Quick start

Once douban is on your PATH:

douban --help       # see the command tree
douban version      # build info

Search across categories, then drill in:

# Search books, movies, music
douban search python
douban search "流浪地球" --type movie

# Fast autocomplete
douban suggest "三体"

# One book in full
douban book 1084336

# Movie charts
douban top250 --limit 50
douban chart
douban nowplaying --city beijing

# A curated list (豆列)
douban doulist 240962

Every command prints a table by default and pipes straight into jq with -o json or -o jsonl:

douban top250 -o jsonl | jq .url

Or build a local mirror you can query offline. Seed the frontier, crawl it, then export:

douban seed ids --type book --from 1084336 --to 1084340
douban crawl --limit 20
douban info
douban export --type book -o jsonl | jq .

The crawl is resumable, so you can stop and resume it freely. See the mirror guide for the full workflow.