Only show changed URLs in the artist history, not unchanged URLs. Makes
the edit history more compact and easier to read by showing only the
things that changed on every edit.
Fix wiki pages like this returning 406 errors:
* https://danbooru.donmai.us/wiki_pages/rnd.jpg
Caused by Rails parsing the .jpg part as a file extension and trying to
return a JPEG in response. This happens deep in Rails' MIME negotiation
code, so it's hard to override. The fix is to pass `format: false` in
the route to disable all special handling of file extensions by Rails,
and then handle it ourselves in the controller. Ugly.
This only affected two tags: `rnd.jpg` and `haru.jpg`.
Remove all unnecessary rules from our CSS reset stylesheet.
Our CSS reset was based on a combination of Eric Meyer's CSS reset [1]
and Nicholas Gallagher's normalize.css [2]. Neither had been updated in
over 10 years, so they both contained a lot of unnecessary cruft. This
included workarounds for bugs in ancient versions of IE, and rules for
elements we don't use. Some of these rules had already been removed
upstream, but we never synchronized our copy of normalize.css with it.
This fixes some minor issues with vertical alignment of form inputs,
caused by unnecessary `vertical-align: middle` rules.
[1]: https://meyerweb.com/eric/tools/css/reset/
[2]: https://nicolasgallagher.com/about-normalize-css/
Adjust heading tags (h1, h2, h3, h4, h5, h6) to use margins instead of
padding so that margins collapse together. Adjust margin sizes to reduce
gaps between headers and paragraphs, and headers and subheaders.
Also adjust paragraph margins so that there's slightly less space
between paragraphs.
Autotag non-web_source on posts that have a non-http:// or https:// URL.
Add a fix script to backfill old posts.
Syntactically invalid URLs are still considered web sources. For
example, `https://google,com` technically isn't a valid URL, but it's
not considered a non-web source.
Fix the paginator not appearing when all posts on the page are hidden,
because of deleted posts, banned artists, censored tags, or non-safe
posts in safe mode. This prevented navigating to the next or previous
page.
Fix BigQuery export jobs failing with:
Google::Cloud::InvalidArgumentError: required: Bucket is requester pays
bucket but no user project provided.
Caused by changing the storage bucket to requester pays. The
`user_project` param must be set to true on requester pays buckets to
bill usage to the current project.
Add a 'Copy Link' action to forum posts and comments. This copies the
full link to the clipboard. The 'Copy ID' action copies just the DText
shortlink (comment #XXX or forum #XXX).
* Add data-is-deleted, data-updater-id, data-topic-id, and
data-topic-is-deleted data attributes to forum posts.
* Change data-forum-post-id to data-id.
* Dim deleted forum posts like comments.
Add a `z=N` param to the URL when doing a search from the search box, or
when clicking a link in the tag list. This is to gather data on how
often the links in the tag list are used, and whether it's more common
to click tags in the tag list or to perform searches manually.
This is temporary and will be removed when enough data is collected.
Don't default the "Final source" field to the current source. This way
the default will be to set the source to the normalized replacement URL.
The "final source" field can be set to override this.
Fix the /posts index controller not logging the normalized search query
to NewRelic when the search failed, either because of a tag limit error,
a search timeout, or a RSS feed rate limit error.
Also don't log the number of search results when it's an API request or
failed search. This is to avoid doing a potentially slow full post count
when it's not otherwise needed.
When a banned artist tag is aliased into a nonbanned artist tag, move
the is_banned flag from the old artist entry to the new artist_entry.
Related to #4940. Fixes a case where a banned artist could lose the
banned status when it was moved.
Disable the ability to click and drag thumbnails. This usually happened
by accident when you clicked on a thumbnail and accidentally dragged it,
especially during tag scripting.
Fix an open redirect exploit where if you went to <https://danbooru.donmai.us/login?url=//fakebooru.com>,
then after you logged in you would be redirected to https://fakebooru.com.
This was actually fixed by the upgrade to Rails 7.0. `redirect_to` now
raises an `UnsafeRedirectError` on redirect to an offsite URL. Before we
tried to prevent offsite redirects by checking that the URL started with
a slash, but this was insufficient - it allowed protocol-relative URLs
like `//fakebooru.com`.
Add a test case for protocol-relative URLs and return a 403 error on an
offsite redirect.
* Update framework files with `bin/rails app:update`.
* Update to use new Rails 7.0 default settings, except for a couple
things regarding new cookie and cache formats that would prevent us
from rolling back to Rails 6.1 if necessary.
Fix the paginator not detecting the first or last page correctly during
sequential pagination.
Caused by the fact that we fetch one more record than needed to detect
whether we're on the last page, then throw that record away by
overriding Rails' internal `records` method. An upstream refactoring
meant that the `size` method now counts the number of records *after*
the extra record is thrown away, where before it counted *before* the
extra record was thrown away.
Previously we overrode `ActiveRecord::Base#columns` to make it ignore
tsvector columns so they wouldn't show up in API responses. These
columns have been removed, so this is no longer necessary.
