* Re: i assume "ptest" is an all-or-nothing selection, yes?
2014-08-05 9:55 ` Richard Purdie
@ 2014-08-05 19:37 ` Robert P. J. Day
0 siblings, 0 replies; 3+ messages in thread
From: Robert P. J. Day @ 2014-08-05 19:37 UTC (permalink / raw)
To: Richard Purdie; +Cc: Yocto discussion list
On Tue, 5 Aug 2014, Richard Purdie wrote:
> On Tue, 2014-08-05 at 05:40 -0400, Robert P. J. Day wrote:
> > dev manual, section 5.13.6, clearly implies that ptest is selected
> > either across your build or not at all -- i see no suggestion that you
> > can enable it with, say, package overrides, correct?
>
> Correct.
i thought as much, just wanted to make sure. and while that might be
obvious to many people, i think a short note at the beginning of that
section wouldn't hurt, but a couple more questions about ptest so i
can make some adjustments all at once, based on what's written here:
http://www.yoctoproject.org/docs/current/dev-manual/dev-manual.html#testing-packages-with-ptest
toward the end of that section, there is the claim:
"However, you still must add a do_compile_ptest function to build the
test suite. Add a function similar to the following to your recipe:
do_compile_ptest() {
oe_runmake buildtest-TESTS
}
well, not really ... i did some perusing and there are numerous
ptest-enabled recipes that don't define their own do_compile_ptest()
routine, obviously because the standard compile step handles the
compilation of the test code.
and, finally, there's this last claim:
"The ptest class automatically copies the file run-ptest to the target
and then runs make install-ptest to run the tests. If this is not
enough, you need to create a do_install_ptest function and make sure
it gets called after the "make install-ptest" completes."
couple things there. rather than define a do_install_ptest()
routine, i notice a number of recipes simply patch their Makefile to
add the appropriate install-ptest: target, so if one is clever enough
to write the proper Makefile target, one could have a ptest-enabled
recipe with no definition whatsoever of *any* of the do_*_ptest()
routines.
oh, and about that last claim:
"If this is not enough, you need to create a do_install_ptest function
and make sure it gets called after the "make install-ptest"
completes."
don't know what you mean by "make sure it gets called ..." since the
code in ptest.bbclass:
if grep -q install-ptest: Makefile; then
oe_runmake DESTDIR=${D}${PTEST_PATH} install-ptest
fi
do_install_ptest
pretty much guarantees that it will be called -- there is no need for
the developer to "make sure" of it, yes?
anyway, i think i have a handle on this. any thoughts on the above?
rday
--
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
^ permalink raw reply [flat|nested] 3+ messages in thread