Oliver Smith authored 4 months ago and Clayton Craft committed 4 months ago

Right now we have both dirs, but contrib seems more fitting. Rename
utils.

[ci:skip-build]: already built successfully in CI

As discussed within the CoC team, it would be useful to display on top
what we changed compared to the standard Contributor Covenant 2.1.

Make clear who will receive e-mails sent to the CoC address.

Add a redirect for pmb MR 2251. If we add more of these in the future,
it would make sense to make this more generic just like how it works
for the wiki pages.

Add a script that creates a draft for the next monthly blog post by
grabbing merged MRs from the gitlab API.

Rename the local html variable of various functions to html_str, so it
does not conflict with the html module.

New post that should accompany: pmaports!5636

[ci:skip-build]: already built successfully in CI

It also seemed to like flagging Drew DeVault's last name so I also added
an exception for that.

Arnav Singh authored 5 months ago and Clayton Craft committed 5 months ago

{core,trusted}-contributors, update contributor gitlab.com profile links to gitlab.postmarketos.org (MR 319)

Co-authored-by: Stefan Hansson <newbyte@postmarketos.org>

* gitlab.com -> gitlab.postmarketos.org
* linux kernel: point to SoC Commuities page, don't link to the very
  outdated archive (3 years and older) of 6 open bug reports anymore
* /download -> /install (it has been renamed earlier)
* Link to "how to report issues" wiki page, which goes into detail on
  where to report what.

Pablo Correa Gomez authored 6 months ago and Oliver Smith committed 6 months ago

Co-Authored-By: Oliver Smith <ollieparanoid@postmarketos.org>

The checks did include ones that are not valid paths for
any page on this site, but crucially it did not include
those ones that are valid. It missed all of them, expect
/team/ itself.

Signed-off-by: Markus Göllnitz <camelcasenick@bewares.it>

This is a workaround, a missing feature or otherwise bad
UX in a browser. It is thrown together with navigation
options inside the website, and not on the specific page.
Not just out of place, it also fills the UI. Especially
on mobile, this precious space in an ideally slim footer.

Signed-off-by: Markus Göllnitz <camelcasenick@bewares.it>

The current footer buttons stuck out strongly, which I
would not want nor expect from a footer.

Further, this harmonises the used button gaps with the
navigation row at the top.

Thirdly, this introduces a hover-style for the inline
links in the footer, which until now did not change
their style when being hovered.

Signed-off-by: Markus Göllnitz <camelcasenick@bewares.it>

When the content is particularly short, one can find
the footer to be directly below the main content even
if that happen to end before the viewport's height.

It looks better if it would always stick to the bottom,
even in those cases.

Link: https://postmarketos.org/edge/2024/02/05/gtk4-new-renderers/



Signed-off-by: Markus Göllnitz <camelcasenick@bewares.it>

I see no way how this flag can affect production systems,
as they use frozen flask, and even if it was plain flask,
ideally the system should not see template changes without
a controlled restart.

In contrast, for development setups this will allow
faster iteration due to the elimination of the need to
restart the whole flask server.

Signed-off-by: Markus Göllnitz <camelcasenick@bewares.it>

The 2018 Akademy slides are broken due to issues with
re-minification of the remark-lates.min.js file.

While excluding that specific file would be an option, as
would any file with a *.min.* indicator of already being
minified (as the issue stems from unusual coding
practices that are the result of the input being minified
already), I suggest—after thinking about it—instead to
exclude the slides from being minified.

The reason for that: I added Flask-Minify to make sure
the website is served fast by reducing its overall size
as well the render blocking components especially. And
I see that as an important feature for most parts of the
website, but the slides serve a different purpose: They
are documentation, more of an archive. I delay of a few
tens to hundreds of milliseconds for those slides is
more than acceptable, I would argue.

I looked at the front page, as well as multiple blog
posts, and most other pages, and intend to continue to do
so. They may continually evolve while the slides are best
served as is. And this argument stands for any future
additions of archive-like content. (And it does so for
static/pdf/* as well, which just is not suspect of HTML,
CSS, or JS minification.)

Fixes: 144a0001 ("app: minify CSS and HTML")
Closes: https://gitlab.com/postmarketOS/postmarketos.org/-/issues/181



Signed-off-by: Markus Göllnitz <camelcasenick@bewares.it>

Caleb Connolly authored 6 months ago and Oliver Smith committed 6 months ago

Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
Tweaked-by: Oliver Smith <ollieparanoid@postmarketos.org>

The on device installer is currently not maintained, and we don't built
images with it. We have ideas to use BuffyBox for this though, see the
related issue.

Related: https://gitlab.com/postmarketOS/buffybox/-/issues/36