We've seen in our error logs browser clients that were constantly
retrying requests to our server after the user logged out in one tab
but didn't close another.
(imported from commit 68dd8d9e618083bc116ae3a96dbcc78fa9301bba)
When we rebuild the user list from scratch, set the unread
counts in the templates to avoid multiple DOM updates.
(imported from commit 2d0c9b0fb99b382332e464ba7c3caad95e05363e)
Features:
* Only shows messages in the narrow
* New messages in the narrow will arrive as they are sent
* Works even for streams you're not subscribed to
* Automatically subscribes you to a stream on send
* Doesn't update your pointer
* All searches etc. automatically have the narrow added
(imported from commit 2e12b76849f6ca0f53dda5985dad477a04f7bbac)
I'd check for bookends, but not daterows. Now, we just check if we can get
a message id out of it, rather than excluding specific types of rows.
(imported from commit 39ebc35e81dcec7fc83b603bf941f816fcd3d38d)
basically this tries to turn scroll-the-world into not-scroll-the-world
This is not very good--maybe Allen has a better idea. The best solution would be to
turn off scroll-the-world. Look for it after the tables->divs change happens.
(imported from commit ae0b6976bca57986f95022f2470bc7117eda7fa3)
Previously, we would slowly accumulate duplicate copies of events that
happened while the user was in the tutorial at a rate of 1 copy per 5
seconds.
(imported from commit 3e3c58aca4b1ba3bfdd1c93f47330a0f4cf4b60f)
We now use window.innerWidth to check for CUSTOMER7's skinny
mode, which empirically seems to be more consistent with
CSS's max-width @media settings.
I tested under FF, Safari, and Chrome.
(imported from commit d440998634633c11b471fe732104be252c979cd4)
It's possible for a message to be considered "long" at one point,
and then if you narrow to it later, it should be considered "short",
because either the screen's wider or taller. This commit makes
sure that we remove the "condensed" flag from short messages,
and it also fixes the "More..." click handler's assumption that
could-be-condensed will always be true for condensed messages.
(imported from commit 77e4a1ad299c09f12e5609a972d5668472bd4a81)
Cache the height of messages to speed up ui.condense_and_collapse()
to make narrows work more quickly. The height of the message
determines whether it is auto-condensed or not. We clear specific
cache entries when messages get edited, and we reset the entire
cache when the window width changes.
(imported from commit 7c12070a3eb3e2e1a2dfeb8d9109f3404a46c032)
Trying to condense messages when they are not yet visible just
leads to wasted effort and wrong results.
This commit makes it so that current_msg_list always points to a
visible list, so the code related to message rendering knows when
to call ui.condense_and_collapse(). For activating narrows, we
now let rendering handle the condense/collapse case.
The home view situation is a little trickier, because we render
new messages in the home view even when we're inside inside of a narrow,
presumably to make it fast to switch into the home view. When
we actually go back to the home view, we need to sweep for messages
that might need to be condensed, since they have been originally
rendered while the home view was not visible.
(imported from commit 4da2d278a4353e9e0c2b98cbf8c9dd03b06cd59b)
Before this change, you could open the Administration page
for a 2nd time, and there would be two copies of each stream.
The simple fix is to remove any existing rows before populating
the table.
(imported from commit 957306d9c6418e59b5b288dad08864547ca63e53)
This is the "Tried to call a Dict method with an undefined key" error
because it tried to look up stream information for `undefined`.
(imported from commit 0187f185f3e424a0c9ea940d9b32f07376ac8952)
There are now 2 cases for narrowing:
1. We narrowed, but only backwards in time (ie no unread were
read). In this case, try to go back to exactly where we were before
narrowing. This behavior is unchanged.
2. We read some unread messages in a narrow. Instead of going back to
where we were before the narrow, go to our first unread message (or
the bottom of the feed, if there are no unread messages). This is new.
This means that after catching up through the sidebar, on returning
home you'll be at the bottom of your feed.
Searching for the first unread message in a message list with 40,000
messages only takes 17ms according to:
function timeit() {
var t0 = new Date().getTime();
_.find(current_msg_list.all(), unread.message_unread);
var t1 = new Date().getTime();
console.log('Find first unread: ' + (t1 - t0) + ' ms');
}
(imported from commit 87c467578a2cced0aa976d8ae2924371b85d2445)
This changes the algorithm slightly for the 2-block case, because
I simplified the logic to just divvy up the space naively based
on the relative size of the blocks.
(imported from commit 9498edd916f65e07fb64d138276691d0d5cc0e55)
This is a functional change. Before this change, the stream
list and user list were allowed to take equal space in the left
sidebar; now, we take the size of each list in deciding which
proportion each block gets.
(imported from commit febedcb0518353825e18a6ebe60d1883b98bc78d)
I don't think share-the-love is turned on for CUSTOMER7 in the
first place, but even if it is, we should hide it when they go
into narrow mode, to make room for streams and users.
(imported from commit 2e80eec0f2ddee06753f48248dca5ac4745db6f1)
Calling $('#foo') is expensive, so if you need it twice, you
should store it in a variable. I did this for stream_filters
and user_presences, and added the expectOne() call for them as
well.
(imported from commit 69e689e28b1248a93ef426a89c14033d2fb36104)
I'd also like to add a database table to actually store the values
that we get out of this and our send message requests for future
inspection, but for now, grepping logs+statsd is good enough.
(imported from commit 99ef179651850217fe6e82c5e928d122ca91101e)
I renamed ui.process_condensing() to ui.condense_and_collapse(),
and, more importantly, it now takes a list of elements, not a single
element, which allows us to do some computations outside the loop.
(imported from commit d5984088030c2a0d4ec8b258c7fcec3e84caf2b1)
This sets us up for the next commit, where we will change
process_condensing to loop over several elements, and we will
not need to recompute height_cutoff every time.
(imported from commit 1cc5b44598b85d1e301bc84492e4dc38f41ec16e)
Without an explicit close, we will continue to get messages on the old
connection, even once we replace the _sockjs property. This leads to
"impossible" situations, such as receiving ACKs after reconnecting.
(imported from commit f8927fdf0551610e5b9d1db29bba00e160e76b6d)
Now that this is no longer a known problem with our product, we want
to hear about it when it happens.
I worry a bit that a 2s fuse may be too aggressive for the case of
customers in Europe, but it might be OK.
(imported from commit d1bd6b85cd8dffab9c0d0fd410de5331736b00af)
We convert sender:me to sender:steve@zulip.com at parsing time,
so users will see the canonicalization in the search bar. Likewise
for pm-with.
(imported from commit aa9951f13d4633cfef85f03e5486d607fdef414f)
R means "I want to send a PM, you can guess the destination"
r means "I want send a stream message, you can guess the destination"
C means "I want to send a PM and specify the destination"
c means "I want to send a stream message and specify the destination"
(imported from commit 4c93cc3029892c21accadd9624da70ee818dec68)
I think we've got plenty of data of user agent strings for when this
warning triggers at this point.
(imported from commit 8bf6601a325b8e35f5127914a447bd522439c531)
If we load a browser window in a narrowed view and then un-narrow
before the home message list has loaded, we end up attempting to
select message ID -1 from home_msg_list even though it is empty,
triggering a traceback.
(imported from commit eb8b686f6e9c1fa518028e5755ac6196781e92d7)
This doesn't actually prevent a user from making the API call into our
servers to actually go and edit a message, so this isn't a bulletproof
solution for realms where messages ABSOLUTELY MUST NOT be edited.
(imported from commit 5bf043a201e2952189b45f93b8c5ca7648f6aee7)
I love press-enter-to-send but find this behavior confusing when
I'm in an edit box -- if we're going to respect it, we should
do so obviously (with a checkbox visible).
This reverts commit 6e3fc6495b7012aa12728a78b8bdd95701bb21e1.
(imported from commit d1ae16110f5504e879e315037c85c211ba3bca9a)
Turn on topic zooming for three realms with users who
requested to see more than five topics (tracked on #1248).
(imported from commit 56c73926a9f3c1006451a3e4a01b7661807908b5)
We need to resize the textarea when it is changed via .val(). By
clearing and resizing the compose box when it is closed, we can
avoid calling autosize_textarea() when the user opens the compose box.
This saves at least 15ms on every compose and might also be a cause
of longer delays.
(imported from commit fe6e092efcd1c4b95a868ee66653448f99af84c0)
R means "I want to send a PM, you can guess the destination"
r means "I want send a stream message, you can guess the destination"
C means "I want to send a PM and specify the destination"
c means "I want to send a stream message and specify the destination"
(imported from commit 755c92aed79ab79089b2e35d2c100582f012736a)
We encountered a strange bug in Chrome on OSX where the initial call
to set scrollTop when the browser window loads did not do anything,
until we first scrolled to some other location on the page. This
patch "fixes" that by automatically scrolling somewhere else before
scrolling to the target place on the page.
(imported from commit 96be6a0016e9e5fd78380613c9587f38406604df)
In the first cut at topic zoom, I was re-rendering the
streams list, but this created glitches with orphaned
list items. The reproducible bug was that unread counts
on unshown streams weren't updating.
In the new approach, I keep the elements more permanent, and
I just hide and show them as needed, either through jQuery
show/hide or permanent CSS selectors.
I got rid of toggle_zoom(), so that we just explicitly zoom
in and zoom out in all situations. In particular, when we
narrow, it's more clear now that only stay zoomed in when
we're narrowing to the same stream as before (including topic
narrows within that stream).
When you zoom in, the number of topics is no longer limited
to 30, since that was kind of arbitrary anyway. (In practice,
the number of topics is usually well under 30, anyway, due to
the way we track them on the client.)
(imported from commit 5b6c143dee9ba9fe557d8cc36335ff28efb4b0de)
This link lets you zoom in to more topics. We only show it if
there are topics that we had to hide to respect the max-5 limit
along with other rules of when you show topics.
This is feature flagged to staging only.
(imported from commit 9915004ec2eb3df7416fe45c0e60cebcd7fecfea)
When your left sidebar is zoomed to show just one stream,
there is a link to to show all streams again.
(imported from commit 92f39b042168c443cbb9f524bf892557ef492551)
If you double-click on a stream that you've narrowed to, it
will either zoom in or zoom out the left sidebar view. Zooming
in shows just that stream; zooming out shows all streams.
This is feature flagged to staging only.
(imported from commit 6fdb3cacd68635f313f2a8a81edf2d6101cce2cb)
This commit doesn't actually add the final UI to zoom/unzoom
topics, because I want to keep those in separate commits, in
case we change how to enable the feature. But this commit
adds a toggle_zoom() function that zooms/unzooms topics.
Zooming is minimally invasive, because we don't really introduce
many extra elements to the UI; instead, we just make the list of
streams be a list of length one (i.e. the active stream). This
gives us a lot of stuff for free, basically, like unread counts, etc.
(imported from commit 814c1361b6210d1591b4174bed1d6e0c98a3f255)
This is a subtle change, to make sure that list items
from stream_filters always go through click handlers, which
ensures, e.g., consistent behavior on opening the compose box.
(imported from commit cd734901def8959c8c8a29c164c7a60730ae0915)
Search suggestions were limiting you to the last 10 topics
in a stream, not the last 10 *matching* topics in a stream.
This was just a bug, where the slice() call was supposed to
come after the filter() call.
(imported from commit 7ad8f8409315d8ca389aa5bb085275a60f6e08c0)
This should make it possible to either open these using middle-click
or copy the links for e.g. putting in a bug tracker ticket.
(imported from commit 0c531453cdd7197f932079c245700948b416a3d5)
`$(message.content)` breaks on /me messages because they are not
wrapped in `<p>` so the message content is interpreted as a selector.
The message text is no longer used, so this line can simply be removed.
(imported from commit ee8d48c1f5fc489cc577cc466f629891ea65d55f)
This seems to only work in Chrome and Safari.
Firefox (at least my version) simply doesn't fire an onclick
event, and our desktop app has its own native code that decides
what to do when a notification is clicked.
(imported from commit 30bacec4726b9e6c022dd2c74f83d37747260dba)
Safari doesn't let you (shift-)tab to buttons, so we have to support
this ourselves in several cases.
(imported from commit 1b8ae28d6950fd9686b442948d71ab300c7328dd)
This is a behavior that basically only administrators can trigger
today, with the exception of the fact that anyone can edit a no-topic
message.
(imported from commit d50eded79ddf3438d87e3dc6a8641fbfb034d50c)
This makes the notification slightly more conservative as we do our
initial roll out. In particular, it avoids cases like being notified
when you are almost at the bottom of your feed for a message in your
narrow.
(imported from commit 9c834b1c344d8c429de92fb3512f32494fc02379)
When you upload a 2nd avatar to Zulip, the URL doesn't actually
change, so even new messages can show the old avatar, if your
browser is caching. We work against the cache by having the
"stamp" argument, which we vary at reload time and also when
we upload the new avatar. The browser still benefits from
cached images as new messages come in.
(imported from commit 84869c8d7f251c9f2498026a5e9e3b2451784879)
UserProfile.show_admin was intended to be a check for users that have
administrative rights in other realms, which we've harmlessly but
erroneously been using to check if they are an admin in their realm.
Use the more straightforward check instead, with a more intuitive
name.
(imported from commit d81050c7dbbb19e59c5e31750be303a4630e1456)
There will be browser errors on staging when this is deployed due to the socket
protocol changing.
(imported from commit f1eda5b5c2ec9c60c23b3ca96277a61debadf5bb)
If you don't call ui.resize_page_components(), then the user
list will have zero height. Now we call it.
(imported from commit ca2e295319550509360e4d2278489f9a328335ff)
This moves the notify-not-in-view notifications into the composebox area.
It also tries to be a bit smarter about what action it links and what it displays.
(imported from commit 1c79bd0d9ef972059a006b17501a09b72e961ee3)
The inactive flag wasn't really supported, and the
activation_toggle_button class was misleading, since
we don't yet toggle in both directions.
(imported from commit 0c8511021dd580c86df4e80092a6dd49e32773f9)
We had people_list-related code left over from before we
started using realm_people_dict.
(imported from commit f77c441269f07db2a8998c2184de81f9d0c053d6)
Aborting a xhr actually calls the error callback, which was
restarting the get_events loop, and we were also restarting it
manually in restart_get_updates.
(imported from commit dcdb4a27539ce35ecd83c2ade5b4edca3fb848e1)
If you used the mousewheel in certain page sub-elements, scrolling
to the bottom would sometimes turn into scrolling the main page.
We had code to make that not happen, but it was sensitive to
integer rounding, so I added a one pixel fudge factor.
(imported from commit 5ed1f4e717e5f9ce3316383812dee27913afdd49)
When GROUP PMs are in the right sidebar, and when there are
lots of users, we set max heights so that both USERS and
GROUP PMs show up.
(imported from commit 4a5a16a8cf03d853769b3a690ac5b0338a2c8461)
When a user squishes the browser window enough to move users
to the left sidebar, we need to move user groups as well. More
importantly, when they widen again, we move both of them back
in a way that users stay on top of groups.
(imported from commit d80722b58190d42a2dfc5247bfffaa7708cde21a)
Leo and I were looking over this code and I couldn't remember what it
was doing. So after I re-derived its purpose, we figured that an
explanatory comment was in order.
(imported from commit ce984362e2b504b40f3d3586777bd73691ad5ea7)
If a user types "/me runs to the store", we put "runs to the store"
in bold after their name.
(imported from commit fbc11e99244e1c8fa1c03e4753e706957fcd449e)
The function show_actions_popover() actually toggles the
message action menu open and closed. I renamed the method
to toggle_actions_popover(), and I check to make sure the
menu is open before trying to focus its first item.
(imported from commit b2c32b6c4e0be6066cd1d41463457b7e991df0ec)
Every time we re-render the huddles section, we need to
update the unread count. We do this is in very similar
fashion to how we update individual users.
(imported from commit 2419365bc602ddaebc609090e119c0dcfad35bb7)
This fixes a mild regression in 6bd2a0315ff56a20027074d65ccaa094bd35e63f,
which fixed a bigger bug. That change added the event parameter to
add_alert_word(); this change updates one of the callers to now
supply it.
(imported from commit c6375abb8bd22d12c538cd7439462aae4665ee38)
Show up to 10 of your recent group PM conversations in the right
sidebar. Clicking on the links narrows to the huddle and opens the
compose box for the huddle. The green circles have opacity
proportional to the number of users present in the huddle.
This is feature flagged to staging only.
Some of this code was written by Allen before commits were squashed.
Known issue: unread counts disappear on certain refresh events.
(imported from commit 3b44665150ba20594d8b0295cb30df03601c1d52)
Add a method that lets us know what percentage of a huddle is
present. (We can use this later to set the opacity of huddles
in the UI.)
(imported from commit 8a2383951807d7bfbf9d730a8980d977cf23b379)
This should substantially mitigate the get_updates failures that we've
been seeing, since users will experience a 2s-slow send, not a 0-60
second slow send.
We should revert this once we resolve the root problem.
(imported from commit b665d0dfe674e1113bdd62cef50e3d9c52758e4c)
The global in question was `event`, set to:
MouseEvent {dataTransfer: null,
toElement: button.btn btn-mini add-alert-word,
fromElement: null, y: 442, x: 763…}
in Chrome. This does seem to be the correct event.
In Chrome, this global variable existed, so the code path didn't error
out. In Firefox, the global variable didn't exist, causing adding
alert words to fail.
(imported from commit 6bd2a0315ff56a20027074d65ccaa094bd35e63f)
This will hopefully help with the send dialog being stuck on
"sending" as well as allowing us to not show errors to the user on
reconnect.
(imported from commit 31ee889853f348e486863073dc130cdfb4e1338d)
Clients can only have one connection at a time, anyway, so we can
just keep track of a client id, instead. This makes reconnections
easier.
It's a little funny to use queue ids for the client id, but we know
they should exist by the time the client is connecting and they are
guaranteed to already be unique and authenticatable. We will also
eventually be integrating the event system and the socket code closer
anyway.
(imported from commit 1f60e06fb16d31d6c121deafd493fb304d19a6c2)
This is unnecessary without summarization and has a performance
impact on scrolling.
This reverts commit 6270a33ff0fe98fc901b687ca533ef0ea127d052.
(imported from commit ac328a9fe344adfaf1cc1822cd58bc9178ae1a33)
This reverts commit c10d9c1a0d23891acce88bf8d79866c08cb75681.
This reverts commit 9259a246946cd968a8725c38ff5ef2d4b4793717.
(imported from commit 50e9e0136c2487cc63d75ae2b78df0c70a1b0be1)
This is the amount of time between when it is sent, and when it is
rendered into the user's home view.
(imported from commit 468c28e77ba16c7256c359e90ab5aacf9d497585)
Activity.js now has the capability to track huddles that
come through in loaded messages and return them in reverse
chronological order by their most recent message. Right
now this only connected to a unit test, not any production
code.
(imported from commit 59957086fa2e454e5711472df091f178217aed2b)
This can be squashed with the prior commit, which inlined this
MIT check into update_users() while trying to avoid other
complications. After inlining the code, it's clear that we
don't need to call the sort_users() line of code for MIT users,
so I moved the MIT guard condition up.
(imported from commit fa5b52e14964ad595b34d40ce6c8450ea93726c5)
Since ui.set_presence_list is only called from activity.js, I am
inlining the code into activity.update_users(). This also allows
us to move ui.presence_descriptions into activity.js, which
is the right home for presence-related things.
(imported from commit 0ff239275c544a86c14d517bc386d06726b81cd9)
The user_info var was mapping users to presence information, and
presence_info is a better name for it. This change is partly tactical,
as it sets us up to inline ui.set_presence_list, which receives
user_list as presence_list and then has a completely different
variable called user_info.
By doing this pre-factoring, the next commit becomes just a pure
code move without more moving parts of renaming variables.
(imported from commit 4b015d19886b43d24905124eb37cd9dd317aa87b)
Narrow.activate was called without an options object in one
obscure location. I'm not sure how it's actually triggered,
since all searches seem to go through the typeahead path with
the "Search for x" option, but someone managed to do it.
(imported from commit 5e0e1bbe7570e5f6a654949547dc164e01125efa)
It's somewhat buggy, and has thus been annoying our internal users, so
better to disable it until we can fix the bugs.
(imported from commit f981791d32d321b0cfe06b4a337e26ab48832bb3)
This should help with determining the prevalence of slow sends as
experienced by users.
(imported from commit f00797679315c928af3c87ad8fdf0112f1dfa900)
Perhaps we should be checking based on an ID rather than the actual
text, but this fixes the proximal issue.
(imported from commit b125415c3a8356255b64c8e22733532491ea0065)
I switched narrow.by_subject and narrow.by_recipient to use the all_msg_list
instead of current_msg_list, since we wanted to be able to narrow to messages
specifically not in the current_msg_list. However, in searches which revealed
old messages outside the range of all_msg_list (which only has a single contiguous range),
this broke narrowing.
Let's use msg_metadata_cache instead.
(imported from commit 427f717484b4ae83d9bb4cc6e51ce17177d037fe)
Displays a notification above the composebox in cases where the message
is not visible (further down), or where it's outside the current narrow or search.
It also offers a link to the appropriate narrow when it makes sense, and offers
timetravel when appropriate. There is currently a bug with timetravel (and you can
see this when using the popover menu) that makes "narrow to messages around this time"
not work for muted messages.
This resolves Trac #1518.
(imported from commit 391ca0b9c07d91496f6585a4fd8e15723d1170e2)
Looking at the historical data, fewer than 50% of active users have
completed the checklist, which means that it is just persistent
clutter. We also have other better ways of encouraging people to send
traffic and get the apps now.
This commit removes both the frontend UI and backend work but leaves
the db row for now for the historical data.
(imported from commit e8f5780be37bbc75f794fb118e4dd41d8811f2bf)
Nobody uses it, and it causes confusion. (How is it different from 'Home'?)
For now, leave in the ability to get to the narrow, although we may
ultimately want to do away with it as well.
(imported from commit 35b3f27e39c4de3391bc5571b32f7242a29f4cfa)
We were using Gravatar for user avatars, but now users can
upload their avatars directly to Zulip, and we will store
their avatar for them. This removes the old Gravatar-related
interface and polling code.
This commit does not attempt to update the avatars in
messages that have already been loaded, either for the user
making the change or other users.
(imported from commit 301dc48f96f83de0136c93de57055638c79e0961)
The "Your Account" and "Notifications" boxes on the Settings
page each had their own border and their own "Save changes"
button, but they were within the same form and sending to the
same back end point.
This commit creates a separate form and endpoint for each
of the two boxes.
(imported from commit 04d4d16938f20749a18d2c6887da3ed3cf21ef74)
Previously, if the user held down the enter button while the socket
was disconnected, the client would try to connect in a very tight
loop. Now we throttle reconnection attempts to 1 per second.
(imported from commit 7b18260b992d5a34f3ea7925cf72b383f84bbabd)
If the user has a flaky connection, we might be in the backoff state
of attempting a reconnect. But when the user regains connectivity
and tries to send, we want to send the message as soon as possible.
(imported from commit 3c5c8e9c3104ff7923258f73c9ab700548518d16)
All local_server JS stuff should ideally be controlled from
feature-flags so it's easy to see at a glance what's been done.
(imported from commit 45b1cdae382679d3fa3b5f02f67e8ab749a89a51)
We leave the HTML in because it's harmless. (We could do a more
thorough expunge, but I don't see the point).
(imported from commit d3d68b0b2df96a9f3de73903b043c18bc6f77563)
The Streams page should only show active public realms, even though
a user might have info about a "retired" stream in their browser.
I regressed this in 69b83d769 for "retired" streams. A retired
stream is a stream that no longer has subscribers. The bug
scenario here was that you could create a stream, regret it,
unsubscribe yourself, and then the stream never went away from
the Streams page.
This diff tries to be a little more explicit about building the
list of streams for the Streams page. Basically you have two
sources:
* Get only the subscribed streams from the internal
data structures.
* Get the unsubscribed streams by calling the back end
for all public streams, and subtract out the subscribed
streams.
I tested the following scenarios:
normal stream with me: in Streams
normal stream without me: in Streams
my invite-only: in Streams
their invite-only with me: in Streams
their invite-only without me: not in Streams
retired stream: not in Streams (but message colors are good)
See the email "custom query to get public stream names" for some
related discussion.
(imported from commit bc9224e68797b26b795399941117faa9d6858b39)
I want to make subscribed_streams() external, but it conflicts with
a legacy name in the same module (stream_data.js), so I have to rename
it in the same commit. The new name conforms better to the current
naming convention, which generally has functions returning objects
use "sub" in the name.
(imported from commit 9f1ed60772c649359a413257e0998857eab3603f)
Trac #1734
This is implemented by bouncing uploaded file links through a view
that checks authentication and redirects to an expiring S3 URL.
This makes file uploads return a domain-relative URI. The client converts
this to an absolute URI when it's in the composebox, then back to relative
when it's submitted to the server.
We need the relative URI because the same message may be viewed across
{staging,www,zephyr}.zulip.com, which have different cookies.
(imported from commit 33acb2abaa3002325f389d5198fb20ee1b30f5fa)
When we add starring of historical messages, they'll get added to the
home view (since we don't filter them out), which isn't a big deal --
but we don't want to make an inaccurate claim that the user subscribed
to this stream and then unsubscribed.
This should most properly be handled by filtering server-side, but at
the moment our lack of an index on UserMessage.flags makes that
impractical from a performance perspective.
(imported from commit 00751a5f9fc20e9da5c09914c11d1579f9e7398e)
There seems to be some sort of bug involving PhantomJS and XHR
streaming messages. When successive pages are loaded that use XHR
streaming, PhantomJS seems to think the second one never finishes
loading and therefore hangs.
(imported from commit db93b4cab816f1fdc3f3f543c9394b1cba8abedb)
Because our authentication system reads cookies from the initial
connection attempt, several SockJS transports can't be used.
(imported from commit 34b9571225d39072985b8223fb12c43c7235841f)
We want to deprecate reload_subscriptions(), which was kind
of a big hammer to use when only a single stream is being
renamed. Now we call stream_list.rename_stream() to update
the sidebar.
(imported from commit a77d09c0433d9b605b7baa7d7c61183bc8c37ba9)
For a large domain like HS, we were pulling back about 100k of
text with subscriber emails when we opened the Streams page.
This was unnecessary, as the subscribers aren't shown until
you expand the stream, and there's already an AJAX call.
(imported from commit 69b83d769030d87318acefc364ac6ff3a2ec3605)
Use the new count_full_messages_between instead of subtraction in
message_list_view.append. By finding a count higher than it should be
when summarized messages are present, it didn't add new messages until
the pointer moved under certain conditions.
(imported from commit c10d9c1a0d23891acce88bf8d79866c08cb75681)
Summary blocks can contain hundreds of messages. When the rendering window
code didn't take this into account, it would lead to all kinds of
unpleasant behavior when you scroll.
Trac #1888
Unfortunately, this replaces a subtraction with a function that iterates
through all the messages.
(imported from commit 9259a246946cd968a8725c38ff5ef2d4b4793717)
* Disable for search-like narrows (whitelist stream and home instead of
blacklisting topics and PMs)
* Use home view summarization flag for All Messages
(imported from commit 48bd10ae5da7c7564c2efe86a40078f1a7e96e20)
Don't warn when @-mentioning a bot on a public stream that it does
not appear to be subscribed to. It may be receiving those messages
anyway.
(imported from commit 4a00694942a721897a01736f48033c71048e0b16)
This doesn't address the more complicated case of someone @-mentioning
you on a muted topic, which consensus is you do want to get
information for, but we need to develop some infrastructure to present
that case to users clearly.
(imported from commit a4bc1e89c108fa8ba6eccc0a198eabf2231326ab)
Add the option "Narrow to just this message" to the chevron
menu. This has two use cases:
* It's an easy way to get a sharable URL for the message.
* It reduces distractions.
For now it is feature flagged to just customer12.invalid and staging.
See #1880.
(imported from commit 897d247176f9024ff825ccd3b338236569eed5ab)
Allow users to open Zulip windows in new tabs with command-click
from the left sidebar narrowing links and recipient bar
narrowing links.
(imported from commit d60c038c7bf1efccd461f5284d513b9cbfbdaebf)
If the user has text in the compose box, don't close or
change the compose box when they narrow.
(imported from commit f9b400f6bac37cb313f1fd87aadb3ba1d3a035ef)
For the two cases where narrowing should open the compose box,
we now put that logic inside of narrow.js.
(imported from commit 570e22e90c2f6d422ba71cce400c075f0b8adf51)
Handle closing the compose box inside of narrow.js, to
ensure consistent behavior for all the narrowing UI options.
(imported from commit f17a687491eb2361c73032cd974cedb2a0a2dd85)
The main user-facing feature here is that users can open narrows
in new tabs or windows. Internally, it makes the HTML more semantic.
One consequence of making these elements into actual anchor tags
is that clicking on them no longer triggers this logic to
close the compose box when you click outside of it:
// Unfocus our compose area if we click out of it. Don't let exits out
// of modals or selecting text (for copy+paste) trigger cancelling.
if (compose.composing() && !$(e.target).is("a") &&
($(e.target).closest(".modal").length === 0) &&
window.getSelection().toString() === "") {
compose.cancel();
}
Instead of patching the above code, I elected to just call
compose.cancel() explicitly in the click handlers for the links
themselves.
We are gonna try to clean up the compose-box behavior globally soon.
(imported from commit c9a01916f1714fe3dd495d25c78cd5e5532105ef)
Trac #1737
Firefox has the slightly broken behavior that it emits a click event on
the document when you right click, but not on any of the child nodes, so
our #compose stopPropagation doesn't stop right-click events from closing
compose. Chrome never emits click events on right click.
(imported from commit 2519c542715c93877b6d38e5dcff1f3e011688e4)
When decoding an operand, a + can be converted to a space
only if the operand is not an email address.
(imported from commit 08fc36a579bbe6409137c60c0fa9579fe3ab2c43)
It's a bit of a contrived use case, but you could make a topic
called "12345" and then mute it from the sidebar, and it would
crash inside Dict. We now call attr(), not data(), to avoid
string-to-integer conversions.
(imported from commit 89666f40d624df154d41077546e1c13a23ee7e67)
Before this change, you could narrow to an old topic, and it would
not show up in the sidebar unless it had unread messages.
(imported from commit f177a7378dac064e46a6417357cc86ada4475936)
The old code was looping through recent_subjects to find the
original subject name, but we already have logic in get_subject_filter_li()
to be case-insensitive. I tried this with various casings of topics,
narrowing to topics, as well as narrowing to a stream (so no active topic)
and narrowing to an old topic.
(imported from commit 1decde13477cb742fd4bc82798f1afb282182bdf)
Create our own objects for handlebars, so that we can add transient
flags like is_zero without worrying about side effects to other
code.
(imported from commit b351a369cb3f36233e108e270c7abdd4ab8c5860)
When we call rerender() on MessageList, it is usually because
something big has happened, so it's reasonable that the pointer
may now be invalid. As an example, the old pointed-to message
may been removed from the home view. We have always just
moved the pointer under the hood, but now we do it in a way
that doesn't generate spurious blueslip errors.
(imported from commit d399a101f36b744a423ea7da80dda8352440c6c9)
There is a scenario where we call process_read_message()
for a message that we haven't recorded as unread before.
I'm not sure how it happens, but I put back code to
guard against crashing. The regression happened in
5752458c821.
(imported from commit 5ce15d2e236b738b445ed88f1733aa0612be0ff3)
This fixes a blueslip error where we were trying to add the same
message into a MessageList twice. Muting complicates our duplicate
detection, because empty() can return true even when we already
are storing messages that just aren't shown (because of muting).
The name empty() should probably be fixed in light of muting,
but filtering with self.get() is not really a problem here.
(imported from commit 83b0890471c9a0aa21996f3d0d3be4a238f23e65)
We have been persisting muting preferences on the back end for
a while, but we haven't been adding them to page_params for the
client to have at reload/startup time.
(imported from commit d9ca68aa0e4d22bfb0e6ce67fc0bc63981175c8b)
With the muting/hiding features, it has become the case that
this._hash can have entries that don't map to actually visible
rows, so this.get() can return true on an id that shouldn't
actually be selected, causing downstream code to explode. Now
we call this.closest_id() regardless of whether the hash is
populated, to be safe, but then we still call blueslip.error
if the pointer moved.
(imported from commit 348e8ff67ce3a6d61aaeb31f80549386518af2d0)
If you have two browsers open for the same account, muting in one
browser will now be reflected in the other browser. This got
regressed when changing the approach from collapsing to hiding.
The new code should be less brittle, as we encapsulate re-rendering
in muting.rerender().
(imported from commit 4e65e265b64513d38f518770453b7436cb92b4ca)
Update get_counts() so that it ignores counts for muted topics
when calculating stream/home unread counts.
(imported from commit 9b4e4da4346c225c535e97d709d3dee032603cc5)
The indirection was more confusing than helpful, especially
since the function had side effects, despite its getter-like
name.
(imported from commit 85d9cf642b4177f62488136f0e0f7f6c9304942e)
After killing off unread_counts.stream, the only field of
unread_counts was "private", so I just made unread_privates.
(imported from commit 9678f5b03524afb883ec4fa638b059e698888e78)
The prior commit makes it so that we no longer use unread_counts.stream
in get_counts(). This commit removes the code that updates the
data structure.
(imported from commit 5752458c8212bf02cf9c8733ce349fc35b204a9b)
These two data structures are kind of redundant:
unread_counts['stream']
unread_subjects
We are deprecating the former. The latter is more flexible for
features like muting.
Now, in get_counts(), we compute home counts and stream counts
in the same loop that computes subject counts.
(imported from commit c8d0ea12a56d0128811e0aa165de9882546906a5)
We are still showing the same data points, but the logic to drill
down on details for a particular realm is now all server side,
not client side, and we are smarter about omitting fields. In
summary mode, we don't show empty Name or Email columns. In
detailed mode, we show the realm as a headline instead of a column.
In this version you do lose the ability to see all system users in
the same view, but Waseem is ok with this.
(imported from commit edd2e646ab4cf5783ea64232d0cd621debece8d4)
When you load the activity report, it will just show summary
counts for realms, but if you click on a realm, you will see
details about users in the realms. You can also click "Show all"
to see an interleaved view of realms and users.
(imported from commit b106557b1fae64d525071afc124b5a8aed319086)
`Cannot read property 'flags' of undefined` in the
_.each callback in expand_summary_row.
Messages loaded when you scroll up in a narrow are not added to
all_msg_list. Because the user just clicked the message, we know
the message is in current_msg_list, so use that instead.
(imported from commit e76449a2a2748b96f69a2ab05d288b708d9e3ac0)
Instead of collapsing muted messages, just hide them altogether
in view where it makes sense to hide them.
(imported from commit 1c2c987ff302ceb135a025753cf421b4de1aea71)
I moved code into MessageList to further encapsulate details
of filtering. The MessageList instances should be their own
gatekeepers for what messages they care about.
(imported from commit ee6cd7f6eabf97962d724a05d7d0b0a3e6ab19e5)
Warn inside these functions when you get data on streams that you
are not subscribed to:
add_subscriber
remove_subscriber
user_is_subscribed
The back end should be smart enough not to spam us with subscriber
info that we don't care about.
(imported from commit b27644be2abc37c11ddff884ef392ea208bd1bd3)
The first menu option supported is to narrow to the topic.
The chevron only shows up if you turn on feature_flags.muting.
(imported from commit 17482f538a6d3e4ff96a36c042bad972d34f4b11)
Use the stream_data API to set up subscribers, so we don't
leak the data structure details into subs.js
(imported from commit e95616f2eb535ecf0e1cef35a143a71ad88de5bc)
This was requested by CUSTOMER28 and would also bring symmetry with
the user sidebar, which already does this.
(imported from commit 4dca3c957d3a710f00bbb34a7cd7dfe6074da8f7)
Though this should not be common, getting a peer subscribed/
unsubscribed notification to a stream we don't yet know about should
not be fatal
(imported from commit ee28b163e0efc9adfad31e1b321e986dfe56271e)
We create a blueslip error for undefined keys in Dict. This led
to a straightforward change in the unit tests for Dict. For the
unread test, to avoid the blueslip error, we had to be more specific
in setting up a user in one place, but this reduced our coverage,
leading to another small test being added.
(imported from commit 33e14795500d9283de2a7c03c4c58aec11cea4b8)
The exceptions were cryptic before, and they were inconsistent with
the fold_case: false behavior.
(imported from commit a40704d1a22bcdc60d91be832ee3c81eb416c6dd)
These engagement data will be useful both for making pretty graphs of
how addicted our users are as well as for allowing us to check whether
a new deployment is actually using the product or not.
This measures "number of minutes during which each user had checked
the app within the previous 15 minutes". It should correctly not
count server-initiated reloads.
It's possible that we should use something less aggressive than
mousemove; I'm a little torn on that because you really can check the
app for new messages without doing anything active.
This is somewhat tested but there are a few outstanding issues:
* Mobile apps don't report these data. It should be as easy as having
them send in update_active_status queries with new_user_input=true.
* The semantics of this should be better documented (e.g. the
management script should print out the spec above)x.
(imported from commit ec8b2dc96b180e1951df00490707ae916887178e)
people_list and people_dict include the feedback bot and anyone you've
cross-realm PM'd with. Useful for autocomplete, but not for admin and
stream settings views.
Fixes the UI part of Trac #1772.
(imported from commit cdefd4e86980447aad5190e7fc8ae3666d66e3c3)
hashchange, which calls admin.setup_page, runs in another onready
event handler before zulip.js main().
(imported from commit 5d4e71f4666baf0a53e3fe9804561ce6cdc06261)
This is feature-flagged to staging only.
There are basically 3 parts to the implementation:
1. In response to typing, fade/unfade the user list.
2. When a compose is aborted, unfade all users.
3. When the presence list is redrawn, fade/unfade the user list again.
(imported from commit cd416de232849a9f69dcacdc8b0fcfc20e3848a2)
This is the last step in getting a consistent client-side picture of who
is on a stream (provided non-MIT realm, and provided the local user is
subscribed to that stream).
(imported from commit 8bca722f169860ad4c1c92fdcb70d62c60f70fed)
acrefoot
Tim Abbott
Waseem Daher
Steve Howell
Kevin Mehall
Jessica McKellar
Allen Rabinovich
Zev Benjamin
Leo Franchi
Jeff Arnold
acrefoot
Jessica McKellar
Tim Abbott
Luke Faraone
Scott Feeney