Possibly breaking changes in rgeos

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Possibly breaking changes in rgeos

Roger Bivand
Administrator
In this sf issue (https://github.com/r-spatial/sf/issues/822), it was
pointed out that error handling in rgeos (and sf) leaked memory at least
when invalid geometries triggered an exception, leading to an R error, but
without the GEOS objects being freed/destroyed. An attempt has been made
to address this problem in rgeos, and is available for testing from
R-Forge (version 0.4-1):

install.packages("rgeos", repos="http://R-Forge.R-project.org")

Please check whether this version (throwing the R error after freeing
memory in calling function rather than throwing the R error immediately on
exception) has missed any important points in your typical workflows - if
you find any, please write directly to me as maintainer with a small
reproducible example for adding to tests/leak_by_exception.R. If you find
anything, also consider using gdb and/or valgrind to backtrack to the
rgeos C function causing the mayhem.

Grateful for any reports,

Roger
(rgeos maintainer)

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
http://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Roger Bivand
Department of Economics
Norwegian School of Economics
Helleveien 30
N-5045 Bergen, Norway
Reply | Threaded
Open this post in threaded view
|

Re: Possibly breaking changes in rgeos

Roger Bivand
Administrator
I have completed reverse dependency checks using rgeos:

rgeos version: 0.4-1, (SVN revision 572)
  GEOS runtime version: 3.7.0rc2-CAPI-1.11.0 2644482d

without seeing any impacts on CRAN packages using rgeos. I can't check
development versions of these packages or packages not on CRAN - so please
do check your workflows as advised earlier before rgeos 0.4-1 is submitted
to CRAN.

Roger

On Tue, 4 Sep 2018, Roger Bivand wrote:

> In this sf issue (https://github.com/r-spatial/sf/issues/822), it was pointed
> out that error handling in rgeos (and sf) leaked memory at least when invalid
> geometries triggered an exception, leading to an R error, but without the
> GEOS objects being freed/destroyed. An attempt has been made to address this
> problem in rgeos, and is available for testing from R-Forge (version 0.4-1):
>
> install.packages("rgeos", repos="http://R-Forge.R-project.org")
>
> Please check whether this version (throwing the R error after freeing memory
> in calling function rather than throwing the R error immediately on
> exception) has missed any important points in your typical workflows - if you
> find any, please write directly to me as maintainer with a small reproducible
> example for adding to tests/leak_by_exception.R. If you find anything, also
> consider using gdb and/or valgrind to backtrack to the rgeos C function
> causing the mayhem.
>
> Grateful for any reports,
>
> Roger
> (rgeos maintainer)
>
>

--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
http://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Roger Bivand
Department of Economics
Norwegian School of Economics
Helleveien 30
N-5045 Bergen, Norway