sqlbot在使用中的一些问题与建议
Nobody has claimed this yet.
- Dominant language
- JavaScript
- Stars
- 6.8k
- Forks
- 882
- Avg merge
- 4h 37m
- Merged PRs (30d)
- 32
Description
使用背景:尝试在项目中使用sqlbot辅助用户通过对话的方式查看感兴趣的指标。
问题1:数据库中存在上百张表,如果选择整个库中的所有表,会导致系统提示词超长且可能溢出,使用成本也会很高。
建议:能否对问题做一个预处理机制,通过预处理筛选出相关度topn的表,当前这个预处理机制可能会很复杂依赖表自身的描述信息,字段信息等因素,需要具体讨论可行的方案。
问题2:同一个数据源中,能否提供主题域的概念?比如电商场景下:一个库中有上百张表,常见的主题域可能是订单主题域、用户主题域、商品主题域......,通过划分主题域并且在主题域中配置好表与表之间的关系,在对话时,先通过问题预测主题域,获取命中主题域对应表的元数据,理论上也能减小系统提示词的长度。当前这个功能不影响现有的对话流程,如果没有主题域,退化为现有的对话逻辑。
问题3:现有的表关系管理模块,是否能够利用大模型的能力自动生成初版表关系,然后人工进行校对?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files, tests, or entry points are named. Start by locating the dialogue flow, metadata/table-relation management, and data-source configuration, then determine whether table preselection, topic domains, and model-assisted relationship generation can be specified independently; done requires an agreed design and documented behavior for each proposal.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- javascript
- Domain
- ai, databases
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100