This decouples from Chrome notifications, which gives us cross-platform
support in at least modern browsers.
We log this action so its replayable in our message logs.
This implements the model change indicated by the previous schema commit.
(imported from commit b21213cdde54f43670bbb0bf1f607147fc732b38)
Previously, we were fetching Message.objects.select_related() from the
database, even if we actually ended up fetching the message dicts from
memcached and thus not actually using them. Especially in the cached
case, this resulted in a lot of overhead where the Django ORM put
together Message objects with lots of data in them that were never
used. This commit switches the model to only fetch the full message
objects from the database for those messages which are not found in
the memcached caches.
Here are the timings for get_old_messages before this patch was applied:
(cached)
127ms (db: 42ms/2q) /json/get_old_messages (starnine@mit.edu via website)
385ms (db: 105ms/1q) /json/get_old_messages (starnine@mit.edu via website)
(uncached)
315ms (mem: 6ms/41) (db: 90ms/22q) /json/get_old_messages (starnine@mit.edu via website)
507ms (db: 94ms/14q) /json/get_old_messages (starnine@mit.edu via website)
Here are the timings for get_old_messages after this patch was applied:
(cached)
80ms (db: 9ms/2q) /json/get_old_messages (starnine@mit.edu via website)
133ms (db: 4ms/1q) /json/get_old_messages (starnine@mit.edu via website)
(uncached)
230ms (mem: 9ms/41) (db: 48ms/23q) /json/get_old_messages (starnine@mit.edu via website)
385ms (db: 55ms/15q) /json/get_old_messages (starnine@mit.edu via website)
(imported from commit c4748513392a906393314aa7cd41d98a69865411)
On my laptop, this saves about 80 milliseconds per 1000 messages
requested via get_old_messages queries. Since we only have one
memcached process and it does not run with special priority, this
might have significant impact on load during server restarts.
(imported from commit 06ad13f32f4a6d87a0664c96297ef9843f410ac5)
It's subtle, but the slice was in the wrong place and wasn't
actually truncating the stream name at all, so the client and
server disagreed about where the tutorial messages should go.
(It might be the case that we should accept the tutorial stream
name from the client directly, rather than computing it in two
places.)
(imported from commit 8273223f182e8ad36eaea1cbf75e1426fcfdfbab)
See PEP 328[1] for details. This feature was introduced in Python 2.5 and
will become mandatory in Python 3.
[1]: http://www.python.org/dev/peps/pep-0328
(imported from commit 7444eeba8a08d5f91b94c7921848f2274979bd76)
This commit will incorrectly list past-online users as active, a shortcoming that is
addressed in the next commit
(imported from commit b018767df686f88c0ca939c067c573e4d7cea357)
This avoids 10s of seconds of delay when you invite several people at
once through the web UI.
(imported from commit 75acdbdb04caf62bbb08affc7796330246d8a00e)
We accidentally lost this when we did the User/UserProfile merge (this
commit also deletes the old code to add the auth_user index in
do-destroy-rebuild-database).
This below is mostly just notes for future reference, but when
deploying this change to staging, we should consider running the
following instead of using the migration directly:
CREATE UNIQUE INDEX CONCURRENTLY zephyr_userprofile_email_uniq ON zephyr_userprofile(email);
ALTER TABLE zephyr_userprofile ADD CONSTRAINT zephyr_userprofile_email_uniq UNIQUE USING INDEX zephyr_userprofile_email_uniq;
CREATE INDEX CONCURRENTLY zephyr_userprofile_email ON zephyr_userprofile(email);
But I think it might be the case that it's fine to just run it
directly, since the ALTER TABLE part seems to hang if there's an open
transaction working on a UserProfile object anyway.
(imported from commit 1bf34ce242de51e97c91c8bab86b6b273e17fb43)
This is preparatory for removing the StreamColor model, so we also set
things up so anything changing the StreamColor model changes the
Subscription model too.
The manual task is to run the copy_colors.py management command after
deployment to each of staging and prod.
(imported from commit 1be7523ca59f5266eb2c4dc2009e31209ed49635)
I think all that one needs to do to deploy this commit is on developer
laptops, run `generate-fixtures --force`.
(imported from commit 34916341435fef0875b5a2c7f53c2f5606cd16cd)
And keep the fields updated, by copying on UserProfile creation and
updating the UserProfile object whenever we're updating the User
object, and add management commands to (1) initially ensure that they
match and (2) check that they still match (aka that the updating code
is working).
The copy_user_to_userprofile migration needs to be run after this is
deployed to prod.
(imported from commit 0a598d2e10b1a7a2f5c67dd5140ea4bb8e1ec0b8)
Our testing code had a number of places where it was using User
objects where it should have been using UserProfile objects --
e.g. using a User id as the type_id in a Recipient table. This commit
addresses this in the filter_by_subscriptions code paths.
(imported from commit e305bc8e2a8bdbfd04c93c59d56955e7971552af)
The previous situation was bad for two reasons:
(1) It had a lot of copies of the code, some of them missing pieces:
UserProfile.objects.get(user__email__iexact=foo)
This was in particular going to be inconvenient since we are dropping
the __user part of that.
(2) It didn't take advantage of our memcached caching.
(imported from commit 2325795f288a7cf306cdae191f5d3080aac0651a)
This can result in a significant performance benefit because we only
need to update the columns that changed..
(imported from commit 42bef1fcc58ad79bd864f89263fe82e90743ee5b)
The policy this implements is:
* 1 week for most persistent data (Clients, etc.)
* 1 day for messages
(imported from commit d57bb2c6b9626ffa2155c6d0ef9b60827d1f2381)
This saves 2 database queries per user in the huddle when sending the
first message to a particular huddle.
(imported from commit f71aa32df846fb4b82651a93ff9608087ffcaa5a)
See http://bugs.python.org/issue5876 for an explanation for why this
is needed -- basically __repr__() needs to return a string, not a
unicode object in Python 2.
This causes problems on Django 1.5 because the more expressive
exception code in model.objects.get() will crash with a __repr__()
containing non-ascii unicode characters.
(imported from commit f44085e67d9d14629b821a29bbf65738f1794d6c)
This fixes an experienced bug where you couldn't subscribe to a stream
with non-ASCII characters (failing with a UnicodeEncodeError), as well
as many other potential bugs.
(imported from commit f084a4b4b597b85935655097a7b5a163811c4d71)
This change could significantly load the database if pushed to prod at
the same time as:
send_message: Save rendered message content to the database.
goes out. So this should go out in a different prod push from that
commit.
(imported from commit 4e5a370bc43defb5083cac3df45c649cf2366abf)
Leo Franchi
Tim Abbott
Luke Faraone
Waseem Daher
Jessica McKellar
Zev Benjamin