[Qgis-developer] standardisation of the editing map tools: modify behaviour of press-pan-release tools

[Bernhard Ströbl]

Am 24.09.2014 15:23, schrieb Denis Rouzaud:
> Hi Richard,
>
> On 24.09.2014 14:00, Richard Duivenvoorde wrote:
>> On 24-09-14 10:56, Denis Rouzaud wrote:
>>
>>> So, the bottom line, any objection to changing the behaviour of:
>>> * edit node tool
>>> * move feature
>>> * rotate feature
>>> * move label
>>> * rotate label
>>> * any other press-pan-release map tool that I am not aware of
>>> ???
>> Hi Denis,
>>
>> though I'm not a hard core digitizer/edittool user, I think removing
>> inconsistency is always good!
>> But I would really wait for some input of a hardcore editing user.
> We are ;)
>>
>> What about writing per tool, the behaviour you propose in a QEP
>> (https://github.com/qgis/QGIS-Enhancement-Proposals) so we can discuss
>> it there?
> yeah, I wonder if this really needs one. It's quite a small change with
> not many consequences.

let me intervene here :-) I have been a GIS user (ESRI) at my first 
contact with CAD (Bentley Microstation in that case) about 10 years ago. 
At that time many tools needed two clicks in Microstation (first = 
select, second = accept selection, or right click = select other). 
Nowadays you can just select a feature with left click and holding down 
the left mouse button you move it by moving the mouse, just like in 
QGIS. The dedicated "move feature" tool still works the way you propose 
(left click = select, left click = drop). It was hard for me to get used 
to that, I am glad I did not do any serious digitizing in CAD.

Bottom line:
1) Intuitive is what you are used to.
2) The way it currently works in QGIS is IMHO more intuitive because I 
am used to it :) and it is the way graphic software works (I tried 
LibreOfficeDraw, Inkscape and GIMP for a start).
3) I wouldn't say this to be a "change with not many consequences".

IMHO QGIS should do it similar to other GIS packages (how is this solved 
in e.g. ArcGIS or MapInfo?) because users coming from similar software 
packages should feel at home, not those coming from CAD.

I am not saying the idea is bad but the consequences might be larger 
than you expect especially for occasional users if the behaviour of the 
tools is different from what they would expect from their prior 
experience (of other GIS or graphic software).

I could surely live with it (am not a hard-core digitizer either).

Bernhard

>>
>> One thing that I cannot fit in your proposal is how to do the multiple
>> selection of features (eg used for rotating and moving a group of
>> features). If you have one tool for both selecting features and
>> dragging/rotating them, when does the tool 'know' when you finished
>> selecting and start doing the action? (trying to answer myself: using
>> the ctrl and/or shift key??). You also miss the different selection
>> options... OR you make the selection via the selection tool optional: so
>> for one feature the 'selecting' is via a click, for a set of features
>> you go via the normal selection tool?
> Click means select features + activate the tool (e.g. rotate).
> You would still have the rectangle tool to select several features. Once
> you released the mouse button, all selected features they are directly
> activated.
> We can think of being able to add/remove features by shift or ctrl +
> click which are common modifiers to extend/modify selections.
>> Last thing: when moving (a set of features) without really 'dragging',
>> is the 'dragging' visible in your proposal? Or can you only see the
>> displacement/move after the second click. I hope I can preview stuff
>> during  my action, but I cannot fit that in your view of being able to
>> do other things then.
> Of course, you see the dragging. Right click will cancel, left click
> will validate.
>
> Doing other things, is mainly using intermediate points. For example,
> you need to move 10m north, 5m west.
> You will first move 10m north using an intermediate point (the left
> click will not be given to the map tool), then go 5m west (left click
> given to the map tool).
>>
>> Shorter version: as long as you make it more consistent and usable: go
>> :-)
> That's the idea. Or at least the goal ;)
>
> Cheers,
>
> Denis
>


__________ Information from ESET Mail Security, version of virus signature database 10458 (20140924) __________

The message was checked by ESET Mail Security.
http://www.eset.com