Summary:
Add specs to test the components of open tracking and link
tracking. Notably does not test the overall functionality, which
still needs specs.
Test Plan: adds specs
Reviewers: juan, evan, bengotow
Reviewed By: evan, bengotow
Differential Revision: https://phab.nylas.com/D2667
Summary:
Fixes an issue with sending where certain conditions could result in a
duplicated message.
Fixes task dependency logic for draft syncback and send. Changes `createdAt`
on tasks to instead be `sequentialId`, assigned when the task is queued, to
track order of enqueueing. Renames `isDependentTask` => `isDependentOnTask`
and adds comments for clarity.
Test Plan:
Specs updated. Might be good to add some later to test this particular
edge case.
Reviewers: juan, evan, bengotow
Reviewed By: evan, bengotow
Differential Revision: https://phab.nylas.com/D2681
Summary:
- Separate gmail's remove-from-view and delete behaviors and write logic
for each of those
- Remove MailboxPerspective::{canArchiveThreads, canTrashThreads,
removeThreads} and some unecessary code in TaskFactory
- Instead, add MailboxPerspective::tasksForRemovingFromPerspective (I
know its a bit of a mouthful)
- I initially tried to put all of the logic for each execution path
inside the TaskFactory by checking perspective types, but it made
more sense to use the polymorphism already in place for the different
perspective types.
- There is a default delete/remove-from-view behavior which is
configurable via simple ruleset objects. The gmail behavior is
configured in this way.
- Update swipe css classes based on destination of threads
- Fixes#1460:
- Update logic to display archive/trash buttons and context menu options correctly
when selected threads can be archived/trashed (not based on
perspective)
- Same for swiping
- Add a bunch of specs
- Convert some code to ES6
- TODO write some docs for new functions
Test Plan: Unit tests
Reviewers: drew, evan, bengotow
Reviewed By: bengotow
Differential Revision: https://phab.nylas.com/D2682
Summary:
- Setting the locale in moment was not sufficient to actually use the
correct localized formats.
- Moment provides localized format via the formats : 'L', 'LL', 'll',
etc. See: http://momentjs.com/docs/#/displaying/format/
- Updates to set our date formats based on localized formats:
- Unfortunately, localized formats always contain the year, so I
manually removed the year from our short format.
- Actually fixes: #1515
- Fixes bug where setting nextWeek or thisWeekend returned incorrect
date if the current day was saturday or sunday
- Add specs
Test Plan: Unit tests
Reviewers: evan, drew, bengotow
Reviewed By: bengotow
Differential Revision: https://phab.nylas.com/D2688
Summary: Adds a new visual theme picker to the menu that allows users to select different themes based on color palettes and then change their themes live.
Test Plan: Test included.
Reviewers: evan, bengotow
Reviewed By: evan, bengotow
Differential Revision: https://phab.nylas.com/D2669
Summary:
This commit changes how N1 parses config.cson very slightly
so that when "env" is set to "local" N1's auth flow can be
tested against local copies of Nylas' auth servers.
Test Plan: I tested it manually against a local auth stack. It worked!
Reviewers: bengotow
Reviewed By: bengotow
Subscribers: kav-ya
Maniphest Tasks: T6690
Differential Revision: https://phab.nylas.com/D2671
The rationale being it's covering all of the cases while still having a good performance.
Tracking a non-standard URL is not a user experience problem in a way. It's embedded in a <a href> tag.
As of now, link-tracking creates link tracking for every <a> tag, no matter the href attribute. This commit only creates link tracking for valid URLs.
The reason I wrote a new regex instead of rewriting linkTagRegex is that this regex is used by message-list.
https://regex101.com/r/jD5zC7/3 vs https://regex101.com/r/cK0zD8/2Resolves: #1525
Summary:
- Also refactors the code a bit for testability and maintainability
- Fixes#1515
Test Plan: - Unit tests
Reviewers: evan, drew, bengotow
Reviewed By: bengotow
Differential Revision: https://phab.nylas.com/D2651
Summary:
This fixes a serious issue where drafts could appear to be sending if
they were located in the same index of the message list as a draft which was
previously sending.
Under the hood this was due to two bad programming choices:
1) State based on props in MessageContainer requires correct implementation of
componentWillReceiveProps. This is definitely an anti-pattern.
2) Using item index rather than clientId as the key for items in the MessageList
caused containers to be given a different message prop when one was inserted,
rather than just shifting the existing ones down and inserting a new one.
Test Plan: Not sure how to test this really...
Reviewers: drew, juan
Differential Revision: https://phab.nylas.com/D2673
Summary:
When we delete drafts, they can sometimes come back because we receive deltas about them,
find that they're not in the cache, and re-save them.
This diff fixes that by permanently incrementing the optimistic change count, which was also do
to prevent deltas from changing threads while we're in the middle of a request that will change
them further. (This may not actually be necessary anymore, because we'd ignore the delta anyway
due to it's out-of-date version?)
Test Plan: Run a couple new tests
Reviewers: drew, juan
Reviewed By: juan
Differential Revision: https://phab.nylas.com/D2665
Addresses #411
The `N1.sh` launch script fails to account for running from
`/usr/{local/}bin/nylas` where it would throw an error saying
it couldn't find the electron binary.
The Atom launch script works just fine and takes that into account.
The installed launch script here is a reflactored Atom launch script
that properly accounts for running the `nylas` executable and locates
the actual binary accordingly.
