Commiting to master, site, and svn is the way to go I think.

It is not a problem if master and site are not in sync all the time, but it
is a problem if
we push directly from the master to svn. Another think to avoid is
commiting directly to svn.

Now in order to align the three branches I think it is best to wait for the
next release (which is supposed to come out inside 2019-02)
where everything should be in sync.

Στις Κυρ, 10 Φεβ 2019 στις 5:11 μ.μ., ο/η Andrei Sereda <[email protected]>
έγραψε:

> Yes static site is tricky. You have to keep 3 branches in sync: master,
> site and svn.
>
> I always publish from site branch (and have fewer diffs).
>
> Sometimes git master and site branches are only synced during release.
>
>
> On Sun, Feb 10, 2019 at 9:17 AM Stamatis Zampetakis <[email protected]>
> wrote:
>
> > @Francis: Whatever I said concerns the site branch (not the master).
> > @Andrei: I did more or less the same thing but without using docker.
> >
> > It appears that the site branch in git and the published site in svn are
> > not sync. Compare for instance:
> >
> > https://github.com/apache/calcite/blob/site/site/_docs/index.md
> > https://calcite.apache.org/docs/
> >
> > and look for "it does not even have a favorite data format"; you will not
> > find it in both places (this is also what the previous diff showed).
> >
> > It appears as things from the master branch (e.g.,
> > fc5af78bf37a4a8b9e89c20e2a478fb4a95dbd4d) were pushed directly to the svn
> > without passing through the site branch.
> >
> >
> > Στις Κυρ, 10 Φεβ 2019 στις 3:49 π.μ., ο/η Andrei Sereda <[email protected]
> >
> > έγραψε:
> >
> > > Try first svn checkout then build site:
> > >
> > > $ rm -rf target
> > > $ svn co https://svn.apache.org/repos/asf/calcite/site target
> > > $ docker-compose run build-site
> > > $ cd target && svn status
> > >
> > >
> > > On Sat, Feb 9, 2019 at 9:43 PM Francis Chuang <
> [email protected]>
> > > wrote:
> > >
> > > > Hey Stamatis,
> > > >
> > > > Are you trying to publish using the site branch? When publishing the
> > > > site for anything other than a release, the site branch should be
> used.
> > > > If you're using the master branch, it's possible there are some
> > > > documentation for a future release that should not be published.
> > > >
> > > > Also, can you check if some of the changes are for making files
> > > > non-executable or executable? I recall that when I last published the
> > > > site, some of the files became executable. I believe I should have
> > fixed
> > > > most of them by making them non-executable, but some might have
> slipped
> > > > through.
> > > >
> > > > Francis
> > > >
> > > > On 10/02/2019 10:49 am, Stamatis Zampetakis wrote:
> > > > > It is not copyrights or headers. I get various different things.
> For
> > > > > instance the following:
> > > > >
> > > > >> svn diff docs/index.html
> > > > > Index: docs/index.html
> > > > > ===================================================================
> > > > > --- docs/index.html (revision 1853282)
> > > > > +++ docs/index.html (working copy)
> > > > > @@ -152,7 +152,7 @@
> > > > >   object as a schema and its collection fields <code
> > > > > class="highlighter-rouge">emps</code> and <code
> > > > > class="highlighter-rouge">depts</code> as
> > > > >   tables.</p>
> > > > >
> > > > > -<p>Calcite does not want to own data; it does not even have a
> > favorite
> > > > data
> > > > > +<p>Calcite does not want to own data; it does not even have
> favorite
> > > > data
> > > > >   format. This example used in-memory data sets, and processed them
> > > > >   using operators such as <code
> > > class="highlighter-rouge">groupBy</code>
> > > > and
> > > > > <code class="highlighter-rouge">join</code> from the linq4j
> > > > >   library. But Calcite can also process data in other data formats,
> > > such
> > > > >
> > > > > Στις Σάβ, 9 Φεβ 2019 στις 11:12 μ.μ., ο/η Andrei Sereda
> > > <[email protected]
> > > > >
> > > > > έγραψε:
> > > > >
> > > > >> What are the differences ? Is it just copyright year / header ?
> > > > >>
> > > > >> On Sat, Feb 9, 2019 at 10:08 AM Stamatis Zampetakis <
> > > [email protected]>
> > > > >> wrote:
> > > > >>
> > > > >>> Hello,
> > > > >>>
> > > > >>> I did a git checkout of the site branch and then bundle exec
> jekyll
> > > > >> build.
> > > > >>>
> > > > >>> When I compare the output with the current version of the site in
> > > svn I
> > > > >> see
> > > > >>> lots of modifications (below the output of svn status in target
> > > > >> directory).
> > > > >>> I was wondering if this is normal (given that I didn't modify
> > > anything)
> > > > >> or
> > > > >>> if I didn't set up correctly my environment.
> > > > >>>
> > > > >>> Best,
> > > > >>> Stamatis
> > > > >>>
> > > > >>> ===============================================
> > > > >>>> svn status
> > > > >>>
> > > > >>> M       docs/file_adapter.html
> > > > >>> M       docs/index.html
> > > > >>> M       docs/reference.html
> > > > >>> M       downloads/index.html
> > > > >>> M       news/2014/06/27/release-0.8.0-incubating/index.html
> > > > >>> M       news/2014/08/19/release-0.9.0-incubating/index.html
> > > > >>> M       news/2014/10/02/release-0.9.1-incubating/index.html
> > > > >>> M       news/2014/11/05/release-0.9.2-incubating/index.html
> > > > >>> !       news/2015/01
> > > > >>> !       news/2015/01/31
> > > > >>> !       news/2015/01/31/release-1.0.0-incubating
> > > > >>> !       news/2015/01/31/release-1.0.0-incubating/index.html
> > > > >>> ?       news/2015/02
> > > > >>> !       news/2015/03/13
> > > > >>> !       news/2015/03/13/release-1.1.0-incubating
> > > > >>> !       news/2015/03/13/release-1.1.0-incubating/index.html
> > > > >>> ?       news/2015/03/14
> > > > >>> !       news/2015/04/07
> > > > >>> !       news/2015/04/07/release-1.2.0-incubating
> > > > >>> !       news/2015/04/07/release-1.2.0-incubating/index.html
> > > > >>> ?       news/2015/04/08
> > > > >>> !       news/2015/04/24
> > > > >>> !       news/2015/04/24/new-committers
> > > > >>> !       news/2015/04/24/new-committers/index.html
> > > > >>> ?       news/2015/04/25
> > > > >>> !       news/2015/05/30
> > > > >>> !       news/2015/05/30/release-1.3.0-incubating
> > > > >>> !       news/2015/05/30/release-1.3.0-incubating/index.html
> > > > >>> ?       news/2015/05/31
> > > > >>> !       news/2015/06/05
> > > > >>> !       news/2015/06/05/algebra-builder
> > > > >>> !       news/2015/06/05/algebra-builder/index.html
> > > > >>> ?       news/2015/06/06
> > > > >>> M       news/2015/07/31/xldb-best-lightning-talk/index.html
> > > > >>> M       news/2015/09/02/release-1.4.0-incubating/index.html
> > > > >>> M       news/2015/10/22/calcite-graduates/index.html
> > > > >>> !       news/2015/11/08
> > > > >>> !       news/2015/11/08/new-committers
> > > > >>> !       news/2015/11/08/new-committers/index.html
> > > > >>> ?       news/2015/11/09
> > > > >>> M       news/2015/11/10/release-1.5.0/index.html
> > > > >>> M       news/2016/01/22/release-1.6.0/index.html
> > > > >>> M       news/2016/02/17/elser-pmc/index.html
> > > > >>> M       news/2016/02/17/streaming-sql-talk/index.html
> > > > >>> M       news/2016/03/22/cassandra-adapter/index.html
> > > > >>> M       news/2016/03/22/release-1.7.0/index.html
> > > > >>> M       news/2016/06/13/release-1.8.0/index.html
> > > > >>> M       news/2016/09/22/release-1.9.0/index.html
> > > > >>> M       news/2016/10/12/release-1.10.0/index.html
> > > > >>> M       news/2017/01/09/release-1.11.0/index.html
> > > > >>> M       news/2017/03/24/release-1.12.0/index.html
> > > > >>> M       news/2017/03/31/new-avatica-repository/index.html
> > > > >>> M       news/2017/06/26/release-1.13.0/index.html
> > > > >>> M       news/2017/10/02/release-1.14.0/index.html
> > > > >>> M       news/2017/12/11/release-1.15.0/index.html
> > > > >>> M       news/2018/03/19/release-1.16.0/index.html
> > > > >>> M       news/2018/07/20/release-1.17.0/index.html
> > > > >>> M       news/2018/12/21/release-1.18.0/index.html
> > > > >>> M       news/index.html
> > > > >>> M       news/releases/index.html
> > > > >>>
> > > > >>
> > > > >
> > > >
> > > >
> > >
> >
>

Reply via email to