I did not see the error again. I guess it was an effect of working late hours.. I messed something up.
I was playing with --incremental on jekyll perhaps that's what caused it. Playing with a single branch would make it a lot easier. It bothers me to keep moving between branches as my IDE gets crazy on indexing files (don't ask.. I prefer IDE than VIM... I use vim emulation on idea though ;) ) If more people could double check this is okay.. perhaps we can replace asf-site by my test branch. On Thu, Oct 3, 2019 at 3:09 PM Clebert Suconic <[email protected]> wrote: > > I copied the folder from the branch. > > Check out asf-siite > Move it. > > > Checkout the test branch > > > Moved it back. > > > > On Thu, Oct 3, 2019 at 7:31 AM Robbie Gemmell <[email protected]> > wrote: >> >> I dont see that happen when I run the build, and hadnt actually ever >> seen the 'typechange' file status before. >> >> The content/schema/core/activemq-core.xsd is actually a symlink, and >> some googling suggests the 'typechange' status mainly happens when >> such link is replaced with an actual file. If I do that deliberately, >> I then see the same status you do. Some posts suggested it can happen >> with certain copy commands, that have dereferenced the link and copied >> the referenced files content. How did you initially populate the >> 'content' dir your subsequent status is showing as being updated? >> Running the build afresh, or copying prior build output? >> >> Robbie >> >> On Wed, 2 Oct 2019 at 22:53, Clebert Suconic <[email protected]> >> wrote: >> > >> > I have a branch where I'm using a single branch: >> > >> > https://github.com/clebertsuconic/activemq-website/tree/test >> > >> > >> > The only concern I have so far is that any time I build, I get a >> > change into a schema: >> > >> > Changes not staged for commit: >> > (use "git add <file>..." to update what will be committed) >> > (use "git checkout -- <file>..." to discard changes in working directory) >> > >> > typechange: content/schema/core/activemq-core.xsd >> > >> > >> > >> > Although I would consider that anyone would check stuff before committing.. >> > >> > Any ideas? >> > >> > On Wed, Oct 2, 2019 at 1:03 PM Justin Bertram <[email protected]> wrote: >> > > >> > > If it's simpler and documented that sounds like a win to me. >> > > >> > > >> > > Justin >> > > >> > > On Wed, Oct 2, 2019 at 11:41 AM Robbie Gemmell <[email protected]> >> > > wrote: >> > > >> > > > On Wed, 2 Oct 2019 at 16:47, Clebert Suconic >> > > > <[email protected]> >> > > > wrote: >> > > > > >> > > > > On Wed, Oct 2, 2019 at 10:44 AM Robbie Gemmell >> > > > > <[email protected]> >> > > > wrote: >> > > > > > Thats what we do at Qpid and what I've seen other >> > > > > > projects do, I find it simpler overall. >> > > > > >> > > > > Tl;DR: >> > > > > I'm all +10000000 on this... what we need to change on infra. >> > > > > >> > > > > Can we keep the history from master >> > > > >> > > > >> > > > We wouldnt need to change much to get everything on the same branch I >> > > > dont think, and it seems like we can do so without infra: I just tried >> > > > on your last commit as a test and was able to force-push the asf-site >> > > > branch in the website repo. >> > > > >> > > > So basically it seems it needs something like: repopulate the asf-site >> > > > branch (or a test branch) with the master history locally, simplify >> > > > the build script to only build and nothing else, fix up the .gitignore >> > > > file appropriately, build things, commit, and [force] push. >> > > > >> > > > Assuming of course others agree that it is what should happen? >> > > > >> > > > Robbie >> > > > >> > >> > >> > >> > -- >> > Clebert Suconic > > -- > Clebert Suconic -- Clebert Suconic
