2016-03-01 1:06 GMT+01:00 Michael Raskin <7c6f4...@mail.ru>:
>
> Last time I breaked a thing (and it was recently), it was my commit and
> I actually used the resulting binary for some time. Didn't notice the
> open dialog was broken.  So the question is about price-vs-quality,
> testing doesn't always give as much additional information as one would
> want…
>

Sure, not every error stands out in a cursory check. Just the other day, I
pushed back on a commit in master, that made the gradle command show a java
help screen (because of it not actually running gradle). An error that
would have been easily overlooked by a reviewer just expecting the program
to run and spit out some usage information.
I actually built a unit test to verify gradle being able to build a test
project. Don't know if that's made it into nixos yet.
Maybe part of a solution is to educate contributors on how to write
automated package tests?

I rank reasonableness of notions used in UI as
> monotone > mercurial > SVN > git
>
> I am currently undecided about fossil, although it is definitely in the
> left half of the chain.
>

I guess, I'll have to take a closer look at fossil and monotone. What do
you think of darcs?
I used mercurial for a little while between SVN and git, back when it was
written in python, crashy and slow (probably still is, haha), I think I
actually managed to destroy data with it.
SVN is just: you know, why would I need a server to do version control?
git to me is just like nixos: I might not appreciate all the curly braces
and semicolons in particular, but the whole thing being arranged around an
immutable core + being super fast + having a large, active community is
just too good to be ignored.

git commands are built around gotchas, so I will not even try to get
> a better frontend, I will just script whatever workflow is considered
> acceptable and use the same two scripts to minimize errors.
>

Do try magit though. People that like to show of their git frontends, get
jealous, when I show it to them. And rightfully so.
Disclaimer: I do find git's concepts quite acceptable and the gotchas to be
manageable, I do think that there would be potential in a tool doing
management of patchsets, like apparently darcs or quilt do.
_______________________________________________
nix-dev mailing list
nix-dev@lists.science.uu.nl
http://lists.science.uu.nl/mailman/listinfo/nix-dev

Reply via email to