I think this is definitely carelessness on my part and I welcome your improvements on the doc side, don't get me wrong.
I was working on the first commit since a few weeks, and I think I missed some changes during the merge phase. I'll revert the change I made on the docs/components/modules/ROOT/pages/index.adoc <https://github.com/apache/camel/commit/be547c1d177aaec8c8520ac1970ef9b055b1deb7#diff-5ca7195ef719af6b4fd2451ccfc96fbe> I suppose the second problem comes from the removal of the modules/eips/nav.adoc line, I'll have a closer look. On the build speed, I've spent a lot of time those past months to get the build down to something usable. Using mvnd (http://github.com/apache/mvnd) on a fast rebuild (i.e. no tests, no assembly, already build with no changes), the build time goes down to 51s on my laptop, so 20s more looks very long to me for something that can be done in 300ms. That amounts to 30% just for xref checks, which just looks way too high. That's why I went ahead some time ago and rewrote the xref checks. When you worked on the docs, you certainly saw that i could not support the new layout, so you replaced it back with the original one. I was just trying to get it back. I think I fixed the checks themselves, but I'll have another look. Le sam. 4 avr. 2020 à 17:38, David Jencks <david.a.jen...@gmail.com> a écrit : > https://issues.apache.org/jira/browse/CAMEL-14846 > > thanks > David Jencks > > > On Apr 4, 2020, at 1:31 AM, Andrea Cosentino <anco...@gmail.com> wrote: > > > > Please open a JIRA issue about missing bits. There no secret vote in the > > Apache Camel community, there is only good will from everyone. > > > > Il sab 4 apr 2020, 01:19 David Jencks <david.a.jen...@gmail.com> ha > scritto: > > > >> I’m all for simpler quicker builds. I think the documentation source > >> generation is a cause of a lot of problems. Should the website get > fixed, > >> I have some more proposals that could considerably reduce the need for > >> generating index tables of documentation. > >> > >> My experience is that the automated website build already does link > >> checking. I would think that’s enough. It’s easy enough to run > locally as > >> well. > >> > >> Looking further into the current state of the website, Guillaumes commit > >> be547c1d177aaec8c8520ac1970ef9b055b1deb7 appears to have undone a > >> substantial portion of the work I did in the components component. Is > this > >> carelessness or a secret vote against my changes? > >> > >> David Jencks > >> > >>> On Apr 3, 2020, at 3:25 PM, Andrea Cosentino <anco...@gmail.com> > wrote: > >>> > >>> I think we could remove xcheck from the camel repo and add the check on > >> the > >>> website build eventually. I don't think it's a problem. We are trying > to > >>> improve the building experience in multiple ways, one of this is trying > >> to > >>> reduce the build time where we can, especially through the fastinstall > >>> profile. > >>> > >>> Il sab 4 apr 2020, 00:11 David Jencks <david.a.jen...@gmail.com> ha > >> scritto: > >>> > >>>> Guillaume’s commit 5ef68ba5e1014612ae5c424e6a5e841a523e5f19 yesterday > >>>> broke the components component in the website by removing the eips > from > >>>> navigation. > >>>> > >>>> Regarding this entire commit, personally I’m not a fan of trying to > >>>> duplicate functionality provided by other tools. I don’t think it’s > >>>> maintainable. I’d just run the xref check when building the entire > >>>> website. In the context of a build that has yet to complete on my > >> system, > >>>> 20 seconds doesn’t seem that long. > >>>> > >>>> David Jencks > >> > >> > > -- ------------------------ Guillaume Nodet
