Glynn, thanks for your answer. But, for instance in GUI if a select a type of color table and then I decide to user a rule file, there is no blank option in "Type of color table". And if I choose rule, it gets an error. So, in this cases I just have to close r.colors window and open one again. This may sound a bit unclear .
But thanks Pedro > > I'm reading r.colors manual webpage ( > > http://grass.itc.it/grass64/manuals/html64_user/r.colors.html) and I > have a > > few questions related with defining new color tables. > > 1- At color (in Parameters) one of the options is rules. But if i select > > rules and insert a path to a rules file I get this error: > > *ERROR: "color", "rules", and "raster" options are mutually exclusive* > > > > * > > * > > > > Was this suppose to happen? > > Yes. If you specify a file for "rules", the "color" option should be > blank. > > [color=rules exists for compatibility with previous versions, and only > works from the command-line, not the GUI. > > 2- About color tables with absolute values (e.g. NDVI) if a NDVI pixel > has > > value between 2 defined values, which color does it get? > > It's interpolated. This is true whether the rules uses absolute values > or percentages (or a mix of both). > > > 3- About aspectcolr*.* To each category a color is assigned (e.g. white, > > yellow bla bla bla). Is there a list of possible colors to assign? > > The list of named colours is: > > white black red green blue yellow magenta cyan aqua grey gray > orange brown purple violet indigo > > You can mix named colours and r:g:b notation freely. > > > 4- Regarding assigning a rules.info to a map (as it's demonstrated in > the > > same manual page). There are two ways. How come r.colors can use, as an > > input, rules.info if it's stated before the r.colors statement. > > cat rules.file | r.colors map=threecats color=rules > > color=rules reads rules from stdin, which in the above example is the > contents of the rules.file via "cat". The following commands will > all achieve the same result: > > cat rules.file | r.colors map=threecats rules=- > r.colors map=threecats color=rules < rules.file > r.colors map=threecats rules=- < rules.file > r.colors map=threecats rules=rules.file > > For reading from a file, the last one is preferable (and is the only > one which will work from the GUI). Beyond that, using rules=- is > preferred to color=rules (apart from anything else, rules=- works in > 7.0 while color=rules doesn't; color=rules is only kept in 6.4 for > backwards compatibility). > > The use of "cat file | ..." rather than "... < file" can be easier to > read if you're creating a long pipeline in a script, as it places the > source file at the far left of the command. The following both have > the same effect: > > cat infile | cmd1 | cmd2 | ... | cmdN > outfile > > cmd1 < infile | cmd2 | ... | cmdN > outfile > > but the former is probbably clearer. > > > 5- One last question :) I tried to display the color table associated > with a > > raster map layer (d.colortable) but I get the following message: > > Command 'd.colortable' not yet implemented > > Odd; you can try d.legend instead, or use r.mapcalc to create a test > map to which you can assign the colour table. > > -- > Glynn Clements <gl...@gclements.plus.com> >
_______________________________________________ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user