Eli Zaretskii <eliz@gnu.org> writes:

>> From: Mauro Aranda <maurooaranda@gmail.com>
>> Date: Sun, 29 Sep 2019 08:44:48 -0300
>>
>> 1) C-x b test
>> 2) M-x text-mode
>> 3) Type something like this:
>> One sentence.
>> Other sentence.
>>
>> 4) Go to the first sentence, after the period.  Type anything and then
>> type C-M-i to complete it.
>> I type "comple" and complete it to "complete".
>>
>> 5) I expect the cursor to be at the end of the word just completed.  But
>> I see it moving to the next line.  With different words completed, the
>> exact position changes, but to me the cursor always ends up in an
>> unexpected position.
>
> Looks like it expects you to invoke the command at the beginning of a
> word.

I see.  That works as I expected, right.

But what I described still looks like a bug to me, though...I mean,  I
think the user shouldn't have to type M-b to then type C-M-i, in order
to be able to continue typing after the completed word.

Furthermore, the documentation under Spelling, in the Emacs Manual says:

     Complete the word before point based on the spelling dictionary
     (‘ispell-complete-word’).

and then:

     Insert the beginning of a word, and then type ‘M-<TAB>’;

Note the "before point".  That's what I did.  The only thing different I
did was that I wanted to complete a word that was not at the end of the
buffer.  And knowing only that much, I expected the cursor to be at the
end of the completed word.