[Qgis-developer] Symbolizing a discrete or a boolean raster

[Stéphane Henriod]

Hi again

it seems that the requirements are related (although not 100% similar) to
this other open (and assigned) ticket: http://hub.qgis.org/issues/14449

I really hope to see this very needed feature in a coming version of QGIS
and happy to help wherever I can!

Cheers

Stéphane


Le mardi 10 janvier 2017, Stéphane Henriod <stephanebek at gmail.com> a écrit :

> Hi
>
> let me move here a request that has been shortly discussed on the
> user-mailing list:
> http://lists.osgeo.org/pipermail/qgis-user/2017-January/038793.html
> http://lists.osgeo.org/pipermail/qgis-user/2017-January/038795.html
> http://lists.osgeo.org/pipermail/qgis-user/2017-January/038801.html
>
> and for which a ticket is open:
> http://hub.qgis.org/issues/14845
>
> I was said that some work has been done in this direction for 3.0 but I
> don't know if it completely covers the feature request
>
> ----------------------
>
> When styling a discrete or boolean raster, there seems to be no direct
> method, only a few workarounds. The feature I dream of would do the
> following:
>
>    - A new render type "Discrete raster"
>    - One button allows to fetch all unique values form the raster (with a
>    warning message if there are more than XXX unique values)
>    - A color ramp is applied by default
>    - I can multi-select values and apply them the same style and / or the
>    same label
>    - I can remove values. The pixels with those values will simply not be
>    displayed
>    - I can save my color map and load an existing color map
>
> So pretty much like the ArcGIS implementation:
> http://resources.arcgis.com/en/help/main/10.1/index.html#/
> representing_unique_categories_such_as_land_use/009t00000074000000/
>
> The current limitations I see (in 2.14):
>
>    - The discrete values must be entered manually, which is a pain for a
>    raster with many unique values (or if I don't know in advance what all
>    values are)
>    - If I click on *Classify*, QGIS will choose X values (based on the
>    number of classes and the classification method (continuous or equal
>    interval)), which no longer match the "exact" values of my raster. It
>    doesn't really make sense to classify a discrete rendering
>    - Although it is currently possible to style discrete rasters, the
>    procedure is not really intuitive for beginners
>
> ----------------------
>
> Is this a feature that is envisioned for 3.0? Is there anything I can do
> to support it? (except coding it myself, since I don't have the skills for
> that...)
>
> Thanks a lot in advance
>
> Cheers
>
> Stéphane
>
>

-- 

“When you travel, remember that a foreign country is not designed to make
you comfortable. It is designed to make its own people comfortable." --
Clifton Fadiman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-developer/attachments/20170111/9d9d2898/attachment.html>