Hacker News new | ask | show | jobs
by jarofgreen 170 days ago
This seems to be about hosting an Sqlite database on a static website like GitHub Pages - this can be a great plan, there is also Datasette in a browser now: https://github.com/simonw/datasette-lite

But that's different from how you collect the data in a git repository in the first place - or are you suggesting just putting a Sqlite file in a git repository? If so I can think of one big reason against that.
1 comments
dleslie 170 days ago
Yes, I'm suggesting hosting it on GitHub, leveraging their git lfs support. Just treat it like a binary blob and periodically update with a tagged release.
link
jarofgreen 170 days ago
It's not clear if you are suggesting accepting contributions to the SQLite file via PR from people (but accepting contributions is generally the point of why people put these on projects on GitHub).

But if you are I wouldn't recommend it.

PR's won't be able to show diff's. Worse, as soon as multiple people send a PR at once you'll have a really painful merge to resolve, and GitHub's tools won't help you at all. And you can't edit the files in GitHub's web UI.

I recommend one file per record, JSON, YAML, whatever non-binary format you want. But then you get:

* PR's with diff's that show you what's being changed

* Files that technical people can edit directly in GitHub's web editor

* If 2 people make PR's on different records at once it's an easy merge with no conflicts

* If 2 people make PR's on the same record at once ... ok, you might now have a merge conflict to resolve but it's in an easy text file and GitHub UI will let you see what it is.

You can of course then compile these data files into a SQLite file that can be served in a static website nicely - in fact if you see my other comments on this post I have a tool that does this. And on that note, sorry, I've done a few projects in this space so I have views :-)

link
dleslie 168 days ago
Nah, git is terrible with binaries. But the SQL database can be rebuilt periodically; the problem being solved is replacing the git querying with SQL.

Could even follow your record model, and use that as data to populate the db.

link