Arc'teryx Outlet Stock Watch (Shoes / Sizes)

This repo monitors Arc'teryx Outlet products and sends a Telegram alert when a target size goes from out of stock → in stock. By default, it also alerts when a product+size is first seen and is already in stock (e.g. first run or newly listed products). The primary entrypoint is watch_stock.py.

What it does

• Scrapes a category page (e.g. men's footwear) to collect product URLs (via Selenium).
• Fetches each product page and parses the embedded __NEXT_DATA__ JSON (via requests).
• Checks variant stock status for your target size(s).
• Sends Telegram alerts on restock events (and optionally repeats).

Quick start (recommended)

./setup.sh
cp stock_watch_config.example.json stock_watch_config.json

# Telegram (optional but recommended)
export TELEGRAM_BOT_TOKEN="123456789:replace-me"
export TELEGRAM_CHAT_ID="123456789"   # or TELEGRAM_CHAT_IDS="id1,id2"

./run.sh

run.sh behavior:

• If you pass --config ..., it forwards args to watch_stock.py.
• Else, if stock_watch_config.json exists, it runs watch_stock.py --config stock_watch_config.json.
• Else, it runs watch_stock.py with CLI defaults.

Requirements

• Python 3.7+
• If you use category_url: Google Chrome / Chromium installed (Selenium will manage the driver automatically on recent Selenium versions).

On a Linux server (Ubuntu example), you typically need Chrome/Chromium and common libraries. If Chrome fails to start, install the missing deps indicated in the error log.

Telegram configuration

Environment variables:

• TELEGRAM_BOT_TOKEN
• TELEGRAM_CHAT_ID (single) or TELEGRAM_CHAT_IDS (comma-separated)

The notifier is implemented in telegram_notifier.py.

JSON config guide (stock_watch_config.json)

Copy the example and edit it:

cp stock_watch_config.example.json stock_watch_config.json

Top-level fields

• data_dir: where state is stored (default data)
• state_file: persisted baseline + history (default data/stock_watch_state.json)
• log_file: log path (default logs/watch_stock.log)
• log_level: DEBUG|INFO|WARNING|ERROR
• show_browser: set true to run a visible browser (debugging category scraping)
• render_wait_seconds: seconds to wait after opening category page
• scroll_times: how many times to scroll the category page (lazy-load)
• notify_on_first_run: true by default (alert when a product+size is first seen in stock; set false to only alert on out-of-stock → in-stock transitions after baseline)
• max_products: global cap (0 = no limit)
• no_category_prefilter: if false, keywords are pre-filtered using category tile text to reduce product-page requests
• repeat.max_notifications_per_item: max alerts per product+size per “restock cycle” (min 1)
• repeat.repeat_interval_seconds: minimum seconds between repeated alerts (0 = no limit)
• error_notify.enabled: send a Telegram alert when scraping fails (network errors, blocking, parsing failures)
• error_notify.repeat_interval_seconds: throttle repeated error alerts across runs (0 = no limit)

watches[]

You can define multiple independent watches. Each watch supports:

• name: label for logs/alerts
• category_url: category page URL to scrape (requires Selenium + Chrome)
• product_urls: optional explicit product URLs; if provided, category scraping is skipped
• keywords: list of keywords to match (case-insensitive)
• sizes: list of size labels to monitor (e.g. ["8","8.5"] or ["M","L"])
• max_products: per-watch cap (0 = no limit)
• no_category_prefilter: override no_category_prefilter for this watch

Example config

{
  "version": 1,
  "data_dir": "data",
  "state_file": "data/stock_watch_state.json",
  "log_file": "logs/watch_stock.log",
  "log_level": "INFO",
  "show_browser": false,
  "render_wait_seconds": 10,
  "scroll_times": 3,
  "notify_on_first_run": true,
  "repeat": {
    "max_notifications_per_item": 3,
    "repeat_interval_seconds": 1800
  },
  "error_notify": {
    "enabled": true,
    "repeat_interval_seconds": 3600
  },
  "watches": [
    {
      "name": "shoes-gtx",
      "category_url": "https://outlet.arcteryx.com/ca/en/c/mens/footwear",
      "keywords": ["waterproof", "gtx", "gore-tex"],
      "sizes": ["8", "8.5"],
      "max_products": 0,
      "no_category_prefilter": false
    }
  ]
}

CLI usage (no config file)

python3 watch_stock.py \
  --category-url "https://outlet.arcteryx.com/ca/en/c/mens/footwear" \
  --size 8 \
  --keyword waterproof --keyword gtx --keyword gore-tex

Useful flags:

• --dry-run: do everything except sending Telegram
• --no-notify-on-first-run: disable alerts for first-seen in-stock items
• --no-category-prefilter: fetch all product pages and match keywords there (more requests, more coverage)
• --show-browser: debug category scraping with a visible browser

Output files

• data/stock_watch_state.json: baseline + per-product state
• logs/watch_stock.log: logs

Scheduling on Linux (cron)

crontab -e

Run every 30 minutes:

*/30 * * * * cd $HOME/arcteryx-monitor && ./run.sh >> logs/watch_stock.log 2>&1

Troubleshooting

• Selenium/Chrome fails to start: install Chrome/Chromium and the missing system libraries; check logs/watch_stock.log and logs/chromedriver.log.
• No products found on category page: increase render_wait_seconds, increase scroll_times, or run with --show-browser to inspect the page.
• Too many requests: keep a reasonable schedule (e.g. 30–60 minutes) and avoid aggressive scraping.

Optional / legacy

• monitor_unified.py can also run catalog-change monitoring, but the repo is optimized for watch_stock.py.
• Other scripts in the repo are legacy/deprecated and not maintained.