Summary: Cleaning up almost all the linting errors in the logger
Test Plan: ... run the build
Reviewers: juan, evan
Reviewed By: evan
Differential Revision: https://phab.nylas.com/D3605
Summary:
refactor multi-send
This diff started off by fixing sending with attachments.
The issue is that our `FileUploadStore` listened for
`Actions.sendDraftSuccess` as its signal to remove the files from the
.nylas temp directory. Unfortunately, the old MultiSend tasks, after
delivery of the message, would try and put the base message in the sent
folder. Since we already deleted the file from our local temp dir,
creating the base message for the sent folder would fail.
This exposed a much bigger issue which is that we don't consistently
distinguish between "delivery" of a message and any post-processing we do
(like filling the sent folder). This was leading to a variety of other
subtle issues.
For example, N1 assumes that if the SendMessage task fails, then we pop
the draft back up and ask the user to try again. Unfortunately, since we
were combining "delivery" and "post processing" it was possible for the
message to actually deliver, but fail when stuffing the sent folder, or
fail due to some other random bug. This would cause the user to send the
message twice.
To help us ensure we never "deliver" twice and handle errors more
intuitively, I separated out the two concepts.
Now there are "send" set of tasks and endpoints, and a
"EnsureMessageInSentFolder" set of tasks and endpoint (the latter used to
be ambiguously known as ReconcileMultiSend, whatever that meant)
The logic for send hasn't changed. This is mostly a renaming and moving
files around.
Test Plan: manual :(
Reviewers: jackie, juan, halla
Reviewed By: juan, halla
Differential Revision: https://phab.nylas.com/D3577
Summary:
If you sent an email to yourself it would not show up in your inbox. This
is because sent messages would never get a lastMessageReceived timestamp.
Since we order the inbox by lastMessageReceived, setting that to null to
on sent mail would mean it never shows up in the thread list.
Also fixed an assertion bug in SFDC that requires transactions to return a
promise.
Finally added extra debug interfaces that will show more info if the delta
stream detects an inconsistency
Test Plan: manual
Reviewers: juan, halla, jackie
Reviewed By: jackie
Differential Revision: https://phab.nylas.com/D3552
Summary: We were creating duplicate `Message` objects because the formatting for the date was different between `buildForSend` and `parseFromImap`. Now, we create the initial hash using the same format that `buildmail` uses to ensure that we generate the same IDs.
Test Plan: Tested locally.
Reviewers: evan, juan, spang
Reviewed By: spang
Differential Revision: https://phab.nylas.com/D3559
Summary:
Save metadata correctly by reassigning an object to value.
Since account IDs are different between N1 and N1 Cloud, use just the message ID, which should be unique.
Test Plan: Tested locally.
Reviewers: evan, juan
Reviewed By: evan, juan
Differential Revision: https://phab.nylas.com/D3524
Summary: This is a test
Test Plan: Testing
Reviewers: juan
Differential Revision: https://phab.nylas.com/D3493
[local-private] add old edgehill src/pro into packages/local-private
