[GRASS-dev] [GRASS GIS] #3767: A better i/o system for the r.li family

[GRASS GIS]

#3767: A better i/o system for the r.li family
---------------------------------------------+-------------------------
 Reporter:  SteGobbi                         |      Owner:  grass-dev@…
     Type:  enhancement                      |     Status:  new
 Priority:  normal                           |  Milestone:  7.8.0
Component:  Imagery                          |    Version:  unspecified
 Keywords:  Landscape analysis, r.li, patch  |        CPU:  All
 Platform:  All                              |
---------------------------------------------+-------------------------
 Hello,
 in my research activity i'm currently working with the "suite" r.li.
 The output given by these commands is different if in the setup
 (g.gui.rlisetup) we chose the "moving window" or the "whole map" option.
 In the first case GRASS will write into the
 Users\user1\AppData\Roaming\GRASS7\rli subfolder a very simple parseable
 file containing the results of the landscape analysis. A first improvement
 would be to modify the daemon.h file of the r.li family and allow GRASS to
 write this file in the group folder of the raster image we are studying.
 Which raises another issue: should the daemon be allowed to create a group
 folder in the current location containing the ref to the analysed map and
 the output file of the r.li module (in case the user hasn't set up a group
 folder)?

 The second case is: whole map is set as count method.
 In this case a raster map is created which structure depends on the r.li
 module we are working on, but the file containing the numerical results is
 not generated. Another enhancement would be to generate anyway this file,
 since the landscape analysis is supposed to give as output a numerical
 index.

 I would propose these two improvements for these modules.

-- 
Ticket URL: <https://trac.osgeo.org/grass/ticket/3767>
GRASS GIS <https://grass.osgeo.org>