update and fix build with latest Lix #14
No reviewers
Labels
No labels
Kind/Breaking
Kind/Bug
Kind/Documentation
Kind/Enhancement
Kind/Feature
Kind/Security
Kind/Testing
Priority
Critical
Priority
High
Priority
Low
Priority
Medium
Reviewed
Confirmed
Reviewed
Duplicate
Reviewed
Invalid
Reviewed
Won't Fix
Status
Abandoned
Status
Blocked
Status
Need More Info
No milestone
No project
No assignees
3 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference: lix-project/nix-eval-jobs#14
Loading…
Reference in a new issue
No description provided.
Delete branch "fixes/build-settings-opeq"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Fixes #13. I am not certain that I've used
setDefault()
vsoverride()
in the right places, but I think it shouldn't matter too much and it seems to work?@ -351,3 +351,3 @@
/* FIXME: The build hook in conjunction with import-from-derivation is
* causing "unexpected EOF" during eval */
settings.builders = "";
settings.builders.setDefault("");
Not sure if this makes sense, if it breaks stuff it should be overridden, and if it doesn't why do it at all?
it appears to be nonsense. we removed it and observed that IFD induced remote builds work, so idk, probably some old nix bug that we fixed??
@ -355,3 +355,3 @@
/* Prevent access to paths outside of the Nix search path and
to the environment. */
evalSettings.restrictEval = false;
evalSettings.restrictEval.setDefault(false);
This does not make sense at all, as the default is
false
already. You either want to override it or delete this entirely. Or judging by the comment,false
is a typo to begin with and this should set the default totrue
(but still allowing the user to override it)?Oh that comment makes this very confusing then. The CLI doesn't seem to have any way to specify this in the first place? Should we just leave it as
false
since that's what it's been, and add a FIXME?Looking at the blame of this code, it seems that the comment was added along with explicitly setting this to
true
, and when it was later set tofalse
some time later (presumably because breakage in some situations was noticed), the comment was left. I'm not really qualified to comment on this because I'm not familiar with this codebase, but my gut feeling is that this code should be deleted entirely and the user can pass--option restrict-eval true
if they really want to enable restricted eval mode.@ -360,3 +360,3 @@
'getEnv', 'currentSystem' etc. */
if (myArgs.impure) {
evalSettings.pureEval = false;
evalSettings.pureEval.setDefault(false);
This wants to be override (and I guess things would be easier to understand here without the weird
else if
).