incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Matt Casters (i-Bridge)" <matt.cast...@ibridge.be>
Subject Re: [VOTE] Release Apache Hop (incubating) 0.99-rc1
Date Sun, 13 Jun 2021 10:25:00 GMT
For me I think it's been too frustrating because of a lack of vacation and
the whole pandemic, more than the technical issues.
In the end, this is not that big of a deal I think:
1) add an option through an environment variable like
HOP_LICENSE_HEADER_FILE
2) create a file with the ASF header in it somewhere under
integration-tests/
3) remove .hpl/.hwf from the RAT exclusions
4) Configure the integration test hop-config.json files to point the
variable to the header file.

I couldn't figure out how to update the copyright in Idea for .hpl/.hwf
files but I'm sure a global first replace should do the trick here.

The issue of the old XML and metadata Injection API still being around as
we push towards 1.0 is still around though.  It's nagging in the back of my
mind.  I would still love to hear your opinions about it.

Cheers,
Matt


Op zo 13 jun. 2021 om 11:12 schreef Bart Maertens <bart.maertens@know.bi>:

> Hi,
>
> What I don't get about this whole discussion is that we already had 169 hpl
> and hwf files in the 0.70 release, and these files weren't even mentioned
> in the reviews.
> Now that we have about 400 hpl/hwf files in the 0.99 release, this
> is/becomes an issue.
> I'm sure we'll sort this out, but if this was detected and discussed sooner
> (0.70), that could have saved us all a lot of time, energy and, quite
> frankly, frustration.
>
> Maybe even more importantly, how do we know our current code base doesn't
> have any other problems that may only pop up in future releases?
>
> Regards,
> Bart
>
> On Sat, Jun 12, 2021 at 10:35 PM Matt Casters
> <matt.casters@neo4j.com.invalid> wrote:
>
> > After thinking about it some more I really believe strongly that our
> > developers should have an easy time adding more integration tests.  It's
> > really critical to our project in the longer term.
> > Manually editing XML files to copy/paste an ASF header in there can not
> be
> > part of that experience.  It's simply out of the question.
> > I don't see the same requirement for other Apache projects that have
> .avro
> > or .parquet files as part of their test-suites.  The obvious reason being
> > that folks are not really opening up these file formats with a text
> > editor... just like we never do this with .hpl/.hwf files.  So we're just
> > being punished for having chosen an XML format.
> >
> > Since we've exhausted all other possibilities I'll start the discussion
> on
> > dev to move to a different file format for our metadata.
> >
> > All the best,
> > Matt
> >
> > On Sat, Jun 12, 2021 at 2:09 PM Justin Mclean <justin@classsoftware.com>
> > wrote:
> >
> > > Hi,
> > >
> > > I'm a little confused, and I may be missing some context here. If the
> > work
> > > a part of an ASF project, why do you need to include a copyright
> > statements
> > > anywhere? If the code is not part of the project then we do need to
> know
> > > the license and copyright owner. While this might be a good place to
> > > include it, it would still also need to go in the LICENSE file, so
> that a
> > > duplication of work and something that could easily get out of sync.
> > >
> > > IMO (and there may be there ways of dealing with this) just get the
> > > contributors to sign ICLAs so there no issue with their contributions
> and
> > > there’s no issues with the contents of the LICENSE file or indeed
> > licensing.
> > >
> > > Kind Regards,
> > > Justin
> >
> >
> >
> > --
> > Neo4j Chief Solutions Architect
> > *✉   *matt.casters@neo4j.com
> > ☎  +32486972937
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message