On 23 May 2016, at 9:08pm, Piyush Shah <piyush.shah at slalom.com> wrote:
> We know about sqldiff https://www.sqlite.org/sqldiff.html and would use it > to generate diffs and patch them but we were hoping there was a way to do do > a binary diff of the database because otherwise we will have to figure out a > way for the client to run the sqldiff utility to patch the database. There's no easy way to integrate Sparkle and sqldiff so that they work together like that. The amount of work it would take would be equivalent to writing your own tool which updated the database across the internet. If the database file is small, just let Sparkle update it with the other components of the application. If the database file is big, it may be worth doing work on it. Does the database file compress well as a ZIP file ? Then distribute it as a .zip. On startup, detect if the decompressed version is not present, or has a datestamp before the .zip version. Do you have roughly equal parts of the file taken up with tables which never change and tables which change with each update ? Then split the file up into two database files and use ATTACH to access them both via one connection. Simon.
