Re: logging and builds

[Dylan Millikin <dy...@gmail.com>]

It might be possible to output to a file and then echo or do something with
the file on test failure. Thought I would through that out there. Don't
really know where this is at atm.

On Mon, Dec 7, 2015 at 12:35 PM, Stephen Mallette <sp...@gmail.com>
wrote:

> I don't think that would help as we need the output to know what went wrong
> and i don't think we can get the file off of travis (even if we could the
> convenience factor of travis would be lost).  Anyway, the logging is well
> under control now and if we clean up a few more odds and ends with the
> printlns, we should be fine.  thanks.
>
> On Mon, Dec 7, 2015 at 4:38 AM, Marvin Froeder <ve...@gmail.com> wrote:
>
> > What about redirecting output?
> >
> >
> >
> http://maven.apache.org/surefire/maven-surefire-plugin/test-mojo.html#redirectTestOutputToFile
> >
> > Doesn't help?
> >
> > On Sat, 5 Dec 2015 03:17 Nghia Tran <ng...@gmail.com> wrote:
> >
> > > Sorry - by "this" I meant
> > > https://issues.apache.org/jira/browse/TINKERPOP-1019.  I can work on
> > that
> > > and offer a pull request.
> > >
> > > On Fri, Dec 4, 2015 at 9:13 AM, Nghia Tran <ng...@gmail.com>
> > > wrote:
> > >
> > > > Hi,
> > > > I'd like to take this on.
> > > >
> > > > Regards,
> > > > Nghia Tran
> > > >
> > > > On Fri, Dec 4, 2015 at 9:10 AM, Stephen Mallette <
> spmallette@gmail.com
> > >
> > > > wrote:
> > > >
> > > >> It was painful and I'm scarred, but the simple task for "make builds
> > log
> > > >> less so that travis will be happy" turned into quite an affair.  In
> > any
> > > >> case, I believe that what I've implemented gives us complete control
> > of
> > > >> logging during maven builds, which makes a lot of things possible:
> > > >>
> > > >> 1. The "ci" profile can now work nicely with Travis and the build
> > output
> > > >> far more readable
> > > >> 2. We can stop doing println type output in tests and get control
> over
> > > >> what
> > > >> gets written when:
> > https://issues.apache.org/jira/browse/TINKERPOP-1019
> > > >> 3. We can easily turn on TRACE level logging, alter logging formats,
> > etc
> > > >> for improved ability to debug builds.
> > > >>
> > > >> By default I've set all log levels in all projects to WARN.  For
> > > >> spark-gremlin, there are some additional restrictions
> > (jetty/akka/etc).
> > > >> If
> > > >> you want to ever modify these settings while debugging, you can do
> so
> > as
> > > >> described here:
> > > >>
> > > >>
> > > >>
> > >
> >
> http://tinkerpop.incubator.apache.org/docs/3.1.1-SNAPSHOT/dev/developer/#logging
> > > >>
> > > >
> > > >
> > >
> >
>

Re: logging and builds

[Stephen Mallette <sp...@gmail.com>]

Travis is working pretty nicely.  The PR for

https://issues.apache.org/jira/browse/TINKERPOP-1019

was received and merged and it worked nicely as everything in the test
cases is consistent now.  I don't think there's much to change anywhere at
this point imo.

On Thu, Jan 7, 2016 at 9:26 AM, Dylan Millikin <dy...@gmail.com>
wrote:

> It might be possible to output to a file and then echo or do something with
> the file on test failure. Thought I would through that out there. Don't
> really know where this is at atm.
>
> On Mon, Dec 7, 2015 at 12:35 PM, Stephen Mallette <sp...@gmail.com>
> wrote:
>
> > I don't think that would help as we need the output to know what went
> wrong
> > and i don't think we can get the file off of travis (even if we could the
> > convenience factor of travis would be lost).  Anyway, the logging is well
> > under control now and if we clean up a few more odds and ends with the
> > printlns, we should be fine.  thanks.
> >
> > On Mon, Dec 7, 2015 at 4:38 AM, Marvin Froeder <ve...@gmail.com>
> wrote:
> >
> > > What about redirecting output?
> > >
> > >
> > >
> >
> http://maven.apache.org/surefire/maven-surefire-plugin/test-mojo.html#redirectTestOutputToFile
> > >
> > > Doesn't help?
> > >
> > > On Sat, 5 Dec 2015 03:17 Nghia Tran <ng...@gmail.com> wrote:
> > >
> > > > Sorry - by "this" I meant
> > > > https://issues.apache.org/jira/browse/TINKERPOP-1019.  I can work on
> > > that
> > > > and offer a pull request.
> > > >
> > > > On Fri, Dec 4, 2015 at 9:13 AM, Nghia Tran <ng...@gmail.com>
> > > > wrote:
> > > >
> > > > > Hi,
> > > > > I'd like to take this on.
> > > > >
> > > > > Regards,
> > > > > Nghia Tran
> > > > >
> > > > > On Fri, Dec 4, 2015 at 9:10 AM, Stephen Mallette <
> > spmallette@gmail.com
> > > >
> > > > > wrote:
> > > > >
> > > > >> It was painful and I'm scarred, but the simple task for "make
> builds
> > > log
> > > > >> less so that travis will be happy" turned into quite an affair.
> In
> > > any
> > > > >> case, I believe that what I've implemented gives us complete
> control
> > > of
> > > > >> logging during maven builds, which makes a lot of things possible:
> > > > >>
> > > > >> 1. The "ci" profile can now work nicely with Travis and the build
> > > output
> > > > >> far more readable
> > > > >> 2. We can stop doing println type output in tests and get control
> > over
> > > > >> what
> > > > >> gets written when:
> > > https://issues.apache.org/jira/browse/TINKERPOP-1019
> > > > >> 3. We can easily turn on TRACE level logging, alter logging
> formats,
> > > etc
> > > > >> for improved ability to debug builds.
> > > > >>
> > > > >> By default I've set all log levels in all projects to WARN.  For
> > > > >> spark-gremlin, there are some additional restrictions
> > > (jetty/akka/etc).
> > > > >> If
> > > > >> you want to ever modify these settings while debugging, you can do
> > so
> > > as
> > > > >> described here:
> > > > >>
> > > > >>
> > > > >>
> > > >
> > >
> >
> http://tinkerpop.incubator.apache.org/docs/3.1.1-SNAPSHOT/dev/developer/#logging
> > > > >>
> > > > >
> > > > >
> > > >
> > >
> >
>