| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Whether a derived copy of a flake or flake-input source needs to be
made again depends on the sources it draws inputs from. For one that
isn't in the store, the key had its hash, so every run read all of it,
e.g. all of a nixpkgs work tree, .git included, which takes long even
when nothing changed. For a Git work tree that Nix gets as git+file:,
whose commit and tracked files are all Nix uses, the key has its
commit instead, if no tracked file differs from it, which Git tells
quickly. git_flake decides what Nix gets as git+file:, for resolve as
well.
Keys of copies drawing from such work trees change once, so those
copies get made again on the first run, also after deploying with
krops 1.
rewrite-cache-git checks that such a copy gets made again for a new
commit, but not for redeploying, nor for an untracked file.
|
|
|
pkgs/populate/README.md documents populate on its own: building it,
what it needs, its command line, the sources as JSON and in Nix with
its Nix library, the target directory, the source types, flake inputs
from sources, what it prints, and its tests. The sections about the
target directory, the source types, and flake inputs from sources move
there from krops' README, which links to them from sections of the same
names, and from the first mention of populate. docs/design.md keeps
how populate works.
krops' README describes krops' own tests, and CHANGELOG what populate
being a program of its own means for krops.
|