[QGIS-trac] [Quantum GIS] #1497: colormap edge case handling wrong for negative valued rasters

[Quantum GIS]

#1497: colormap edge case handling wrong for negative valued rasters
------------------------------------------------+---------------------------
   Reporter:  Barryrowlingson                   |              Owner:  nobody       
       Type:  bug                               |             Status:  new          
   Priority:  major: does not work as expected  |          Milestone:  Version 1.0.1
  Component:  Rasters                           |            Version:  HEAD         
   Keywords:                                    |   Platform_version:               
   Platform:  Linux                             |           Must_fix:  No           
Status_info:  0                                 |  
------------------------------------------------+---------------------------
 I have a single-band floating-point raster. It's 20x20 so I can see the
 cells.

 From layer properties choose 'Colormap', on the Colormap tab use Discrete
 Color Interpolation, in generate new color map section choose 4 entries,
 equal interval color map, hit Classify, hit apply.

 The colormap values define inclusive upper values. For a positive-valued
 raster I get three areas of color and a single cell coloured in the shade
 of the smallest value, since that's the only cell which has the minimum
 value.

 But for a negative-valued raster I get three areas of colour and a blank
 cell at the highest value.

  Let's explain with some colormaps. My positive raster has a range of
 0.563 to 8.38, and I get this colormap:

 0.563: blue
 3.169: yellow
 5.775: green
 8.38 : red

 I see one blue cell (at the single pixel with the minimum value) and some
 yellow, green and red cells. All cells are shaded.

 With my negative raster (which is just (* -1 pos) in RasterLang) I get the
 following colormap:

 -8.38 : blue
 -5.775: yellow
 -3.169: green
 -0.563: red

  and I see one transparent cell, no blue cells, and some yellow, green and
 red cells. The transparent cell is at the -0.563 value location. See
 attachment.

 If things were symmetric with the positive raster, it should be red, but
 it doesn't get shaded.

 As it is, it seems that a positive floating point raster will always have
 cells of the single max colour done in one shade, and a negative fp raster
 will always have cells of the minimum colour set transparent...

 Also note the 'Exact' interpolation doesn't match any cells - I thought it
 might find equality with the min and max cells, but doesn't. Probably only
 useful for integer rasters.

-- 
Ticket URL: <https://trac.osgeo.org/qgis/ticket/1497>
Quantum GIS <http://qgis.org>
Quantum GIS is an Open Source GIS viewer/editor supporting OGR, PostGIS, and GRASS formats