[GRASS-dev] Re: [GRASS GIS] #520: r.terraflow/iostream does not respect STREAM_DIR option

GRASS GIS trac at osgeo.org
Wed Apr 8 05:17:28 EDT 2009


#520: r.terraflow/iostream does not respect STREAM_DIR option
---------------------+------------------------------------------------------
  Reporter:  mmetz   |       Owner:  grass-dev at lists.osgeo.org
      Type:  defect  |      Status:  new                      
  Priority:  major   |   Milestone:  6.4.0                    
 Component:  Raster  |     Version:  svn-develbranch6         
Resolution:          |    Keywords:                           
  Platform:  All     |         Cpu:  All                      
---------------------+------------------------------------------------------
Comment (by hamish):

 the bugfix part of it looks good:
 {{{
 -flowStream = new AMI_STREAM<waterWindowBaseType>("/var/tmp/flowStream");
 +flowStream = new AMI_STREAM<waterWindowBaseType>(streamdir->answer);
 }}}


 For the change of the default placement from /var/tmp/ to $MAPSET/.tmp/, I
 don't think using G_tempfile() in the option definition section is allowed
 (ie before G_gisinit() has run), as it could end in G_fatal_error() which
 requires that. Also the call as written would leave behind an empty
 tempfile each time you ran the module, even if just --help or --html-
 description.

 if it is changed away from /var/tmp/, docs would need updating too, but
 I'm not sure if it is a good idea to change the default behaviour in the
 stable branch.

 It has been left in a state "contrary to GRASS policy" due to the large
 file sizes. there was previous discussion on the mailing list about that
 some years ago.

 tip: if you increase the memory= option to (e.g.) 75% of your system RAM,
 the needed tempfile space goes down a lot.


 Hamish

-- 
Ticket URL: <https://trac.osgeo.org/grass/ticket/520#comment:5>
GRASS GIS <http://grass.osgeo.org>


More information about the grass-dev mailing list