[REF] Use CRM_Utils_Mail::send for sending emails for confirming unsu…

[seamuslee001]

…bscribe resubscribe auto replies and subscribing

Overview

This refactors the mail sending code for handling auto replies, unsubscribe/resubscribe/subscription confirmations to using the CRM_Utils_Mail::send method

Before

Multiple separate invocations of pear mail without running through a hook

After

Single invocation of pear mail and passing through the alterParams and alter content hooks

ping @eileenmcnaughton @mattwire

[civibot]

(Standard links)

• If this is your first pull-request for CiviCRM, please browse CONTRIBUTING.md for information about the development and testing processes.
• If you are reviewing this pull-request, you may wish to consult the test sites and the Review Standards (long template, short template).

[mattwire]

@seamuslee001 Test fails relate

[mattwire]

@seamuslee001 This is great! I had not realised that these messages bypassed hooks etc.

[seamuslee001]

@mattwire fixed the test failures now

[mattwire]

@seamuslee001 I tested this on a system with mail set to "redirect to database" and one that can actually send mail. Clicking an unsubscribe link (eg https://example.org/civicrm/mailing/unsubscribe?reset=1&jid=59&qid=29&h=64c9d6be9bd2bb44) works without the patch (an email is sent) but does NOT seem to work with the patch - no email is sent?

[seamuslee001]

@mattwire do you have any logs with errors?

[mattwire]

@seamuslee001 No logs - it is because you are passing $params['to'] and CRM_Utils_Mail::send() requires $params['toEmail']

[seamuslee001]

@mattwire that looks like to have been it, I have fixed that on all emails changed here and also expanded a mailing subscription test to check that the email is sent and that a sensible message id header is set as well

[eileenmcnaughton]

@seamuslee001 still failing :-(

[mlutfy]

In case the test result disappears from jenkins:

Stacktrace

api_v3_MailingGroupTest::testMailerProcess
Message-ID: <s .  not found in  MIME-Version: 1.0
Content-Type: multipart/alternative;
 boundary="=_3fda462e069d3aa3be34e24d37f2c065"
From: "FIXME" <info@EXAMPLE.ORG>
To: <test@test.test>
Subject: Subscription Confirmation Request
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
Return-Path: do-not-reply@setting.com
Reply-To: civicrm+c.4.2.acb3c277fc3b61a7@setting.com
Date: Sat, 13 Jun 2020 22:52:25 +0000
Message-ID: <civicrm+s.4.2.acb3c277fc3b61a7@setting.com>

[seamuslee001]

@eileenmcnaughton @mattwire @mlutfy I believe I have fixed the test now

[eileenmcnaughton]

@seamuslee001 @mlutfy perhaps you could trade this for #17305

[demeritcowboy]

jenkins test this please

[demeritcowboy]

• General standards
  • [r-explain] PASS
  • [r-user] PASS
  • [r-doc] PASS
  • [r-run] PASSed for the parts tested
    • Tested optout, unsubscribe, resubscribe.
    • Didn't test the others.
    • Tested with and without the generate Message-ID setting.
    • Didn't test hooks.
• Developer standards
  • [r-tech] PASS
  • [r-code] PASS
  • [r-maint] PASS
  • [r-test] PASS

[seamuslee001]

@eileenmcnaughton can you merge based on @demeritcowboy 's review

[eileenmcnaughton]

thanks @demeritcowboy - really appreciate all the review work you do and I have a tonne of confidence in your ability to ferrit out any gotchas

[mattwire]

Thanks @demeritcowboy @eileenmcnaughton Sorry @seamuslee001 this fell off my radar!