search

grambank / grambank-analysed

3 stars 0 forks source link

simplify the data generation #73

Closed SimonGreenhill closed 2 years ago

SimonGreenhill commented 2 years ago

It's painful to have to install every package under the moon to just make the wide data files. This PR fixes that.

Also fixes a bug in the makefile

HedvigS commented 2 years ago

Okay. Thank you Simon I understand.

Do you mind if i revert some smaller things here and there and change some library() to p_load() or h_load() to keep it consistent?

HedvigS commented 2 years ago

Thank you for setting up the Makefile correctly. I don't tend to use Makefiles that much, I'm more likely to use shell scripts and check with if() if dirs exists inside R-scripts instead of create them in the Makefile. I'll just check that things run in harmony now and that we don't get warnings for creating dirs that already exists etc.

HedvigS commented 2 years ago

I've changed so that we're now using h_load() instead of library, meaning it'll install packages if they're not currently installed. This makes life on new computers and on clusters easier in my experience.

Is that okay with you @SimonGreenhill ?

HedvigS commented 2 years ago

If you want to simplify every script in the make data rule in the makefile, it'd also involve the various coverage plots scripts. i'll adjust those for you.

HedvigS commented 2 years ago

Can I merge in @SimonGreenhill ?

SimonGreenhill commented 2 years ago

done

HedvigS commented 2 years ago

Thanks!