Put functional tests in `tests/functional`

I think it is bad for these reasons when `tests/` contains a mix of
functional and integration tests

 - Concepts is harder to understand, the documentation makes a good
   unit vs functional vs integration distinction, but when the
   integration tests are just two subdirs within `tests/` this is not
   clear.

 - Source filtering in the `flake.nix` is more complex. We need to
   filter out some of the dirs from `tests/`, rather than simply pick
   the dirs we want and take all of them. This is a good sign the
   structure of what we are trying to do is not matching the structure
   of the files.

With this change we have a clean:
```shell-session
$ git show 'HEAD:tests'
tree HEAD:tests

functional/
installer/
nixos/
```

(cherry picked from commit 68c81c737571794f7246db53fb4774e94fcf4b7e)

1 files changed, 256 insertions, 0 deletions

diff --git a/tests/functional/fetchGit.sh b/tests/functional/fetchGit.sh
new file mode 100644
index 000000000..418b4f63f
--- /dev/null
+++ b/tests/functional/fetchGit.sh
@@ -0,0 +1,256 @@
+source common.sh
+
+requireGit
+
+clearStore
+
+# Intentionally not in a canonical form
+# See https://github.com/NixOS/nix/issues/6195
+repo=$TEST_ROOT/./git
+
+export _NIX_FORCE_HTTP=1
+
+rm -rf $repo ${repo}-tmp $TEST_HOME/.cache/nix $TEST_ROOT/worktree $TEST_ROOT/shallow $TEST_ROOT/minimal
+
+git init $repo
+git -C $repo config user.email "foobar@example.com"
+git -C $repo config user.name "Foobar"
+
+echo utrecht > $repo/hello
+touch $repo/.gitignore
+git -C $repo add hello .gitignore
+git -C $repo commit -m 'Bla1'
+rev1=$(git -C $repo rev-parse HEAD)
+git -C $repo tag -a tag1 -m tag1
+
+echo world > $repo/hello
+git -C $repo commit -m 'Bla2' -a
+git -C $repo worktree add $TEST_ROOT/worktree
+echo hello >> $TEST_ROOT/worktree/hello
+rev2=$(git -C $repo rev-parse HEAD)
+git -C $repo tag -a tag2 -m tag2
+
+# Fetch a worktree
+unset _NIX_FORCE_HTTP
+path0=$(nix eval --impure --raw --expr "(builtins.fetchGit file://$TEST_ROOT/worktree).outPath")
+path0_=$(nix eval --impure --raw --expr "(builtins.fetchTree { type = \"git\"; url = file://$TEST_ROOT/worktree; }).outPath")
+[[ $path0 = $path0_ ]]
+export _NIX_FORCE_HTTP=1
+[[ $(tail -n 1 $path0/hello) = "hello" ]]
+
+# Fetch the default branch.
+path=$(nix eval --impure --raw --expr "(builtins.fetchGit file://$repo).outPath")
+[[ $(cat $path/hello) = world ]]
+
+# Fetch a rev from another branch
+git -C $repo checkout -b devtest
+echo "different file" >> $TEST_ROOT/git/differentbranch
+git -C $repo add differentbranch
+git -C $repo commit -m 'Test2'
+git -C $repo checkout master
+devrev=$(git -C $repo rev-parse devtest)
+out=$(nix eval --impure --raw --expr "builtins.fetchGit { url = file://$repo; rev = \"$devrev\"; }" 2>&1) || status=$?
+[[ $status == 1 ]]
+[[ $out =~ 'Cannot find Git revision' ]]
+
+[[ $(nix eval --raw --expr "builtins.readFile (builtins.fetchGit { url = file://$repo; rev = \"$devrev\"; allRefs = true; } + \"/differentbranch\")") = 'different file' ]]
+
+# In pure eval mode, fetchGit without a revision should fail.
+[[ $(nix eval --impure --raw --expr "builtins.readFile (fetchGit file://$repo + \"/hello\")") = world ]]
+(! nix eval --raw --expr "builtins.readFile (fetchGit file://$repo + \"/hello\")")
+
+# Fetch using an explicit revision hash.
+path2=$(nix eval --raw --expr "(builtins.fetchGit { url = file://$repo; rev = \"$rev2\"; }).outPath")
+[[ $path = $path2 ]]
+
+# In pure eval mode, fetchGit with a revision should succeed.
+[[ $(nix eval --raw --expr "builtins.readFile (fetchGit { url = file://$repo; rev = \"$rev2\"; } + \"/hello\")") = world ]]
+
+# Fetch again. This should be cached.
+mv $repo ${repo}-tmp
+path2=$(nix eval --impure --raw --expr "(builtins.fetchGit file://$repo).outPath")
+[[ $path = $path2 ]]
+
+[[ $(nix eval --impure --expr "(builtins.fetchGit file://$repo).revCount") = 2 ]]
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit file://$repo).rev") = $rev2 ]]
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit file://$repo).shortRev") = ${rev2:0:7} ]]
+
+# Fetching with a explicit hash should succeed.
+path2=$(nix eval --refresh --raw --expr "(builtins.fetchGit { url = file://$repo; rev = \"$rev2\"; }).outPath")
+[[ $path = $path2 ]]
+
+path2=$(nix eval --refresh --raw --expr "(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 --impure --raw --expr "(builtins.fetchGit $repo).outPath")
+[[ $path = $path2 ]]
+
+# Using an unclean tree should yield the tracked but uncommitted changes.
+mkdir $repo/dir1 $repo/dir2
+echo foo > $repo/dir1/foo
+echo bar > $repo/bar
+echo bar > $repo/dir2/bar
+git -C $repo add dir1/foo
+git -C $repo rm hello
+
+unset _NIX_FORCE_HTTP
+path2=$(nix eval --impure --raw --expr "(builtins.fetchGit $repo).outPath")
+[ ! -e $path2/hello ]
+[ ! -e $path2/bar ]
+[ ! -e $path2/dir2/bar ]
+[ ! -e $path2/.git ]
+[[ $(cat $path2/dir1/foo) = foo ]]
+
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit $repo).rev") = 0000000000000000000000000000000000000000 ]]
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit $repo).dirtyRev") = "${rev2}-dirty" ]]
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit $repo).dirtyShortRev") = "${rev2:0:7}-dirty" ]]
+
+# ... unless we're using an explicit ref or rev.
+path3=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = $repo; ref = \"master\"; }).outPath")
+[[ $path = $path3 ]]
+
+path3=$(nix eval --raw --expr "(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 --impure --refresh --raw --expr "(builtins.fetchGit file://$repo).outPath")
+[[ $path2 = $path4 ]]
+
+[[ $(nix eval --impure --expr "builtins.hasAttr \"rev\" (builtins.fetchGit $repo)") == "true" ]]
+[[ $(nix eval --impure --expr "builtins.hasAttr \"dirtyRev\" (builtins.fetchGit $repo)") == "false" ]]
+[[ $(nix eval --impure --expr "builtins.hasAttr \"dirtyShortRev\" (builtins.fetchGit $repo)") == "false" ]]
+
+status=0
+nix eval --impure --raw --expr "(builtins.fetchGit { url = $repo; rev = \"$rev2\"; narHash = \"sha256-B5yIPHhEm0eysJKEsO7nqxprh9vcblFxpJG11gXJus1=\"; }).outPath" || status=$?
+[[ "$status" = "102" ]]
+
+path5=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = $repo; rev = \"$rev2\"; narHash = \"sha256-Hr8g6AqANb3xqX28eu1XnjK/3ab8Gv6TJSnkb1LezG9=\"; }).outPath")
+[[ $path = $path5 ]]
+
+# 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 --expr "builtins.fetchGit { url = $repo; rev = \"$rev3\"; }" >/dev/null
+
+# Update 'path' to reflect latest master
+path=$(nix eval --impure --raw --expr "(builtins.fetchGit file://$repo).outPath")
+
+# Check behavior when non-master branch is used
+git -C $repo checkout $rev2 -b dev
+echo dev > $repo/hello
+
+# File URI uses dirty tree unless specified otherwise
+path2=$(nix eval --impure --raw --expr "(builtins.fetchGit file://$repo).outPath")
+[ $(cat $path2/hello) = dev ]
+
+# Using local path with branch other than 'master' should work when clean or dirty
+path3=$(nix eval --impure --raw --expr "(builtins.fetchGit $repo).outPath")
+# (check dirty-tree handling was used)
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit $repo).rev") = 0000000000000000000000000000000000000000 ]]
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit $repo).shortRev") = 0000000 ]]
+# Making a dirty tree clean again and fetching it should
+# record correct revision information. See: #4140
+echo world > $repo/hello
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit $repo).rev") = $rev2 ]]
+
+# Committing shouldn't change store path, or switch to using 'master'
+echo dev > $repo/hello
+git -C $repo commit -m 'Bla5' -a
+path4=$(nix eval --impure --raw --expr "(builtins.fetchGit $repo).outPath")
+[[ $(cat $path4/hello) = dev ]]
+[[ $path3 = $path4 ]]
+
+# Using remote path with branch other than 'master' should fetch the HEAD revision.
+# (--tarball-ttl 0 to prevent using the cached repo above)
+export _NIX_FORCE_HTTP=1
+path4=$(nix eval --tarball-ttl 0 --impure --raw --expr "(builtins.fetchGit $repo).outPath")
+[[ $(cat $path4/hello) = dev ]]
+[[ $path3 = $path4 ]]
+unset _NIX_FORCE_HTTP
+
+# Confirm same as 'dev' branch
+path5=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = $repo; ref = \"dev\"; }).outPath")
+[[ $path3 = $path5 ]]
+
+
+# Nuke the cache
+rm -rf $TEST_HOME/.cache/nix
+
+# Try again, but without 'git' on PATH. This should fail.
+NIX=$(command -v nix)
+(! PATH= $NIX eval --impure --raw --expr "(builtins.fetchGit { url = $repo; ref = \"dev\"; }).outPath" )
+
+# Try again, with 'git' available.  This should work.
+path5=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = $repo; ref = \"dev\"; }).outPath")
+[[ $path3 = $path5 ]]
+
+# Fetching from a repo with only a specific revision and no branches should
+# not fall back to copying files and record correct revision information. See: #5302
+mkdir $TEST_ROOT/minimal
+git -C $TEST_ROOT/minimal init
+git -C $TEST_ROOT/minimal fetch $repo $rev2
+git -C $TEST_ROOT/minimal checkout $rev2
+[[ $(nix eval --impure --raw --expr "(builtins.fetchGit { url = $TEST_ROOT/minimal; }).rev") = $rev2 ]]
+
+# Fetching a shallow repo shouldn't work by default, because we can't
+# return a revCount.
+git clone --depth 1 file://$repo $TEST_ROOT/shallow
+(! nix eval --impure --raw --expr "(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 --impure --raw --expr "(builtins.fetchTree { type = \"git\"; url = \"file://$TEST_ROOT/shallow\"; ref = \"dev\"; shallow = true; }).outPath")
+[[ $path3 = $path6 ]]
+[[ $(nix eval --impure --expr "(builtins.fetchTree { type = \"git\"; url = \"file://$TEST_ROOT/shallow\"; ref = \"dev\"; shallow = true; }).revCount or 123") == 123 ]]
+
+# Explicit ref = "HEAD" should work, and produce the same outPath as without ref
+path7=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = \"file://$repo\"; ref = \"HEAD\"; }).outPath")
+path8=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = \"file://$repo\"; }).outPath")
+[[ $path7 = $path8 ]]
+
+# ref = "HEAD" should fetch the HEAD revision
+rev4=$(git -C $repo rev-parse HEAD)
+rev4_nix=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = \"file://$repo\"; ref = \"HEAD\"; }).rev")
+[[ $rev4 = $rev4_nix ]]
+
+# The name argument should be handled
+path9=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = \"file://$repo\"; ref = \"HEAD\"; name = \"foo\"; }).outPath")
+[[ $path9 =~ -foo$ ]]
+
+# Specifying a ref without a rev shouldn't pick a cached rev for a different ref
+export _NIX_FORCE_HTTP=1
+rev_tag1_nix=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = \"file://$repo\"; ref = \"refs/tags/tag1\"; }).rev")
+rev_tag1=$(git -C $repo rev-parse refs/tags/tag1)
+[[ $rev_tag1_nix = $rev_tag1 ]]
+rev_tag2_nix=$(nix eval --impure --raw --expr "(builtins.fetchGit { url = \"file://$repo\"; ref = \"refs/tags/tag2\"; }).rev")
+rev_tag2=$(git -C $repo rev-parse refs/tags/tag2)
+[[ $rev_tag2_nix = $rev_tag2 ]]
+unset _NIX_FORCE_HTTP
+
+# should fail if there is no repo
+rm -rf $repo/.git
+(! nix eval --impure --raw --expr "(builtins.fetchGit \"file://$repo\").outPath")
+
+# should succeed for a repo without commits
+git init $repo
+path10=$(nix eval --impure --raw --expr "(builtins.fetchGit \"file://$repo\").outPath")
+
+# should succeed for a path with a space
+# regression test for #7707
+repo="$TEST_ROOT/a b"
+git init "$repo"
+git -C "$repo" config user.email "foobar@example.com"
+git -C "$repo" config user.name "Foobar"
+
+echo utrecht > "$repo/hello"
+touch "$repo/.gitignore"
+git -C "$repo" add hello .gitignore
+git -C "$repo" commit -m 'Bla1'
+cd "$repo"
+path11=$(nix eval --impure --raw --expr "(builtins.fetchGit ./.).outPath")