market-table

# Market Table Canvas Use `market-table` with scenario `quotes`. Never put quote rows or invented market data directly in `config`. Choose exactly one of the following data paths. ## Fixed presets Pass one of these exact English strings; do not create other preset names: - `a-share-top-gainers`: A-share gainers Top 20 - `a-share-top-losers`: A-share losers Top 20 - `a-share-most-active`: A-share turnover amount Top 20 - `a-share-highest-turnover-rate`: A-share turnover-rate Top 20 - `a-share-largest-market-cap`: A-share market-cap Top 20 ```typescript canvas_spawn({ kind: 'market-table', scenario: 'quotes', config: JSON.stringify({ preset: 'a-share-top-gainers' }), }); ``` ## Local FTShare result For custom filters, boards, sorting, or result sizes, call the FTShare `stock-daec-stocks` Skill through Python and save its stdout JSON to a local file. Do not copy the returned `items` into `canvas_spawn`. Pass only the absolute file path: ```bash python <FTSHARE_RUN_PY> stock-daec-stocks \ --board xshg --filter 'name.contains("银行")' \ --page 1 --page_size 50 --order_by "change_rate desc" \ > /absolute/workspace/path/bank-quotes.json ``` ```typescript canvas_spawn({ kind: 'market-table', scenario: 'quotes', config: JSON.stringify({ data_file: '/absolute/workspace/path/bank-quotes.json', }), }); ``` `dataFile` is an alias of `data_file`; never provide both. The file must be an FTShare-style object with a non-empty `items` array and must not exceed 16 MiB. Prefer a bounded page over all 5000+ stocks. On config/data failure, repair the file or config and update the same Canvas ID. ## Controls - Arrow/Page/Home/End: navigate; Space: toggle row; `c`: clear selection. - `a`: attach selected rows or cursor row; Enter: request analysis; `e`: export JSON. - `q` or Escape: close through the socket protocol. When `canvas_wait` is available, call it with this Canvas ID while waiting for an interactive selection or analysis request. Otherwise, rely on the host's automatic Canvas event delivery.