[GRASS-dev] [GRASS GIS] #2409: last call for options keys consolidation

GRASS GIS trac at osgeo.org
Sun Nov 30 06:48:39 PST 2014


#2409: last call for options keys consolidation
----------------------------------+-----------------------------------------
 Reporter:  martinl               |       Owner:  grass-dev@…              
     Type:  task                  |      Status:  new                      
 Priority:  blocker               |   Milestone:  7.0.0                    
Component:  Default               |     Version:  unspecified              
 Keywords:  standardized options  |    Platform:  Unspecified              
      Cpu:  Unspecified           |  
----------------------------------+-----------------------------------------

Comment(by glynn):

 Replying to [comment:73 martinl]:

 > another option could to my mind. To use 'raster3d' and modify the parser
 in the way that it would support also 'rast' and 'rast3d' with appropriate
 warning

 I don't think that "rast" should generate a warning. We want the user to
 be able to abbreviate option names when running commands interactively. At
 the same time, it's preferable for scripts to avoid abbreviations, which
 they can't do if the actual option name is itself an abbreviation.

 Perhaps we should relax the rule relating to ambiguous abbreviations for
 the case where one option name is a prefix of another?

 At present, if a module has options named "raster" and "raster3d", "rast"
 (or any other abbreviation of "raster") will be rejected as ambiguous. The
 option-matching function has to explicitly check for the case where an
 option name is given in full, otherwise even the unabbreviated option name
 would be rejected as ambiguous.

 The change wouldn't be trivial; the function can no longer simply count
 matches, it has to record them. But as a side-effect of that, the error
 message for ambiguous options could list all of the matches.

 If we don't make that change, then we should try to avoid the situation
 where one option name is a prefix of another. In particular, that
 precludes "raster3d" or "rast3d".

-- 
Ticket URL: <http://trac.osgeo.org/grass/ticket/2409#comment:75>
GRASS GIS <http://grass.osgeo.org>



More information about the grass-dev mailing list