[GRASS5] Re: [Pkg-grass-general] testing debiangis

Hamish hamish_nospam at yahoo.com
Wed Jan 19 16:32:57 EST 2005


> > - grass depends on xterm, but the package does not require this
> > dependency 
> 
> Probably an xterm | x-terminal-emulator dependency is better.
> Considering if it qualifies for a recommends is appropriate, too


as Markus noted, the following files have xterm hardcoded:
(my grep)

display/d.m/script/r.colors.rules
display/d.m/script/r.reclass.rules
display/d.m/script/r.recode.rules
display/d.m/d.m.tcl
gui/tcltkgrass/main/gui.tcl
gui/tcltkgrass/script/r.colors.rules
gui/tcltkgrass/script/r.reclass.rules
gui/tcltkgrass/script/r.recode.rules
lib/external/bwidget/entry.tcl
lib/init/epsg_option.tcl
lib/init/grass-run.src
lib/vask/V_init.c
scripts/r.mapcalculator/r.mapcalculator
scripts/r3.mapcalculator/r3.mapcalculator



Those are mostly using 'xterm -e ....'. As Glynn has pointed out before
(I believe), the problem with changing that to $TERM is that usage of
the "-e" flag is not standardized throughout all shells. We'd have to
maintain a database of terms and execute commands.. we already sort of
do that with web browsers for the help system, maybe it wouldn't be too
bad. In what cases would xterm not be installed? and what would be there
to replace it? (I'm talking about functionality, not preference of term)
Embedded/handheld distributions?



Hamish




More information about the grass-dev mailing list