ok great, thanks. I have to step aside for an hour or so. I will check it after that
On Thu, Sep 29, 2016 at 6:11 PM, Alex Harui <aha...@adobe.com> wrote: > The Ant build doesn't run flexjs.dependent.tests automatically. You have > to run ant with that target. It could be my setup, but I think those > files I mentioned need to be updated for the new reflection data. > > Thanks, > -Alex > > On 9/28/16, 10:03 PM, "Greg Dove" <greg.d...@gmail.com> wrote: > > >oh. I thought I had this sorted, I can run the ant and maven builds > >locally, and all the examples were building for me. > > > >I will check again. > > > >On Thu, Sep 29, 2016 at 6:00 PM, Alex Harui <aha...@adobe.com> wrote: > > > >> Hi Greg, > >> > >> The build failed for some strange reason. Might be a threading issue > >> unrelated to your changes. I pulled everything and was able to build. > >> I've restarted the CI build to see if the same error comes up again. > >> > >> I also ran "ant sdk.dependent.tests flexjs.dependent.tests" and there > >>were > >> a couple of failures in the compiler-ix tests. There are files in > >> compiler-jx/src/test/resources/flexjs that need updating but it also > >>looks > >> like an extra new-line is being added after goog.inherits in some cases. > >> Let us know if you get the same results. > >> > >> Results are available in compiler-jx/target/junit-results. If > >> TestFlexJSFile is reporting errors, there will be an xml file that ends > >> with TestFlexJSFIle.xml. It will contain the string "<failure" for each > >> failure. You will see an xml-escaped string with the error, in this > >>case, > >> comparing the expected vs actual results. You can use an unescape tool > >> like http://www.freeformatter.com/xml-escape.html#ad-output to get > >> readable text, then I have a utility hosted on at > >> http://home.apache.org/~aharui/TestResultCompare/TestResultCompare.html > >> that you can use to help you compare the results. Paste the readable > >> results and hit the button, then scroll down the other TextAreas. The > >> text will turn to bold where the first difference is found. There is no > >> way to see all differences, just the first one so you have to make and > >> adjustment, re-run the test, re-compare the results. > >> > >> Hopefully that's enough info so you can figure it out. > >> > >> Thanks, > >> -Alex > >> > >> > >
