Page MenuHomeFreeBSD

Defer evaluation of modified until after it's set
ClosedPublic

Authored by imp on May 31 2019, 3:01 PM.
Tags
None
Referenced Files
F103327599: D20480.diff
Sat, Nov 23, 2:49 PM
Unknown Object (File)
Thu, Nov 21, 3:34 PM
Unknown Object (File)
Thu, Nov 21, 1:38 PM
Unknown Object (File)
Thu, Nov 21, 3:46 AM
Unknown Object (File)
Mon, Nov 18, 9:54 AM
Unknown Object (File)
Mon, Nov 18, 3:27 AM
Unknown Object (File)
Sun, Nov 17, 4:05 PM
Unknown Object (File)
Sun, Nov 17, 9:46 AM
Subscribers
None

Details

Summary

With the reorg r348175, we now look at modified before it is
set. Rearrange things so that we can set include_metadata to either
yes, no or some variable to be evaluated in the future (in this case
modified). Use the eval echo trick to expanded it. This should fix the
-R flag that was broken in r348175, which broke
WITH_REPRODUCIBLE_BUILD for kernels.

Diff Detail

Lint
Lint Passed
Unit
No Test Coverage
Build Status
Buildable 24635
Build 23413: arc lint + arc unit

Event Timeline

No objection but I'm generally not a fan of the deferred evaluation; what about just setting include_metadata=yes/no/if-modified and making the test something like
if [${include_metadata} = yes || (${include_metadata} = if-modified && ${modified} = yes)]

Anyway if you prefer this approach that's fine with me

Fix per ed (though a little different construct)
Fix backwards test after converting from -z notation...

No objection but I'm generally not a fan of the deferred evaluation; what about just setting include_metadata=yes/no/if-modified and making the test something like
if [${include_metadata} = yes || (${include_metadata} = if-modified && ${modified} = yes)]

Anyway if you prefer this approach that's fine with me

Due to the funky way that test works, I opted for a slightly different way to say this...

This revision was not accepted when it landed; it landed in state Needs Review.May 31 2019, 10:57 PM
This revision was automatically updated to reflect the committed changes.