[GRASS-dev] [GRASS GIS] #2067: PyGRASS grid calls g.region at import time
GRASS GIS
trac at osgeo.org
Mon Sep 9 20:33:02 PDT 2013
#2067: PyGRASS grid calls g.region at import time
--------------------------+-------------------------------------------------
Reporter: huhabla | Owner: grass-dev@…
Type: defect | Status: closed
Priority: blocker | Milestone: 7.0.0
Component: Python | Version: svn-trunk
Resolution: fixed | Keywords: wingrass, PyGRASS, Grid, Python, temporal framework, t.remove, t.rast.neighbors
Platform: MSWindows 7 | Cpu: Unspecified
--------------------------+-------------------------------------------------
Comment(by glynn):
Replying to [comment:4 zarch]:
> We are only run the module with the --interface-description and then
parsing the xml
Even that is too work on import.
An import should only execute function and class definitions, and variable
assignments where the value is "simple": no potentially-large loops, no
executing external commands, preferably no I/O (beyond other imports,
although even those should be deferred for third-party modules which don't
abide by these rules).
Don't assume that importing your module implies that any of the
functionality it provides will actually be used, let alone all of it. It's
often the case that A imports B with intent to use some of its
functionality, and B imports C because some other functionality provided
by B depends upon C, but A never actually makes use of that so it
shouldn't be paying a non-trivial price for it.
Also, if a module is only used by a few functions, consider making the
import local to the function(s) which actually use it, particularly for
modules which have significant dependency trees and which are often
otherwise unused.
--
Ticket URL: <http://trac.osgeo.org/grass/ticket/2067#comment:5>
GRASS GIS <http://grass.osgeo.org>
More information about the grass-dev
mailing list