http://blog.gmane.org/gmane.emacs.ess.general hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://comments.gmane.org/gmane.emacs.ess.general/7383 <pre>... another suggestion: If you debug a function foo() in main.R, a buffer (say, foo.R) showing the source of foo() is opened. When leaving debug mode via 'Q', I most certainly would like to leave foo.R and go back to main.R. Since foo.R is automatically opened on debug, I'd expect it to be closed automatically as well. Cheers, Marius ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/ess-help </pre> Marius Hofert 2013-06-18T21:17:48 http://comments.gmane.org/gmane.emacs.ess.general/7382 <pre>Hi, everytime I debug a function (latest ESS version), I get a buffer which shows the current line (highlighted) and the R buffer ("Browse[.]&gt;"). However, the point is always thrown in former buffer and thus requires me to manually put it in the R buffer... Can you change this behavior, please? [it's really annoying] Cheers, Marius ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/ess-help </pre> Marius Hofert 2013-06-18T21:03:16 http://comments.gmane.org/gmane.emacs.ess.general/7378 <pre>Loops no longer executed properly in a browser() session within ESS, when the loop includes curly braces {}. This does not happen when using the R console directly, but happens with both R-3.0.1 and R-2.15.1. See sessionInfo() and $dpkg -l ess at the end of this message. + browser() + for (i in 1:n) ##normal + print(i) + i + for (j in 1:n){ ##broken + print(j) + } + j + for (k in 1:n){print(k)} + k } Called from: testFun(3) Browse[1]&gt; debug at /tmp/ESS-region/test.R&lt; at &gt;2#3: for (i in 1:n) print(i) Browse[2]&gt; for (i in 1:n) ##normal + print(i) [1] 1 [1] 2 [1] 3 Browse[2]&gt; i [1] 3 Browse[2]&gt; for (j in 1:n){ ##broken + print(j) + } debug at /tmp/ESS-region/test.R&lt; at &gt;2#3: j Browse[3]&gt; j NULL Browse[3]&gt; for (k in 1:n){print(k)} debug at /tmp/ESS-region/findDupSamplesPairwise.R&lt; at &gt;2#3: k Browse[4]&gt; k NULL Browse[4]&gt; sessionInfo() R version 3.0.1 (2013-05-16) Platform: x86_64-unknown-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.UTF-8 </pre> Levi Waldron 2013-06-18T00:36:57 http://comments.gmane.org/gmane.emacs.ess.general/7370 <pre>Hi all, the problem I had with 'visible' seems to continue. When I return a table in orgmode with a :session header argument set, I see the visible column again. I report here as I think this is the better place to report this, but feel free to (tell me to) take this over to the orgmode mailing list. Here is an example orgmode file, that demonstrates my problem: --8&lt;---------------cut here---------------start-------------&gt;8--- * Test #+begin_src R :colnames yes :results table :var origin="dtrn" names &lt;- c("drtn", "uiae", "xvlc") values &lt;- 1:3 cbind(name=names, val=values) #+end_src #+results: | name | val | |------+-----| | drtn | 1 | | uiae | 2 | | xvlc | 3 | #+begin_src R :session *R* :colnames yes :results table :var origin="dtrn" names &lt;- c("drtn", "uiae", "xvlc") values &lt;- 1:3 cbind(name=names, val=values) #+end_src #+results: | value.name | value.val | visible | |------------+-----------+---------| | drtn | 1 | TRUE | | uiae | 2 | TRUE | | x</pre> Andreas Leha 2013-06-17T11:15:02 http://comments.gmane.org/gmane.emacs.ess.general/7365 <pre>I would like yo point you to this question in stackoverflow that I believe is an ESS problem: http://stackoverflow.com/questions/16901564/ess-does-not-deal-well-with-data-table-and-knitr Thanks, Rodrigo Belo [[alternative HTML version deleted]] ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/ess-help </pre> Rodrigo Belo 2013-06-14T18:05:05 http://comments.gmane.org/gmane.emacs.ess.general/7357 <pre>If I have an .R file written using another editor, how do I re-indent it, following the ESS "rules". Is there an easy way to replace = assignments with &lt;-? Best, Kasper [[alternative HTML version deleted]] ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/ess-help </pre> Kasper Daniel Hansen 2013-06-14T08:40:45 http://comments.gmane.org/gmane.emacs.ess.general/7354 <pre>Hi all, I have a strange issue with ess-eval-buffer and directlabels. The example below works as expected (i.e. produces a plot) when I send the code line-by-line or region-by-region, but it does fail when I send the whole buffer via C-c C-b (ess-eval-buffer). This is really bad, as that also means that my orgmode src block fails to produce a result. --8&lt;---------------cut here---------------start-------------&gt;8--- library("ggplot2") library("directlabels") haha &lt;- data.frame(x=1:10, y=1:10, c=rep(c("a", "b"),each=5)) myplot &lt;- ggplot(haha, aes(x=x, y=y, colour=c)) + geom_line() direct.label(myplot) --8&lt;---------------cut here---------------end---------------&gt;8--- Regards, Andreas PS: ess-version : 13.05-1 [&lt;unknown&gt;] ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/ess-help </pre> Andreas Leha 2013-06-13T21:08:39 http://comments.gmane.org/gmane.emacs.ess.general/7340 <pre>I believe the quit statement should be indented like the run statement. Attached is the patch. Hopefully this gets incorporated. diff --git a/lisp/ess-sas-l.el b/lisp/ess-sas-l.el index 912e469..b9ff685 100644 --- a/lisp/ess-sas-l.el +++ b/lisp/ess-sas-l.el &lt; at &gt;&lt; at &gt; -811,7 +811,7 &lt; at &gt;&lt; at &gt; number." (back-to-indentation) (or (bobp) (looking-at - "data[ ;]\\|proc[ ;]\\|run[ ;]\\|endsas[ ;]\\|g?options[ ;]\\|%macro[ ;]\\|%mend[ ;]"))) + "data[ ;]\\|proc[ ;]\\|run[ ;]\\|quit[ ;]\\|endsas[ ;]\\|g?options[ ;]\\|%macro[ ;]\\|%mend[ ;]"))) ;; Case where current statement is DATA, PROC, etc... (setq prev-end (point)) (goto-char (point-min)) </pre> Vinh Nguyen 2013-06-12T20:05:43 http://comments.gmane.org/gmane.emacs.ess.general/7332 <pre>Using system.time with a long-running (~ 30 - 60 minutes) command did not produce any timing information. &gt; system.time(imp &lt;- mice(sp2a, m=1, maxit=1, method=ini$method, pred=ini$pred, seed=1776)) iter imp variable 1 1 pAnyUAVI HIVTest &gt; The command produced the output shown while it ran. The little progress indicator on the status bar twirled throughout, or at least every time I checked. Does ESS give up looking for output after awhile? This was after using tracebug and ess developer, though I had removed all breakpoints before the run shown above. ESS git as of May 21, 56e277b089d3b780d766c4f0805fef9e37a91022. Windows 7, emacs 24.2.1, R 2.15.3. Ross Boylan ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/ess-help </pre> Ross Boylan 2013-06-06T18:56:26 http://comments.gmane.org/gmane.emacs.ess.general/7321 <pre>Hi I see that there are two repos for ESS - which one is the master repo one should choose to get the actual development version from, or ae they synced regularly? Thanks, Rainer </pre> Rainer M Krug 2013-06-05T14:46:49 http://comments.gmane.org/gmane.emacs.ess.general/7320 <pre> Hello, I am running R and Splus from ESS for years on a Redhat Linux server, thanks to Vitalie Spinu a few wrinkles when running Splus were ironed out not too long ago. I just installed TERR, TIBCO Enterprise Runtime R and RevolutionAnalytics' Community R. Standard R and Splus are root installed in standard locations. TERR and Revo64 are non-root installations in a directory where I have write permission. My question: how do I modify my ".emacs" file so I can run TERR and Revo64 from ESS? Thanks for any support, Gérald (Embedded image moved to file: pic21153.gif) Gerald Jean, M. Sc. en statistiques Conseiller senior en statistiques Lévis (siège social) </pre> gerald.jean< at >dgag.ca 2013-06-05T13:46:56 http://comments.gmane.org/gmane.emacs.ess.general/7303 <pre>Hi I am using org-mode for programming, therefore editing my .R files directly is futile, as the edits will be overwritten by the next tangle. Nevertheless, it happens to me from time to time, mainly when I am debugging and open the .R file from the ess-traceback buffer. Would it be possible, to have an option to open the .R file from the ess-tracebak buffer as read-only? Or is there another way of achieving this? Thanks Rainer </pre> Rainer M Krug 2013-06-03T11:53:00 http://comments.gmane.org/gmane.emacs.ess.general/7302 <pre>Please ignore </pre> Rainer M Krug 2013-06-05T07:12:22 http://comments.gmane.org/gmane.emacs.ess.general/7301 <pre>OK - next try. Hi I am debugging several packages at the same time, and when an error occurs I would like to open the *.R file. Is there a way of specifying a "search path" for .R files, so that simply clicking on the file in the ess-traceback opens the file? At the moment I have to open them manually and as for some to me unknown reason auto revert does not work, I have to enter "yes" each time the source code is changed in org and tangled. (golbal-auto-revert-mode is enabled) Cheers, Rainer </pre> Rainer M Krug 2013-06-03T11:07:21 http://comments.gmane.org/gmane.emacs.ess.general/7300 <pre>Sorry for the noise - I have the feeling my emails dont arrive at the mailing list. Cheers, Rainer </pre> Rainer M Krug 2013-06-05T06:31:54 http://comments.gmane.org/gmane.emacs.ess.general/7298 <pre>Hi List, hi Henning, I would like to ask whether there is a switch or configuration such that when a long line of roxy documentation is 'fill'ed it creates a new line starting with "##' ". Currently, with my configuration it produces: "##", which I find quite annoying. Thanks for the tips! Brian My truncated configuration: ... ;; Roxy-conf (setq ess-roxy-fill-param-p t) (setq ess-roxy-template-alist (list (cons "description" "") (cons "details" "") (cons "param" "") (cons "return" "") (cons "examples" ""))) ... (load "~/.emacs.d/ess-svn/lisp/ess-site.elc") (add-hook 'ess-mode-hook 'turn-on-auto-fill) ... ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/ess-help </pre> Brian 2013-06-04T13:48:03 http://comments.gmane.org/gmane.emacs.ess.general/7295 <pre>Hi all, I am experiencing strange interactions between orgmode and ess-developer-mode. This is a test file with a test result: --8&lt;---------------cut here---------------start-------------&gt;8--- * Test #+begin_src R :session *test* :colnames yes data.frame(a=1:3, b=1:3) #+end_src #+results: | a | b | |---+---| | 1 | 1 | | 2 | 2 | | 3 | 3 | --8&lt;---------------cut here---------------end---------------&gt;8--- And here is the test file, when ess-developer-mode is activated in *test*: --8&lt;---------------cut here---------------start-------------&gt;8--- * Test #+begin_src R :session *test* :colnames yes data.frame(a=1:3, b=1:3) #+end_src #+results: | value.a | value.b | visible | |---------+---------+---------| | 1 | 1 | TRUE | | 2 | 2 | TRUE | | 3 | 3 | TRUE | --8&lt;---------------cut here---------------end---------------&gt;8--- Is this a known issue? Best, Andreas ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https</pre> Andreas Leha 2013-06-03T21:53:36 http://comments.gmane.org/gmane.emacs.ess.general/7292 <pre>Over the years, my emacs init has used several complicated schemes to get the shortcut S-Return to send lines into the *R* session. 2 weeks ago, Vitalie said this would work with ESS 13.05, and I just tried it. Works great with Emacs 24.3. ;;Vitalie Spinu ESS-help email 2013-05-15 (define-key ess-mode-map [(control return)] nil) (define-key ess-mode-map [(shift return)] 'ess-eval-region-or-line-and-step) Awesome. Now CUA mode C-Return is safe for rectangular selection. This shortens the startup file by about 25 lines. Thanks again, I really appreciate the help. pj </pre> Paul Johnson 2013-06-03T03:54:44 http://comments.gmane.org/gmane.emacs.ess.general/7280 <pre>After I updated to 13.05 on linux I am having two problems. First, when I start an R console with the R icon things seem to work but when I click on a line of R code and click on the icon to send the line to the console, I get "Searching for program: no such file or directory, Splus". Secondly, I am getting a larger number of spaces with auto-indenting than before. How do I revert to the old behavior on both issues? Thanks Frank Harrell ______________________________________________ ESS-help&lt; at &gt;r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/ess-help </pre> Frank Harrell 2013-06-02T12:56:20 http://comments.gmane.org/gmane.emacs.ess.general/7275 <pre>Sorry if this is already posted, but I could not find it. Hi I am writing a simulation and am thinking of using a Reference Class But I tried a simple example to set a breakpoint in a method in a class, and it was not working (M-t b) as I did not see the methods in the class xx (from the ReferenceClasses help). So my question: is it possible to debug methods in reference classes with breakpoints et al? Thanks, Rainer </pre> Rainer M. Krug 2013-05-31T09:17:01 http://comments.gmane.org/gmane.emacs.ess.general/7274 <pre>Sorry if this has been posted, but I could not find it. Hi I am using at the moment devtools and the function load_all() instead of library() to load the packages on which I am working on. Now I saw ESS developer and it looks like a very handy tool in package development. But I have one problem: when I add a package (say asm) to the development list, it gets loaded by a executing library(asm). But I am editing in the package directory of asm and have no package asm installed which can be loaded (actually, this is not true, but the version installed is old and I could actually delete it...). Is there any way that I can set the command executed in R to load_all("/packages/asm")? And this is where it gets tricky - the path. Any suggestions? Thanks, Rainer </pre> Rainer M. Krug 2013-05-31T09:34:56 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.emacs.ess.general