[O] [PATCH] Trouble updating some RSS feeds with org-feed

[Hiroshi Saito]

Hi all,

I noticed that the org-feed could not properly handle RSS feeds which do not
contain  element. The value of  element is used as a key of an
association list to manage entry statuses. The keys become `nil' when a 
element not found. Then no entries are added anymore after first update since a
key of new entry (`nil') is already included in the association list.


Here is an example .emacs:
--
(setq org-feed-alist
  '(("Hacker News"
 "https://news.ycombinator.com/rss;
 "~/feed.org" "Hacker News"
 )))
--

After running `org-feed-update-all', keys of feed status in ~/feed.org
are `nil' like this:
--
:FEEDSTATUS:
((nil t "4e939ac25cb5b7c825c0894c364a220d5a98a7bf")
 (nil t "2eac7fd17ae277ba6ad6fd658da663bdf2a28586")
 (nil t "4939903fe5796ea1b5132209c5ab983e0558b5fd")
 ...
:END:
--

After that, `org-feed-update-all' no longer adds new entries in above reason.


It is possible to work around this issue via `:parse_feed' option. But, I think
it would be reasonable that org-feed handles -less RSS feeds.

So, I wrote a small patch that uses a value of  as a key if  is
missing. It's simple and not too bad since there's certain consistency to
 and  except  is also optional. Another option could be using
a hash of  or  but I feel it's excessive.

--
Sincerely,
Hiroshi Saito


0001-org-feed.el-Use-a-value-of-link-as-guid-if-guid-is-m.patch
Description: Binary data

[O] babel for sml?

[Lawrence Bottorff]

Is there no babel language support for Standard ML? I see a ob-sml at elpa.
. .

LB

Re: [O] Show all PROPERTIES COLUMNS without an explicit definition

[Michael Brand]

Hi Shakthi

On Wed, Sep 9, 2015 at 7:37 PM, Shakthi Kannan  wrote:
> === test.org ===
>
> * Test
>   :PROPERTIES:
>   :COLUMNS: %35ITEM %5ID
>   :ID: 1
>   :END:
> * Exam
>   :PROPERTIES:
>   :COLUMNS: %35ITEM %15APPROVED
>   :APPROVED: Yes
>   :END:
>
> === END ===

The usual way would be

=== test.org ===

#+COLUMNS: %35ITEM %5ID %15APPROVED
* Test
  :PROPERTIES:
  :ID:   1
  :END:
* Exam
  :PROPERTIES:
  :APPROVED: Yes
  :END:

=== END ===

Michael

Re: [O] babel for sml?

[Ista Zahn]

I don't understand. You see ob-sml, but you ask if there is no
support? From the looks of it, ob-sml provides babel language support
for Standard ML, no?

Best,
Ista

On Wed, Sep 23, 2015 at 9:21 AM, Lawrence Bottorff  wrote:
> Is there no babel language support for Standard ML? I see a ob-sml at elpa.
> . .
>
> LB

Re: [O] babel for sml?

[Thomas S . Dye]

Aloha Lawrence,

Lawrence Bottorff  writes:

> The babel language page does in fact list SML, but no specific ob- package
> is mentioned -- as if support is "built-in."

When the entry in the Documentation column is empty it means that there
is no documentation outside of the source code (or that the
documentation wasn't found by the person editing the table).  Nowadays,
it is the "built-in" languages in core that mostly lack documentation,
although the documentation for languages outside core is sometimes
minimal.

Volunteer ob-* documentation writers welcome!

hth,
Tom
-- 
Thomas S. Dye
http://www.tsdye.com

[O] Commit 6d2ab40 breaks table input in R

[Charles C. Berry]

This ECM fails with an error starting with the above commit.

#+NAME: numeric
| 1 | 2 | 3 |
| 4 | 5 | 6 |


#+BEGIN_SRC R :var df=numeric
df
#+END_SRC

Similar tests in test-ob-R.el also fail.

orgtbl-to-tsv now returns a propertized string.

---

Should this be fixed in org-table.el or downstream in ob-R.el?

Maybe like this:

diff --git a/lisp/ob-R.el b/lisp/ob-R.el
index 4bd6eea..4f9fb91 100644
--- a/lisp/ob-R.el
+++ b/lisp/ob-R.el
@@ -239,7 +239,8 @@ This function is called by 
`org-babel-execute-src-block'."

 (min (if lengths (apply 'min lengths) 0)))
 ;; Ensure VALUE has an orgtbl structure (depth of at least 2).
 (unless (listp (car value)) (setq value (list value)))
-   (let ((file (orgtbl-to-tsv value '(:fmt org-babel-R-quote-tsv-field)))
+   (let ((file (org-no-properties
+(orgtbl-to-tsv value '(:fmt org-babel-R-quote-tsv-field
  (header (if (or (eq (nth 1 value) 'hline) colnames-p)
  "TRUE" "FALSE"))
  (row-names (if rownames-p "1" "NULL")))




Chuck

Re: [O] [PATCH] ox-latex: Add scrartcl LaTeX class

[Sebastian Boehm]

Hi Nicolas,

On 21 September 2015 at 21:46, Nicolas Goaziou  wrote:
> But can't users needing a non-default class customize their
> own `org-latex-classes? Or is there an entitling reason to offer it out
> of the box?

you are of course right; users can always customise their own setup.
However, when sharing Org files with other people this kind of tight
coupling between the Org file and the user's Emacs configuration can
become quite problematic.

Granted, this is not a problem if you only work alone on all of your
projects, but it effectively prevents users from using document
classes other than the ones provided by Org mode for shared documents
(especially when sharing documents with less experienced users).

My intention was not to start a debate on "article" vs. "scrartcl",
but it would be great to see some more variety in the default document
classes or maybe even a way to specify the desired document class as
well as the desired sectioning style directly in an Org file without
having to rely on the document class definition being available in
another user's Emacs config.

Btw, since I am not on the list, it would be great if everybody could
CC me on replies.

Best,
Sebastian

Re: [O] DONE repeating events become TODO, not OTHER

[Bernt Hansen]

Nick Dokos  writes:

> Ken Mankoff  writes:
>
>> I have repeating events of type OTHER, as in:
>>
>> * OTHER Foo
>>   <2015-09-16 +1w>
>>
>> When I have finished the task for today, I mark it done. Because it is
>> a repeating event, it doesn't actually change to DONE, instead it
>> changes to TODO and adds a :LAST_REPEAT: item in the :PROPERTIES:
>> drawer.
>>
>> How do I tell this specific item, or all OTHER items in general, that
>> after completion it should remain OTHER rather than changing to TODO?
>>
>
> Do you *have* to have a TODO state? If not,
>
> #+TODO: OTHER | DONE
>
> should do the trick. Or you could use multiple
> sets:
>
> #+TODO: TODO INPROGRESS | DONE
> #+TODO: OTHER | COMPLETED
>
> In the latter case, the keywords all have to be different.
> See
>
>(info "(org) TODO items")


Or after defining OTHER has a todo keyword add a property specifying
which next state you want

* OTHER Foo
  :PROPERTIES:
  :REPEAT_TO_STATE: OTHER
  :END:
  <2015-09-16 +1w>

Regards,
Bernt

Re: [O] babel for sml?

[Lawrence Bottorff]

The babel language page does in fact list SML, but no specific ob- package
is mentioned -- as if support is "built-in." And ob-sml -- at least in my
elpa listing -- says only

 Requires: sml-mode-6.4
 Summary: org-babel functions for template evaluation

I was confused by the summary. I did install it and it works, BTW.

On Wed, Sep 23, 2015 at 2:10 PM, Ista Zahn  wrote:

> I don't understand. You see ob-sml, but you ask if there is no
> support? From the looks of it, ob-sml provides babel language support
> for Standard ML, no?
>
> Best,
> Ista
>
> On Wed, Sep 23, 2015 at 9:21 AM, Lawrence Bottorff 
> wrote:
> > Is there no babel language support for Standard ML? I see a ob-sml at
> elpa.
> > . .
> >
> > LB
>

Re: [O] babel for sml?

[Ista Zahn]

I've added a link from
http://orgmode.org/worg/org-contrib/babel/languages.html to the ob-sml
github project.

Best,
Ista

On Wed, Sep 23, 2015 at 11:49 AM, Lawrence Bottorff  wrote:
> The babel language page does in fact list SML, but no specific ob- package
> is mentioned -- as if support is "built-in." And ob-sml -- at least in my
> elpa listing -- says only
>
>  Requires: sml-mode-6.4
>  Summary: org-babel functions for template evaluation
>
> I was confused by the summary. I did install it and it works, BTW.
>
> On Wed, Sep 23, 2015 at 2:10 PM, Ista Zahn  wrote:
>>
>> I don't understand. You see ob-sml, but you ask if there is no
>> support? From the looks of it, ob-sml provides babel language support
>> for Standard ML, no?
>>
>> Best,
>> Ista
>>
>> On Wed, Sep 23, 2015 at 9:21 AM, Lawrence Bottorff 
>> wrote:
>> > Is there no babel language support for Standard ML? I see a ob-sml at
>> > elpa.
>> > . .
>> >
>> > LB
>
>

[O] [gnorb] error when calling `gnorb-report-tracking-usage'

[Thomas Holst]

Hello,

more and more I like gnorb. It perfectly fits my working habits.

Today I wanted to see how the registy is filled. I called
`gnorb-report-tracking-usage'. And got following error:

  gnorb-registry-tracked-headings: Symbol's 
  function definition is void: hash-table-keys

After a little search on the web I added:

  #+begin_src emacs-lisp
(require 'subr-x)
  #+end_src

to my gnorb config. Again calling `gnorb-report-tracking-usage' I get
another error:

  eieio-default-superclass: Invalid slot name:
  "#", max-size

Well I gave up on this. Couldn't dive deeper :-)

My configuration is:

gnorb: recent git-version
org:   recent git-version
emacs: 25.5.1 on Ubuntu 14.04

Thanks for looking into this.

-- 
Bis neulich ...
  Thomas

Re: [O] [PATCH] ox-latex: Add scrartcl LaTeX class

[Thomas S . Dye]

Aloha Sebastian,

Sebastian Boehm  writes:

> Hi Nicolas,
>
> On 21 September 2015 at 21:46, Nicolas Goaziou  wrote:
>> But can't users needing a non-default class customize their
>> own `org-latex-classes? Or is there an entitling reason to offer it out
>> of the box?
>
> you are of course right; users can always customise their own setup.
> However, when sharing Org files with other people this kind of tight
> coupling between the Org file and the user's Emacs configuration can
> become quite problematic.
>
> Granted, this is not a problem if you only work alone on all of your
> projects, but it effectively prevents users from using document
> classes other than the ones provided by Org mode for shared documents
> (especially when sharing documents with less experienced users).
>
> My intention was not to start a debate on "article" vs. "scrartcl",
> but it would be great to see some more variety in the default document
> classes or maybe even a way to specify the desired document class as
> well as the desired sectioning style directly in an Org file without
> having to rely on the document class definition being available in
> another user's Emacs config.

Sharing documents that are highly configured, but don't rely on the
user's Emacs configuration, is one goal of "reproducible research"
carried out with Org mode.

In general, it is possible to move the configuration into the shared Org
mode document and then use Local Variables to load the configuration
when the document is opened, or when the user refreshes the setup.

The first step is to put the configuration into one or more Babel source
code blocks in the shared document and give each source code block a
name.  Then set up Local Variables to called the named source code
block(s) using the org-sbe (formerly, sbe) function.

Here is a fragment of the Local Variables section I'm using on a
self-contained Org mode document.

# Local Variables: 
# eval: (require 'ox-latex)
# org-hide-macro-markers: t
# org-entities-user: nil
# eval: (and (fboundp 'org-sbe) (not (fboundp 'sbe)) (fset 'sbe 'org-sbe))
# eval: (sbe "user-entities")
# End:

Where the source code block named "user-entities" is:

*** User Entities
The following source code block sets up user entities that are used frequently
in my work. I use the various =*macron= commands to typeset Hawaiian
language words with what is known in Hawaiian as a /kahak\omacron{}/.

#+name: user-entities
#+begin_src emacs-lisp
  (add-to-list 'org-entities-user '("amacron" "\\={a}" nil "" "a" "a" 
"ā"))
  (add-to-list 'org-entities-user '("emacron" "\\={e}" nil "" "e" "e" 
"ē"))
  (add-to-list 'org-entities-user '("imacron" "\\={\\i}" nil "" "i" "i" 
"ī"))
  (add-to-list 'org-entities-user '("omacron" "\\={o}" nil "" "o" "o" 
"ō"))
  (add-to-list 'org-entities-user '("umacron" "\\={u}" nil "" "u" "u" 
"ū"))
  (add-to-list 'org-entities-user '("Amacron" "\\={A}" nil "" "A" "A" 
"Ā"))
  (add-to-list 'org-entities-user '("Emacron" "\\={E}" nil "" "E" "E" 
"Ē"))
  (add-to-list 'org-entities-user '("Imacron" "\\={I}" nil "" "I" "I" 
"Ī"))
  (add-to-list 'org-entities-user '("Omacron" "\\={O}" nil "" "O" "O" 
"Ō"))
  (add-to-list 'org-entities-user '("Umacron" "\\={U}" nil "" "U" "U" 
"Ū"))
#+end_src

Let me know if you have questions.

hth,
Tom

-- 
Thomas S. Dye
http://www.tsdye.com

Re: [O] [PATCH] ox-latex: Add scrartcl LaTeX class

[Ista Zahn]

On Wed, Sep 23, 2015 at 2:16 PM, Eric S Fraga  wrote:
> On Wednesday, 23 Sep 2015 at 18:59, Sebastian Boehm wrote:
>> Hi Nicolas,
>>
>> On 21 September 2015 at 21:46, Nicolas Goaziou  
>> wrote:
>>> But can't users needing a non-default class customize their
>>> own `org-latex-classes? Or is there an entitling reason to offer it out
>>> of the box?
>>
>> you are of course right; users can always customise their own setup.
>> However, when sharing Org files with other people this kind of tight
>> coupling between the Org file and the user's Emacs configuration can
>> become quite problematic.
>
> Yes, this is a real problem.
>
> One solution, but not an ideal one, is to use file local variables to
> define org-latex-classes (and any other variables) as appropriate for
> the document.  It's not ideal because unfortunately most of org's
> variables (as well as many of emacs's own) are "global" and hence affect
> the functionality of org visiting other buffers.  I find myself having
> separate emacs instances for some org files for this reason.

Doesn't "setq-local" solve this problem?

>
> I had thought that maybe using #+bind: would solve this problem but the
> export engine looks for the LaTeX class before binding takes effect.
>
> --
> : Eric S Fraga (0xFFFCF67D), Emacs 25.0.50.2, Org release_8.3.1-270-g256fef
>

Re: [O] http address for cloning org-mode git

[Suvayu Ali]

On Wed, Sep 23, 2015 at 03:03:42PM -0400, Kaushal Modi wrote:
> Hi all,
> 
> I cannot access the git: repo address and so had been using the http:
> version till now ( http://repo.or.cz/r/org-mode.git/ ).

That is a mirror, a relic from the old days when that used to be the
official repo.  Maybe try:  http://orgmode.org/org-mode.git

http://orgmode.org/worg/org-faq.html#keeping-current-with-Org-mode-development

-- 
Suvayu

Open source is the future. It sets us free.

Re: [O] http address for cloning org-mode git

[Achim Gratz]

Kaushal Modi writes:
> Today git pull gives the below error:
>
> Fetching origin
> fatal: unable to access 'http://repo.or.cz/r/org-mode.git/': Couldn't
> resolve host 'repo.or.cz'
> error: Could not fetch origin

That looks like a problem with DNS resolution on your side.

> ​Looks like ​
> http://repo.or.cz/r/org-mode.git/
> ​ is down? Is anyone else seeing the same thing? Is there an alternative
> http: address that I can use for cloning?

With a new enough Git you _can_ pull from

http://orgmode.org/orgmode.git

But it's probably not nice to the server.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Wavetables for the Waldorf Blofeld:
http://Synth.Stromeko.net/Downloads.html#BlofeldUserWavetables

Re: [O] http address for cloning org-mode git

[Achim Gratz]

Suvayu Ali writes:
> On Wed, Sep 23, 2015 at 03:03:42PM -0400, Kaushal Modi wrote:
>> I cannot access the git: repo address and so had been using the http:
>> version till now ( http://repo.or.cz/r/org-mode.git/ ).
>
> That is a mirror, a relic from the old days when that used to be the
> official repo.  Maybe try:  http://orgmode.org/org-mode.git

No, that never was the official repo.  I've set up that mirror.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Samples for the Waldorf Blofeld:
http://Synth.Stromeko.net/Downloads.html#BlofeldSamplesExtra

Re: [O] org-collector unable to handle macros

[Achim Gratz]

Nicolas Goaziou writes:
> Fixed. Thank you.

That commit 6d2ab4071960f8a7f20138291424b468722ab522 breaks three R
tests for me on Linux (not that other R tests do run correctly):

3 unexpected results:
   FAILED  test-ob-R/colnames-nil-header-argument
   FAILED  test-ob-R/colnames-no-header-argument
   FAILED  test-ob-R/colnames-yes-header-argument

It seems that there is no table returned or the interpretation of the
returned data as a table fails.

Test test-ob-R/colnames-nil-header-argument condition:
(user-error "Empty table - created default table")
   FAILED  130/599  test-ob-R/colnames-nil-header-argument
Test test-ob-R/colnames-no-header-argument condition:
(ert-test-failed
 ((should
   (equal '...
(org-babel-execute-src-block)))
  :form
  (equal
   (("col")
("a")
("b"))
   nil)
  :value nil :explanation
  (different-types
   (("col")
("a")
("b"))
   nil)))
   FAILED  131/599  test-ob-R/colnames-no-header-argument
Test test-ob-R/colnames-yes-header-argument condition:
(user-error "Empty table - created default table")
   FAILED  132/599  test-ob-R/colnames-yes-header-argument


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Wavetables for the Waldorf Blofeld:
http://Synth.Stromeko.net/Downloads.html#BlofeldUserWavetables

Re: [O] http address for cloning org-mode git

[Suvayu Ali]

On Wed, Sep 23, 2015 at 09:29:06PM +0200, Achim Gratz wrote:
> Suvayu Ali writes:
> > On Wed, Sep 23, 2015 at 03:03:42PM -0400, Kaushal Modi wrote:
> >> I cannot access the git: repo address and so had been using the http:
> >> version till now ( http://repo.or.cz/r/org-mode.git/ ).
> >
> > That is a mirror, a relic from the old days when that used to be the
> > official repo.  Maybe try:  http://orgmode.org/org-mode.git
> 
> No, that never was the official repo.  I've set up that mirror.

Didn't we used to have the Org repo hosted on repo.or.cz way, way back
in the old days, pre-2010?  Maybe I'm just confusing a different repo
because it was on the same domain?

-- 
Suvayu

Open source is the future. It sets us free.

Re: [O] [PATCH] ox-latex: Add scrartcl LaTeX class

[Thomas S . Dye]

Hi Eric,

Eric S Fraga  writes:

> One solution, but not an ideal one, is to use file local variables to
> define org-latex-classes (and any other variables) as appropriate for
> the document.  It's not ideal because unfortunately most of org's
> variables (as well as many of emacs's own) are "global" and hence affect
> the functionality of org visiting other buffers.  I find myself having
> separate emacs instances for some org files for this reason.

I've noticed this, too, but never investigated.

Is there some way to know which variables are "global" and which can be
set "locally"?

Also, (completely naively) is it difficult to make "global" variables
"local", or is it something simple, like making a function interactive?  

All the best,
Tom

-- 
Thomas S. Dye
http://www.tsdye.com

Re: [O] http address for cloning org-mode git

[Kaushal Modi]

Thanks all.  Strange enough, git pull from repo.or.cz/r/org-mode.git
started working fine once again.

But I cannot yet clone from  http://orgmode.org/org-mode.git. I had tried
the same earlier with the same result.. the clone will get stuck at
"Fetching origin".
My current git version is 2.6.0.rc0.24.gec371ff

In any case, sorry for the noise. Some intermittent issue was failing the
git pull from the repo.or.cz domain and it is working fine once again.


--
Kaushal Modi

On Wed, Sep 23, 2015 at 3:18 PM, Suvayu Ali 
wrote:

> On Wed, Sep 23, 2015 at 03:03:42PM -0400, Kaushal Modi wrote:
> > Hi all,
> >
> > I cannot access the git: repo address and so had been using the http:
> > version till now ( http://repo.or.cz/r/org-mode.git/ ).
>
> That is a mirror, a relic from the old days when that used to be the
> official repo.  Maybe try:  http://orgmode.org/org-mode.git
>
>
> http://orgmode.org/worg/org-faq.html#keeping-current-with-Org-mode-development
>
> --
> Suvayu
>
> Open source is the future. It sets us free.
>
>

[O] http address for cloning org-mode git

[Kaushal Modi]

Hi all,

I cannot access the git: repo address and so had been using the http:
version till now ( http://repo.or.cz/r/org-mode.git/ ).

Today git pull gives the below error:

Fetching origin
fatal: unable to access 'http://repo.or.cz/r/org-mode.git/': Couldn't
resolve host 'repo.or.cz'
error: Could not fetch origin

​Looks like ​
http://repo.or.cz/r/org-mode.git/
​ is down? Is anyone else seeing the same thing? Is there an alternative
http: address that I can use for cloning?

Thanks.
​

--
Kaushal Modi

Re: [O] BEAMER_ACT ignored during export in column environment

[Julien Cubizolles]

Nicolas Goaziou  writes:

> Julien Cubizolles  writes:
>
>> Nothing in my user's guide about overlay specifications for column
>> you're right. However the mode ( ) has an effect:
>> the environment is ignored in  mode at least. Moreover, a pure
>> overlay specification like  to a column environment doesn't
>> prevent a beamer document from being compiled. Couldn't org export it
>> so that we recover the  usage ?
>
> We would be playing with undocumented stuff. 
>
> If you think that's fine, please provide a patch for it, ideally with
> comments in the source, and an entry about it in ORG-NEWS.

I'll consider it but I discovered that it's not behaving the way I
expect it to so, I'll wait until I've cleared that up.

Julien.

Re: [O] http address for cloning org-mode git

[Kaushal Modi]

You're right!

This time I let it stay stuck for more than few minutes and I can confirm
that the http://orgmode.org/org-mode.git link works too.
Somehow the other link is much faster for me when cloning/pulling.


--
Kaushal Modi

On Wed, Sep 23, 2015 at 4:53 PM, Robert Klein  wrote:

> Hi,
>
> Kaushal Modi  wrote:
>
> > Just to clarify,
> >
> > git clone http://orgmode.org/org-mode.git org-mode
> >
> > gets stuck at "Cloning into 'org-mode' ..."
>
>
> That's the point where it typically take a couple of minutes to
> complete for me.
>
> Best regards
> Robert
>
> >
> > But git cloning the mirror http repo works fine.
> >
> >
> >
> > --
> > Kaushal Modi
> >
> > On Wed, Sep 23, 2015 at 3:29 PM, Achim Gratz 
> > wrote:
> >
> > > Suvayu Ali writes:
> > > > On Wed, Sep 23, 2015 at 03:03:42PM -0400, Kaushal Modi wrote:
> > > >> I cannot access the git: repo address and so had been using the
> > > >> http: version till now ( http://repo.or.cz/r/org-mode.git/ ).
> > > >
> > > > That is a mirror, a relic from the old days when that used to be
> > > > the official repo.  Maybe try:  http://orgmode.org/org-mode.git
> > >
> > > No, that never was the official repo.  I've set up that mirror.
> > >
> > >
> > > Regards,
> > > Achim.
> > > --
> > > +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+
> > >
> > > Samples for the Waldorf Blofeld:
> > > http://Synth.Stromeko.net/Downloads.html#BlofeldSamplesExtra
> > >
> > >
> > >
>
>
>

Re: [O] [PATCH] ox-latex: Add scrartcl LaTeX class

[Eric S Fraga]

On Wednesday, 23 Sep 2015 at 18:59, Sebastian Boehm wrote:
> Hi Nicolas,
>
> On 21 September 2015 at 21:46, Nicolas Goaziou  wrote:
>> But can't users needing a non-default class customize their
>> own `org-latex-classes? Or is there an entitling reason to offer it out
>> of the box?
>
> you are of course right; users can always customise their own setup.
> However, when sharing Org files with other people this kind of tight
> coupling between the Org file and the user's Emacs configuration can
> become quite problematic.

Yes, this is a real problem.

One solution, but not an ideal one, is to use file local variables to
define org-latex-classes (and any other variables) as appropriate for
the document.  It's not ideal because unfortunately most of org's
variables (as well as many of emacs's own) are "global" and hence affect
the functionality of org visiting other buffers.  I find myself having
separate emacs instances for some org files for this reason.

I had thought that maybe using #+bind: would solve this problem but the
export engine looks for the LaTeX class before binding takes effect.

-- 
: Eric S Fraga (0xFFFCF67D), Emacs 25.0.50.2, Org release_8.3.1-270-g256fef

Re: [O] http address for cloning org-mode git

[Kaushal Modi]

Just to clarify,

git clone http://orgmode.org/org-mode.git org-mode

gets stuck at "Cloning into 'org-mode' ..."

But git cloning the mirror http repo works fine.



--
Kaushal Modi

On Wed, Sep 23, 2015 at 3:29 PM, Achim Gratz  wrote:

> Suvayu Ali writes:
> > On Wed, Sep 23, 2015 at 03:03:42PM -0400, Kaushal Modi wrote:
> >> I cannot access the git: repo address and so had been using the http:
> >> version till now ( http://repo.or.cz/r/org-mode.git/ ).
> >
> > That is a mirror, a relic from the old days when that used to be the
> > official repo.  Maybe try:  http://orgmode.org/org-mode.git
>
> No, that never was the official repo.  I've set up that mirror.
>
>
> Regards,
> Achim.
> --
> +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+
>
> Samples for the Waldorf Blofeld:
> http://Synth.Stromeko.net/Downloads.html#BlofeldSamplesExtra
>
>
>

Re: [O] [PATCH] ox-latex: Add scrartcl LaTeX class

[Rasmus]

Nicolas Goaziou  writes:

> Hello,
>
> Rasmus  writes:
>
>> The long-term solution to this is a new "bundled org" file format
>> containing dependencies, such as org files, init.el etc.
>>
>> https://lists.gnu.org/archive/html/emacs-orgmode/2015-02/msg00765.html
>
> This was just an idea, not a long-term solution or a plan. Actually,
> using Babel, you may not even need to introduce a new file format, since
> a document can generate its own set-up.

I know.  I still think it's a great idea.  It also nice for pictures, css
etc.

-- 
Dung makes an excellent fertilizer

Re: [O] a patch to org.el of git version

[Nicolas Goaziou]

Tokuya Kameshima  writes:

> But it looks like `org-show-children' is a replacement of `show-children'.
> Actually the key bind for `show-children' (C-c TAB in my case) is remapped
> to `org-show-children' as below.
>
> org.el:5733:
>> (define-key org-mode-map [remap show-children] 'org-show-children)
>
> When I typed C-C TAB on an org-mode buffer, I got the following error:
>
> Wrong type argument: commandp, org-show-children

You are right. This is now fixed. Thank you.

Regards,

[O] html export list of figures

[Cook, Malcolm]

Hi,

I see know what to generate a list of figures to appear in my html export in 
addition to my current

#+TOC: listings
#+TOC: tables

Exporting latex provides 

#+LATEX: \listoffigures

To complement

#+LATEX: \tableofcontents
#+LATEX: \listoftables

But I see nothing analogous for HTML

Am I missing something?

Thanks,

Malcolm Cook

Re: [O] [PATCH] ox-latex: Add scrartcl LaTeX class

[Nicolas Goaziou]

Hello,

Rasmus  writes:

> The long-term solution to this is a new "bundled org" file format
> containing dependencies, such as org files, init.el etc.
>
> https://lists.gnu.org/archive/html/emacs-orgmode/2015-02/msg00765.html

This was just an idea, not a long-term solution or a plan. Actually,
using Babel, you may not even need to introduce a new file format, since
a document can generate its own set-up.


Regards,

-- 
Nicolas Goaziou

Re: [O] org-collector unable to handle macros

[Nicolas Goaziou]

Hello,

Achim Gratz  writes:


> That commit 6d2ab4071960f8a7f20138291424b468722ab522 breaks three R
> tests for me on Linux (not that other R tests do run correctly):
>
> 3 unexpected results:
>FAILED  test-ob-R/colnames-nil-header-argument
>FAILED  test-ob-R/colnames-no-header-argument
>FAILED  test-ob-R/colnames-yes-header-argument
>
> It seems that there is no table returned or the interpretation of the
> returned data as a table fails.

I just applied a patch that may fix these issues.


Regards,

-- 
Nicolas Goaziou

Re: [O] Commit 6d2ab40 breaks table input in R

[Nicolas Goaziou]

Hello,

"Charles C. Berry"  writes:

> This ECM fails with an error starting with the above commit.
>
> #+NAME: numeric
> | 1 | 2 | 3 |
> | 4 | 5 | 6 |
>
>
> #+BEGIN_SRC R :var df=numeric
> df
> #+END_SRC
>
> Similar tests in test-ob-R.el also fail.
>
> orgtbl-to-tsv now returns a propertized string.

Indeed. This is a bug, which is now fixed. Thank you.

Regards,

-- 
Nicolas Goaziou

Re: [O] [PATCH] ox-latex: Add scrartcl LaTeX class

[Rasmus]

Sebastian Boehm  writes:
> My intention was not to start a debate on "article" vs. "scrartcl",
> but it would be great to see some more variety in the default document
> classes or maybe even a way to specify the desired document class as
> well as the desired sectioning style directly in an Org file without
> having to rely on the document class definition being available in
> another user's Emacs config.

The long-term solution to this is a new "bundled org" file format
containing dependencies, such as org files, init.el etc.

https://lists.gnu.org/archive/html/emacs-orgmode/2015-02/msg00765.html

Rasmus

-- 
Got mashed potatoes. Ain't got no T-Bone. No T-Bone

Re: [O] http address for cloning org-mode git

[Robert Klein]

Hi,

Kaushal Modi  wrote:

> Just to clarify,
> 
> git clone http://orgmode.org/org-mode.git org-mode
> 
> gets stuck at "Cloning into 'org-mode' ..."


That's the point where it typically take a couple of minutes to
complete for me.

Best regards
Robert

> 
> But git cloning the mirror http repo works fine.
> 
> 
> 
> --
> Kaushal Modi
> 
> On Wed, Sep 23, 2015 at 3:29 PM, Achim Gratz 
> wrote:
> 
> > Suvayu Ali writes:
> > > On Wed, Sep 23, 2015 at 03:03:42PM -0400, Kaushal Modi wrote:
> > >> I cannot access the git: repo address and so had been using the
> > >> http: version till now ( http://repo.or.cz/r/org-mode.git/ ).
> > >
> > > That is a mirror, a relic from the old days when that used to be
> > > the official repo.  Maybe try:  http://orgmode.org/org-mode.git
> >
> > No, that never was the official repo.  I've set up that mirror.
> >
> >
> > Regards,
> > Achim.
> > --
> > +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+
> >
> > Samples for the Waldorf Blofeld:
> > http://Synth.Stromeko.net/Downloads.html#BlofeldSamplesExtra
> >
> >
> >

Re: [O] org-babel-execute:dot -- why doesn't this work?

[Nick Dokos]

Matt Price  writes:

> I would, however, like to avoid the clumsy intermediate step and use 
> something like this instead:
> #+name: graph-from-tables
> #+HEADER: :var nodes=students-table graph=students-graph horiz='t
> #+BEGIN_SRC emacs-lisp :file ~/example-diagram.png :colnames yes :exports 
> results
>   (org-babel-execute:dot
>(concat
> "graph {\n"
> (when horiz "rankdir=LR;\n")   ;up-down or left-right
> (mapconcat
>  (lambda (x)
>(format "%s [label=\"%s\" shape=%s style=\"filled\" fillcolor=\"%s\"]"
>(car x)
>(nth 1 x)
>(if (string= "" (nth 2 x)) "box" (nth 2 x))
>(if (string= "" (nth 3 x)) "none" (nth 3 x))
>)) nodes "\n")
> "\n"
> (mapconcat
> (lambda (x)
>   (format "%s -- %s;"
>   (car x) (nth 1 x) )) graph "\n")
> "}\n") params)   
> #+END_SRC
>

There are a couple of problems, one minor (the :colnames yes part
truncates the graph so you have only two edges), and one major: the
"params" argument is not defined.

When a dot source block is passed to org-babel-execute:dot, the params
argument I get [fn:1] is:

((:comments . #1="") (:shebang . #1#) (:cache . "no") (:padline . #1#)
 (:noweb . "no") (:tangle . "no") (:exports . "results")
 (:results . "file replace") (:session . "none") (:hlines . "no")
 (:file . "Images/foobar.png") (:result-type . value)
 (:result-params "file" "replace") (:rowname-names) (:colname-names))

So I tried passing it to org-babel-execute:dot and somewhat
to my surprise it worked [fn:2] - try evaluating the following in your
*scratch* buffer (make sure there is an Images subdir under the
current directory of that buffer):

--8<---cut here---start->8---

(org-babel-execute:dot
 "graph {
rankdir=LR;
a [label=\"Omar\" shape=ellipse style=\"filled\" fillcolor=\"green\"]
b [label=\"Hindia\" shape=ellipse style=\"filled\" fillcolor=\"orange\"]
c [label=\"Yuvrai\" shape=ellipse style=\"filled\" fillcolor=\"purple\"]
a -- b;
a -- c;
b -- c;
}
"
 '((:comments . #1="") (:shebang . #1#) (:cache . "no") (:padline . #1#)
 (:noweb . "no") (:tangle . "no") (:exports . "results")
 (:results . "file replace") (:session . "none") (:hlines . "no")
 (:file . "Images/foobar.png") (:result-type . value)
 (:result-params "file" "replace") (:rowname-names) (:colname-names)))
--8<---cut here---end--->8---

But I would hardly call this method less clumsy than your earlier
attempt.

Footnotes:

[fn:1]  I made a simple dot source block with your graph

#+BEGIN_SRC dot :file Images/foobar.png
graph {
rankdir=LR;
a [label="Omar" shape=ellipse style="filled" fillcolor="green"]
b [label="Hindia" shape=ellipse style="filled" fillcolor="orange"]
c [label="Yuvrai" shape=ellipse style="filled" fillcolor="purple"]
a -- b;
a -- c;
b -- c;
}
#+END_SRC

and instrumented org-babel-execute:dot under edebug - when I executed the code
block, the debugger kicked in when the function got called and I could
get the "params" argument.

[fn:2] I thought the #1 and #1# constructs (which I don't understand at
   all) would make it blow up.

-- 
Nick

Re: [O] funny bug if shifting scheduled items

[Detlef Steuer]

Am Tue, 22 Sep 2015 21:24:59 +0200
schrieb Nicolas Goaziou :

> Hello,
> 
> Detlef Steuer  writes:
> 
> > Met this funny bug in the morning.
> >
> > (orgmode version: git from today)
> >
> > Given the file
> >
> > * Testappointment
> >   SCHEDULED: <2015-09-21 Mo 23:00-23:55>
> >
> > If I now shift the *starttime* using shift-uparrow
> > the *endtime* goes to 24:00, 24:05 etc.
> >
> > If I keep shifting the *endtime* goes from
> > 24:55 to 24:00 again.
> >
> > The *starttime* correctly rolls over to 0:00 .
> 
> Fixed, somewhat. Thank you.
> 
> Note that it can still produce meaningless timestamps, e.g.,
> 
>   <2015-09-21 Mo 23:00-23:55> => <2015-09-21 Mo 23:05-00:00>
> 
> but at least, it is predictable.
> 

Thx!

I don`t consider that time-stamp meaningless. At least for humans :-)
Using the context you know it´t 0:00 the next day.

I doubt it were better to split a time intervall at midnight into a
regular time span, say

<2015-09-21 Mo 23:00-23:55> => (+ 5min)
<2015-09-21 Mo 23:05>--<2015-09-22 Di 00:00>

Not worth the trouble I´d say. The time span given in an intervall
HH:MM-HH:MM always is < 24h and endtime > startime, so even a program could
interpret it correctly.

As always, thx for fixing so fast!

Detlef


> 
> Regards,
> 



-- 
Dr. Detlef Steuer
Helmut-Schmidt-Universität
Fakultät WiSo
Holstenhofweg 85
22043 Hamburg

Tel:  040/6541-2819
mail: ste...@hsu-hh.de

Re: [O] html export list of figures

[John Kitchin]

I am not aware of anything in html for this. In latex, org is just
passing the baton to Latex to generate the list of tables/figures. For
HTML, you will probably have to generate the code yourself.

In org-ref I provide a list-of-tables and list-of-figures link, which
has the latex export code you show, but it is also clickable and
generates an org buffer with a clickable list of tables or figures. See:

https://github.com/jkitchin/org-ref/blob/master/org-ref.el#L932

while there is no html export code there, the clicking functions lay out
how to get the information. I guess you would need one more detail of
putting some kind of anchor or something on the tables/figures so you
could click on the list and jump to the table. it might get you stated anyway.

Cook, Malcolm writes:

> Hi,
>
> I see know what to generate a list of figures to appear in my html export in 
> addition to my current
>
>   #+TOC: listings
>   #+TOC: tables
>
> Exporting latex provides
>
>   #+LATEX: \listoffigures
>
> To complement
>
>   #+LATEX: \tableofcontents
>   #+LATEX: \listoftables
>
> But I see nothing analogous for HTML
>
> Am I missing something?
>
> Thanks,
>
> Malcolm Cook

--
Professor John Kitchin
Doherty Hall A207F
Department of Chemical Engineering
Carnegie Mellon University
Pittsburgh, PA 15213
412-268-7803
@johnkitchin
http://kitchingroup.cheme.cmu.edu

[O] org-babel-execute:dot -- why isn't this working?

[Matt Price]

I'm trying to draw a very simple, silly graph using dot. I'm following the
code here:
http://irreal.org/blog/?p=2866
and here:
http://orgmode.org/worg/org-tutorials/org-dot-diagrams.html

I have tables like this (the real ones are substantially longer):

#+name: students-graph
| a | b |
| a | m |
| a | f |
| b | t |

#+name: students-table
| *node* | *label*   | *shape* | *fillcolor* |
|+---+-+-|
| a  | Omar  | ellipse | green   |
| b  | Hindia| ellipse | orange  |
| c  | Yuvrai| ellipse | purple  |


I can generate the diagram I'm looking for with this code, mostly stolen
from another :
#+name: make-dot
#+BEGIN_SRC emacs-lisp :var table=students-table graph=students-graph
:results output :exports none
  (mapcar #'(lambda (x)
  (princ (format "%s [label=\"%s\" shape=%s style=\"filled\"
fillcolor=\"%s\"];\n"
 (car x) (second x) (nth 2 x) (nth 3 x) )))
table)
  (mapcar #'(lambda (x)
  (princ (format "%s -- %s;\n"
 (first x) (second x graph)


#+END_SRC

[O] org-babel-execute:dot -- why doesn't this work?

[Matt Price]

I'm trying to draw some silly diagrams with dot, based on code stolen from
tutorials here:
http://irreal.org/blog/?p=2866
and here:
http://orgmode.org/worg/org-tutorials/org-dot-diagrams.html

The code won't work, though I can generate the diagram using a somewhat
clumsier method from here:
http://orgmode.org/worg/org-contrib/babel/languages/ob-doc-dot.html

Here are my two tables, one for nodes and one for edges (the real ones are
longer, with something like 30 or 40 nodes):

#+name: students-table
| *node* | *label*   | *shape* | *fillcolor* |
|+---+-+-|
| a  | Omar  | ellipse | green   |
| b  | Hindia| ellipse | orange  |
| c  | Yuvrai| ellipse | purple  |

#+name: students-graph
| a | b |
| a | c |
| b | c |

I can generate my table using these two code blocks:

#+name: make-dot
#+BEGIN_SRC emacs-lisp :var table=students-table graph=students-graph
:results output :exports none
  (mapcar #'(lambda (x)
  (princ (format "%s [label=\"%s\" shape=%s style=\"filled\"
fillcolor=\"%s\"];\n"
 (car x) (second x) (nth 2 x) (nth 3 x) )))
table)
  (mapcar #'(lambda (x)
  (princ (format "%s -- %s;\n"
 (first x) (second x graph)


#+END_SRC

#+BEGIN_SRC dot :file Images/test-dot.png :var input=make-dot :exports
results
graph {
rankdir=TB
$input
}
#+END_SRC



I would, however, like to avoid the clumsy intermediate step and use
something like this instead:
#+name: graph-from-tables
#+HEADER: :var nodes=students-table graph=students-graph horiz='t
#+BEGIN_SRC emacs-lisp :file ~/example-diagram.png :colnames yes :exports
results
  (org-babel-execute:dot
   (concat
"graph {\n"
(when horiz "rankdir=LR;\n")   ;up-down or left-right
(mapconcat
 (lambda (x)
   (format "%s [label=\"%s\" shape=%s style=\"filled\"
fillcolor=\"%s\"]"
   (car x)
   (nth 1 x)
   (if (string= "" (nth 2 x)) "box" (nth 2 x))
   (if (string= "" (nth 3 x)) "none" (nth 3 x))
   )) nodes "\n")
"\n"
(mapconcat
(lambda (x)
  (format "%s -- %s;"
  (car x) (nth 1 x) )) graph "\n")
"}\n") params)
#+END_SRC


However, this just creates a file ~/example-diagram.png whose content is
"nil%".  I'm not sure how to debug the problem.  Any suggestions? Does this
code work for you guys? I am using an uptodate arch linux, emacs 25, and a
recent graphviz.

thanks as always for the help.

Also -- WISHLIST: instead of a manually-entered list of edges, I'd like to
randomly create a  random set of connections between the nodes. I can
easily create a list of nodes:

 (let ((names () ))
   (dolist (x nodes)
 (push (nth 1 x ) names)
 )
... do some stuff here)

but generating a list of unique edges -- that is, node pairs where
(a . b) and (b . a) are considered equivalent -- is somewhat beyond me.

Again, many thanks,
Matt

Re: [O] org-babel-execute:dot -- why isn't this working?

[Matt Price]

sorry, I must have sent this by accident, rather than deleting it by
accident as I'd thought. please see the finished email, which you've
already received.

On Wed, Sep 23, 2015 at 8:51 PM, Matt Price  wrote:

> I'm trying to draw a very simple, silly graph using dot. I'm following the
> code here:
> http://irreal.org/blog/?p=2866
> and here:
> http://orgmode.org/worg/org-tutorials/org-dot-diagrams.html
>
> I have tables like this (the real ones are substantially longer):
>
> #+name: students-graph
> | a | b |
> | a | m |
> | a | f |
> | b | t |
>
> #+name: students-table
> | *node* | *label*   | *shape* | *fillcolor* |
> |+---+-+-|
> | a  | Omar  | ellipse | green   |
> | b  | Hindia| ellipse | orange  |
> | c  | Yuvrai| ellipse | purple  |
>
>
> I can generate the diagram I'm looking for with this code, mostly stolen
> from another :
> #+name: make-dot
> #+BEGIN_SRC emacs-lisp :var table=students-table graph=students-graph
> :results output :exports none
>   (mapcar #'(lambda (x)
>   (princ (format "%s [label=\"%s\" shape=%s style=\"filled\"
> fillcolor=\"%s\"];\n"
>  (car x) (second x) (nth 2 x) (nth 3 x) )))
> table)
>   (mapcar #'(lambda (x)
>   (princ (format "%s -- %s;\n"
>  (first x) (second x graph)
>
>
> #+END_SRC
>
>
>