This has been disabled since forever, but we've mostly been assuming it's
enabled. Let's actually turn it on and see what breaks.
(imported from commit a194ba4b77b5290c31b46401d400bece68a33e6d)
There are other focus issues related to switching between a class
message and huddle during composition, but this fixes the immediate
problem.
(imported from commit ec7ebc1c68e52942e9eaa0ea61e0de2f6ccae521)
This also helps us manage checking, case-insensitive, for
subscriptions while preserving the casing used by the class creator
for display.
It also fixes a bug where the class_list would become out of sync with
your true subscriptions, allowing you to appear to send messages to
classes to which you had unsubscribed.
(imported from commit 5e8d017bcfb27a71c52f7517733eda7b926d721b)
This also prevents the "bounciness" associated with the fact
that the stream/huddle selector was an absolutely positioned
div relative to the bottom of the compose window.
(imported from commit 413003a83c187efd45d0281f7cb6c9d0bd550674)
Potential downside: If you were narrowed and went to, e.g.,
settings, you would lose your place. If we wanted to be more
fancy we could have this code fire only if you were already
on the #home tab.
(imported from commit 5e8d90ba255076a11aa6f7a58c447156729ca23f)
href="#" will cause the browser to jump to the top of the page, which
breaks everything in the scrolltheworld world, unless you suppress the
event.
(imported from commit ae03cbb9e8a82f28e8536f29485e9e947e7e76c2)
This actually is a bit cleaner than our last approach,
because (I believe) we run our scroll handler code
even after the final scroll event, which prevents the
pointer from getting hidden by rapid scrolls.
(imported from commit 3bde4e8f067cd2406f90c04425c6e4ffb81ea784)
The issue was that our 'button' was actually a link with href='#',
and we didn't actually cancel the event; so it jumped to the
top of the page.
(imported from commit e66decb3a202ea22176ec539a1ec1d405faf0117)
I'd like it to ultimately be the case that if you make the window
too narrow, the menubar goes to the very top and has small icons.
(imported from commit 0a5fe548d4d848c0358c9780df7af1fe1abc93fa)
Try to adhere to Bootstrap grid system a bit more consistently.
Unstyle subscriptions page for consistency.
Still not sure why the "affix" jumps when we switch to "Subscriptions."
(imported from commit 1c28de4e2ef3d6ed3d1bdcfd143eadefba2b46cb)
It causes problems when [un]narrowing, and we already do the rest of formatting
on the server side.
This reverts commit 90af0192b37bbebbf56d5e7c50f182485ddbca10.
This reverts commit df7e355648d2c4d6319de049933547ed96402fd8.
(imported from commit 99d87f0826ec2f49741f86fad6524ed93e76723f)
scrolling-tab used to be used to target which tab would be a recipient
of the scroll events generated by PgUp, PgDn, etc. Since we now let
the browser handle these natively, we don't need to worry about it.
(Instead, when we hook the 'scroll' event, we should make sure that
home screen.)
(imported from commit c555d960da995a09b370867c96d17f4ce4f2171f)
I would like to have split this into several commits, but it's hard to do
incrementally, because functions need to be defined before their use in global
data.
(imported from commit 31533396b236f3b657950b073982c317256f082a)
My expectation is that you may or may not get to "Search" by clicking
something like this.
(imported from commit 3f09286b1a8fa80455d6f6d55997f6f829fbad15)
Previously, when we narrowed on a super-long class/instance or
a huddle with lots of people, the box overflowed. This is an
attempt to fix that, even if it isn't the prettiest.
(imported from commit 4eb58726a4c4714bd5435a791ad8fea0eabb58ed)
This required some serious retooling of the table,
and some thinking about the interactions between
table-layout: fixed and colspan.
And some of it is still a little magic-number-y.
(Like that 97% width on zephyr_compose_box -- without
it, the stream compose box looks weird in Chrome,
but not Firefox.)
(imported from commit 20c426ad2dae5efa3107890b28976a957bb3d1e3)
We add a few templates for django-confirmation. We define a
"PreregistrationForm" which is validated by accounts_home, which then
generates a confirmation object and emails the user. This required creating
a new table for a PreregistrationUser with an email and status (confirmed)
field.
The register function now no longer accepts a "email" field in the form
and deals only with confirmation IDs to determine the email used to sign
up a user.
(imported from commit 4fcde04530aa7ad4de84579668daee7290b424ac)
But do keep the pointer on the screen as you scroll, and smoothly
scroll all the way up to the top or all the way down to the bottom.
(imported from commit 6acad2a16339ee5eeeebd62a853000c21e4b7152)
Previously, we were trying to expand a variable which wasn't defined. We
correct that here. Furthermore, we don't right-align the gravatar if
its being shown in the settings panel by removing the "profile_picture"
class.
(imported from commit 1ae09d1a44aca87962f971af82393bf4ece6d16c)
We will probably re-style this eventually.
Also, the animation freezes during template rendering. And the HTML is a "give
up and use tables" situation.
(imported from commit 847374b616dc7ce909834f23d5ed9522aa457254)
Split off some content previously placed into base back into index.html.
Move content previously localised in accounts_home to a new content_base
for use in the non-app version of the site.
We wrote up some copy and styled it according to taste.
In our new signup form, we no longer collect a short name and instead
extrapolate from the email.
(imported from commit cd734b13a4bda7f304129379524c4cc96b77b24b)
Benchmarks [1] show it's much faster than ICanHaz, though we will have to work
to realize those gains.
[1] http://jsperf.com/jquery-template-table-performance/77
(imported from commit b037a6f15e27fd49af88f814ff949811dbcd48f2)
In this commit we also remove the navbar from base.html for implementing
further changes as requested by Waseem.
(imported from commit 85a8ea82ff0e271120adde83e15eb671b5584722)
Put all the 'API'-ish URLs in the /json/ namespace so that
/subscriptions and /subscriptions/add, etc. can be real
user-facing URLs if we want them to be.
(imported from commit 1455a32d27e44547ab1874b6289243cb40c85dde)
Now that 'manage_subscriptions' only removes subscriptions,
let's name it that. Also, since both functions now only
act on one subscription at a time, use them in the singular.
(imported from commit 52cfc7b36eb1533da7aec1bc3a6e460151a56da8)
I don't love some of these icons, so I view them as pretty temporary.
(I'd like, e.g., settings to be a gear.)
(imported from commit 0433f226082acc37392e97b20648be8eee704b7b)
We no longer break random things! Its pretty grand, actually.
This reworks and reverts commit fbadd6e854722e41cccd2535748ee47f4efd657b.
Conflicts:
zephyr/static/js/zephyr.js
(imported from commit 534a120290855d3bf2cf979ac174267c2d07bf68)
For now the idea is that if you're in the search view (or the
subscriptions view) and you click 'New Message', we should
snap you back to the main view.
This may or may not be the right behavior in the long term.
(imported from commit 43c24e1af9e80225ff9be8a62f4c8c925960436a)
Previously we were iterating through all zephyrs twice on the server, now
we are moving part of that to the client so the client has an index of
Zephyrs that it generates itself as part of add_message.
(imported from commit c07a6c36911fa2eae6f216c34331be34b2aa8178)
Take a look with 'git show -w' to ignore whitespace changes,
there are some tiny tweaks to try to make sure we close
tags that we weren't properly closing.
(imported from commit 44cba1b8859842fd85a39a3062963af617556ca8)
You don't care about that content anymore. We probably want to clear
even more aggressively than this.
(imported from commit 29f6910a12e1b722f5801db644b04f54cf5bfd63)
I still don't love it visually, but it's getting there.
Some obvious issues:
Personals window is totally unstyled
Clicking 'new message' should perhaps give you a fresh new window
not something prepopulated with old stuff. Think about this.
(imported from commit 8b28fd084d550db404eabbe63c056fa6866c0697)
Put a little 'x' by the class or class-instance indicator, to make
it more analogous to the planned behavior of the view-in-context
search box.
(imported from commit fa01001cffa6a6094ba5fbdcbdc965addb2efa1c)
The purpose of this is to ensure that we can reference a zephyr by its ID
and get all the data about the zephyr; in the future we may not have all
such data stored in TRs as we do now.
(imported from commit 9128ddd01b46396fd276124ca1e6430538d3dd63)
Today I'm of the mindset that searching is a fundamentally
distinct operation than narrowing, so this reflects that.
(There would be a separate screen and UI for searching,
I guess.)
(imported from commit 432a4088612dafd06184bec228b63056961325dd)
unhide -> show_all_messages
narrow_indicator -> currently_narrowed_to
narrow -> searchbox
This is a little clearer about what these buttons/functions
do, which will be useful as we add a little more complexity
associated with searching.
(imported from commit b51e745ea71c212d6735e04011eaea5e71bf6d5a)
I'm not exactly sure that this is the interface we'll want for
this, but it's the start of an experiment.
(imported from commit ea18a9b05106546475bc151229955ddafd8e7b8e)
This was a historical artifact of how that used
to be a search icon right by a textbox.
(imported from commit 6f73bc8b5d174e4c07f1503ffd08650110e80420)
I have some serious concerns with this implementation,
but I think it's still an incremental improvement.
(imported from commit 6ed8d2545c727e25bf85b98a1528dbf3d155bc92)
This is mostly working; things that need to be implemented:
- Ask users for their Real Name
- Personal replies need skinning
- Miscl UI changes to match mockups
(imported from commit 3638cf5ec2ba2306004ba6db6fa4c25c47f76517)
Here we check if a class exists. If not, we prompt the user to create, sub,
and send his message to the class. If the class exists but we're not subbed
we prompt the user to sub.
This commit also added a decorator to views.py and refactored out some
redundant code.
(imported from commit 7234ef6c080f2a6de6ff0922635dddd90032f7fe)
Note that realms aren't actually used for anything, aren't prompted
for, etc, so this is mostly just a schema change and the refactoring
needed to allow setting the realm in the future.
(imported from commit b8b483dcc9601b288702284879e9c99707a50a5c)
Here we pull in the jqueryui library and a style sheet to provide a widget
to do autocomplete for various fields. We populate a list of all ever-seen
classes, instances, and people (C/I/P) into JS arrays and then pass them to
jqueryui.
Each time a new C/I/P appears, we reinitialise the autocomplete functions
so they will notice the new data.
(imported from commit f1ab69c99d21f68ccf40c015bd43960f463c6ff2)
Previously, if hamlet wrote to iago, and then hit "r" on his own message,
the form would prompt hamlet to reply to himself. Now, Humbug will DTRT and
prefill the form with iago instead.
(imported from commit cb3260d1d0bc89b184dac84ebf1e5642d0bc1606)
Ideally this would be expressed as an event handler bound to the
in every place that the pointer changes, but I haven't found a way to
do that yet (this is easy for *inputs* but not arbitrary DOM
elements).
(imported from commit c190e23f5ed5fd3e3a4823b763682d611ec5b446)