You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Sometimes there is a reason for unit-tests to use the JUnit4 "@before" annotation (that is, run before each test).
Most of the time, an least in the javalib tests I recently had cause to examine, "@BeforeClass" (that is, run once
before each class) is intended. Similar concern with "@after" & "@afterclass" annotation.
posixlib unit-tests should be more consistent with using the "@BeforeClass", if that is what the logic of the
test intends.
In some of the files, there is only one test. So now, the two are equivalent. Once some diligent future maintainer
adds a test, the execution strategy will change/fall_apart.
In the files with more than one Test, there should be some minor improvement in execution speed because the
setup/teardown is done only once per class.
I have a number of irons in the fire, each requiring tending. I did not do a sweep through javalib, etc. to see
if a similar situation exists there. As I am in those projects from time to time, I will take a look when I am there.
The text was updated successfully, but these errors were encountered:
Sometimes there is a reason for unit-tests to use the JUnit4 "@before" annotation (that is, run before each test).
Most of the time, an least in the javalib tests I recently had cause to examine, "@BeforeClass" (that is, run once
before each class) is intended. Similar concern with "@after" & "@afterclass" annotation.
posixlib unit-tests should be more consistent with using the "@BeforeClass", if that is what the logic of the
test intends.
In some of the files, there is only one test. So now, the two are equivalent. Once some diligent future maintainer
adds a test, the execution strategy will change/fall_apart.
In the files with more than one Test, there should be some minor improvement in execution speed because the
setup/teardown is done only once per class.
I have a number of irons in the fire, each requiring tending. I did not do a sweep through javalib, etc. to see
if a similar situation exists there. As I am in those projects from time to time, I will take a look when I am there.
The text was updated successfully, but these errors were encountered: