70275c9f04
Yank Solid Queue from development
...
It clogs up the test log and occasionally causes the puma to hang when
you want to quit.
2025-04-18 16:30:30 +02:00
1026408736
Need this for the draft header to be visible
2025-04-18 15:22:59 +02:00
f3000a02c5
Follow render_card_replacement pattern
...
And clean up the rest of the implementation a tad.
2025-04-18 15:20:38 +02:00
7395d4e034
Removing the image can also just be done as a card replacement
2025-04-18 15:14:36 +02:00
de23b4a866
Not needed for a single dependency
2025-04-18 15:10:47 +02:00
3533fd1cd4
Merge branch 'main' into notifs-hiding
...
* main:
Fix failing test (we're no longer redirecting here
2025-04-17 17:04:02 -05:00
c61fa77d6b
Fix failing test (we're no longer redirecting here
...
See: #411
2025-04-17 17:03:18 -05:00
e16dc859e3
Display the empty message via CSS when there are no unread notifications
2025-04-17 16:57:24 -05:00
b37a8e3f33
Hide Previously Seen section when empty with CSS
2025-04-17 16:54:11 -05:00
09e544a881
Run beamer accessories in production only
2025-04-17 16:31:09 +01:00
36e9ff47fb
Less noisy logging in production
2025-04-17 16:27:01 +01:00
b8775d41f5
Inline perma to container partial
...
Reveals all the container elements and perma wasnt used elsewhere.
2025-04-17 16:06:28 +02:00
0e442ac0d0
Use card replacement for card updates too
2025-04-17 15:56:56 +02:00
316ec76a1d
Merge pull request #411 from basecamp/card-changes-without-redirect
...
Card changes without redirect
2025-04-17 15:43:31 +02:00
461c104ac2
Extract helper method to keep partial at a uniform level of abstraction
2025-04-17 15:42:07 +02:00
63b2abe020
Remove duplicated conditional
2025-04-17 15:38:47 +02:00
b508a93aba
Style
2025-04-17 15:36:11 +02:00
182d1c934b
Note that we dont want to keep reloading these frames
2025-04-17 15:35:56 +02:00
ac9cbad4f9
I like this even better
2025-04-17 15:32:09 +02:00
62088b66d1
Suffix qualifier is unnnecessary
2025-04-17 15:30:55 +02:00
df1f623435
Group partials for container
2025-04-17 15:25:11 +02:00
b788ef1638
Match domain language
2025-04-17 15:23:40 +02:00
4dbd8205ce
Fix trailing space
2025-04-17 15:21:14 +02:00
7adbd8d634
Use card rerendering instead of redirects
2025-04-17 15:20:27 +02:00
d019e282ae
Better name
2025-04-17 15:00:14 +02:00
770029c26d
Stick with domain language
2025-04-17 14:59:15 +02:00
06da295f44
Fix tests
2025-04-17 14:56:53 +02:00
547e5ea24f
Rerender instead of redirect
2025-04-17 14:54:29 +02:00
ab1835294d
Style
2025-04-17 14:53:45 +02:00
f24287915f
Extract partials that can be rerendered
2025-04-17 14:52:35 +02:00
fcb3b1f2f1
Merge pull request #410 from basecamp/pin-without-redirect
...
Pin without redirect
2025-04-17 14:33:09 +02:00
7b5a1c0232
Fix test
2025-04-17 14:32:26 +02:00
4d0dddec46
No need to loop through a job when this is the sole point of the action
2025-04-17 14:31:28 +02:00
87ce4a623b
Even clearer about where these pins are going
2025-04-17 14:31:02 +02:00
1437bcb3b7
Clearer about where these are being sent
2025-04-17 14:29:12 +02:00
8fe8241069
Reuse create template
2025-04-17 14:14:08 +02:00
bddd624e41
Match template name to container id
2025-04-17 14:14:03 +02:00
fc0cf3917f
Pin without redirect
...
Use turbo stream response to save on the redirect latency
2025-04-17 14:05:26 +02:00
07e801b74c
Fix pinning to tray
...
They need to have a complete dom element that is unique
2025-04-17 13:52:55 +02:00
0c743d27a8
There is no more/less flicker with morphing
...
Where did you see the flickering that this solved, @jorgemanrubia?
2025-04-17 13:42:49 +02:00
b5d6d40462
Use domain language termn
2025-04-17 13:42:23 +02:00
bf99da8914
Style
2025-04-17 13:33:45 +02:00
a5de4863fa
Allow for multiple notifications to be read
...
cc @kevinmcconnell
2025-04-17 11:39:37 +02:00
6fd5168c21
Merge pull request #409 from basecamp/read-zero-or-many-notifs-at-once
...
Read a card should dismiss all notifications
2025-04-17 09:49:24 +01:00
6c8fe164b3
Read a card should dismiss all notifications
...
Previous code assumed that there was a single notification to dismiss
for that card, but in practice there can be zero or many. If it's zero,
we need to avoid the crash. If it's many, we should dismiss them all.
Also re-rendering the notification stack after dismissing enables us to
remove all the removed cards, as well as reveal any older ones that were
previously outside the visible page size.
This reverts commit 6135909b7e
2025-04-17 09:40:59 +01:00
94ccbdb601
Fix close card notch state
2025-04-16 19:06:18 -05:00
91df76928b
Add fizzy-app-102 as a replica
...
The setup here is a bit rough while we settle some of the details, but
the gist of it is:
- Define accessories to run the primary/replica processes
- Add a web-replica role, which is similar to web but has the
replication-side accessory
- Load the beamer extension on the app's primary tenanted database
Note that we aren't sending any traffic into the replica yet, but it
would be functional to do so. This stage is mainly about testing the
database replication.
2025-04-16 17:46:44 +01:00
622d7c396a
Merge branch 'main' into morning-updates
...
* main: (29 commits)
Copy edit
Change alignment
Clearer assignment icon
No longer used
Note that this need to be able to dynamically appear
Style
Slim it down
Trailing spaces
Clarify a bit
More natural to fetch that which will be shown first
Rely on caching of notifications
Card readings just need to remove the notification
Dont render inline
Explain this
Hide implementation detail of reading
Fixed up controller test
Allow for the footer frames to be hidden on the notifications page
Simplify the containers
Only broadcast creations
Dont turbo stream when coming from a read_all redirect
...
2025-04-16 10:40:18 -05:00
b5090c4b21
Copy edit
2025-04-16 10:39:24 -05:00
00dd312cd6
Change alignment
2025-04-16 10:39:24 -05:00
David Heinemeier Hansson