[python-uk] Static site generators with Python
stestagg at gmail.com
Fri Sep 2 19:36:59 EDT 2016
On Fri, Sep 2, 2016 at 7:57 PM Andy Robinson <andy at reportlab.com> wrote:
> On 2 September 2016 at 19:30, Stestagg <stestagg at gmail.com> wrote:
> > Hi Andy
> > We use travis to build the site. You can see the travis yaml file and
> > scripts in our repo:
> > If you can make your repo public, I would recommend travis as a great
> way to
> > build your site.
> We can
> > If you've any questions, please ask on or off list.
> Thanks for this. I have never touched Travis, but it's time I learned.
> If I understand correctly, it's a hosted service which can do pretty
> much anything you want when it sees a commit; in this case
> - pull from jobs repo
> - generate output files
> - push back the finished HTML up to the GitHub repo
> (Yes, I've been under a rock for about 10 years)
> I am embarking on a rather ambitious project to modernise an entire
> sport. Thousands of clubs have the problem of somewhere between zero
> and one webmasters at any one time, and the dynamic things they need
> to do - fixture lists, race results - can just about all he handled by
> no need for database server. And GitHub solves the problem of letting
> multiple people access the repo and write/edit stories with no coding
> Speaking of which, is or was anyone on this site involved with
> athletics or running?
> - Andy
Yes, that's right, travis (and circleci, and the others..) basically just
spin up a container, check out the repo, and run the commands listed in the
travis.yml (or equivalent) in a full environment.
This allows us to pull, and even push, using secret tokens, to other git
repos (although that's not a very common use-case), install software, run
tests, and build the site.
One thing I would caution about is that github is not a CMS, and while it's
awesome and does DCVS really well, there is definitely a minimum level of
experience required to use it to submit a good pull request. Simple things
like naming files correctly, editing online, cloning repos etc, are not
intuitive to people not used to these sort of workflows.
If your users are willing to learn/already know enough to understand what's
happening, then it's fine, but I would expect a certain amount of
frustration from the less technically oriented editors. (Something of a
feature for pythonjobs, but I don't know if that's the case for athletics)
Finally, if you're thinking of doing this at scale, I'd recommend reaching
out to the github support team fairly early on. We did for pythonjobs, just
to check that what we were doing was a good idea, and they were very keen
on the idea, but opening a channel early on may help with any issues down
> python-uk mailing list
> python-uk at python.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the python-uk