Hacker News Reader

This is a great idea. There are so many APIs and text dumps from websites and databases that are XML, and it would be nice to be able to query them with regular relational operations. I'll be following your project! I will say that I believe a streaming backend would be more suitable, since most of the XML dumps I'm thinking of are humongous and won't always fit in memory.

View on HN

Thanks! About fitting into memory; the way the library currently works should technically work even on very big files, as the XML is iteratively converted into a SQL database. You can choose whether you want that database to be in a file or in memory. The only thing that could break right now is synchronizing back to XML after having made changes with UPDATE or INSERT statements, because I'm not paging results.

Reply on news.ycombinator.com