and set
+    asyncio.new_event_loop.assert_called_once()
+    set_loop.assert_called_once_with(new_loop_mock)

tests/test_none_filename_handling.py

@@ -0,0 +1,259 @@
+"""
+Tests for handling Media objects with None filename.
+
+When download_from_url fails, it returns None. Various enrichers and
+the metadata deduplication logic must gracefully handle Media objects
+where filename is None, rather than crashing with TypeError.
+"""
+
+from datetime import datetime, timezone
+from unittest.mock import MagicMock
+
+import pytest
+
+from auto_archiver.core.metadata import Metadata, Media
+from auto_archiver.modules.hash_enricher import HashEnricher
+from auto_archiver.modules.meta_enricher import MetaEnricher
+
+
+# ── HashEnricher ──────────────────────────────────────────────────────
+
+
+class TestHashEnricherNoneFilename:
+    """hash_enricher should skip media with None filename without crashing."""
+
+    @pytest.fixture(autouse=True)
+    def setup(self, setup_module):
+        self.enricher = setup_module(HashEnricher, {"algorithm": "SHA-256", "chunksize": 100})
+
+    def test_skips_none_filename(self):
+        m = Metadata().set_url("https://example.com")
+        media = Media(filename=None)
+        media.set("src", "https://example.com/video.mp4")
+        m.add_media(media)
+
+        # Should not raise
+        self.enricher.enrich(m)
+        # No hash should be set
+        assert m.media[0].get("hash") is None
+
+    def test_hashes_valid_skips_none(self, tmp_path):
+        """Mix of valid and None-filename media: only valid ones get hashed."""
+        valid_file = tmp_path / "test.txt"
+        valid_file.write_text("hello world")
+
+        m = Metadata().set_url("https://example.com")
+        m.add_media(Media(filename=str(valid_file)))
+        m.add_media(Media(filename=None))
+
+        self.enricher.enrich(m)
+
+        assert m.media[0].get("hash") is not None
+        assert m.media[1].get("hash") is None
+
+    def test_all_none_filenames(self):
+        """All media have None filename – enricher should not crash."""
+        m = Metadata().set_url("https://example.com")
+        m.add_media(Media(filename=None))
+        m.add_media(Media(filename=None))
+
+        self.enricher.enrich(m)
+
+        assert len(m.media) == 2
+        for media in m.media:
+            assert media.get("hash") is None
+
+
+# ── MetaEnricher ──────────────────────────────────────────────────────
+
+
+class TestMetaEnricherNoneFilename:
+    """meta_enricher should skip media with None filename without crashing."""
+
+    @pytest.fixture(autouse=True)
+    def setup(self, setup_module):
+        self.enricher = setup_module(MetaEnricher, {})
+
+    def test_skips_none_filename(self):
+        m = Metadata().set_url("https://example.com")
+        m.set("_processed_at", datetime.now(timezone.utc))
+        media = Media(filename=None)
+        media.set("src", "https://example.com/video.mp4")
+        m.add_media(media)
+
+        # Should not raise
+        self.enricher.enrich(m)
+        assert m.get("total_bytes") == 0
+
+    def test_sizes_valid_skips_none(self, tmp_path):
+        """Mix of valid and None-filename media: only valid ones get sized."""
+        valid_file = tmp_path / "test.txt"
+        valid_file.write_text("A" * 500)
+
+        m = Metadata().set_url("https://example.com")
+        m.set("_processed_at", datetime.now(timezone.utc))
+        m.add_media(Media(filename=str(valid_file)))
+        m.add_media(Media(filename=None))
+
+        self.enricher.enrich(m)
+
+        assert m.media[0].get("bytes") == 500
+        assert m.media[1].get("bytes") is None
+        assert m.get("total_bytes") == 500
+
+
+# ── Metadata.remove_duplicate_media_by_hash ───────────────────────────
+
+
+class TestRemoveDuplicateMediaNoneFilename:
+    """remove_duplicate_media_by_hash should keep media with None filename."""
+
+    def test_none_filename_kept(self):
+        m = Metadata().set_url("https://example.com")
+        none_media = Media(filename=None)
+        none_media.set("src", "https://example.com/video.mp4")
+        m.add_media(none_media)
+
+        m.remove_duplicate_media_by_hash()
+
+        assert len(m.media) == 1
+        assert m.media[0].filename is None
+
+    def test_none_and_valid_mixed(self, tmp_path):
+        """None-filename media is kept alongside valid-filename media."""
+        valid_file = tmp_path / "test.txt"
+        valid_file.write_text("content")
+
+        m = Metadata().set_url("https://example.com")
+        m.add_media(Media(filename=str(valid_file)))
+        none_media = Media(filename=None)
+        none_media.set("src", "https://example.com/video.mp4")
+        m.add_media(none_media)
+
+        m.remove_duplicate_media_by_hash()
+
+        assert len(m.media) == 2
+
+    def test_multiple_none_filename_all_kept(self):
+        """Multiple None-filename media are all kept (can't deduplicate without file)."""
+        m = Metadata().set_url("https://example.com")
+        m.add_media(Media(filename=None))
+        m.add_media(Media(filename=None))
+
+        m.remove_duplicate_media_by_hash()
+
+        assert len(m.media) == 2
+
+
+# ── Twitter dropin create_metadata ────────────────────────────────────
+
+
+class TestTwitterDropinNoneFilename:
+    """Twitter dropin should skip media when download_from_url returns None."""
+
+    @pytest.fixture
+    def twitter_dropin(self):
+        from auto_archiver.modules.generic_extractor.twitter import Twitter
+
+        return Twitter()
+
+    def test_create_metadata_skips_failed_photo_download(self, twitter_dropin):
+        """When download_from_url returns None for a photo, it's not added to media."""
+        tweet = {
+            "user": {"name": "Test User"},
+            "created_at": "Sun Feb 08 18:45:00 +0000 2026",
+            "full_text": "Test tweet with photo",
+            "entities": {
+                "media": [
+                    {"type": "photo", "media_url_https": "https://pbs.twimg.com/media/test.jpg"},
+                ]
+            },
+        }
+
+        mock_archiver = MagicMock()
+        mock_archiver.download_from_url.return_value = None  # simulate failed download
+
+        result = twitter_dropin.create_metadata(tweet, None, mock_archiver, "https://x.com/test/status/123")
+
+        # The result should have no media since the download failed
+        assert len(result.media) == 0
+
+    def test_create_metadata_skips_failed_video_download(self, twitter_dropin):
+        """When download_from_url returns None for a video, it's not added to media."""
+        tweet = {
+            "user": {"name": "Test User"},
+            "created_at": "Sun Feb 08 18:45:00 +0000 2026",
+            "full_text": "Test tweet with video",
+            "entities": {
+                "media": [
+                    {
+                        "type": "video",
+                        "video_info": {
+                            "variants": [
+                                {
+                                    "url": "https://video.twimg.com/vid/1280x720/test.mp4",
+                                    "content_type": "video/mp4",
+                                },
+                            ]
+                        },
+                    },
+                ]
+            },
+        }
+
+        mock_archiver = MagicMock()
+        mock_archiver.download_from_url.return_value = None
+
+        result = twitter_dropin.create_metadata(tweet, None, mock_archiver, "https://x.com/test/status/123")
+
+        assert len(result.media) == 0
+
+    def test_create_metadata_keeps_successful_download(self, twitter_dropin, tmp_path):
+        """When download_from_url succeeds, media is added."""
+        tweet = {
+            "user": {"name": "Test User"},
+            "created_at": "Sun Feb 08 18:45:00 +0000 2026",
+            "full_text": "Test tweet with photo",
+            "entities": {
+                "media": [
+                    {"type": "photo", "media_url_https": "https://pbs.twimg.com/media/test.jpg"},
+                ]
+            },
+        }
+
+        test_file = tmp_path / "test.jpg"
+        test_file.write_text("fake image data")
+
+        mock_archiver = MagicMock()
+        mock_archiver.download_from_url.return_value = str(test_file)
+
+        result = twitter_dropin.create_metadata(tweet, None, mock_archiver, "https://x.com/test/status/123")
+
+        assert len(result.media) == 1
+        assert result.media[0].filename == str(test_file)
+
+    def test_create_metadata_mixed_downloads(self, twitter_dropin, tmp_path):
+        """One download succeeds, one fails – only successful one is kept."""
+        tweet = {
+            "user": {"name": "Test User"},
+            "created_at": "Sun Feb 08 18:45:00 +0000 2026",
+            "full_text": "Test tweet with two photos",
+            "entities": {
+                "media": [
+                    {"type": "photo", "media_url_https": "https://pbs.twimg.com/media/test1.jpg"},
+                    {"type": "photo", "media_url_https": "https://pbs.twimg.com/media/test2.jpg"},
+                ]
+            },
+        }
+
+        test_file = tmp_path / "test1.jpg"
+        test_file.write_text("fake image data")
+
+        mock_archiver = MagicMock()
+        # First call succeeds, second fails
+        mock_archiver.download_from_url.side_effect = [str(test_file), None]
+
+        result = twitter_dropin.create_metadata(tweet, None, mock_archiver, "https://x.com/test/status/123")
+
+        assert len(result.media) == 1
+        assert result.media[0].filename == str(test_file)