Thanks for putting that together. Does this mean you've volunteered to
referee bikeshedding?

On Mon, Jan 30, 2017 at 5:21 PM Eugene Kirpichov
<[email protected]> wrote:

> The initial PR has been merged and the style guide is live
> https://beam.apache.org/contribute/ptransform-style-guide/ - let us
> continue discussing and tweaking on this thread and via smaller PRs
> modifying the document.
>
> On Mon, Jan 30, 2017 at 7:50 AM Aljoscha Krettek <[email protected]>
> wrote:
>
> > Wow, that's a long read. But quite informative +1
> >
> > On Sat, 28 Jan 2017 at 06:54 Jean-Baptiste Onofré <[email protected]>
> wrote:
> >
> > > Hi Eugene,
> > >
> > > As said in the PR: great work and thanks a lot !
> > >
> > > I will take a complete look during the week end. I'm pretty sure it's a
> > > great guide as it's basically the result of our discussions and reviews
> > ;)
> > >
> > > Thanks again !
> > > Regards
> > > JB
> > >
> > > On 01/28/2017 06:21 AM, Eugene Kirpichov wrote:
> > > > Hello all,
> > > >
> > > > I just sent a pull request with a style guide for developers of new
> > > > PTransforms - intended for library writers, e.g. people who
> contribute
> > > new
> > > > connectors and other transforms to Beam. The guide is mainly based on
> > > > experience from reviewing connectors contributed by JB and others,
> but
> > > it's
> > > > intended to be generally applicable.
> > > >
> > > > It covers a variety of points - from code organization, to overall
> API
> > > > design, to error handling and so on. I expect most of it to be
> > > > non-controversial and just reflect the style of existing transforms
> in
> > > Beam
> > > > - however all of it is, of course, up to debate.
> > > >
> > > > https://github.com/apache/beam-site/pull/134/
> > > >
> > > > I'm hoping that this documentation will help guide new transform
> > authors
> > > in
> > > > the right direction from the start, as well as make the job of
> > reviewers
> > > > easier by providing a source they can link to and helping focus the
> > > review
> > > > on resolving more ambiguous points.
> > > >
> > > > (Note that, like all other documentation, this will evolve, so the
> goal
> > > of
> > > > the current PR is not to be complete, but to be a starting point)
> > > >
> > > > When the guide is ratified, I think it'll make sense to file JIRAs to
> > > bring
> > > > Beam in accordance with it - there are a few transforms that were
> > written
> > > > before the best practices shaped up.
> > > >
> > > > Thanks!
> > > >
> > >
> > > --
> > > Jean-Baptiste Onofré
> > > [email protected]
> > > http://blog.nanthrax.net
> > > Talend - http://www.talend.com
> > >
> >
>

Reply via email to