max/nix-super
1
0
Fork 0
mirror of https://github.com/privatevoid-net/nix-super.git synced 2024-11-24 06:56:14 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions
nix-super/tests/functional/check-refs.sh

54 lines
1.5 KiB
Bash
Raw Permalink Normal View History

* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
source common.sh
Restore old (dis)allowedRequisites behaviour for self-references stdenv relies on this. So ignore self-references (but only in legacy non-structured attributes mode).
2018-10-27 16:40:09 +03:00
clearStore
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
RESULT=$TEST_ROOT/result
* Refactoring: remove unnecessary variables from the tests.
2011-10-11 00:32:34 +03:00
dep=$(nix-build -o $RESULT check-refs.nix -A dep)
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
# test1 references dep, not itself.
* Refactoring: remove unnecessary variables from the tests.
2011-10-11 00:32:34 +03:00
test1=$(nix-build -o $RESULT check-refs.nix -A test1)
Harden tests' bash Use `set -u` and `set -o pipefail` to catch accidental mistakes and failures more strongly. - `set -u` catches the use of undefined variables - `set -o pipefail` catches failures (like `set -e`) earlier in the pipeline. This makes the tests a bit more robust. It is nice to read code not worrying about these spurious success paths (via uncaught) errors undermining the tests. Indeed, I caught some bugs doing this. There are a few tests where we run a command that should fail, and then search its output to make sure the failure message is one that we expect. Before, since the `grep` was the last command in the pipeline the exit code of those failing programs was silently ignored. Now with `set -o pipefail` it won't be, and we have to do something so the expected failure doesn't accidentally fail the test. To do that we use `expect` and a new `expectStderr` to check for the exact failing exit code. See the comments on each for why. `grep -q` is replaced with `grepQuiet`, see the comments on that function for why. `grep -v` when we just want the exit code is replaced with `grepInverse, see the comments on that function for why. `grep -q -v` together is, surprise surprise, replaced with `grepQuietInverse`, which is both combined. Co-authored-by: Robert Hensing <roberth@users.noreply.github.com>
2021-12-09 17:26:46 +02:00
nix-store -q --references $test1 | grepQuietInverse $test1
nix-store -q --references $test1 | grepQuiet $dep
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
# test2 references src, not itself nor dep.
* Refactoring: remove unnecessary variables from the tests.
2011-10-11 00:32:34 +03:00
test2=$(nix-build -o $RESULT check-refs.nix -A test2)
Harden tests' bash Use `set -u` and `set -o pipefail` to catch accidental mistakes and failures more strongly. - `set -u` catches the use of undefined variables - `set -o pipefail` catches failures (like `set -e`) earlier in the pipeline. This makes the tests a bit more robust. It is nice to read code not worrying about these spurious success paths (via uncaught) errors undermining the tests. Indeed, I caught some bugs doing this. There are a few tests where we run a command that should fail, and then search its output to make sure the failure message is one that we expect. Before, since the `grep` was the last command in the pipeline the exit code of those failing programs was silently ignored. Now with `set -o pipefail` it won't be, and we have to do something so the expected failure doesn't accidentally fail the test. To do that we use `expect` and a new `expectStderr` to check for the exact failing exit code. See the comments on each for why. `grep -q` is replaced with `grepQuiet`, see the comments on that function for why. `grep -v` when we just want the exit code is replaced with `grepInverse, see the comments on that function for why. `grep -q -v` together is, surprise surprise, replaced with `grepQuietInverse`, which is both combined. Co-authored-by: Robert Hensing <roberth@users.noreply.github.com>
2021-12-09 17:26:46 +02:00
nix-store -q --references $test2 | grepQuietInverse $test2
nix-store -q --references $test2 | grepQuietInverse $dep
nix-store -q --references $test2 | grepQuiet aux-ref
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
# test3 should fail (unallowed ref).
Fix tests So all these years I was totally deluded about the meaning of "set -e". You might think that it causes statements like "false && true" or "! true" to fail, but it doesn't...
2014-08-21 22:50:19 +03:00
(! nix-build -o $RESULT check-refs.nix -A test3)
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
# test4 should succeed.
* Refactoring: remove unnecessary variables from the tests.
2011-10-11 00:32:34 +03:00
nix-build -o $RESULT check-refs.nix -A test4
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
# test5 should succeed.
* Refactoring: remove unnecessary variables from the tests.
2011-10-11 00:32:34 +03:00
nix-build -o $RESULT check-refs.nix -A test5
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
# test6 should fail (unallowed self-ref).
Fix tests So all these years I was totally deluded about the meaning of "set -e". You might think that it causes statements like "false && true" or "! true" to fail, but it doesn't...
2014-08-21 22:50:19 +03:00
(! nix-build -o $RESULT check-refs.nix -A test6)
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
# test7 should succeed (allowed self-ref).
* Refactoring: remove unnecessary variables from the tests.
2011-10-11 00:32:34 +03:00
nix-build -o $RESULT check-refs.nix -A test7
* Checks for allowedReferences and some other features. * Use nix-build in a test.
2006-10-19 20:44:51 +03:00
# test8 should fail (toFile depending on derivation output).
Fix tests So all these years I was totally deluded about the meaning of "set -e". You might think that it causes statements like "false && true" or "! true" to fail, but it doesn't...
2014-08-21 22:50:19 +03:00
(! nix-build -o $RESULT check-refs.nix -A test8)
Add disallowedReferences / disallowedRequisites For the "stdenv accidentally referring to bootstrap-tools", it seems easier to specify the path that we don't want to depend on, e.g. disallowedRequisites = [ bootstrapTools ];
2014-08-28 19:57:13 +03:00
# test9 should fail (disallowed reference).
(! nix-build -o $RESULT check-refs.nix -A test9)
# test10 should succeed (no disallowed references).
nix-build -o $RESULT check-refs.nix -A test10
`unsafeDiscardReferences` Adds a new boolean structured attribute `outputChecks.<output>.unsafeDiscardReferences` which disables scanning an output for runtime references. __structuredAttrs = true; outputChecks.out.unsafeDiscardReferences = true; This is useful when creating filesystem images containing their own embedded Nix store: they are self-contained blobs of data with no runtime dependencies. Setting this attribute requires the experimental feature `discard-references` to be enabled.
2022-09-24 00:30:29 +03:00
if isDaemonNewer 2.12pre20230103; then
Stabilize `discard-references` It has been there for a few releases now (landed in 2.14.0), doesn't seem to cause any major issue and is wanted in a few places (https://github.com/NixOS/nix/pull/7087#issuecomment-1544471346).
2023-05-12 10:56:39 +03:00
if ! isDaemonNewer 2.16.0; then
enableFeatures discard-references
restartDaemon
fi
`unsafeDiscardReferences` Adds a new boolean structured attribute `outputChecks.<output>.unsafeDiscardReferences` which disables scanning an output for runtime references. __structuredAttrs = true; outputChecks.out.unsafeDiscardReferences = true; This is useful when creating filesystem images containing their own embedded Nix store: they are self-contained blobs of data with no runtime dependencies. Setting this attribute requires the experimental feature `discard-references` to be enabled.
2022-09-24 00:30:29 +03:00
# test11 should succeed.
test11=$(nix-build -o $RESULT check-refs.nix -A test11)
[[ -z $(nix-store -q --references "$test11") ]]
fi
Reference in a new issue Copy permalink