User:StockCrawlerBot

About this account

This is a bot account for a free desktop video creation tool that builds a semantic search index of multimedia assets for video creators.

Operator

Independent developer. Contact via this bot's talk page. The operator monitors and responds within a few days.

Task details

Read-only API queries to MediaWiki Action API:

  • action=query, list=search — keyword search
  • action=query, prop=imageinfo — file metadata + thumbnail URL

Builds an internal semantic search index for in-app stock asset discovery. This bot does NOT modify any Wikimedia data (no edit / upload / delete operations).

Operation mode

  • Automatic (no human assistance per request)
  • Runs intermittently during the indexing phase (a few hours per day,
 spread over 2–6 months)
  • Runs on-demand afterwards (only when a user query hits a cache miss)

Maximum request rate

  • Initial indexing: ~5,000–20,000 requests/day (≈1 req/sec spread)
  • Steady state: ~1,000 requests/day
  • Concurrency: single-threaded, sequential requests
  • Respects maxlag=5 parameter (auto-pause on server lag)

Language / program

  • Language: Python 3.11
  • HTTP library: httpx (direct HTTP, no MediaWiki client libs)
  • User-Agent: StockCrawler/0.1 (contact via this account's talk page) python-httpx

Compliance

  • Read-only (no edit / upload / delete operations)
  • Respects maxlag=5 parameter
  • Single-threaded; sequential requests
  • Caches all API responses per Wikimedia guidelines (24h+ where applicable)
  • Falls back gracefully on HTTP 429 (exponential backoff)
  • Full attribution preserved per CC license requirements
 (photographer / source URL / license metadata stored)