Closes #10611 - Fix exit status of Flank

[AaronMT]

Closes #10611

[rpappalax]

I'm not sure this will work. The reason why we use a function to set the exit code is that we still want some artifacts to be written regardless of pass or fail outcome. Most of the terminate on fail bash functions will end the script abruptly. I think we still need to capture the return code from the Java execution in a variable and hold onto it til the end. Not sure if pipefail is doing that

[AaronMT]

I'm not sure this will work. The reason why we use a function to set the exit code is that we still want some artifacts to be written regardless of pass or fail outcome. Most of the terminate on fail bash functions will end the script abruptly. I think we still need to capture the return code from the Java execution in a variable and hold onto it til the end. Not sure if pipefail is doing that

Setting pipefail will get us the the exit code of the process that's piped to another (sets the exit status $? to the exit code of the last program to exit non-zero (or zero if all exited successfully)).

E.g, what's happening now

$ false | true; echo $?
0

and after (see the ui-test-x86 debug failed above showing the right result now)

$ set -o pipefail
$ false | true; echo $?
1

[rpappalax]

I'm not sure this will work. The reason why we use a function to set the exit code is that we still want some artifacts to be written regardless of pass or fail outcome. Most of the terminate on fail bash functions will end the script abruptly. I think we still need to capture the return code from the Java execution in a variable and hold onto it til the end. Not sure if pipefail is doing that

Setting pipefail will get us the the exit code of the process that's piped to another (sets the exit status $? to the exit code of the last program to exit non-zero (or zero if all exited successfully)).

E.g, what's happening now

$ false | true; echo $?
0

and after (see the ui-test-x86 debug failed above showing the right result now)

$ set -o pipefail
$ false | true; echo $?
1

I'm not sure this will work. The reason why we use a function to set the exit code is that we still want some artifacts to be written regardless of pass or fail outcome. Most of the terminate on fail bash functions will end the script abruptly. I think we still need to capture the return code from the Java execution in a variable and hold onto it til the end. Not sure if pipefail is doing that

Setting pipefail will get us the the exit code of the process that's piped to another (sets the exit status $? to the exit code of the last program to exit non-zero (or zero if all exited successfully)).

E.g, what's happening now

$ false | true; echo $?
0

and after (see the ui-test-x86 debug failed above showing the right result now)

$ set -o pipefail
$ false | true; echo $?
1

per our zoom sesion LGTM 👍

[codecov-io]

Codecov Report

Merging #10612 into master will increase coverage by 0.20%.
The diff coverage is n/a.

@@             Coverage Diff              @@
##             master   #10612      +/-   ##
============================================
+ Coverage     19.20%   19.40%   +0.20%     
- Complexity      599      602       +3     
============================================
  Files           360      359       -1     
  Lines         14801    14638     -163     
  Branches       2001     1967      -34     
============================================
- Hits           2842     2841       -1     
+ Misses        11699    11537     -162     
  Partials        260      260              

Impacted Files	Coverage Δ	Complexity Δ
...g/mozilla/fenix/settings/logins/SavedLoginsView.kt	0.00% <0.00%> (-16.22%)	0.00% <0.00%> (ø%)
...ix/home/sessioncontrol/SessionControlController.kt	71.17% <0.00%> (-0.91%)	0.00% <0.00%> (ø%)
...pp/src/main/java/org/mozilla/fenix/FeatureFlags.kt	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
...rc/main/java/org/mozilla/fenix/FenixApplication.kt	12.24% <0.00%> (ø)	4.00% <0.00%> (ø%)
...g/mozilla/fenix/addons/AddonsManagementFragment.kt	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
...x/home/sessioncontrol/viewholders/TabViewHolder.kt	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
...me/sessioncontrol/viewholders/TopSiteViewHolder.kt	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
.../sessioncontrol/viewholders/TabHeaderViewHolder.kt	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
...sessioncontrol/viewholders/CollectionViewHolder.kt	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
...oncontrol/viewholders/TabInCollectionViewHolder.kt	0.00% <0.00%> (ø)	0.00% <0.00%> (ø%)
... and 22 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7993f10...34cd540. Read the comment docs.