Clean-up docs in MigrationInterface

I really only have one reaction to this sort of doxygen.

(The migration is currently importing for STATUS_IMPORTING, The destination plugin for getDestinationPlugin() etc)

Oh well, at least phpcs is happy. I guess it's not the role of every @return to explain what's what.

Moving to jhodgdon's queue.

This is pretty good... and thanks for making the patch! A few errors:

a)

+  /**
+   * The migration is currently been rolled back.
+   */
   const STATUS_ROLLING_BACK = 2;

...
+
+  /**
+   * The migration is been stopped.
+   */
   const STATUS_STOPPING = 3;

been => being

b) This section seems inconsistent:

+   * Migration error.
    */
   const MESSAGE_ERROR = 1;
+
+  /**
+   * Migration warning.
+   */
   const MESSAGE_WARNING = 2;
+
+  /**
+   * Migration notice.
+   */
   const MESSAGE_NOTICE = 3;
+
+  /**
+   * Represents a condition that did not prevent the operation from succeeding.
+   */
   const MESSAGE_INFORMATIONAL = 4;

If you're going to be all verbose on the last one, be verbose on the others... or make the last one more concise? I don't think the description on Informational really makes it different from Notice either?

c)

+  /**
+   * The process has interrupted itself (e.g., the memory limit is approaching)
+   */
+  const RESULT_INCOMPLETE = 2;

Really? interrupted itself?... OK, but needs to end in .

d)

+
+  /**
+   * The process was stopped externally (e.g., via drush migrate-stop)
+   */
+  const RESULT_STOPPED = 3;

Also needs to end in . and can this be more consistently worded with the previous one? The other one says the process "interrupted itself" (??) and this one says it "was stopped" (seems much better). But really, what's the difference?

e)

    * @return \Drupal\migrate\Plugin\MigrateIdMapInterface
+   *   The idMap.
    */
   public function getIdMap();

Can this say "The ID map"?

f)

    * @return int
+   *   A unix timestamp representing the highwater mark.
    */
   public function getHighwater();

What's a highwater mark, and shouldn't it be "high water" (two words) anyway?

Regarding "highwater" vs. "high water", the text you cited has it as two words. :) But I agree, the text above that in the documentation for this method has two words, so one word is fine, I guess...

The methods are also inconsistent. Some of them are CamelCased with HighWater and some with Highwater. This should all be fixed, but I agree it's a separate issue: #2320245: High water should consistently be two words -- so let's leave it as-is for now.

We should probably capitalize Unix, though? (in that same line about the high water mark). Everything else looks good; sorry for missing that last time through.

Thanks, looks good to me!

Patch doesn't apply. Looks like the interface moved?

wait, I was trying to apply to wrong branch. :)

Committed to 8.0.x.