Configure sending options prior typing in ComposePageRecipe

[rrrooommmaaa]

This PR first sets up sending options, then types the message, invoking draft save.

close #5052

---

Tests (delete all except exactly one):

• internal test refactoring

---

To be filled by reviewers

I have reviewed that this PR... (tick whichever items you personally focused on during this review):

• addresses the issue it closes (if any)
• code is readable and understandable
• is accompanied with tests, or tests are not needed
• is free of vulnerabilities
• is documented clearly and usefully, or doesn't need documentation

[sosnovsky]

I tried to reproduce bug which you described in #5052, but it works well for me - Saved label appears even after clicking checkbox for Rich text. Were you able to reproduce it?

Also for this case I think we should fix extension behavior, not change test, as users should see Saved label even when they click sending options checkboxes after composing message.

[rrrooommmaaa]

I tried to reproduce bug which you described in #5052, but it works well for me - Saved label appears even after clicking checkbox for Rich text. Were you able to reproduce it?

Yes, in Firefox, after typing something, when "Saved" appears, I switch "Rich Text" from "off" to "on" -- "Saved" gets immediately cleared and doesn't re-appear.
Important: If switching "Rich text" from "on" to "off", then "Saved" appears because this switch results in message re-formatting with draft save.

Also for this case I think we should fix extension behavior, not change test, as users should see Saved label even when they click sending options checkboxes after composing message.

Maybe they should, but switching the options takes some time, and as "Saved" may appear before (or during) the switching, there will be less time left to check that "Saved" is "continuously visible", potentially leading to unstable test.

[rrrooommmaaa]

How can I see the actual failed assertion?

[sosnovsky]

How can I see the actual failed assertion?

https://flowcrypt.semaphoreci.com/workflows/c6f72f99-cc1e-4b59-b8f4-101d97f25a5e/summary?pipeline_id=e4cefd3d-9a70-4a52-b77c-7201a721cc00&report_id=4ae71336-e44b-39bf-b9d2-752e234818a5

[sosnovsky]

And exact assertion is shown here, I think - https://storage.googleapis.com/p8ejr437e40a1hc10jpmsp5q2wfx83/artifacts/jobs/c309e55f-8e74-493c-9e53-38874c19f3c4/debugHtmlAttachment-0.html?Expires=1680609800&GoogleAccessId=artifacthub%40semaphore2-prod-artifacts.iam.gserviceaccount.com&Signature=b3Lzm%2BOOJCUbQGXGfYL2xnElyvbvY7qYKQ6lLolZAP4L%2FhKMqQ71x4q4EgxvXZtU%2BWDcfZmROjalQBj6knP1PCIPku%2FWBoCoYFyu5liTOKl9HDIZv9Mzxg79y6BakwHrLQ9t8m8vqcmayvhQCQ7jUs%2B8WeLqHMwD898DhyH3b5Ql9FSi0GADNOJ0k8rOEEWhd1H1BwxSJ1se2EWVPLH0Oxvcxh9I2QcnlxRB10vF0d9%2FDZ4GL%2Fc%2FMp7uWk7%2B1cLaiDw88AygcUZUpxKZI0FnhwlNV8TxaoIh%2FlzgxgqxHkXOmkr7PGzERU2VnCisxTSwoC8xuE2R4TtPZ95a0jKCGQ%3D%3D&response-content-disposition=inline&response-content-type=text/html;%20charset=utf-8

[rrrooommmaaa]

We need to extract this meaningful piece into the test result:

  ✘ [fail]: compose - saving and rendering a draft with RTL text (plain text) [ALL RETRIES FAILED for compose - saving and rendering a draft with RTL text (plain text)]
    ℹ (attempt 1 of 1) Failed:   AssertionError: expected '<div><br></div><div><br></div><div>--…' to include '<div dir="rtl">مرحبا<br></div>'
          at <anonymous> (/home/osboxes/flowcrypt-browser/test/source/tests/compose.ts:1508:62)
          at <anonymous> (/home/osboxes/flowcrypt-browser/test/source/test.ts:81:11)
          at BrowserPool.BrowserPool.withNewBrowserTimeoutAndRetry (/home/osboxes/flowcrypt-browser/test/source/browser/browser-pool.ts:113:11)
          at <anonymous> (/home/osboxes/flowcrypt-browser/test/source/test.ts:75:7)
          at <async> <anonymous> (/test/source/tests/compose.ts:1495:39)
          at <async> <anonymous> (../test/source/test.ts:75:55)

The line with (attempt x of x) Failed (last attempt)

[rrrooommmaaa]

This toast appears only once, doesn't reappear on further edits, is this behavior expected?

[sosnovsky]

Yes, it's expected behavior - we set disableSendingDrafts to true in case of such error (as there is no reason to retry draft saving)

https://github.com/FlowCrypt/flowcrypt-browser/blob/master/extension/chrome/elements/compose-modules/compose-draft-module.ts#L183

[sosnovsky]

Well done 👍