Re: Get Grades Done: the joys of Org's simple power
Oh, my. I would not have imagined that setting up Emacspeak could be so complicated. Was your initial thought that if you could export the mechanisms in your assignment to interactive HTML (of whatever form) then you could let the browsers and their various accessibility APIs (of which I should confess right now I have almost no knowledge) to present that interactive content in whatever way the consumer of said content has requested? I'm far from being an expert, but I've spent my share of time hacking around and I think there are a variety of ways that one could embed something like JavaScript (I know, not an ideal choice of languages) in a variety of ways at several stages in the composition-to-export process that could render the output "live" if it was sufficiently simple. And while I'm especially bad at this part, a long time ago I had limited success wrapping some transformations in lisp within Org-mode and I've seen others perform similar trickery with greater success. On Fri, Jun 12, 2020 at 6:22 PM Devin Prater wrote: > Well, some teachers are blind, which means Emacspeak, and Spacemax does > have visual stuff, so Emacspeak may not work well with that. I’ll have to > try Doom Emacs though, maybe that’ll work better. The bigger problem > though, is that Emacspeak relies on speech servers, and the one for Windows > hasn’t been updated in… quite a while. I just want not only myself and > technically inclined sighted teachers to be able to access this, plus its > good to have accessible teaching tools no matter what, because you never > know when another blind person may want to use it later on. > > On Jun 12, 2020, at 6:23 PM, Phil Regier wrote: > > A friend showed me Org-mode running in spacemacs a few years back, and I > was pretty impressed with how well it seemed to be working, though I > haven't messed with it much myself. Especially not sure how much sugar it > offers as far as sharing a particular experience with new users, but at the > very least you should be able to include basic usage instructions within > the Org file? > > On Wed, Jun 10, 2020 at 9:56 PM George Mauer wrote: > >> You know...I believe some people have gotten emacs running in browser... >> You could do it by compiling it to wasm. So in theory you could create a >> completely in-browser emacs which is optimized primarily for org mode usage. >> >> Would be kind of an awesome thing for someone to tackle as it would >> greatly increase the reach of org. Not easy though. Could probably be a >> whole thesis project. >> >> Not sure how well it would work with screen readers and other >> accessibility tech though. That would be even more work >> >> On Wed, Jun 10, 2020, 10:24 PM Russell Adams >> wrote: >> >>> On Wed, Jun 10, 2020 at 03:38:43PM -0500, Devin Prater wrote: >>> > Now, I do wish I could share these “self-grading” performance tests >>> with >>> > others. I’ve tried exporting one to HTML, but the grade doesn’t seem >>> to update >>> > automatically like it does in Org-mode. >>> >>> Unfortunately updating the count is performed by a hook in Org when you >>> use C-c >>> C-c to check/uncheck a box. That information is static in the text, and >>> static >>> in html. >>> >>> I'm not aware of a built-in way to handle that case. Sorry. >>> >>> -- >>> Russell Adamsrlad...@adamsinfoserv.com >>> >>> PGP Key ID: 0x1160DCB3 http://www.adamsinfoserv.com/ >>> >>> Fingerprint:1723 D8CA 4280 1EC9 557F 66E8 1154 E018 1160 DCB3 >>> >>> >
Re: Get Grades Done: the joys of Org's simple power
A friend showed me Org-mode running in spacemacs a few years back, and I was pretty impressed with how well it seemed to be working, though I haven't messed with it much myself. Especially not sure how much sugar it offers as far as sharing a particular experience with new users, but at the very least you should be able to include basic usage instructions within the Org file? On Wed, Jun 10, 2020 at 9:56 PM George Mauer wrote: > You know...I believe some people have gotten emacs running in browser... > You could do it by compiling it to wasm. So in theory you could create a > completely in-browser emacs which is optimized primarily for org mode usage. > > Would be kind of an awesome thing for someone to tackle as it would > greatly increase the reach of org. Not easy though. Could probably be a > whole thesis project. > > Not sure how well it would work with screen readers and other > accessibility tech though. That would be even more work > > On Wed, Jun 10, 2020, 10:24 PM Russell Adams > wrote: > >> On Wed, Jun 10, 2020 at 03:38:43PM -0500, Devin Prater wrote: >> > Now, I do wish I could share these “self-grading” performance tests with >> > others. I’ve tried exporting one to HTML, but the grade doesn’t seem to >> update >> > automatically like it does in Org-mode. >> >> Unfortunately updating the count is performed by a hook in Org when you >> use C-c >> C-c to check/uncheck a box. That information is static in the text, and >> static >> in html. >> >> I'm not aware of a built-in way to handle that case. Sorry. >> >> -- >> Russell Adamsrlad...@adamsinfoserv.com >> >> PGP Key ID: 0x1160DCB3 http://www.adamsinfoserv.com/ >> >> Fingerprint:1723 D8CA 4280 1EC9 557F 66E8 1154 E018 1160 DCB3 >> >>
Re: [O] org-agenda-todo obliterates my windows
Well, that sure was easy, and I certainly had missed it. This selection does indeed produce the behavior I was hoping for; thanks so much for the help! PR On Thu, Aug 29, 2019 at 12:17 PM Jack Kamm wrote: > > Is there any way to just open my org-agenda-todo items in the /current/ > > window, or even just any window that already happens to be open, without > > modifying, splitting, or above all deleting any others? > > Does customizing the variable org-agenda-window-setup (e.g. by setting > it to 'current-window) solve this? >
[O] org-agenda-todo obliterates my windows
I've been ignoring this for a long time, but it's getting more disruptive as more of my daily life lives in emacs/org; I've tried a little blind hacking and only managed to make things worse, and I feel like I'm missing something easy and obvious. Is there any way to just open my org-agenda-todo items in the /current/ window, or even just any window that already happens to be open, without modifying, splitting, or above all deleting any others? Sorry if this has been discussed already; I tried searching my list folder (and admit I don't do a good job of following the list in the first place) but keep coming up empty.
Re: [O] Force center alignment in LaTeX table export?
See inline below and attached. If exporting to PDF the first table gets a
prefix of "ced!10" and the second is the pale red highlight I was wanting.
Here are the important lines in the LaTeX buffer export:
\(\left[\begin{array}{>{\cocumncococ{ced!10}}c|ccc|ccc}
vs.
\(\left[\begin{array}{a|ccc|ccc}
This latter only works because in my header I have
\newcolumntype{a}{>{\columncolor{red!10}}c}
I'm not sure I'm reading the original issue, the patch, or the surrounding
code right, but is it possible the [lr]->c substitution was intended for
the case where no :align was provided by the author and org generated the
alignment automatically? If so, would it be appropriate to just move the
substitution to the ";; Extract column groups and alignment ..." block in
org-latex--align-string, which is to say the case (if I'm reading
correctly) where the author has not provided an override?
ox-latex.el-3154-(defun org-latex--align-string (table info)
ox-latex.el-3155- "Return an appropriate LaTeX alignment string.
ox-latex.el-3156-TABLE is the considered table. INFO is a plist used as
ox-latex.el-3157-a communication channel."
ox-latex.el:3158: (or (org-export-read-attribute :attr_latex table :align)
ox-latex.el-3159- (let (align)
ox-latex.el-3160-;; Extract column groups and alignment from first
(non-rule)
Org file to demonstrate problem and workaround:
#+latex_header: \usepackage{xcolor}
#+latex_header: \usepackage{colortbl}
* Non-Working Example
#+ATTR_LaTeX: :mode inline-math :environment array
#+attr_latex: :math-prefix \left[ :math-suffix \right]
#+attr_latex: :align >{\columncolor{red!10}}c|ccc|ccc
| | | * | * | | |
| * | | | * | | |
| \hline* | | | | * | |
| | | | | * | * |
* Working Example
Just define a column type that doesn't use the characters 'l' or 'r'
#+latex_header: \newcolumntype{a}{>{\columncolor{red!10}}c}
#+ATTR_LaTeX: :mode inline-math :environment array
#+attr_latex: :math-prefix \left[ :math-suffix \right]
#+attr_latex: :align a|ccc|ccc
| | | * | * | | |
| * | | | * | | |
| \hline* | | | | * | |
| | | | | * | * |
On Sun, Mar 26, 2017 at 5:14 AM, Eric S Fraga wrote:
> Can you please post an ECM (small example) of the org table with
> attributes that you are having problems with?
>
> --
> : Eric S Fraga (0xFFFCF67D), Emacs 26.0.50, Org release_9.0.5-385-g72fc2d
>
ecm.org
Description: Lotus Organizer
[O] Force center alignment in LaTeX table export?
Apologies in advance if this has been discussed already; I lost track of
the list for a while.
That said, since updating to ELPA org-20170210, I can no longer (I'm pretty
sure this was working previously but I can't prove it -- maybe my document
was broken and I just didn't notice) use \columncolor{red} in the :align
argument to #+latex+attr: for tables any more; this gets changed to
\cocumncococ{ced} by the following ox-latex code:
3361 ;; Make sure cells are always centered while preserving
3362 ;; vertical separators.
3363 (let ((align (replace-regexp-in-string
3364 "[lr]" "c" (org-latex--align-string table
info
at
http://orgmode.org/w/org-mode.git?p=org-mode.git;a=blob;f=lisp/ox-latex.el;hb=9faf000e6a6977135a1d09dba4a078cdd9e04b38#l3361
I'm not certain whether I should be surprised that all non-centered
alignments are changed to centered, or that in general any text with an 'l'
or 'r' isn't valid in :align, or both.
This seems to have been commit 975260cb2596fb8cf0660cc91f2207c48b845699 to
preserve column groups, but I'm not sure I understand the original problem
or the fix. What does replacing l's and r's with c's achieve? Should I be
defining new column types and/or just using a \newcommand that doesn't use
the characters "l" or "r" instead of trying to directly set \columncolor
directly in the alignment string?
Thanks...
Phil
Re: [O] Is there an environment for Org syntax?
Sorry, Fletcher, but this is beyond me. If what you want is essentially a
buffer-based screenshot, then it may in theory be easier, relatively speaking,
to build elisp routines to copy the contents of buffers, but that would be even
further beyond my abilities.
Phil
- Original Message -
From: Fletcher Charest fletcher.char...@gmail.com
To: Phil Regier preg...@ittc.ku.edu
Cc: emacs-orgmode@gnu.org
Sent: Monday, March 17, 2014 8:58:18 AM
Subject: Re: [O] Is there an environment for Org syntax?
Hello Phil,
When I use the edit buffer (by pressing C-c '), my Org syntax is indented
properly (in accordance with org-indent-mode, which is set to t). The code
block itself looks like this after editing:
#+BEGIN_SRC org :exports code
,* Top level headline
,** Second level
,*** Third level
some text
,*** Third level
more text
#+END_SRC
As you can see, the text ('some text' and 'more text') is not indented.
This is a little bit annoying, but not too much - I can add the spaces
myself.
The problem is with the HTML export: I can't see the HTML output as I would
see it in my Emacs buffer, that is to say, without the leading stars. In my
output (see attached), leading stars are visible, although not colored. I
would like them to be completely invisible.
Do you observe the same thing? My org-version is 8.2.5h-30-gdd810b-elpaplus
under GNU Emacs 24.3.1.
Thanks,
FC
[O] Named latex blocks for inline execution?
Hi, all; sorry to pester again, but I haven't been able to find a way to do
something I would have thought would be relatively simple.
What I want to do is declare a named latex block in Org which I can use to
reproduce markup patterns which I reuse very frequently; the usual reasons
include readability, brevity, and maintainability.
The reasons I don't want to use Latex macros for this include:
1. Inflexibility of macro names in Latex (no digits or special characters
allowed)
2. Unavailability of short macro names (most of the good ones are taken, so my
schemes end up inconsistent or self-defeatingly long or I get subtle mysterious
errors from overriding short macro names)
3. By mixing Org calls with Latex macros I can debug problems at multiple
levels (by examining my Org source and generated Latex source separately, which
is messier with Latex macros alone)
4. I can shield my advisor from my worst coding habits if he wants to read my
Latex source
5. I can pass parameters/variables between Org blocks, performing computations
on them when needed, more readily than Latex macros (and I would dearly love to
follow real examples of numerical algorithm execution inline with the
discussion of said algorithms), perhaps even marking Octave results up as Latex
invisibly if I'm lucky.
So I thought I would begin with a simple block to create a small inline
bracketed matrix:
#+name: Smx
#+begin_src latex :var Contents=pi
\(
\left[
\begin{smallmatrix}
Contents
\end{smallmatrix}
\right]
\)
#+end_src
What I get in a Latex export buffer from call_Smx() invariably is some
variant of the following:
=$\backslash$(
\left[
\begin{smallmatrix}
\Pi
\end{smallmatrix}
\right]
$\backslash$)=
There are several particularly stubborn issues here:
1. The surrounding '=' symbols have been impossible to eliminate. I assume
they are part of an attempt to add code markdown, but Latex does not interpret
them as such; they just get passed through.
2. The \( and \) directives get literalized no matter what I try. Wrapping
the call in math mode delimiters hasn't been terribly fruitful either, as this
seems to confuse the export (either inline math mode never gets entered, or the
call_Smx() becomes literal text within math mode).
3. For some reason, my \pi gets capitalized in the Latex output.
I'm not sure exactly what I expect from any particular syntax, but what I want
is just an interpretable stamp of the Latex code inside the original block with
variables substituted exactly. I've tried extra backslashes in the code, tried
a variety of [:results ...] and [:exports ...] before and after the call (as
well as prologues and epilogues), exporting to raw and latex and org, and even
declaring the block as org instead of latex. Is there a correct way to achieve
this, or am I trying to make Org do something it is not meant to do?
Thanks for any ideas...
Phil
Re: [O] Named latex blocks for inline execution?
Many thanks, Tom; these are all variations I had not thought to try. I will
try them out and see what combination works best for me.
Phil
- Original Message -
From: Thomas S. Dye t...@tsdye.com
To: Phil Regier preg...@ittc.ku.edu
Cc: org mode emacs-orgmode@gnu.org
Sent: Friday, March 14, 2014 11:54:04 AM
Subject: Re: Named latex blocks for inline execution?
Aloha Phil,
Org mode has its own macros, which might work in this instance.
The variable org-babel-inline-result-wrap is set to =%s= by
default. Setting it as follows will get rid of the surrounding =
symbols:
#+begin_src emacs-lisp
(setq org-babel-inline-result-wrap %s)
#+end_src
Alternatively, on my setup at least, call_Smx()[:results raw] will also
get rid of the surrounding = symbols.
The capitalization of pi seems to be related to capitalization of the
variable name Contents. This is new to me and I don't recall reading
about it in the documentation, but I haven't looked lately. At any rate,
if I name the variable contents then I get a lowercase pi in the
output.
My guess about the backslash part is that Org mode is expecting an
inline LaTeX math snippet to be on one line. At any rate, this seems to
work for me:
#+name: Smx
#+begin_src latex :var contents=pi
\begin{equation}
\left[
\begin{smallmatrix}
contents
\end{smallmatrix}
\right]
\end{equation}
#+end_src
hth,
Tom
Re: [O] Named latex blocks for inline execution?
Thanks so much, Eric; during my recent updates/testing I had wiped my drawer export settings and had not attempted to use drawer output. I will be sure to try it out. I'm not sure whether the positive results will carry over to the inline call_Smx() syntax, but this will be enlightening even if they do not. Phil - Original Message - From: Eric S Fraga e.fr...@ucl.ac.uk To: Phil Regier preg...@ittc.ku.edu Cc: org mode emacs-orgmode@gnu.org Sent: Friday, March 14, 2014 12:09:15 PM Subject: Re: [O] Named latex blocks for inline execution? Hello, The following works for me with your example: #+call: Smx[:exports results](Contents=Interesting) :results drawer latex I am not sure why because I don't really understand the semantics of drawer for results and didn't see anything in the docmentation about this option (but also didn't look that hard for it). -- : Eric S Fraga (0xFFFCF67D), Emacs 24.3.1, Org release_8.2.5h-660-gef207f
Re: [O] Is there an environment for Org syntax?
I had thought that '#+begin_src org :exports code' would do just this, though I believe you have to then begin each line with a comma inside the block. For me, the following: #+begin_src org :exports code ,#+begin_src org :exports code ,#+end_src #+end_src produces Org code via the HTML and Latex export backends. I think there are syntax highlighting export extensions in the wild (just search the web for something like 'org export syntax highlighting' to see a diverse collection of proposed solutions; Pygments appears to be a common theme, though I know nothing about it), but AFAIK they are not distributed with Org at this time. Does that at least address the first part of your inquiry? Sorry I couldn't offer anything actionable for the second; maybe someone else can pick up where I left off. Phil - Original Message - From: Fletcher Charest fletcher.char...@gmail.com To: emacs-orgmode@gnu.org Sent: Friday, March 14, 2014 4:41:12 PM Subject: [O] Is there an environment for Org syntax? Dear all, I am currently writing a practical tutorial about Org for absolute beginners. In the end, I would like to export it in multiple formats, HTML for sure, but maybe LaTeX too (I am, of course, writing it with org-mode). I am not aware of a functionality, or hack, which would allow me to include Org syntax in the document, without it being taken into account by Org itself and the exporter as being part of the document structure. Instead, I'd like it to be exported as it appears in my Emacs Org file, including colors, etc. I hope I am explaining this clearly. Is it possible? Thank you very much for you help, FC
Re: [O] Is there an environment for Org syntax?
I spent hours trying to figure out recursive org syntax, and only got it working with the help of the list. :) I realized after my last message that Worg uses some limited syntax highlighting on some of the newer pages; I'm not very good with javascript and css, but you might be able to borrow some code for highlighting by inspecting the org and html source... Phil - Original Message - From: Fletcher Charest fletcher.char...@gmail.com To: Phil Regier preg...@ittc.ku.edu Cc: emacs-orgmode@gnu.org Sent: Friday, March 14, 2014 5:59:03 PM Subject: Re: [O] Is there an environment for Org syntax? Yes, thank you very much, that was it. I was thinking about doing this, but I didn't know how to escape the syntax. This is quite embarrassing since I just noticed that it is explained in the manual, and even more, since the comma is added automatically when you edit the org code by hitting C-c ' anyway... Thanks again! Best regards, FC
Re: [O] Is there an environment for Org syntax?
I'm not sure whether I follow you here, because I'm accustomed to seeing indents passed through. If I export the following: #+begin_src org :exports code ,#+begin_src org :exports code indented text ,#+end_src #+end_src ...then my HTML export adds as many spaces as there are in the block edit buffer. Is that not what you see, or are you looking for something more sophisticated? Phil - Original Message - From: Fletcher Charest fletcher.char...@gmail.com To: emacs-orgmode@gnu.org Sent: Friday, March 14, 2014 6:21:35 PM Subject: Re: [O] Is there an environment for Org syntax? Thank you, this sounds interesting; I'm not very good either, but I will have a look at this. And I actually have a related question (I hope it's not in the manual this time): is it possible to indent the Org syntax in the HTML export as it is indented in an Emacs buffer when org-indent-modeis activated? The indentation in the HTML file looks like the one you get in Org by default, even if my org-indent-mode variable is set to t in my Org buffer. Thank you, FC
[O] Org plus contrib (and ditaa.jar)
An issue I've repeatedly encountered when trying to use Babel with ditaa is that ditaa.jar never seems to be available despite a few claims in the documentation that it is distributed with Org. Since I'm already trying to find the source of my octave issues from my previous post, I thought I may have installed Org improperly and decided to wipe out my .emacs.d and start over. I see that now Org has its own elpa repository (so far, so good), so I added it as my second (well, first in customize) package-archives entry after removing my .emacs.d directory to run my 24.3.1 build from my home directory (with .emacs modified only using Custom, and only containing standard org- and package- settings). I refreshed the package list, and installed the org-plus-contrib. Everything seemed successful (Total of 161 files compiled, 1 failed, 4 skipped); I got some errors in the install output buffer, but nothing mentioning ditaa explicitly except Compiling ... ob-ditaa.el. So now when I exit Emacs I expect the following command to produce output: find ~/.emacs.d/ -name ditaa.jar but it still does not. Lowering my expectations slightly, I hope for output from the following: find ~/.emacs.d/ -name contrib but still nothing. I assume I am installing Org improperly somehow, despite the fact that my version shows correctly: Org-mode version 8.2.5h (8.2.5h-32-g9639ed-elpaplus @ /home/pregier/.emacs.d/elpa/org-plus-contrib-20140310/) Am I missing something obvious? Phil
Re: [O] Octave/Matlab Code Export Issue?
Wow; I'm embarrassed. Thanks so much for the assist; terribly sorry to spam the list with such a simple error. X( Phil - Original Message - From: Eric S Fraga e.fr...@ucl.ac.uk To: Phil Regier preg...@math.ku.edu Cc: emacs-orgmode@gnu.org Sent: Thursday, March 13, 2014 11:08:36 AM Subject: Re: [O] Octave/Matlab Code Export Issue? Remove octave from the #+end_src line. There should be nothing else on that line. -- : Eric S Fraga (0xFFFCF67D), Emacs 24.3.50.2, Org release_8.2.5h-709-g99f1ff
Re: [O] Org plus contrib (and ditaa.jar)
This actually does help; thanks! I had tried copying ditaa.jar from ditaa SVN, but that still did not work; copying from Org git, however, works like a charm. Not sure why the first method failed, but I like this approach better anyway, so I'm happy. Thanks again! Phil - Original Message - From: Eric S Fraga e.fr...@ucl.ac.uk To: Phil Regier preg...@math.ku.edu Cc: emacs-orgmode@gnu.org Sent: Thursday, March 13, 2014 11:10:54 AM Subject: Re: [O] Org plus contrib (and ditaa.jar) I have no idea whether it is included in the elpa distribution or not. If you track the git version, it is most definitely there under .../contrib/scripts/ditaa.jar. Just in case this helps! -- : Eric S Fraga (0xFFFCF67D), Emacs 24.3.50.2, Org release_8.2.5h-709-g99f1ff
[O] Octave/Matlab Code Export Issue?
I'm having trouble with Octave export, and I am not quite sure where to look. In an Org file I have the following: #+BEGIN_SRC octave a = [1;3;5;7;9] b = [2;4;6;8] ans=a; #+END_SRC octave When exporting to LaTeX I get the following: \#+BEGIN\_SRC octave a = [1;3;5;7;9] b = [2;4;6;8] ans=a; \#+END\_SRC octave The style is consistent whether I export to LaTeX (buffer or PDF) or HTML; I've tried 'matlab' in lieu of 'octave', and various :results and :exports combinations, and I always get the literal output of the org syntax itself for the code, though I get a proper verbatim block of output contents when I export results. 'sh' and 'latex' blocks work properly (as do 'org' and 'awk' in other documents), exporting code and results alike without passing Org markup through, but this issue persists with octave on older and newer (emacs 24.3.1 with org 8.2.5h) builds. I've set the usual execution confirmation options and added matlab and octave to the loaded languages, and even disabled super/subscript rendering, all through customize, and even tried wiping out my .emacs and starting over just in case. I've tried poking around the source but I don't know what I'm looking for; I don't see direct references to export/publish in ob-octave, and ox is way beyond me, I'm afraid. Can anyone guess either what I'm doing wrong or where I might start looking look for such a language-specific, non-option-specific, non-backend-specific export glitch? Thanks for any suggestions... Phil
[O] Babel #+CALL: results?
I'm new to Babel -- I've been using Org for the last few years just to organize and typeset simple LaTeX for PDF and MathJax export -- and furthermore have had to compile Emacs from scratch and install into my home directory (installing Org as an ELPA package) to get my versions to sync up with the Worg documentation, so I may have broken something, but I've had no end of trouble trying to get even the simplest named block calls to produce output. As I understand Worg, and some old list messages from gmane, the following should produce meaningful output: #+NAME: testfun #+BEGIN_SRC sh :var Var1=Val1 echo Var1: $Var1 #+END_SRC #+CALL: testfun[:results output](Var1=Val3) :results output verbatim #+RESULTS: : *Once* I got output, but I could not reproduce this after making and undoing a few formatting changes; I would chalk this up to my own cluelessness, except that I can get what I *think* is statefully incorrect output from the first block: Iteration 1 (control case): Begin with only the #+NAME: ... #+END_SRC with :results verbatim and execute block to get #+NAME: testfun #+BEGIN_SRC sh :var Var1=Val1 :results output verbatim echo Var1: $Var1 #+END_SRC #+RESULTS: testfun : Var1: Val1 Iteration 2: Replace :results output verbatim with :results output raw and execute block to get #+NAME: testfun #+BEGIN_SRC sh :var Var1=Val1 :results output raw echo Var1: $Var1 #+END_SRC #+RESULTS: testfun Var1: Val1 Iteration 3: Change back to :results output verbatim and execute block to get #+NAME: testfun #+BEGIN_SRC sh :var Var1=Val1 :results output verbatim echo Var1: $Var1 #+END_SRC #+RESULTS: testfun =Var1: Val1 =Var1: Val1 Iteration 3+n: Execute the same block to get #+NAME: testfun #+BEGIN_SRC sh :var Var1=Val1 :results output verbatim echo Var1: $Var1 #+END_SRC #+RESULTS: testfun =Var1: Val1 ==Var1: Val1 ==Var1: Val1 (prev. line repeated n times total) =Var1: Val1 Is this intended behavior? I will admit I don't understand Babel's output modes sufficiently well to know for sure; *however*, I also believe my one isolated success in getting output from #+CALL: was a side effect of this stateful behavior. If this is likely to be due to a bad install/configuration, can anyone clarify the right way to get a current org-mode installation (or at least, a version that corresponds to one or more official online documentation repositories) as an unprivileged user on a system that already has a prepackaged emacs installation? If this is intended behavior, where can I find the right combination of :results properties to pass output from a named block to a later call for display and/or export? Running from my current home directory installation, here are my emacs and Org versions: GNU Emacs 24.3.2 (x86_64-unknown-linux-gnu, GTK+ Version 2.18.9) of 2013-11-10 on host Org-mode version 8.2.2 (8.2.2-dist @ /homedir/.emacs.d/elpa/org-20131108/) Any guidance would be greatly appreciated... Phil Regier preg...@ittc.ku.edu
Re: [O] Babel #+CALL: results?
Oops; forgot to reply-all. Aaron's advice did set me straight, and #+CALL: is working fine for me now without my ill-advised debugging artifacts. Thanks to Aaron for the assistance, and to the Org list/maintainers for all the great Org tools and documentation. Phil - Original Message - From: Aaron Ecay aarone...@gmail.com To: Phil Regier preg...@ittc.ku.edu, emacs-orgmode@gnu.org Sent: Tuesday, November 12, 2013 5:13:34 PM Subject: Re: [O] Babel #+CALL: results? Hi Phil, I’m far from an expert myself, but I think I see what is going on with your testcases. 2013ko azaroak 12an, Phil Regier-ek idatzi zuen: [...] *Once* I got output, but I could not reproduce this after making and undoing a few formatting changes; I would chalk this up to my own cluelessness, except that I can get what I *think* is statefully incorrect output from the first block: Iteration 1 (control case): Begin with only the #+NAME: ... #+END_SRC with :results verbatim and execute block to get #+NAME: testfun #+BEGIN_SRC sh :var Var1=Val1 :results output verbatim echo Var1: $Var1 #+END_SRC #+RESULTS: testfun : Var1: Val1 So far, as expected. Iteration 2: Replace :results output verbatim with :results output raw and execute block to get #+NAME: testfun #+BEGIN_SRC sh :var Var1=Val1 :results output raw echo Var1: $Var1 #+END_SRC #+RESULTS: testfun Var1: Val1 Also as expected. Iteration 3: Change back to :results output verbatim and execute block to get #+NAME: testfun #+BEGIN_SRC sh :var Var1=Val1 :results output verbatim echo Var1: $Var1 #+END_SRC #+RESULTS: testfun =Var1: Val1 =Var1: Val1 Now org would like to remove the previous output. However, it cannot do so, since it does not know where it begins and ends (since raw output could contain in principle anything, or nothign at all). It’s best to use “drawer” instead of “raw” in most cases, because the drawer wrapper bounds the result and lets subsequent calls see it and remove it. Without removing the previous result, org tries to add the “Var1: Val1” as verbatim text (before the instance of that text which is left over from the previous call). It looks like even though there is a newline in the string org chooses to use the single-line =verbatim= syntax, rather than the multiline : verbatim I don’t know why it does this, and it may be a bug. As for your original example: #+CALL: testfun[:results output](Var1=Val3) :results output verbatim You don’t need the second :results output. That applies to an invisible elisp source block which wraps the evaluation of the #+call line, and which produces no output. I get the right result consistently with #+CALL: testfun(Var1=Val3) #+RESULTS: : Var1: Val3 Or (note the addition of quotes in the result): #+CALL: testfun(Var1=Val3) :results verbatim #+RESULTS: : Var1: Val3 -- Aaron Ecay
