[KPhotoAlbum] Token setting in thumbnail view

[Robert Krawitz]

On Sat, 03 Aug 2013 15:16:17 +0200, Johannes Zarl wrote:
> This is a multi-part message in MIME format.
>
> --===============5423907013158183624==
> Content-Type: multipart/alternative; boundary="nextPart2078532.fMp2JVumiN"
> Content-Transfer-Encoding: 7Bit
>
> This is a multi-part message in MIME format.
>
> --nextPart2078532.fMp2JVumiN
> Content-Transfer-Encoding: 7Bit
> Content-Type: text/plain; charset="us-ascii"
>
> Hi,
>
> On Friday 02 August 2013 22:07:32 Robert Krawitz wrote:
>> The first two steps in my workflow are:
>> 
>> 1) Eliminate the bad frames. [...]
>> 
>>    What I actually do is run quickly through the frames, applying a "k"
>>    token to each.  Prior to the uppercase token, I had to refrain from
>>    stacking the images at first.  That means I have 2000 images to go
>>    through, and have to hit 'k' 400 times.  With uppercase tokens, I can
>>    prestack the images, and run through 1000 images and hit 'K' 200
>>    times.  The caps lock key is definitely good for *something*.
>> 
>>    After this, I delete everything tagged 'k'.
>
> How about adding "Add stacked images to current view"?
>
> This way you wouldn't need to tag the whole stack, but could just add the "missing" images after tagging.

That would sound a bit more complicated.

>> 2) Take a second pass through everything to select the frames I want to
>>    use.  This is a subset of the first pass frames -- how big of a
>>    subset depends.  Sometimes it might be as little as 10%, sometimes it
>>    might be 90%.  But having seen everything in the first pass makes it
>>    easier for me to cut things down in the second pass, since I have an
>>    idea what other frames I have that have what I want.
>> 
>>    Again, having the uppercase tags mean that I can justt tag everything
>>    I want 'A' once.
>
> Admittedly, this pass would need some more steps with the proposed "add stacked images". You'd tag one image of the stack in one pass, select all tagged images, add the stacked images, expand all stacks, and tag all images again.

I'd rather not add unnecessary steps.

> Another solution could be to add some sort of selector to the gui so that you can change the behaviour as needed. The selector would then have the states "affect visible", "affect stack in view", and "affect whole stack". The difference between the last two would be whether only those images in the stack that are also in the current view are affected by tagging, or if one really wants to affect the whole stack.

When I'm viewing photos, I'm in full screen mode, so there's nothing to
tell me how that's set.

>> It should be very apparent that I do *not* want to be prompted when I
>> try to tag a stacked image.  I want to tag the entire stack, no
>> questions asked.
>
> I still think there should be a warning message so that newbies are made aware of the fact that KPA might behave differently than they think. Obviously, most people would disable the warning after it has fulfilled its purpose.

If it's a one time warning, OK.

> All in all I'd say KPA is already vim-like enough(*). That's why my first question on the patch was how we could document this in a way that people actually get it. Also, I think Miika has a point when he says that the behaviour with the capital letter should ideally be the same everywhere in KPA.

Ideally it should, yes.

> (*) Note that I'm a big fan of vim myself and use it everyday. I just think that nobody can ever hope to truly master all aspects of it.

Sorry, I use emacs.

-- 
Robert Krawitz                                     <rlk at alum.mit.edu>

MIT VI-3 1987 - Congrats MIT Engineers 5 straight men's hoops tourney
Tall Clubs International  --  http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom  --  http://ProgFree.org
Project lead for Gutenprint   --    http://gimp-print.sourceforge.net

"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton