ICRAR / daliuge / 2317 / 1
81%
master: 81%

LAST BUILD BRANCH: LIU-401
DEFAULT BRANCH: master
Ran
Files 37
Run time 2s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 64.002%. First build
NO_DLG_RUNTIME=1

push

travis-ci-com

rtobar 
Simplify wait_until function and friends

The previous version of these functions worked, but I found its design
was a bit over-complicated. I believe this also led to using it
incorrectly, with some calls doing a "double wait_until" invocation,
where an outer wait_until call in the test code used as a condition a
function that itself used wait_until.

This commit re-designs the wait_until function and its callers. The
first difference in this new version is that the timeout and interval
values always use their default values (although they can still be set
by on each invocation), as in the previous version all invocations
passed the same value explicitly. Secondly, instead of passing an update
function and a set of arguments to call it with, we simply pass an
update function, which in the caller site is bound to local arguments
(using a lambda in our cases, but could be a functools.partial if
required). These two changes simplified the function calling convention
a bit.

Finally, I've removed the "double wait_until" instances with what I
believe is a correct replacement. While I'm not particularly sure I've
done the right thing, I still see the same test code coverage, and all
tests are still passing. This leads me to believe I did the right thing,
but I will need confirmation. As a bonus, this change reduced the test
walltime from ~110 seconds to ~< 60.

Signed-off-by: Rodrigo Tobar <rtobar@icrar.org>

3595 of 5617 relevant lines covered (64.0%)

0.64 hits per line

Source Files on job 2317.1 (NO_DLG_RUNTIME=1)