[GRASS5] [bug #2597] (grass) Some binary files not cleaned

Glynn Clements glynn.clements at virgin.net
Wed Sep 1 19:54:50 EDT 2004


Request Tracker wrote:

> this bug's URL: http://intevation.de/rt/webrt?serial_num=2597

> Subject: Some binary files not cleaned

> GRASS Version: 5.7.0
> 
> While trying to package grass for Debian I get the following errors. 
> Basically this means these files are not removed by 'make distclean'.
> 
> pkg-source: cannot represent change to lib/proj/nad2bin: binary file contents changed

nad2bin is no longer built.

> dpkg-source: cannot represent change to tools/timer/main.o: binary file contents changed
> dpkg-source: cannot represent change to raster/r.le/r.le.setup/polytocell/ply_t_bmif.o: binary file contents changed
> dpkg-source: cannot represent change to raster/r.le/r.le.setup/polytocell/yadjust.o: binary file contents changed
> dpkg-source: cannot represent change to raster/r.le/r.le.setup/polytocell/bmf_to_cll.o: binary file contents changed
> dpkg-source: cannot represent change to raster/r.le/r.le.setup/polytocell/save_area.o: binary file contents changed
> dpkg-source: cannot represent change to raster/r.le/r.le.setup/polytocell/wrte_recrd.o: binary file contents changed
> dpkg-source: cannot represent change to raster/r.le/r.le.setup/polytocell/do_dots.o: binary file contents changed
> dpkg-source: cannot represent change to raster/r.le/r.le.setup/polytocell/find_area.o: binary file contents changed
> dpkg-source: cannot represent change to raster/r.le/r.le.setup/polytocell/do_line.o: binary file contents changed

Both of these are due to similar Makefile bugs. I.e. the default rule
changes to a subdirectory then builds something, but there isn't a
matching "clean" rule (the default "clean" rule only deletes object
files in the current directory).

I'll fix these specific cases, but ultimately there needs to be some
changes to the build system. Every directory with subdirectories
should define SUBDIRS, and the default clean rule should recurse into
those subdirectories.

-- 
Glynn Clements <glynn.clements at virgin.net>




More information about the grass-dev mailing list