"incorrect" here means rejected by a bot's validate_config() method.
A common scenario for this is validating API keys before the bot is
created. If validate_config() fails, the bot will not be created.
The former "the-bot-is-being-created" indicator was useless,
since it only edited the value attribute of the "create-bot" button;
that attribute only sets the text of a button when it is initially created.
This commit replaces the old code with a spinning loading indicator,
like the one used for editing bots.
This effectively changes a condition from `=== 0` to `<= 0`,
but that probably doesn't matter (and would only help). More
importantly, it just makes the code more consistent.
This mostly moves code from ui.js.
We change the arguments to `message_fetch.load_more_messages()`
to be `opts` with callbacks for `show_loading` and `hide_loading`.
We also defer starting the scroll handler until `message_fetch.js`
has been initialized.
Adds realm_bot delete event. On bot ownership change, add event is
sent to the bot_owner(if not admin) and delete event to the
previous bot owner(if not admin). For admin, update event is sent.
if the test fails, the 'output_dir' would not be deleted and
hence it would give an error when we run the tests next time,
as 'do_convert_data' expects an empty 'output_dir'.
Also the unzipped data file should be removed if the test fails
at 'do_convert_data'.
To prevent specifying notifications individually for global
updates, automate handle_global_notification_updates using
settings_notifications.notification_settings where we have
the keys of all the types of notifications.
If we have bugs in our compose code, we don't want them to
cause other strange side effects during reload. (We recently
had a test deployment that was a bit buggy, and the reload
problem was tough to chase down due to the misleading
symptoms.)
@brockwhittaker wrote the original prototype for having
pills in the recipient box when users compose PMs (either
1:1 or huddle). The prototype was test deloyed on our
main realm for several weeks.
This commit includes all the original CSS and HTML from
the prototype.
After some things changed with the codebase after the initial
test deployment, I made the following changes:
* In prior commits I refactored out a module called
`user_pill.js` that implemented some common functions
against a more streamlined version of `input_pill.js`,
and this commit largely integrates with that.
* I made changes in a prior commit to handle Zephyr
semantics (emails don't get validated) and tested
this commit with zephyr.
* I fixed a reload bug by extracting code out to
`compose_pm_pill.js` and re-ordering some
calls to `initialize`.
There are still two flaws related to un-pill-ified text in the
input:
* We could be more aggressive about trying to pill-ify
emails when you blur or tab away.
* We only look at the pills when you send the message,
instead of complaining about the un-pill-ified text.
(Some folks may consider that a feature, but it's
probably surprising to others.)
The main point of this change is to streamline the core
code for input pills, and we use also modify user groups.
The main change to input_pill.js is that you now
configure a function called `create_item_from_text`, and
that can return an arbitrary object, and it just needs
a field called `display_value`.
Other changes:
* You now call `input.create(opts)` to create the
widget.
* There is no longer a cache, because we can
write smarter code in typeahead `source` functions
that exclude ids up front.
* There is no value/optinalKey complexity, because
the calling code can supply arbitrary objects and
do their own external data management on the pill
items.
* We eliminate `prependPill`.
* We eliminate `data`, `keys`, and `values`, and just
have `items`.
This split notification settings to separate settings for streams,
PM/mention and other notifications settings.
Previously we have to treat them combined, but with this refactor,
we can treat them separately for future changes.
Fixes#8587.