Re: [rkward-devel] rkwardtests package: done :-)

2010-10-15 Thread Thomas Friedrichsmeier
Hi, On Thursday 14 October 2010, meik michalke wrote: well, i'm a bit surprised myself, but i'm done already. there are several new internal functions that hanlde the creation and removal of temp dirs, and as far as i can tell they work really fine :-) looks good to me, but I have only tested

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-15 Thread meik michalke
hi, Am Freitag, 15. Oktober 2010, um 11:36:49 schrieb Thomas Friedrichsmeier: looks good to me, but I have only tested on Linux so far, too. btw, i wonder whether these changes should already be documented in the wiki (automated plugin testing)? if someone started working on a new set of

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-14 Thread Thomas Friedrichsmeier
Hi, On Wednesday 13 October 2010, meik michalke wrote: well, something like path - tempfile(paste(rktests., su...@id, sep=)) if(!dir.create(path, recursive=TRUE)) { stop(simpleError(Couldn't create temporary directory!)) } else { assign(.rktest.temp.dir, path,

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-14 Thread meik michalke
hi, am Mittwoch 13 Oktober 2010 (20:02) schrieb Thomas Friedrichsmeier: indeed, using a temporary directory could be a clean solution. well, i'm a bit surprised myself, but i'm done already. there are several new internal functions that hanlde the creation and removal of temp dirs, and as far

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-13 Thread meik michalke
Am Dienstag, 12. Oktober 2010, um 20:17:56 schrieb Thomas Friedrichsmeier: Hm, it's certainly less dangerous, now. However, some tests/suites create files other than messages, output, and commands, esp. import_export_plugins.R, and may potentially be thrown off track, if those files already

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-12 Thread meik michalke
Am Donnerstag, 7. Oktober 2010, um 22:25:39 schrieb Thomas Friedrichsmeier: Still, it would be good (but not a high priority), if rktest.resetEnvironment() could be used to restore a normal environment after rktest.initializeEnvironment(). ok, i've tried something there: storing the status

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-12 Thread Thomas Friedrichsmeier
Hi, On Tuesday 12 October 2010, meik michalke wrote: ok, i've tried something there: storing the status quo in a temporal object and restoring it from there afterwards. yes, that's a clean solution. Great. also, i tried to prevent the cleanup function from doint too much damage by limiting

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-07 Thread meik michalke
hi, am Donnerstag 07 Oktober 2010 (18:30) schrieb Thomas Friedrichsmeier: Hm, do you have the file /usr/lib/R/bin/roxygen ? Possibly a problem with multiple installations of R? Check wether which R is the one you expect (this one keeps tricking me again and again...). i didn't install

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-07 Thread Prasenjit Kapat
On Thu, Oct 7, 2010 at 2:59 PM, meik michalke meik.micha...@uni-duesseldorf.de wrote: hi, am Donnerstag 07 Oktober 2010 (18:30) schrieb Thomas Friedrichsmeier: Hm, do you have the file   /usr/lib/R/bin/roxygen ? Possibly a problem with multiple installations of R? Check wether   which R is

Re: [rkward-devel] rkwardtests package: done :-)

2010-10-07 Thread Thomas Friedrichsmeier
Hi, On Thursday 07 October 2010, Thomas Friedrichsmeier wrote: You might consider assigning explicitly to globalenv() (as is done with date()), and then allowing to remove them with rktest.resetEnvironment, as before. The idea was to provide a convenient way to restore a mostly normal