Add a get_data and store_data functions

[tohanss]

We're looking at creating a RepoBee plugin that will require data that we probably want to store inside a repo based on some sort of key. To test a first draft of how to do this, a simple solution is to just create a get_data function, that first gets the repo if we don't have it locally, where #672 is probably helpful, and then reads a file or part of a file from it

The idea is to create a repobee plugin that fetches some data from a student repo, say that we want to know the average amount of Lines of Code in a/all student repos, then repobee already has the functionality to pull/update all repos, so we run something like (names WIP)

$ repobee scrape lines (defaults to all students)

Which writes/updates the number of lines for each students as its own entry in some file format (like JSON) inside a cache repository

A more advanced example would be to get a bunch of data from commits, like so:

$ repobee scrape commits

Which saves information for each commit under each student, the data could be commit times/dates and number of additions/deletions, whatever is useful.

A separate command/plugin can then use that data. An example is building a bot or automated system that keeps track of students that start working on assignments very late, giving the TA valuable feedback that they can relay to the student

[marceloFA]

Can I work on this?

[slarse]

Hi @marceloFA,

You're very welcome to come up with a draft, but this issue is not well defined, and the scraper that's suggested to use the functionality isn't something that's being worked on afaik. If we were to actually add this to RepoBee, we'd first need to come up with at least one driving use case!

Was there something about this issue in particular that intrigued you? Otherwise, there are some better options. I will be putting up new issues in the coming days for not very exciting but necessary maintenance and refactoring stuff. There's also some low-hanging fruit in some of the plugin repositories, for example repobee-feedback.

[marceloFA]

Thanks! I'll start with one of the low hanging fruits or one of the upcoming issues then :)

[marceloFA]

Hi @marceloFA,

You're very welcome to come up with a draft, but this issue is not well defined, and the scraper that's suggested to use the functionality isn't something that's being worked on afaik. If we were to actually add this to RepoBee, we'd first need to come up with at least one driving use case!

Was there something about this issue in particular that intrigued you? Otherwise, there are some better options. I will be putting up new issues in the coming days for not very exciting but necessary maintenance and refactoring stuff. There's also some low-hanging fruit in some of the plugin repositories, for example repobee-feedback.

Please let me know when those new issues come up 😄

[slarse]

Hi @marceloFA,

I'm currently a bit preoccupied with a major rewrite of the CLI to accommodate an auto-generated GUI (see #912). This is something I don't think you can help with as the code for the current CLI is rather convoluted and hard to understand. So easy-to-tackle issues will take another week or two to appear.

In the mean time, there are a bunch of relatively standalone issues in repobee-csvgrades. I think repobee/repobee-csvgrades#18 would be a relatively easy thing for you to start with given what you did on repobee-feedback :)

[marceloFA]

Oh thanks! Have a good work with the CLI and also happy new year!

[slarse]

Happy new year to you as well!