The leading-CSS strip emptied or cut ordinary previews:
"@media team, the brief is attached", "@Page can you check?" and
"@import the CSV ..." became empty, and "Reminder: your appointment
{date: 2026-10-01} is confirmed" became "is confirmed".
At-rule names now match in lower case only, only `@import` and
`@charset` end at a semicolon, an at-rule word with no block after it
is prose, and a selector colon must not be followed by a space.
The conversation view's collapsed cards, contact activity, the pro
folder tab and the list's row-height estimate now use `cleanPreview`
too, so they agree with the list rows. The test's invisible characters
are written as escapes.
An aria-label on a plain span is not reliably read out, so the count
now comes with a visually hidden label and the digits themselves are
hidden from screen readers. Also reuses the sidebar's existing accounts
selector instead of subscribing to the same list twice.
A forward carries the original message's attachments, and the quoted
body's inline images, as part blobIds of the account that holds it
(replyTo.accountId, #1104). When the forward goes out through another
account - a From picked from another account's identities, or the
holding account has none of its own - the draft was saved and sent
through that other account, which cannot resolve those blobs: Email/set
failed with blobNotFound and the forward could not be sent.
Parts of the original are now tagged with the holding account. Before a
draft save or a send, any part owned by another account is downloaded
through that account's client and uploaded through the composing one,
once; later saves reuse the copy. Blob/copy cannot do it, as every login
is its own JMAP session. Quoted inline images are loaded through the
holding account's client as well. Forwards that go out from the holding
account keep referencing the original blobs exactly as before.
After a load the logins other than the one on screen reconnect in the
background, and until they are all back the unified counts cover only
some of them. With many logins that lasts seconds, and a partial total
looks exactly like the whole one.
While that restore runs, the "All accounts" header shows how many logins
are already in the scope, e.g. "4/10", with a translated label for screen
readers. It goes away when the scope is complete or the restore is over -
a login that failed to reconnect does not keep it up.
`restoringAccounts` in the auth store marks the background restore; the
count itself is a pure function with its own tests.
The values of a multi-value condition were glued together as one
unbreakable run - the space around each "or" was a margin, not a
character the line could break at - so a rule with several addresses ran
out of its card. Each value is now an item of its own: they wrap between
each other under the first value, with "or" leading the next line and
still inside the condition's chip, apart from the "and"/"or" between
conditions. A single value too long for any line, in a condition or an
action, breaks inside itself instead of overflowing.
The list's scroll container outlives a switch of folder, so the next
folder opened wherever the last one was scrolled to - often halfway down,
or past its end. Opening another folder, tag, account, unified view or
plugin tab, or running another search, now starts the list at the top.
Refreshes, new mail and loading more keep the view and the position.
Some servers build a message's preview from its HTML body without skipping
the <style> element, so a marketing mail's preview - in the list and in its
push notification - can read "@media screen and (min-width:600px){.hide{
display:none!important;…" and never reach the text.
Drop whole leading rules from the preview before showing it; a preview that
ends inside the style sheet was nothing but CSS and becomes empty, so the
usual "no preview" text shows instead. A rule only counts as CSS when its
block holds declarations or further rules, so prose with a brace in it
("the meeting is at {time} tomorrow") is left alone.
A notification link now waits up to 20 s for its login to reconnect.
If the user opened another folder or message, switched account or left
Mail in the meantime, the late link used to switch account anyway and
take them away from it. It now stays put.
When the account switch was overtaken by another one or failed, a
folder link was resolved against whichever folders were on screen, so
an Inbox link could open the wrong account's Inbox. It now stops there.
Clicking a folder's unread count turns on the unread filter and runs
the search. Since the search scope stopped following the open folder
(#788) its default is "all folders", and since #1082 that fans out over
every shared account too, so the Inbox badge listed unread mail from
every folder of every account the login reaches.
The badge now scopes the search to its folder. That scope moves with
the open folder, the way the search itself carries across folders
(#553); picking a scope in the dropdown or clearing the filters stops
it. Clicking the badge from a tag view or the Scheduled list now also
leaves that view, as clicking the folder does.
The unified scope is rebuilt on every connection while a browser restores
its logins, on every background push, and on every unified browse,
load-more and search - and each build asked every login for its folders
again, one login after the other. With ten logins that was ten full
Mailbox/get per build and well over a hundred in the first seconds, and the
slowest build waited on the sum of every login's round trip.
Fetch the logins in parallel, and fetch each login's list once: a build
shares a fetch that is already in flight and reuses a finished one until
that login reports a change - the push handlers invalidate the login they
hear from, so a background push now costs one Mailbox/get instead of one
per login. A 60 s limit backstops a change whose push was lost.
A reused list is never written back into `accountMailboxes`: optimistic
updates land there directly, and publishing an older list over them would
snap a just-changed counter back.
The push effect re-runs every time a login connects, which while a browser
with many logins restores means once per account. Each run closed every
client's push and set it up again: the live streams were torn down and
reopened and every slow poll re-primed its baseline with a full state read
(Mailbox, Email, Calendar, CalendarEvent, SieveScript) - quadratic in the
number of accounts. With ten logins that was 21 stream openings and about
180 state-poll methods in the first seconds.
Reconcile instead: bind the clients that are new, unbind the ones that are
gone, and on an account switch re-bind only the two logins whose role
changed, handing the live stream to the newly active one when it has
none. Handlers are read through a ref at event time, so a binding never has
to be re-created because a callback changed identity.
The reconciliation is a pure function (lib/push-bindings.ts) with its own
tests; the effect only supplies the callbacks.
Once a link had waited for its login and switched to it, the folder was
looked up at once, in a store that still held the previous mailbox's
folders or none at all: an inbox link opened on a mailbox that was just
switched to reported "this folder is no longer available". Wait for the
new mailbox's folder list to arrive before resolving the link against it.
A notification tapped with the app closed reopens it on a link that names
its mailbox (`?slot=`). The browser reconnects its logins one at a time,
so that mailbox is often not connected yet when the link is read: with ten
logins, one on the Gmail bridge came back at 3.3 s and the link was judged
at 1.8 s. It was declared unavailable - "sign in to that account to open
this link" - and the app stayed on whichever mailbox was active.
Wait for the login instead, and give up only when it is not one of this
browser's logins, its restore fails, or 20 s go by.
The preview hands the attachment's blob to the browser's own viewer,
which picks the viewer from the blob's type and downloads anything it
cannot show - under the blob's UUID. It only corrected the exact
application/octet-stream declaration from the file name. Other
declarations kept the preview from working: IBM i (AS/400) systems send
every attachment as application/x-as400attachment, some senders use
application/x-pdf or force-download, and the octet-stream check was
case-sensitive.
Move the decision into previewBlobType(): a PDF, image, audio or video
preview takes the first type its viewer can show - declared, then
served, then the one the file name implies. Other previews keep the
declared type unless it is generic, and the name never yields a
script-capable type.
The expand/collapse toggle referenced calendar.events.show_more and
show_less, which were missing from every locale (translations test failed),
and passed a defaultMessage option next-intl does not support. Add both keys
to all 27 locales, reusing each locale's existing email_composer.show_less
wording, and drop the unsupported option.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The all-day/tasks area in the week view took its height from
(allDayRowCount + taskRowCount) across every loaded day. Since free
scrolling loads 37 to 180 days, one crowded day anywhere in that range
blew the header up to 250+ px for every week and pushed the time grid down;
tasks were also stacked below the global all-day row count, leaving a gap
on days without all-day events.
- Stack each day's tasks directly below that day's all-day events.
- Size the area from the days in view only: track the first visible day by
date (so columns prepended while loading do not shift it) plus the number
of visible columns (7 on desktop, by width on mobile).
- Cap the area at 3 rows with an expand/collapse toggle when a visible day
has more entries.
Tests cover per-day task stacking, the 3-row cap and toggle, an empty week
next to a crowded one, and the area growing when scrolling into it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Link Mobile App step-up marked the tab before asking the server to
start the provider round trip. When that request failed, or the user
backed out at the provider, the mark stayed, and the next sign-in's
callback in that tab went to the pairing route and failed. The mark is
now set only once the round trip starts, holds that round trip's state,
and the callback drops one that does not match.
Linus Rath
lucamzanon
dealerweb
Linus Rath
Sebastian SanitzandClaude Opus 5.5