>git rebase

>>60239695
Rebase is easy once you understand the concept.

Starting from commit X through commit Y, rebuild the timeline according to the provided options.

The options can be reordering or inserting commits, or changing them outright.

It's time surgery. Being a time lord isn't for everyone.

>>60239721

how about I just use pijul and then I dont have to worry about a million complex weird commands

It's just a way to see commits for what they are: patches. And rebase just mean apply patches on another source tree.

File: tosmart35.png (19KB, 500x590px) Image search: [Google]

19KB, 500x590px

>>60239695
>upload files separately to github via browser

File: b4a.jpg (1MB, 3840x2160px) Image search: [Google]

1MB, 3840x2160px

>git commit -i

>>60239695
>git stash -u --keep-index

File: 1490047134668.jpg (21KB, 170x170px) Image search: [Google]

21KB, 170x170px

>>60239695
>understanding basic graph theory makes you smart
You're not that special, honeybunch

File: muh_branches.jpg (15KB, 236x324px) Image search: [Google]

15KB, 236x324px

>>60239721
Why would you do that? Just use merge for fucks sake.

Are you afraid of the linus police coming to inspect the linearity of your branches or what?

Let branches be actual branches instead of linear paths. The whole point of using Git is that it is so easy to merge branches that you don't care about the branching factor.

Reposting from /sqt/:

Question about using git repos on GoboLinux:

Cloning remote repos is easy, of course, but once I have it on my GoboLinux machine, how should I go about compiling the cloned program?

Can I just use Gobo's Compile command? Do I need to make a recipe first? Or just do the usual configure/make/make install routine and hope no funky errors occur?

>>60245866
b-but the linear history looks so much nicer than branching

>manipulating time instead of using git revert

>>60244166
>calling it graph theory