lix-project/lix-releng-staging
0
0
Fork 0
Code Packages Activity
lix-releng-staging/tests/fetchGit.sh

160 lines
5.5 KiB
Bash
Raw Normal View History

fetchGit: Add a test
2017-11-03 12:33:50 +00:00
source common.sh
if [[ -z $(type -p git) ]]; then
echo "Git not installed; skipping Git tests"
Show when tests are skipped Also, don't depend on tput (ncurses). It's really not needed since ANSI escape sequences have been standardized for 35 years or so.
2017-11-07 11:06:58 +00:00
exit 99
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
fi
clearStore
repo=$TEST_ROOT/git
Misc changes from the flakes branch
2020-03-24 13:26:13 +00:00
export _NIX_FORCE_HTTP=1
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
rm -rf $repo ${repo}-tmp $TEST_HOME/.cache/nix $TEST_ROOT/worktree $TEST_ROOT/shallow
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
git init $repo
git -C $repo config user.email "foobar@example.com"
git -C $repo config user.name "Foobar"
echo utrecht > $repo/hello
fetchGit/fetchMercurial: Fix directory inclusion check E.g. the existence of .gitignore would cause .git to be included.
2017-11-21 18:12:47 +00:00
touch $repo/.gitignore
git -C $repo add hello .gitignore
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
git -C $repo commit -m 'Bla1'
rev1=$(git -C $repo rev-parse HEAD)
echo world > $repo/hello
git -C $repo commit -m 'Bla2' -a
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
git -C $repo worktree add $TEST_ROOT/worktree
echo hello >> $TEST_ROOT/worktree/hello
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
rev2=$(git -C $repo rev-parse HEAD)
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
# Fetch a worktree
unset _NIX_FORCE_HTTP
path0=$(nix eval --raw "(builtins.fetchGit file://$TEST_ROOT/worktree).outPath")
export _NIX_FORCE_HTTP=1
[[ $(tail -n 1 $path0/hello) = "hello" ]]
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
# Fetch the default branch.
path=$(nix eval --raw "(builtins.fetchGit file://$repo).outPath")
[[ $(cat $path/hello) = world ]]
Add pure evaluation mode In this mode, the following restrictions apply: * The builtins currentTime, currentSystem and storePath throw an error. * $NIX_PATH and -I are ignored. * fetchGit and fetchMercurial require a revision hash. * fetchurl and fetchTarball require a sha256 attribute. * No file system access is allowed outside of the paths returned by fetch{Git,Mercurial,url,Tarball}. Thus 'nix build -f ./foo.nix' is not allowed. Thus, the evaluation result is completely reproducible from the command line arguments. E.g. nix build --pure-eval '( let nix = fetchGit { url = https://github.com/NixOS/nixpkgs.git; rev = "9c927de4b179a6dd210dd88d34bda8af4b575680"; }; nixpkgs = fetchGit { url = https://github.com/NixOS/nixpkgs.git; ref = "release-17.09"; rev = "66b4de79e3841530e6d9c6baf98702aa1f7124e4"; }; in (import (nix + "/release.nix") { inherit nix nixpkgs; }).build.x86_64-linux )' The goal is to enable completely reproducible and traceable evaluation. For example, a NixOS configuration could be fully described by a single Git commit hash. 'nixos-rebuild' would do something like nix build --pure-eval '( (import (fetchGit { url = file:///my-nixos-config; rev = "..."; })).system ') where the Git repository /my-nixos-config would use further fetchGit calls or Git externals to fetch Nixpkgs and whatever other dependencies it has. Either way, the commit hash would uniquely identify the NixOS configuration and allow it to reproduced.
2018-01-16 17:50:38 +00:00
# In pure eval mode, fetchGit without a revision should fail.
[[ $(nix eval --raw "(builtins.readFile (fetchGit file://$repo + \"/hello\"))") = world ]]
(! nix eval --pure-eval --raw "(builtins.readFile (fetchGit file://$repo + \"/hello\"))")
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
# Fetch using an explicit revision hash.
path2=$(nix eval --raw "(builtins.fetchGit { url = file://$repo; rev = \"$rev2\"; }).outPath")
[[ $path = $path2 ]]
Add pure evaluation mode In this mode, the following restrictions apply: * The builtins currentTime, currentSystem and storePath throw an error. * $NIX_PATH and -I are ignored. * fetchGit and fetchMercurial require a revision hash. * fetchurl and fetchTarball require a sha256 attribute. * No file system access is allowed outside of the paths returned by fetch{Git,Mercurial,url,Tarball}. Thus 'nix build -f ./foo.nix' is not allowed. Thus, the evaluation result is completely reproducible from the command line arguments. E.g. nix build --pure-eval '( let nix = fetchGit { url = https://github.com/NixOS/nixpkgs.git; rev = "9c927de4b179a6dd210dd88d34bda8af4b575680"; }; nixpkgs = fetchGit { url = https://github.com/NixOS/nixpkgs.git; ref = "release-17.09"; rev = "66b4de79e3841530e6d9c6baf98702aa1f7124e4"; }; in (import (nix + "/release.nix") { inherit nix nixpkgs; }).build.x86_64-linux )' The goal is to enable completely reproducible and traceable evaluation. For example, a NixOS configuration could be fully described by a single Git commit hash. 'nixos-rebuild' would do something like nix build --pure-eval '( (import (fetchGit { url = file:///my-nixos-config; rev = "..."; })).system ') where the Git repository /my-nixos-config would use further fetchGit calls or Git externals to fetch Nixpkgs and whatever other dependencies it has. Either way, the commit hash would uniquely identify the NixOS configuration and allow it to reproduced.
2018-01-16 17:50:38 +00:00
# In pure eval mode, fetchGit with a revision should succeed.
[[ $(nix eval --pure-eval --raw "(builtins.readFile (fetchGit { url = file://$repo; rev = \"$rev2\"; } + \"/hello\"))") = world ]]
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
# Fetch again. This should be cached.
mv $repo ${repo}-tmp
path2=$(nix eval --raw "(builtins.fetchGit file://$repo).outPath")
[[ $path = $path2 ]]
[[ $(nix eval "(builtins.fetchGit file://$repo).revCount") = 2 ]]
[[ $(nix eval --raw "(builtins.fetchGit file://$repo).rev") = $rev2 ]]
# Fetching with a explicit hash should succeed.
path2=$(nix eval --tarball-ttl 0 --raw "(builtins.fetchGit { url = file://$repo; rev = \"$rev2\"; }).outPath")
[[ $path = $path2 ]]
path2=$(nix eval --tarball-ttl 0 --raw "(builtins.fetchGit { url = file://$repo; rev = \"$rev1\"; }).outPath")
[[ $(cat $path2/hello) = utrecht ]]
mv ${repo}-tmp $repo
# Using a clean working tree should produce the same result.
path2=$(nix eval --raw "(builtins.fetchGit $repo).outPath")
[[ $path = $path2 ]]
# Using an unclean tree should yield the tracked but uncommitted changes.
fetchGit/fetchMercurial: Filter out directories with untracked files
2017-11-03 12:48:02 +00:00
mkdir $repo/dir1 $repo/dir2
echo foo > $repo/dir1/foo
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
echo bar > $repo/bar
fetchGit/fetchMercurial: Filter out directories with untracked files
2017-11-03 12:48:02 +00:00
echo bar > $repo/dir2/bar
git -C $repo add dir1/foo
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
git -C $repo rm hello
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
unset _NIX_FORCE_HTTP
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
path2=$(nix eval --raw "(builtins.fetchGit $repo).outPath")
[ ! -e $path2/hello ]
[ ! -e $path2/bar ]
fetchGit/fetchMercurial: Filter out directories with untracked files
2017-11-03 12:48:02 +00:00
[ ! -e $path2/dir2/bar ]
fetchGit/fetchMercurial: Fix directory inclusion check E.g. the existence of .gitignore would cause .git to be included.
2017-11-21 18:12:47 +00:00
[ ! -e $path2/.git ]
fetchGit/fetchMercurial: Filter out directories with untracked files
2017-11-03 12:48:02 +00:00
[[ $(cat $path2/dir1/foo) = foo ]]
fetchGit: Add a test
2017-11-03 12:33:50 +00:00
[[ $(nix eval --raw "(builtins.fetchGit $repo).rev") = 0000000000000000000000000000000000000000 ]]
# ... unless we're using an explicit ref or rev.
path3=$(nix eval --raw "(builtins.fetchGit { url = $repo; ref = \"master\"; }).outPath")
[[ $path = $path3 ]]
path3=$(nix eval --raw "(builtins.fetchGit { url = $repo; rev = \"$rev2\"; }).outPath")
[[ $path = $path3 ]]
# Committing should not affect the store path.
git -C $repo commit -m 'Bla3' -a
path4=$(nix eval --tarball-ttl 0 --raw "(builtins.fetchGit file://$repo).outPath")
[[ $path2 = $path4 ]]
fetchGit: Ignore tarballTtl if rev is set and not in the repo. Fixes #1697.
2017-11-24 02:50:01 +00:00
# tarball-ttl should be ignored if we specify a rev
echo delft > $repo/hello
git -C $repo add hello
git -C $repo commit -m 'Bla4'
rev3=$(git -C $repo rev-parse HEAD)
nix eval --tarball-ttl 3600 "(builtins.fetchGit { url = $repo; rev = \"$rev3\"; })" >/dev/null
fetchGit: Fix handling of local repo when not using 'master' branch Add tests checking this behavior.
2017-12-22 20:18:29 +00:00
# Update 'path' to reflect latest master
path=$(nix eval --raw "(builtins.fetchGit file://$repo).outPath")
# Check behavior when non-master branch is used
git -C $repo checkout $rev2 -b dev
echo dev > $repo/hello
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
# File URI uses dirty tree unless specified otherwise
fetchGit: Fix handling of local repo when not using 'master' branch Add tests checking this behavior.
2017-12-22 20:18:29 +00:00
path2=$(nix eval --raw "(builtins.fetchGit file://$repo).outPath")
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
[ $(cat $path2/hello) = dev ]
fetchGit: Fix handling of local repo when not using 'master' branch Add tests checking this behavior.
2017-12-22 20:18:29 +00:00
# Using local path with branch other than 'master' should work when clean or dirty
path3=$(nix eval --raw "(builtins.fetchGit $repo).outPath")
# (check dirty-tree handling was used)
[[ $(nix eval --raw "(builtins.fetchGit $repo).rev") = 0000000000000000000000000000000000000000 ]]
# Committing shouldn't change store path, or switch to using 'master'
git -C $repo commit -m 'Bla5' -a
path4=$(nix eval --raw "(builtins.fetchGit $repo).outPath")
[[ $(cat $path4/hello) = dev ]]
[[ $path3 = $path4 ]]
# Confirm same as 'dev' branch
path5=$(nix eval --raw "(builtins.fetchGit { url = $repo; ref = \"dev\"; }).outPath")
[[ $path3 = $path5 ]]
fetchGit.sh: Test we don't "corrupt" cache if invoke w/o git avail
2018-01-09 14:58:19 +00:00
# Nuke the cache
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
rm -rf $TEST_HOME/.cache/nix
fetchGit.sh: Test we don't "corrupt" cache if invoke w/o git avail
2018-01-09 14:58:19 +00:00
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
# Try again, but without 'git' on PATH. This should fail.
fetchGit.sh: Test we don't "corrupt" cache if invoke w/o git avail
2018-01-09 14:58:19 +00:00
NIX=$(command -v nix)
# This should fail
(! PATH= $NIX eval --raw "(builtins.fetchGit { url = $repo; ref = \"dev\"; }).outPath" )
# Try again, with 'git' available. This should work.
path5=$(nix eval --raw "(builtins.fetchGit { url = $repo; ref = \"dev\"; }).outPath")
[[ $path3 = $path5 ]]
Backport libfetchers from the flakes branch This provides a pluggable mechanism for defining new fetchers. It adds a builtin function 'fetchTree' that generalizes existing fetchers like 'fetchGit', 'fetchMercurial' and 'fetchTarball'. 'fetchTree' takes a set of attributes, e.g. fetchTree { type = "git"; url = "https://example.org/repo.git"; ref = "some-branch"; rev = "abcdef..."; } The existing fetchers are just wrappers around this. Note that the input attributes to fetchTree are the same as flake input specifications and flake lock file entries. All fetchers share a common cache stored in ~/.cache/nix/fetcher-cache-v1.sqlite. This replaces the ad hoc caching mechanisms in fetchGit and download.cc (e.g. ~/.cache/nix/{tarballs,git-revs*}). This also adds support for Git worktrees (c169ea59049f861aaba429f48b828d0820b74d1d).
2020-03-30 14:04:18 +00:00
# Fetching a shallow repo shouldn't work by default, because we can't
# return a revCount.
git clone --depth 1 file://$repo $TEST_ROOT/shallow
(! nix eval --raw "(builtins.fetchGit { url = $TEST_ROOT/shallow; ref = \"dev\"; }).outPath")
# But you can request a shallow clone, which won't return a revCount.
path6=$(nix eval --raw "(builtins.fetchTree { type = \"git\"; url = \"file://$TEST_ROOT/shallow\"; ref = \"dev\"; shallow = true; }).outPath")
[[ $path3 = $path6 ]]
[[ $(nix eval "(builtins.fetchTree { type = \"git\"; url = \"file://$TEST_ROOT/shallow\"; ref = \"dev\"; shallow = true; }).revCount or 123") == 123 ]]
Copy permalink