That is in fact a good point raised to revisit our excludes to make sure we
do not have any unnecessary ones. Me and Jarcec did some brainstorming
around this and looks like most of the excludes we have are for better
developer usability, which mostly exclude some generated files.
1) Git specific files that might not contain headers:
<exclude>.git/</exclude>
<exclude>.gitignore</exclude>
2) IDE specific files that we are ignoring to make live easier for
developers:
<exclude>.idea/</exclude>
<exclude>**/*.iml</exclude>
<exclude>**/nb-configuration.xml</exclude>
<exclude>**/.settings/**</exclude>
<exclude>**/.classpath</exclude>
<exclude>**/.project</exclude>
3) Maven working directory:
<exclude>**/target/</exclude>
4) Pre commit testing generated files
<exclude>maven-repo/</exclude>
<exclude>test-output/</exclude>
5) Derby files that can be created after test run
<exclude>*.log</exclude>
<exclude>**/*.lck</exclude>
<exclude>**/service.properties</exclude>
6) Patch files that can be easily lying around
<exclude>**.patch</exclude>
7) Our readme file
<exclude>README*</exclude>
8) Test data with exact format
<exclude>**/kv1.dat</exclude>
9) Self explaining:
<!-- exclude generated solr config files -->
<exclude>**/solr/collection1/conf/**</exclude>
<!-- exclude generated thrift files →
<exclude>**/gen/**</exclude>
X) Not sure what this is for, to be removed?
<exclude>**/.metadata/</exclude>
So I feel confident that we are in good shape with respect to excludes.
David, does this break down look ok?
Regards,
On Fri, Apr 4, 2014 at 10:37 AM, David Nalley <[email protected]> wrote:
> On Fri, Apr 4, 2014 at 11:15 AM, Jarek Jarcec Cecho <[email protected]>
> wrote:
> > Hi David,
> > thank you very much for your feedback, greatly appreciated! You've
> raised couple of really good points.
> >
> > The rat issues is interesting - indeed we are not setting the
> excludeSubProjects to false but at the same time rat is checking all
> subprojects. I've just tried to put incorrectly licensed file into one of
> the sub modules and it has been correctly detected, so I'm assuming that
> this one should be covered.
> >
>
> The bigger issue isn't rat configuration, it's how do we answer the
> question of why so many files don't have license headers.
>
> > Looking at the rest of the feedback it seems that nothing is a release
> blocker per say, so I'm wondering if you are fine with us moving forward
> with the release?
> >
>
> My vote is just a single vote, nothing says you have to stop. -1's
> aren't vetoes in release votes. I personally want to get to the point
> where I can vote +1 but am not there yet.
>
> --David
>
--
Sravya Tirukkovalur
