[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
[wtp-releng] Status of our Helios builds ... and actions required
|
Apologies for the many "WTP 3.3" build breaks, caused by me, but I think
I've finally tweaked our (main) build to use our same 3.2 source, but with
a different set of pre-reqs.
It appears its we are fairly compatible with the Platform's 3.6 M1-ish
code, except ...
1) a small tweak was required to widen our range of jdt pre-req in
jsp.core ... a good reminder we should all strive to get off all use of
non-API code so we can use wide ranges.
2) A large number of JSF JUnits are consistently failing (all for same
reason, it appears). It'd be good to check these out in next day or two
just in case it really has uncovered a change/bug in Platform's code --
their M1 deadline is this Friday, the 8th -- but offhand it could just be
a test problem, or perhaps they improved error reporting? But, if they
have actually "broken us", we should let them know before they promote
their M1 -- I know they want to maintain compatibility if they can.
Otherwise, we should fix our code to work equally well with each set of
pre-reqs.
3) Some, not completely consistent, JUnit errors in Dali's unit tests. As
above, let us know if that's a change in behavior from Platform, or if
it's something we can fix in our one code base.
Remember that we won't have a formal smoke-test and formal "promote vote"
for our Helios M1 -- we will use Galileo as the base for our testing for
our 3.2 plans -- but I'd still encourage everyone to do your own informal
smoke tests and try and catch any obvious bugs or incompatibilities. Our
plan is to have our code base to be compatible with both streams (Galileo
and Helios).
Thanks all,