Yes, with some some config fiddling I've seen this crm-based tagging
wotkflow you describe, but problems were the same.

Suddenly, I tried to remove some customization code for Helm (namely
this -
https://github.com/search?q=helm-hide-minibuffer-maybe&type=Code&utf8=✓)
and it seems proper completion starts working to some extent.

Anyway, thanks for prompt reply and mind seeding! :)


John Kitchin <jkitchin@andrew.cmu.edu> writes:

> I can't reproduce that behavior on my end. I see different behavior though.
> My helm window is titled crm-complete, and the buffer is called
> *helm-mode-crm-complete*
>
> This seems to be the helm core package I am using.
>
> helm-core-20170112.917
>
> John
>
> -----------------------------------
> Professor John Kitchin
> Doherty Hall A207F
> Department of Chemical Engineering
> Carnegie Mellon University
> Pittsburgh, PA 15213
> 412-268-7803
> @johnkitchin
> http://kitchingroup.cheme.cmu.edu
>
>
> On Tue, Mar 14, 2017 at 12:24 PM, <aaermolov@gmail.com> wrote:
>
>> Hi John,
>>
>> no, these do not seem to fix anything. :(
>>
>> Namely, the following is still happens:
>>
>> 1) I'm pressing C-c C-q then TAB to select freeform input
>> 2) then there is "org-set-tags" above candidates list and buffer title
>>    is "*helm-mode-org-set-tags*"
>> 3) after any input candidates list looks like "[?]<helm query chars>"
>>    (initial candidates disappeared)
>>
>> regards,
>> Alex
>>
>> John Kitchin <jkitchin@andrew.cmu.edu> writes:
>>
>> > have you tried these settings:
>> >
>> > (setq helm-mode-fuzzy-match t
>> >       helm-completion-in-region-fuzzy-match t)
>> >
>> > According to https://github.com/emacs-helm/helm/wiki/Fuzzy-matching, the
>> > make fuzzy matching work everywhere.
>> >
>> > It seems to work ok for me.
>> >
>> > aaermolov@gmail.com writes:
>> >
>> >> Hi,
>> >>
>> >> I've spotted a problem with completing Org tags with Helm. While Helm is
>> >> not at least about [fuzzy] completion, when I try to input any
>> >> completion subsequence, all candidates are gone and variants are
>> >> shrinked to whatever will be typed.
>> >>
>> >> Is there any widely known mistake, that I've stepped into with this
>> >> case? Can anyone point me to some comprehensible direction?
>> >>
>> >> regards,
>> >> Alex
>> >
>> >
>> > --
>> > Professor John Kitchin
>> > Doherty Hall A207F
>> > Department of Chemical Engineering
>> > Carnegie Mellon University
>> > Pittsburgh, PA 15213
>> > 412-268-7803
>> > @johnkitchin
>> > http://kitchingroup.cheme.cmu.edu
>>