Sherief N. Farouk wrote: > I'm currently trying to port an application that does a lot of work based on > two NTFS features: Streams, and Transactional NTFS. Since TxF is a HUGE > feature, with a lot of potential pitfalls, I thought I might do this as a > layer over SQLite: a db file with one table, multiple rows: one for each > stream. I'd love to hear what everyone thinks, and if there's a more > obvious/performance oriented way to do this. How should I go around storing > the stream data, as the files are not text (and indeed, lots of NULLs lie > within). I'll mostly need to read/write sub-regions of the file from > multiple apps, all running at the same time. >
That was pretty vague, at least to me. Can you explain what NTFS streams and transactions do and how you use them? Can you tell us how your application uses these features? Without more information about what you are trying to do I don't think I can comment on the efficacy of using SQLite. Dennis Cote _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
