Removing banner from mtxrun output
Greetings all! In the modernized version of the R module (attached), which makes it possible to interact with the statistical programming environment R (http://www.r-project.org), mtxrun is used to call R on saved buffers containing snippets of R code. Optionally, the output from R can be included using \typefile. However, mtxrun includes a banner in the output file, example below. MTXrun | MTXrun | executing: R -q --save --restore MTXrun | MTXrun |> sample(1:10) [1] 5 3 1 9 2 6 10 8 4 7
sample(1:10, replace=TRUE) [1] 1 10 7 1 7 5 2 2 10 5
If the results of the computations are to be included in the text, this banner is ugly and distracting. Question: is it possible to instruct mtxrun not to make such a banner? Or, alternatively, is there an easy way to remove it before the text is included? I am using MKIV exclusively. Regards, Johan -- Johan Sandblom, MD PhD m +46735521477 Sweden "What is wanted is not the will to believe, but the will to find out, which is the exact opposite" - Bertrand Russell
On Mon, Sep 20, 2010 at 10:26 AM, Johan Sandblom
Greetings all!
In the modernized version of the R module (attached), which makes it possible to interact with the statistical programming environment R (http://www.r-project.org), mtxrun is used to call R on saved buffers containing snippets of R code. Optionally, the output from R can be included using \typefile. However, mtxrun includes a banner in the output file, example below.
MTXrun | MTXrun | executing: R -q --save --restore MTXrun | MTXrun |> sample(1:10) [1] 5 3 1 9 2 6 10 8 4 7
sample(1:10, replace=TRUE) [1] 1 10 7 1 7 5 2 2 10 5
If the results of the computations are to be included in the text, this banner is ugly and distracting.
Question: is it possible to instruct mtxrun not to make such a banner? Or, alternatively, is there an easy way to remove it before the text is included?
I am using MKIV exclusively.
Regards, Johan A bit off-topic: do you know aditya module t-filters ? See http://meeting.contextgarden.net/2010/talks/2010-09-15-aditya-filters/
-- luigi
On Mon, 20 Sep 2010, luigi scarso wrote:
A bit off-topic: do you know aditya module t-filters ? See http://meeting.contextgarden.net/2010/talks/2010-09-15-aditya-filters/
Also see http://randomdeterminism.wordpress.com/2010/06/10/replacing-r-module/ Aditya
A bit off-topic: do you know aditya module t-filters ? See http://meeting.contextgarden.net/2010/talks/2010-09-15-aditya-filters/
I didn't, but I'll investigate! Thank you, Johan
-- luigi ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________
-- Johan Sandblom, MD PhD m +46735521477 Sweden "What is wanted is not the will to believe, but the will to find out, which is the exact opposite" - Bertrand Russell
The filters module works great! The R module should be abandoned in
favor of it. The only problem I have now is the problem with the
unwanted lines at the end of R output. Is there a way to instruct
\typefile to ignore the last four lines of the outputfile? I assume
defining an alternative command which does just that is not too
difficult, but sadly beyond me.
Johan
2010/9/20 Johan Sandblom
A bit off-topic: do you know aditya module t-filters ? See http://meeting.contextgarden.net/2010/talks/2010-09-15-aditya-filters/
I didn't, but I'll investigate!
Thank you, Johan
-- luigi ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________
-- Johan Sandblom, MD PhD m +46735521477 Sweden "What is wanted is not the will to believe, but the will to find out, which is the exact opposite" - Bertrand Russell
-- Johan Sandblom, MD PhD m +46735521477 Sweden "What is wanted is not the will to believe, but the will to find out, which is the exact opposite" - Bertrand Russell
On Tue, Sep 21 2010, Johan Sandblom wrote:
Is there a way to instruct \typefile to ignore the last four lines of the outputfile? I assume defining an alternative command which does just that is not too difficult, but sadly beyond me.
It seems, that it's not possible at the moment. There is a tracker item for this feature: http://tracker.luatex.org/view.php?id=475 Cheers, Peter -- Contact information: http://pmrb.free.fr/contact/
On Tue, 21 Sep 2010, Peter Münster wrote:
On Tue, Sep 21 2010, Johan Sandblom wrote:
Is there a way to instruct \typefile to ignore the last four lines of the outputfile? I assume defining an alternative command which does just that is not too difficult, but sadly beyond me.
It seems, that it's not possible at the moment. There is a tracker item for this feature: http://tracker.luatex.org/view.php?id=475
Not hard to do, just no interface. We can change buffers.range so that it accepts {1,-4} to mean from first to last-4 lines, and then just set \typefile[lines={1,-4}]{filename} In any case, it is easy to define a lua function that does this. \startluacode function buffers.stripfromend(name, count) local str = buffers.loaddata(name) if str and str~= "" then local lines = str:splitlines() local line, n, action = 0, 0, buffers.typeline local first, last, m = buffers.strip(lines) last = last - count m = m - count for i=first,last do n, line = action(lines[i], n, m, line) end end end \stopluacode \unprotect \def\strippedtyping#1#2% count file {\begingroup \tttf \let\obs\specialobeyedspace \ctxlua{buffers.stripfromend(\!!bs#2\!!es, #1)} \endgraf \endgroup} \protect \starttext \strippedtyping{4}{filename} \stoptext To use this with t-filter, you can just say readcommand=\strippedtyping{4}, Aditya
Perfect, thank you.
Johan
2010/9/21 Aditya Mahajan
On Tue, 21 Sep 2010, Peter Münster wrote:
On Tue, Sep 21 2010, Johan Sandblom wrote:
Is there a way to instruct \typefile to ignore the last four lines of the outputfile? I assume defining an alternative command which does just that is not too difficult, but sadly beyond me.
It seems, that it's not possible at the moment. There is a tracker item for this feature: http://tracker.luatex.org/view.php?id=475
Not hard to do, just no interface. We can change buffers.range so that it accepts {1,-4} to mean from first to last-4 lines, and then just set
\typefile[lines={1,-4}]{filename}
In any case, it is easy to define a lua function that does this.
\startluacode
function buffers.stripfromend(name, count) local str = buffers.loaddata(name) if str and str~= "" then local lines = str:splitlines() local line, n, action = 0, 0, buffers.typeline local first, last, m = buffers.strip(lines) last = last - count m = m - count for i=first,last do n, line = action(lines[i], n, m, line) end end end \stopluacode
\unprotect \def\strippedtyping#1#2% count file {\begingroup \tttf \let\obs\specialobeyedspace \ctxlua{buffers.stripfromend(\!!bs#2\!!es, #1)} \endgraf \endgroup} \protect
\starttext
\strippedtyping{4}{filename}
\stoptext
To use this with t-filter, you can just say
readcommand=\strippedtyping{4},
Aditya ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________
-- Johan Sandblom, MD PhD m +46735521477 Sweden "What is wanted is not the will to believe, but the will to find out, which is the exact opposite" - Bertrand Russell
On 21-9-2010 10:18, Aditya Mahajan wrote:
Not hard to do, just no interface. We can change buffers.range so that it accepts {1,-4} to mean from first to last-4 lines, and then just set
indeed a few lines patch ... will be in next beta \starttext \starttyping line 1 ! line 2 ! line 3 ! line 4 ! line 5 ! line 6 ! \stoptyping \starttyping[range={2,3}] line 1 line 2 ! line 3 ! line 4 line 5 line 6 \stoptyping \starttyping[range={2,-3}] line 1 line 2 ! line 3 ! line 4 line 5 line 6 \stoptyping \stoptext ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On Mon, 20 Sep 2010, Johan Sandblom wrote:
In the modernized version of the R module (attached), which makes it possible to interact with the statistical programming environment R (http://www.r-project.org), mtxrun is used to call R on saved buffers containing snippets of R code. Optionally, the output from R can be included using \typefile. However, mtxrun includes a banner in the output file, example below.
Why are you using input output redirection R .... < \Rfile\space > \Rfile.out rather than a simple R .... \Rfile\space \Rfile.out
MTXrun | MTXrun | executing: R -q --save --restore MTXrun | MTXrun |> sample(1:10) [1] 5 3 1 9 2 6 10 8 4 7
sample(1:10, replace=TRUE) [1] 1 10 7 1 7 5 2 2 10 5
See http://archive.contextgarden.net/message/20100611.141914.fd6567be.en.html but the banner still appears the first time mtxrun --ifchanged is run. @Hans, mtxrun be completely silent when used to launch external programs and only give some output if --verbose is passed. BTW, there is a serious bug in the R module: each code chunck carries a state with it (--save --restore), but this state can depend on which chunck was changed. Consider (I don't know the R syntax on top of my head, but this gives the idea of the bug) Chunk 1 \startR a = 10 ; b = 10 ; \stopR Chunk 2 \startR a = 20 ; b = 20 \stopR Chunck 3 \startR print(a+b) ; \stopR Run this once. The output will be 40. Now, add c = 10 ; in Chunk 1. Also change Chunk 3 to (add spaces so that md5 sum is different) print( a + b ) ; Now run the file again. Since Chunk 2 is not changed, it will not be run. Chunk 3 will see the state set by Chunk 1, and hence the output will be 20. This "bug" can be avoided by doing some book keeping for the external state (saving the file to a different name and restoring it), or by not using --save and --restore. Aditya
2010/9/20 Aditya Mahajan
On Mon, 20 Sep 2010, Johan Sandblom wrote:
Why are you using input output redirection
R .... < \Rfile\space > \Rfile.out
rather than a simple
R .... \Rfile\space \Rfile.out
Originally, I think it was because it was the only way that worked on both Windows and Linux, but lately it is because R CMD BATCH produces produces a couple of lines of processing times at the end, which I do not want.
@Hans, mtxrun be completely silent when used to launch external programs and only give some output if --verbose is passed.
That would make more sense.
BTW, there is a serious bug in the R module: each code chunck carries a state with it (--save --restore), but this state can depend on which chunck was changed.
Indeed. But I have learned to live with it.
This "bug" can be avoided by doing some book keeping for the external state (saving the file to a different name and restoring it), or by not using --save and --restore.
Yes. But the book keeping gets a little complicated and ugly since it needs to be done from within R (--save takes no arguments). Or possibly all chunks after an altered chunk should be run. Thanks for your comments, Johan
Aditya ___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________
-- Johan Sandblom, MD PhD m +46735521477 Sweden "What is wanted is not the will to believe, but the will to find out, which is the exact opposite" - Bertrand Russell
participants (5)
-
Aditya Mahajan -
Hans Hagen -
Johan Sandblom -
luigi scarso -
Peter Münster