Hi, This is now fixed at the low-level (C) interface level. The fix is part of the bugfix-release 2.2.3.
L. On 2011-09-22 08:12, Laurent Gautier wrote: > On 2011-09-22 00:08, Christian Marquardt wrote: >> Laurent, >> >> is there an example for such a custom cleaner somewhere, or a >> documentation on the interface? > > For rpy2, the documentation is currently the unit tests (this is > tested) and the (C) source (grep for "cleanup"). Since this is mostly > a wrapper around R, the R documentation for it may come handy if after > a more advance usage. > >> Also, is it possible to obtain the path to the temporary R >> directory on the python side in order not to delete temporary >> directories of other, parallel R sessions? > > This is an R question. I think that it is returned by the function > tempdir(). > >> I've come across the same issue with several recent versions of R >> and rpy2, though I haven't checked the most recent one yet. >> >> Thanks, >> >> Chris. >> >> >> >> On 21 Sep 2011, at 23:17, Laurent Gautier<lgaut...@gmail.com> wrote: >> >>> R itself _does_ create a temp directory each time it starts. >>> >>> Rpy2 has placeholders for custom callback cleaners, and it might >>> interfere with the default cleanup made by an R console; this is >>> currently a little-used (and little-documented) features and I can't >>> tell with looking more into it. You could quickly fix it by writing a >>> custom cleaner that does delete that directory. >>> >>> L. >>> >>> PS: The current release for rpy2 is 2.2.2. You do want to upgrade. >>> >>> >>> >>> On 2011-09-21 17:52, Christian Hudon wrote: >>>> Hi, >>>> >>>> I'm running a server process that uses rpy2, on a VM with limited disk >>>> space. Every time the process is restarted, a Rtmp directory is >>>> created >>>> under /tmp. (Example name for the directory: "RtmpFThW0A"... this >>>> looks >>>> similar to the result of calling the R tempdir() function.) This is a >>>> problem for me, as said temporary directories accumulate and slowly >>>> fill >>>> up the disk space. >>>> >>>> I've narrowed it down to the following simple Python code, which will >>>> create said temporary directory for me (with rpy2 2.1.3): >>>> >>>> import rpy2.rinterface >>>> rpy2.rinterface.initr() >>>> >>>> However, I can't find what's causing the temporary directory to be >>>> created in the rpy2 source code. (Starting R by itself doesn't >>>> create a >>>> temporary directory.) >>>> >>>> I have a couple of questions... Is this directory created by rpy2? If >>>> so, where? Is it possible to disable its creation, or at least >>>> clean it >>>> up on exit? If not, can I remove said tempdir without interfering >>>> with rpy2? >>>> >>>> Thanks in advance, >>>> >>>> Christian >>>> >>>> >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> >>>> >>>> All the data continuously generated in your IT infrastructure >>>> contains a >>>> definitive record of customers, application performance, security >>>> threats, fraudulent activity and more. Splunk takes this data and >>>> makes >>>> sense of it. Business sense. IT sense. Common sense. >>>> http://p.sf.net/sfu/splunk-d2dcopy1 >>>> _______________________________________________ >>>> rpy-list mailing list >>>> rpy-list@lists.sourceforge.net >>>> https://lists.sourceforge.net/lists/listinfo/rpy-list >>> >>> ------------------------------------------------------------------------------ >>> >>> >>> All the data continuously generated in your IT infrastructure >>> contains a >>> definitive record of customers, application performance, security >>> threats, fraudulent activity and more. Splunk takes this data and makes >>> sense of it. Business sense. IT sense. Common sense. >>> http://p.sf.net/sfu/splunk-d2dcopy1 >>> _______________________________________________ >>> rpy-list mailing list >>> rpy-list@lists.sourceforge.net >>> https://lists.sourceforge.net/lists/listinfo/rpy-list >> ------------------------------------------------------------------------------ >> >> >> All the data continuously generated in your IT infrastructure contains a >> definitive record of customers, application performance, security >> threats, fraudulent activity and more. Splunk takes this data and makes >> sense of it. Business sense. IT sense. Common sense. >> http://p.sf.net/sfu/splunk-d2dcopy1 >> _______________________________________________ >> rpy-list mailing list >> rpy-list@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/rpy-list > ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2dcopy2 _______________________________________________ rpy-list mailing list rpy-list@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rpy-list
