When someone posts a new topic, this will automatically remove any tag
that matches the group name. So for example, if the topic is being
posted in ~music, a tag of "music" will be removed.
If there turns out to be some edge case where this is important, it can
be re-added through the tag editing interface, since this only applies
when posting a new topic.
When a user interacts with a comment (by voting, replying, labeling, or
bookmarking), any unread notifications they have from that comment will
now be marked as read.
This behavior is on by default, but can be disabled in Settings if the
user would rather mark all notifications read manually.
Unfortunately, right now Chrome only supports "pixelated", and Firefox
only supports "crisp-edges", so this should give the best result across
browsers.
This updates the clean_private_data script to delete more data
associated with users that deleted their accounts at least 30 days ago,
including all of their votes, subscriptions, bookmarks, and
notifications.
Due to hiding the overflow on topic sources to deal with longer domain
names, descenders (in 'y', 'g', 'j', etc.) were being hidden if they
went outside of the box. This increases the line-height and removes the
top margin - the result should be almost exactly the same, but won't cut
off the descenders.
This increases the touch size of the nav links, gives them a bit more
spacing on desktop, and center-aligns the header items vertically, which
looks better when they wrap (but that's still pretty bad overall).
This looked a bit off because the comment always has its own left
border, so the border was 1px thicker next to the button (and sometimes
a different color).
Previously, this was set as "same-origin" which will only send a
referrer to Tildes itself. This changes so that it will continue sending
the full referrer to Tildes, but will send only the domain to external
sites if they use HTTPS (and no referer to HTTP ones).
This can be useful because there are often situations where an article
author sees traffic coming from a site and will come to check it out and
be able to participate in the discussion.
This was a little more strict before and would only skip linkification
if the entire path was digits and/or periods. However, I've seen it
still hitting some people if they write things like "~100k". It's very
unlikely that we're ever going to have a top-level group with a name
starting with a number, so let's just skip linkification for all
instances where a number is the first thing.
Previously, the site's main content would be pushed down a little when
you have new comment notifications or messages. This moves them to the
left of the username instead, and uses a different flexbox arrangement
inside the sidebar when it's collapsed on smaller screens.
Using bootstrap() seems to cause issues with re-declaring the Prometheus
metrics (which happens in the tweens that we don't really need or want
anyway). There might be better ways to do this including not attaching
the tweens for scripts, but this seems to work fine (and was already
being done this way in the YouTube API consumer).
This is a bit ugly (and probably the wrong spot to do it), but we need
to wipe the old metadata before adding in new stuff from the
re-scrape(s). Otherwise, if we do something like change a YouTube video
to an article, the duration of the video will still be left in the
metadata even though it's no longer relevant.
This was only being done when a new topic was created before, which
meant that if a topic's link was edited, the new link wouldn't go
through the transformation process (and we do want it to).
I was seeing some strange behavior from this tween on non-GET requests,
where a huge number of Set-Cookie headers were being added. I'm not sure
exactly what was causing this, but it's not necessary to do on non-GET
requests anyway, so this should be safer.
The site-icons spritesheet has already become unwieldy - it's almost
1MB, is mostly rarely-needed icons, and needs to be fully replaced and
re-downloaded whenever a new icon is added. With HTTP/2 now being widely
supported, spritesheets seem to be mostly obsolete, and I probably never
should have done it that way in the first place.
This commit changes over to simply using individual icon images, and
rebuilds the CSS file whenever new icons are downloaded. This new CSS
file will probably be somewhat large, but should gzip extremely well.
This probably still needs some work to support cache-busting on the CSS
file.
This is mostly motivated by recently enabling the themes from the main
Tildes site on the Docs and Blog. To support users maintaining their
theme between sites, we need to set the domain field on the cookie,
which we weren't doing previously. This tween will automatically convert
"old" cookies to "new" ones, but unfortunately there's no way to
determine whether their cookie has domain set or not, so we just need to
set a new-style cookie every time.
This also will start setting the cookie if they don't already have one,
but have a default theme set on their account. This is necessary to be
able to have the default theme carry over to Docs/Blog.
In the future (maybe in a month or so), we can change this so that it
only does the default-theme function.
We're going to make the static sites depend on the main site's
stylesheet, but as part of that we need to be able to do a little bit of
customization specific to them - specifically, being able to bring over
the rules for setting up how lists and links look. Hopefully we
shouldn't need to use this much, but this is reasonable for now.
This will allow individual groups to always show the username on topics,
overriding the usual behavior of only showing username when it's a text
topic. On Tildes itself, this will be useful for groups like ~creative.
Markdown won't merge subsequent quoted paragraphs into a single
blockquote unless the blank line between them also has a ">" on it. Most
people don't expect this behavior when quoting a multi-paragraph
section, and end up with a bunch of separated blockquotes.
This should fix that issue by default, but still allows people to keep
their blockquotes separated by adding at least one more newline between
the two quoted paragraphs (so they have at least two blank lines), among
various other methods.