Re: [O] ID property generated by org-mobile

2015-03-21 Thread Alexis


On 2015-03-20T22:45:02+1100, Randomcoder randomcod...@gmail.com 
said:


R Does the ID property have any use? I know I can disable it, 
but R where


R is it being used ?  (the ID property that org-mobile generates 
for R each heading)


R Are there any drawbacks to just disabling it with (setq R 
org-mobile-force-id-on-agenda-items nil) ?


R Is it being used somewhere in particular ?

i think it might be used for syncing purposes, to uniquely 
identify items that need to be added, removed etc.



Alexis.



Re: [O] Possible bug: Can not search for text in links - only description

2015-03-21 Thread Alexis


On 2015-03-21T01:59:07+1100, Richard Lawrence 
richard.lawre...@berkeley.edu said:


RL I am not sure if this counts as a bug or not, so someone else 
RL should still address this question.  Maybe this is the 
desired RL behavior, given that the link text is hidden?  Or 
maybe it's just RL not possible to search in hidden text?


Cf. the `search-invisible` variable:

   search-invisible is a variable defined in `isearch.el'.  Its 
   value is open 

   Documentation: If t incremental search/query-replace can match 
   hidden text.  A nil value means don't match invisible text. 
   When the value is `open', if the text matched is made 
   invisible by an overlay having an `invisible' property and 
   that overlay has a property `isearch-open-invisible', then 
   incremental search will show the contents.  (This applies when 
   using `outline.el' and `hideshow.el'.) 

   To temporarily change the value for an active incremental 
   search, use M-s i. 



Alexis.



Re: [O] [Org-Caldav] Trouble connecting to a caldav server other Google calendar

2015-03-13 Thread Alexis


On 2015-03-14T01:11:35+1100, Charles-H. Schulz c...@adocentyn.io 
said:


CS Browsing this list's archives I did find, however, that a 
CS few other people had problems outside Google Calendar.


Perhaps this might be of use?

https://github.com/dengste/org-caldav/issues/45#issuecomment-68527518

(i've not yet tried it, myself, though i hope to get to doing so 
in the near future )


There might also be a connection with this Emacs issue:

http://debbugs.gnu.org/cgi/bugreport.cgi?bug=18290


Alexis.



Re: [O] Bleeding edge in elpa

2015-03-10 Thread Alexis


On 2015-03-10T06:21:10+1100, T.F. Torrey tftor...@tftorrey.com 
said:


TFT The biggest obstacle I to this that I see is that developers 
tend TFT to hate the package manager and love git.


This developer appreciates both, and thus particularly appreciates 
MELPA as well. :-)



Alexis.



Re: [O] Bug: `org-edit-special` doesn't respect dedicated windows

2015-03-07 Thread Alexis


On 2015-03-08T07:34:36+1100, Nicolas Goaziou 
m...@nicolasgoaziou.fr said:


NG Alexis flexibe...@gmail.com writes:

 Thanks for bringing my attention to this variable! However, 
 setting it to 'current-window seems to have no effect; rather 
 than the Org buffer being replaced by an Emacs Lisp buffer, 
 either a new window is created for the latter, or an existing 
 (non-dedicated) window is taken over by the latter. Am i 
 missing something / doing something else wrong?
 
 (Btw, i'm using version 20150223 of Org from the orgmode.org 
 ELPA - sorry i didn't mention that in my previous message!)


NG I cannot reproduce the problem with `current-window' on NG 
development branch.


Okay, it turns out that the problem was the 
`display-buffer-function` variable being set elsewhere, which 
prevented `org-src-window-setup` from working correctly. Sorry for 
the hassle.



Alexis.



[O] Setting `org-src-window-setup` has no effect? [was: Re: Bug: `org-edit-special` doesn't respect dedicated windows]

2015-03-06 Thread Alexis


Ping? 


On 2015-03-01T20:31:33+1100, Alexis flexibe...@gmail.com said:

A On 2015-03-01T20:20:06+1100, Nicolas Goaziou said: NG Org 
uses `org-src-window-setup' to control the display, which NG 
overrides dedicated windows. You may want to customize the A 
former.


A Thanks for bringing my attention to this variable! However, A 
setting it to 'current-window seems to have no effect; rather 
than A the Org buffer being replaced by an Emacs Lisp buffer, 
either a A new window is created for the latter, or an existing 
A (non-dedicated) window is taken over by the latter. Am i 
missing A something / doing something else wrong?


A (Btw, i'm using version 20150223 of Org from the orgmode.org 
ELPA A - sorry i didn't mention that in my previous message!)



A Alexis.




Re: [O] Emacs unresponsive while executing Sh code block

2015-03-05 Thread Alexis


On 2015-03-06T05:40:33+1100, Giacomo M jackja...@gmail.com said:

GM Dear all,

GM when I C-c C-c in: #+BEGIN_SRC sh gnome-terminal #+END_SRC

GM a gnome-terminal window appears, but Emacs hangs until I 
close GM it.  In the *Messages* I get:


GM executing Sh code block...  Wrote GM 
/tmp/babel-2307H-J/ob-input-2307i3c (here Emacs hangs... then 
GM I close gnome-terminal) Error reading results: GM 
(beginning-of-buffer) Code block produced no output.


My guess is that Emacs is waiting for the final output from 
`gnome-terminal`, which doesn't get sent until you close 
`gnome-terminal`.



Alexis.



Re: [O] refiling with helm

2015-03-05 Thread Alexis


On 2015-03-06T02:13:41+1100, Leo Ufimtsev lufim...@redhat.com 
said:


LU Hello Xebar,

LU I had the same issue. I used the file-expand-wildcards 
function LU to make a list of all my org-mode files.


LU The only thing is that I have to reload my .emacs when adding 
org LU files for refile to work properly.


You shouldn't need to do that; once you've added a new Org file, 
you can just move point to the end of the `(setq myvar/org-files 
...` s-expression, and press C-x C-e (`eval-last-sexp`). That 
should cause the `setq` to be re-evaluated, such that the new file 
becomes part of the value of the `myvar/org-files` variable.



Alexis.



Re: [O] org-calendar-holiday and local holidays

2015-03-02 Thread Alexis


On 2015-03-03T08:58:20+1100, Jorge A. Alfaro-Murillo 
jorge.alfaro-muri...@yale.edu said:


JAA holidays.el appends holiday-local-holidays to 
calendar-holidays JAA via a defcustom, so if you set 
holiday-local-holidays in your JAA .emacs, restart emacs and the 
local holidays are not in JAA calendar-holidays, it is because 
you are calling something that JAA loads holidays.el before you 
set holiday-local-holidays.


JAA If you add the code above to your .emacs and later modify 
your JAA configuration and remove or move the part that loads 
JAA holidays.el, then either your code will fail (because JAA 
calendar-holidays is not yet defined) or calendar-holidays will 
JAA have your local holidays twice and they will show twice in 
your JAA agenda.


JAA I think that you should look for whatever calls holidays.el 
and JAA set holiday-local-holidays before that.


i just tried moving my `(setq holiday-local-holidays ...)` to the 
very first line of my config setup, and lo, that does result in 
local holidays appearing in my Org agenda. However, my config 
setup is a 3000+ line Org Babel file, in which i group together 
things that are related in my mind, and the setup for the calendar 
is about a third of the way through this. Thus 


JAA If not, then at least use eval-after-load so that JAA 
calendar-holidays is already defined when the code is run, and 
JAA add-to-list so that the entries do not get added twice if 
they are JAA already there:


JAA #+BEGIN_SRC emacs-lisp (eval-after-load 'holidays '(dolist 
JAA (holiday holiday-local-holidays) (add-to-list 
'calendar-holidays JAA holiday)) #+END_SRC


 works better in my context, and is more robust, longer-term, 
than my original suggestion to use


   (setq calendar-holidays (append calendar-holidays 
   holiday-local-holidays)) 

So, thank you! Although i do note that my suggestion was 
nevertheless within the guidelines of the documentation for 
`calendar-holidays`:


   Note that these variables have no effect on 
   `calendar-holidays' after it has been set (e.g. after the 
   calendar is loaded).  In that case, customize 
   `calendar-holidays' directly.


i feel the above documentation could be improved by adding that 
`eval-after-load` should probably by used in this context, e.g.:


   Note that these variables have no effect on 
   `calendar-holidays' after it has been set (e.g. after the 
   calendar is loaded).  In that case, customize 
   `calendar-holidays' directly, for example by using 
   `eval-after-load':


   (eval-after-load 'holidays
'(dolist (holiday holiday-local-holidays)
  (add-to-list 'calendar-holidays 
  holiday)))


i'll open a GNU Emacs issue to that effect. :-)

Thanks again!


Alexis.



Re: [O] org-calendar-holiday and local holidays

2015-03-02 Thread Alexis


On 2015-03-03T02:26:37+1100, Jorge A. Alfaro-Murillo said:

JAA Alexis writes:

 When i scroll down to look at the current value of 
 `calendar-holidays`, however, i see that neither the current 
 value nor the original value makes any reference to the 
 `holiday-local-holidays` variable. And indeed, when i examine 
 my agenda for next Monday, which is a local holiday i've 
 specified in `holiday-local-holidays`, i can't see that local 
 holiday. To fix this, i use M-: to evaluate:
 
 (setq calendar-holidays (append calendar-holidays 
 holiday-local-holidays))
 
 after which the local holiday next Monday appears in my Org 
 agenda.


JAA You do not need to add that, calendar-holidays appends JAA 
holiday-local-holidays when holidays.el is loaded, just restart 
JAA emacs.


Not in my Emacs (manually compiled 24.4.1, the most recent 
official stable release). My `local-holidays` variable was set for 
years, such that only as part of trying to help the OP did i 
notice that it's been obsoleted; the documentation for it says:


 This variable is an alias for `holiday-local-holidays'. 
 This variable is obsolete since 23.1; use 
 `holiday-local-holidays' instead.


So i changed my init to refer to `holiday-local-holidays` instead 
of `local-holidays`, and restarted Emacs, and the issue persisted: 
the value of `holiday-local-holidays` is /not/ included in 
`calendar-holidays` by default. The `(setq calendar-holidays ...` 
line i described above is necessary to work around this.


JAA It is also not a documentation bug, at least in my emacs 
JAA (25.0.50.1) the documentation of calendar-holidays says 
clearly:


JAA Note that these variables [`holiday-other-holidays', JAA 
`holiday-general-holidays', `holiday-local-holidays', JAA 
`holiday-christian-holidays', `holiday-hebrew-holidays', JAA 
`holiday-islamic-holidays', `holiday-bahai-holidays', JAA 
`holiday-oriental-holidays' and `holiday-solar-holidays'] have 
JAA no effect on `calendar-holidays' after it has been set JAA 
(e.g. after the calendar is loaded). In that case, customize JAA 
`calendar-holidays' directly.


In 24.4.1, the documentation is phrased differently; it says:

   Additional holidays are easy to add to the list, just put them 
   in the list `holiday-other-holidays' in your init file. 
   Similarly, by setting any of `holiday-general-holidays', 
   `holiday-local-holidays', `holiday-christian-holidays', 
   `holiday-hebrew-holidays', `holiday-islamic-holidays', 
   `holiday-bahai-holidays', `holiday-oriental-holidays', or 
   `holiday-solar-holidays' to nil in your init file, you can 
   eliminate unwanted categories of holidays.


   The aforementioned variables control the holiday choices 
   offered by the function `holiday-list' when it is called 
   interactively.


   They also initialize the default value of `calendar-holidays', 
   which is the default list of holidays used by the function 
   `holiday-list' in the non-interactive case.  Note that these 
   variables have no effect on `calendar-holidays' after it has 
   been set (e.g. after the calendar is loaded).  In that case, 
   customize `calendar-holidays' directly.


   The intention is that (in the US) `holiday-local-holidays' be 
   set in site-init.el and `holiday-other-holidays' be set by the 
   user.


It's the fact that, despite the above docstring, and that, as i 
described above, setting the value of `holiday-local-holidays` has 
no direct effect on `calendar-holidays` /even after a restart of 
Emacs/, that led me to suggest there might be a code bug or a 
documentation bug (e.g. maybe some variable needed to be set to 
`t` to ensure the value of `holiday-local-holidays` gets included 
in `calendar-holidays`).


Since things work for you, and the phrasing for the documentation 
for `calendar-holidays` has changed between the most recent stable 
release and the development version of Emacs you're using, my 
guess is that there is indeed a bug in 24.4.1 and earlier that has 
subsequently been fixed. Later today i'll try building from the 
first 24.5 pretest and the master branch, and examine what happens 
with `holiday-local-holidays` / `calendar-holidays` in both 
instances.



Alexis.



Re: [O] Bug: `org-edit-special` doesn't respect dedicated windows

2015-03-01 Thread Alexis


On 2015-03-01T20:20:06+1100, Nicolas Goaziou said:

NG Org uses `org-src-window-setup' to control the display, which 
NG overrides dedicated windows. You may want to customize the 
former.


Thanks for bringing my attention to this variable! However, 
setting it to 'current-window seems to have no effect; rather than 
the Org buffer being replaced by an Emacs Lisp buffer, either a 
new window is created for the latter, or an existing 
(non-dedicated) window is taken over by the latter. Am i missing 
something / doing something else wrong?


(Btw, i'm using version 20150223 of Org from the orgmode.org ELPA 
- sorry i didn't mention that in my previous message!)



Alexis.



Re: [O] org-calendar-holiday and local holidays

2015-03-01 Thread Alexis


[Crossposted to the help-gnu-emacs list, for possible advice on 
whether or not this involves a bug in GNU Emacs.]


On 2015-03-02T09:29:09+1100, Melleus said:

M I'm afraid to ask. But... Anyway. Does 
%%(org-calendar-holiday) M know about holiday-local-holidays? 
I'm not programmer, sorry. I've M set up those local holidays 
but cannot see them in my agenda.


You can examine the definition of an ELisp function by:

1. typing C-h f whilst on a function;
2. typing RET to take you to the documentation for that function;
3. typing TAB then RET to take you to the function definition.

Starting with point on `org-calendar-holiday`, we find that:

- `org-calendar-holiday` calls (if available) 
 `calendar-check-holidays` or (otherwise) 
 `check-calendar-holidays`;


- `calendar-check-holidays` calls `calendar-holiday-list`;

- `calendar-holiday-list` makes use of the `calendar-holidays` 
 variable.


We can examine the documentation for the `calendar-holidays` 
variable by moving point onto and typing C-h v RET. On my setup 
(manually compiled Emacs 24.4.1 on Debian Wheezy(+updates) x86_64 
together with Org 20150223), the documentation suggests that 
`calendar-holidays` makes use of the `holiday-local-holidays` 
variable; and the documentation for `holiday-local-holidays` 
merely refers us back to the documentation for 
`calendar-holidays`.


When i scroll down to look at the current value of 
`calendar-holidays`, however, i see that neither the current value 
nor the original value makes any reference to the 
`holiday-local-holidays` variable. And indeed, when i examine my 
agenda for next Monday, which is a local holiday i've specified in 
`holiday-local-holidays`, i can't see that local holiday. To fix 
this, i use M-: to evaluate:


   (setq calendar-holidays (append calendar-holidays 
   holiday-local-holidays))


after which the local holiday next Monday appears in my Org 
agenda.


Given the documentation for the `calendar-holidays` variable, the 
fact that i need to manually add the value of the 
`holiday-local-holidays` variable to `calendar-holidays` seems to 
me like it might be a coding or documentation bug in Emacs  ?



Alexis.



Re: [O] Set a property or take action on state change to done

2015-02-28 Thread Alexis


On 2015-02-19T10:31:05+1100, Subhan Michael Tindall said:

SMT I know the hooks are in there somewhere, but is there any 
SMT relatively straightforward way to set a particular property, 
or SMT run a function to do so, when a TODO changes status to a 
DONE SMT state?


Perhaps use a combination of `org-after-todo-state-change-hook` 
and the `org-state` variable?



Alexis.



[O] Bug: `org-edit-special` doesn't respect dedicated windows

2015-02-28 Thread Alexis


Context:

Manually compiled Emacs 24.4.1 on Debian Wheezy(+updates) x86_64.

ECM:

1. emacs -Q
2. Split frame into two windows via `C-x 3`.
3. Make one *scratch* buffer window dedicated via:
  (a) `M-:`
  (b) (set-window-dedicated-p (selected-window) t)
4. Select the other window.
5. Visit file `test.org`, containing:

   * Test
   #+begin_src emacs-lisp
 (message Test.)
   #+end_src

6. Move point within source block, do `org-edit-special`.
7. The *scratch* buffer window will have its contents changed to 
  contain an Org buffer.



Alexis.



Re: [O] Citation syntax and ODT

2015-02-23 Thread Alexis


On 2015-02-24T10:25:39+1100, Richard Lawrence said:

RL Vaidheeswaran C vaidheeswaran.chinnar...@gmail.com writes:

 But whatever style is chosen, I would still think that the 
 fact that the citation is in-text rather than parenthetical, 
 and that it has a prefix and suffix, should be represented in 
 the output.
 
 1. When you choose 'style' (Chicago etc.) wouldn't be one of 
 in-text or parenthetical already chosen for you?  Stated other 
 way, is the choice between parenthetical or in-text 
 document-wide or is it that one could intermix the two styles 
 in the same document.


RL These could be intermixed in the same document.  The RL 
document-level style determines how each type ultimately looks, 
RL but the choice of style is (mostly) independent of using RL 
parenthetical vs. in-text citations.


Fwiw, it seems to me that there might be some confusion here arising
from two separate usages of the word 'style':

(a) 'style' to mean writing style, i.e. which words are used, 
how they're put together, etc. For example, Plain English.


(b) 'style' to mean presentation style, i.e. how words, symbols, 
glyphs etc. are presented visually or (for vision-impaired people) 
aurally. For example, the sort of things specified by Cascading 
Style Sheets.


Thus, the citation 'style' can be independent of the presentation 
style used for that citation style.


For examle, one might have a citation style like:

[Smith 2001]

which in certain contexts is expected to have a presentation style 
of 'bolded'.


So what i understand Vaidheeswaran to be asking is: Please don't 
code things such that presentation style is /necessarily/ carried 
along with citation style. Make it so that exporting a document 
faithfully reproduces the citation style in the target format, but 
don't /force/ the presentation style used in the source format for 
citation style to be the presentation style used in the 
destination format.


Vaidheeswaran, is that correct?


Alexis.



Re: [O] bug#19887: 24.4; Cannot kill buffer; Wrong type argument: overlayp, nil

2015-02-20 Thread Alexis


On 2015-02-20T18:32:12+1100, Damian Nadales said:

DN That's weird. Which buffer are you killing, the DN Org 
buffer, or the new buffer that is created when pressing C-c ' DN 
inside the source block? Is the latter the one I cannot kill.


DN I've only installed packages through the emacs DN package 
manager, and Org mode was not one of them. The command DN 
``org-version`` reports:


DN Org-mode version 8.2.10 (release_8.2.10 @ DN 
/home/dnadales/opt/share/emacs/24.4/lisp/org/)


Having thought about it a bit, i seem to remember that a while 
back, i had the same issue as the one you're currently 
experiencing. (Even though, as i noted above, i can't reproduce it 
my current setup, a manually compiled 24.4.1 on Debian 
Wheezy(+updates) x86_64.)  Unfortunately, i can't remember what 
made the problem go away; the only two things that come to mind 
are:


* installing and using Org from the orgmode.org ELPA, i.e.

   (add-to-list 'package-archives 
'(org . http://orgmode.org/elpa/;) t) 


or

* removing existing Org-related byte-compiled files (i.e. *.elc 
 files), 
 so that they got regenerated.



Alexis.



[O] bug#19887: 24.4; Cannot kill buffer; Wrong type argument: overlayp, nil

2015-02-17 Thread Alexis


On 2015-02-18T06:25:57+1100, Glenn Morris said:

GM Damian Nadales wrote:

 - Run emacs -Q
 
 - Create an org-mode file (i.e. ``myorgfile.org``)
 
 - Insert the following text:
 
 o #+BEGIN_SRC C++
 
 #+END_SRC
 
 - Edit the source block by placing the cursor inside the SRC 
 block.
 
 - Start C++ mode (c++-mode)
 
 - Try to kill the buffer.


GM Please do

GM M-x toggle-debug-on-error

GM repeat the problem, and post the resulting backtrace.  (I 
can't GM reproduce it.)


Running a manually compiled Emacs 24.4.1 on Debian 
Wheezy(+updates) x86_64, and following the above steps, i can't 
reproduce this either.


From the above description, i assume by the Start C++ mode line, 
you're not moving the cursor inside the source block and then 
doing:


   C-c '

(i.e. `org-edit-special`) in order to open a c++-mode buffer for 
editing the block contents?



Alexis.





Re: [O] orgmode-contacts wrong type arguments

2015-02-17 Thread Alexis


On 2015-02-17T00:17:59+1100, Tory S. Anderson said:

TSA In my efforts to improve my elisp, can anyone tell me why 
the code TSA doesn't work, and what might have changed to cause 
it to break?


TSA Error: completion-in-region: Wrong type argument: listp, 
#(NAME TSA addr...@gmail.com, NAME addr...@aol.com, NAME 
TSA addr...@gmail.com, NAME addr...@gatech.edu 0 15 
(fontified TSA nil org-category contacts) 44 65 (fontified nil 
org-category TSA contacts) 88 99 (fontified nil org-category 
contacts) 127 TSA 141 (fontified nil org-category contacts))


It looks like something in `completion-in-region` wanted a plain 
old list of items, but instead got a propertized string. For 
further information, check out section 31.19.2, Changing Text 
Properties of the Emacs Lisp Reference Manual - in particular, 
the entry for the `propertize` function.



Alexis.



[O] org-carddav [was: Re: ANN: org-vcard. Export/import vCards. Backwards-compatible with org-contacts.el.]

2015-02-13 Thread Alexis

On Wed 21 January 2015, Rasmus wrote:

 Gour address@hidden writes:

 On Sub, 2014-08-02 at 11:12 +1000, Alexis wrote:

 2. Using org-vcard as a library, create org-carddav (which i hope to
 start working on shortly) in order to be able to synchronise contacts
 stored in Org with arbitrary CardDAV servers.

 Cool stuff!

Sorry i've not replied to this sooner; i've had a lot happening, and
haven't been able to follow this list.

Unfortunately my work on org-carddav got stymied at an early stage by
this Emacs issue:

http://debbugs.gnu.org/cgi/bugreport.cgi?bug=18290

Iirc, this not only presented an obstacle to interaction with Google's
CardDAV service, but also with a DAViCal server i set up on my LAN.

If someone familiar with the source of url-http.el reviewed this, and
amended its code along the lines i suggested in my bug report - even if
only in the Emacs master branch - i would happily resume working on
org-carddav. :-)


Alexis.



Re: [O] org-plus-contrib yes, org no!

2014-12-10 Thread Alexis

Sharon Kimble writes:

 My emacs version is in the sig, and my org-version is Org-mode
 version 8.2.10 (8.2.10-23-g1ec416-elpaplus @
 /home/boudiccas/.emacs.d/elpa/org-plus-contrib-20141208/) so we're
 in agreement, but still it happens. I've checked everywhere in my
 init.org and I don't have a spare (require 'org) floating
 around, its not in there at all!

 I'm puzzled!

My guess is that the version of Org that comes bundled with Emacs is
being loaded. If so, you could try renaming the `org` directory in your
Emacs installation's `lisp` directory to e.g. `.org` so that it's not
found by default, then restart Emacs.


Alexis.



Re: [O] Getting org-mobile-sync to work.

2014-10-17 Thread Alexis

Sharon Kimble writes:

 I'm trying to get mobileorg set up and working, which has worked. But I
 also want to use org-mobile-sync.el from ELPA, the actual package
 being org-mobile-sync-20131118.1116. Looking at the source file it
 says -

 --8---cut here---start-8---
 ;;; Commentary:

 ;; Adds delayed `org-mobile-push' upon saving files that are part of
 ;; `org-mobile-files-alist'. Watches the `org-mobile-capture-file' for
 ;; changes with `file-notify.el' and then invokes `org-mobile-pull'.

 ;;; Requirements:

 ;; Emacs 24.3.50 with `file-notify-support' is required for it to work.
 --8---cut here---end---8---

 But I can't find file-notify.el on my system, nor by googling for it
 either. Can somebody help me to get the file please?

 How do I know if file-notify-support is part of my Emacs please?

If i understand correctly (and if i don't, please correct me!), 24.3.50
was the base for what has become the 24.4 pre-release / release
candidate series. i'm running version 24.3.94.2, via the emacs-24
branch, which contains a package `filenotify.el` (note lack of dash!),
and which i can successfully `require`  Perhaps this is what
`org-mobile-sync` is wanting?


Alexis.



Re: [O] Enforcing newlines in plain text export

2014-09-27 Thread Alexis

Richard Lawrence writes:

 Does anyone know why the behavior of \\ is presently restricted to
 appearing at the end of the line in a paragraph?  Does anyone need it
 to be exported literally when it appears elsewhere?

When exporting code listings in any programming language which uses '\'
for escaping characters in strings, and subsequently uses '\\' to mean
literal backslash, literal export is required. For example, one might
have an Org document with Emacs Lisp source:

  -
 ' * A
 | ** B
 | #+begin_src emacs-lisp
 |   (setq my-regex (.+))
 | #+end_src
 `-


Alexis.



Re: [O] Emacs server and org-protocol

2014-09-17 Thread Alexis

Fabrice Popineau writes:

 Yes seconds or even much more.
 This is the reason I don't use Emacs to read my mail/news anymore :-/

If i may ask, which email front-end were you using? (Gnus, perhaps?) i
used to use notmuch.el, and currently use mu4e, and basically don't have
this issue 


Alexis.



Re: [O] Emacs server and org-protocol

2014-09-17 Thread Alexis

Eric S Fraga writes:

 The issue is not the email front-end per se but the email servers (IMAP,
 POP, whatever).  A couple of years ago, I ended up having to use an
 email server that would take many seconds, often minutes, to access,
 even just to query to find out if there was any new email.

Well, i run getmail+procmail to fetch and sort my mail, so if a server
is running slowly, that doesn't affect Emacs. Emacs is only affected by
e.g. mu indexing newly-arrived emails, which is fast enough that i don't
notice it.


Alexis.



Re: [O] Emacs server and org-protocol

2014-09-17 Thread Alexis

Stefan Huchler writes:

 What I have more a problem with is that it often hangs have to cancel
 it at least every seond time.

Sounds like something that might need to be reported as a bug?


Alexis.



Re: [O] crypt and sync

2014-09-06 Thread Alexis

Hannes Schulz writes:

 Shouldn't org-crypt disregard property drawers?

The current org-contacts format stores contact details in property
drawers; were it ignored by org-crypt, all those details would be lost
in the encryption process. So ignoring of property drawers by org-crypt
would probably best be controlled by an in-buffer setting or an
org-crypt variable.


Alexis.



[O] Typo in org guide

2014-09-03 Thread Alexis Roda

Hi all,
the org guide Release 8.2.7c says in page 32, 12.1 Export options:

  The whole set of lines can be inserted into the buffer with C-c C-e t.

It should say:

  The whole set of lines can be inserted into the buffer with C-c C-e #.


Regards



Re: [O] Triggering clock in/out from org state change and progress logging

2014-09-01 Thread Alexis Roda

Hi list,
after struggling a bit with edebug (too many things to learn) I have 
traced what the problem is. For the archives:


State-change logging is deferred until the command finishes and is 
implemented by the function 'org-add-log-setup' adding the function 
'org-add-log-note' to the hook 'post-command-hook'. The details of the 
note are stored in global variables 'org-log-note-xxx'.


That implementation prevents recursive calls to 'org-todo' (that's what 
my code does in order the pause the active task) from logging both the 
paused and the started log messages.


'org-add-log-note' will be called only once since the hook won't accept 
duplicated entries. Anyway, the innermost call to 'org-add-log-setup' 
will overwrite the variables 'org-log-note-xxx' with the details of the 
note being paused.


The outcome is that only the state change for the paused note is being 
logged.



Now that I know what the problem is I only have to figure a workaround. 
Any clue?




Regards




[O] Triggering clock in/out from org state change and progress logging

2014-08-26 Thread Alexis Roda

Hi list,
as part of learning org I'm trying to configure it so that state changes 
trigger clocking in/out tasks.


Why testing workflow is:

#+TODO: TODO(t) STRT(s!) PAUS(p!) WAIT(w!) | DONE(d!) CANC(c!)

What I try to accomplish:

* entering STRT pauses the active task (if any) and clocks-in the 
current task.


* entering PAUS or WAIT clocks-out the current task (if clocking the 
current task).


* entering TODO probably should reset the current task but I don't care 
at that point.



Changing state and clocking in/out works great but logging is weird:

* Tasks

** STRT first task
   - State STRT   from TODO   [2014-08-26 dt 12:27]
   :20:
   CLOCK: [2014-08-26 dt 12:27]
   :END:
** TODO second task


Now I change second task from TODO to STRT:


* Tasks

** PAUS first task
   - State PAUS   from STRT   [2014-08-26 dt 12:28]
   - State STRT   from TODO   [2014-08-26 dt 12:27]
   :20:
   CLOCK: [2014-08-26 dt 12:27]--[2014-08-26 dt 12:28] =  0:01
   :END:
** STRT second task
   :20:
   CLOCK: [2014-08-26 dt 12:28]
   :END:


first taks gets paused and second task is started and clocking but 
no logging State 'STRT' from 'TODO' ... is added to second task.


Starting a task only logs in the other task.

After 6 hours struggling with that (clocked with org :) ) I can't figure 
why that's happening. Any clue?



Thanks in advance

Here's the code:

(defvar arv/org-todo-entering-state-clocking-actions
  '((STRT . start)
(PAUS . stop)
(WAIT . stop)))

(defvar arv/org-todo-state-change-triggers-clocking-enabled t)
(defvar arv/org-todo-paused-state PAUS)

(defun arv/org-todo--pause-other-task ()
  (when (org-clock-is-active)
(save-excursion
  (org-clock-goto)
  (org-clock-out)
  (let ((arv/org-todo-state-change-triggers-clocking-enabled nil)) 
; avoid recursion

(org-todo arv/org-todo-paused-state)

(defun arv/org-todo--state-change (from to)
  (when (and arv/org-todo-state-change-triggers-clocking-enabled
 (not (string= from to)))
(let ((action (cdr (assoc to 
arv/org-todo-entering-state-clocking-actions

  (unless (null action)
(cond
 ((eq action 'start)
  (arv/org-todo--pause-other-task)
  (org-clock-in))
 ((eq action 'stop)
  (let ((org-state DONE) ; hackish, review
(org-clock-out-when-done t))
(org-clock-out-if-current)))
 (t (user-error Unknown action.)))

(defun arv/org-todo--state-change-trigger (p)
  (let ((type (plist-get p :type))
(from (plist-get p :from))
(to   (plist-get p :to)))
(when (eq type 'todo-state-change)
  (arv/org-todo--state-change from to


(add-hook 'org-trigger-hook 'arv/org-todo--state-change-trigger)



[O] Capture template expansion

2014-08-24 Thread Alexis Roda

Hi all,
org newbie here, neither proficient with org nor english, please be patient.

I'm having a hard time figuring how to gather property values with 
completion from a capture template.


8-[ .emacs ]
  (setq org-capture-templates
  '((t Todo entry (file+headline gtd.org Tasks)
 * TODO %?
  :PROPERTIES:
  :TIPUS: %^{TIPUS}p
  :END:
)))
8---

8-[ gtd.org ]---
#+PROPERTY: TIPUS_ALL foo bar

* Tasks
8---

gtd.org defines the allowed values for TIPUS but M-x org-capture does 
not provides completion when evaluating %^{TIPUS}p. After expanding the 
template, C-c C-x p does provide completion.


That's intentional? a bug in my setup?



TIA



[O] Error in org guide

2014-08-24 Thread Alexis Roda

Hi all,
the org guide Release 8.2.7c says in page 20, 8.4 Clocking work time:

C-c C-x C-x
  Cancel the current ...

It should say:

C-c C-x C-q


Regards



Re: [O] Some thoughts on MobileOrg and its development ....

2014-08-13 Thread Alexis

Carlos Sosa writes:

 There's a difference feature wise between MobileOrgNG[1] and
 MobileOrg-Android[2], which I believe should be merged into one project.
 For instance, I don't know of MobileOrgNG having a SSH synchronizer
 which many people make use of, including me, but with that said,
 MobileOrgNG has a far greater and friendlier interface.

 I would suggest merging both projects for the sake of a better user
 experience.

Well, again, the fundamental problem here is that none of these three
MobileOrg* projects are being actively worked on / maintained. The last
commit to MobileOrgNG was on 13 February last year (and is 149 commits
ahead, 863 commits behind MobileOrg-Android). And despite
not-insignificant discussion here on the Org mailing list about the
MobileOrg* projects over the last few months, none of the lead devs for
any of these projects has chimed in, which suggests work on the Org
ecosystem is a very low priority for them right now. So it seems to me
that the likelihood of getting the two projects to work on a merge
process is similarly very low.

It's certainly true that someone could create a third Android-specific
version of MobileOrg that combines what they feel to be the best of both
Android apps. However, assuming that new app took off, we would still
have two different implementations on two different platforms where
fixes and improvements to code and documentation on one platform
wouldn't be immediately useful and/or appropriate for the other
platform. (Not to mention there'd still not be even theoretical support
for platforms other than iOS and Android.)

From a purely selfish point of view, as an Android user, a new Android
app combining MobileOrg-Android and MobileOrgNG would probably meet my
needs in the short term. In the medium to long term, however, i'm not
yet convinced that it would make the Org-on-mobile-devices situation
much more sustainably active than it is now.


Alexis.



Re: [O] Some thoughts on MobileOrg and its development ....

2014-08-12 Thread Alexis

Ashton Kemerling writes:


 I can say for certain that we would have to figure out the handoff of
 various credentials from the old maintainers (who I am assuming would
 not like to continue being maintainers) for the respective app-stores
 and Dropbox tokens.

Not necessarily. One could, for example, create an entirely new project
on GitHub called 'MobileOrgRebooted', and create entirely new apps in
the respective stores using that name. (As it is, there's not a
uniformly named app in any case - we have 'MobileOrg' for iOS, and
'MobileOrg-Android' for, well, Android.) And it certainly seems to me
that it would be best to start the actual coding of the reboot /first/,
and only worry about naming rights issues if and when it takes
off. Doing otherwise is likely to bring into play another possible
obstacle to getting actual implementation happening.


Alexis.



Re: [O] Some thoughts on MobileOrg and its development ....

2014-08-12 Thread Alexis

David Wagle writes:

 What's involved in 'rebooting' the project?

My thought involves starting from scratch. Say we use Cordova (i can't
speak to the pros and cons of Cordova vs. Kivy in terms of things like
functionality provided, possible barriers to contribution etc.). Cordova
takes care of a certain amount of the (semi-)boilerplate code needed for
each platform; and of the remainder of the code, the core functionality
would need to get reimplemented in HTML/CSS/JavaScript, with 'shims'
being created between this code and the surrounding platform-specific
code. Consequently, i think there would likely be very little use of
most of the code in either current MobileOrg project.


Alexis.



[O] Some thoughts on MobileOrg and its development ....

2014-08-08 Thread Alexis

Hi all,

In light of recent discussions about 'MobileOrg' - which seemingly
actually constitutes two distinct projects for two different platforms -
together with the apparent relative lack of activity of both projects,
despite demand for them, i can't help but wonder if the 'MobileOrg'
endeavour needs a reboot.

More specifically, it seems to me that rebuilding MobileOrg as a single
project on top of Apache Cordova:

https://cordova.apache.org/

might be a way forward, for several reasons:

* It would help ensure that basically the same set of functionality is
  available across platforms, modulo specific limitations/issues with
  specific platforms. And when people add or modify core functionality,
  that functionality would more easily become available to people across
  platforms, rather than the functionality being initially implemented
  on platform X, and people on platform Y having to wait for it to be
  implemented in its entirety.

* Overall, only one lot of end-user documentation would need to be
  maintained.

* It would enable MobileOrg to be made available for mobile platforms
  other than Android and iOS, such as Windows Phone and Blackberry.

* The number of people available to assist with development might well
  be greater, due to the core development environment involving HTML,
  CSS and JavaScript. Barriers to entry for both regular and occasional
  committers would be much lower.

i'm aware that Cordova has various limitations,
including-but-not-limited-to the lack of native 'feel' of Cordova-based
applications. However, i feel that the above advantages, combined with
the my notion that Emacs users are probably less concerned with a
perfectly slick UI than with having access to functionality they
need/want, probably outweighs such limitations.

Unfortunately, due to other existing commitments, i wouldn't be able to
take point on such a reboot. But i'd definitely be willing and able to
help out!  Particularly in the area of contact management and syncing,
of course. :-)

Thoughts/comments/criticisms?


Alexis.



Re: [O] MobileOrg documentation?

2014-08-07 Thread Alexis

David Masterson writes:

 Unfortunately, this looks like Android documentation where I have an
 iPhone.

i didn't even realise MobileOrg was available for iOS until this moment!
It looks like there are two quite separate codebases (neither of which
seems to be actively maintained) which complicates things further; i can
easily imagine the provided functionality between both versions
gradually drifting apart 


Alexis.




Re: [O] ANN: org-vcard. Export/import vCards. Backwards-compatible with org-contacts.el.

2014-08-07 Thread Alexis

Charles Philip Chan writes:

 Personally I find no MUA as usable and feature rich as Gnus. ;-)

Heh! i didn't mind Gnus as a news reader, despite some initial
challenges in getting it set up for that; but i gave up trying to get it
working to my satisfaction as an MUA. It seems to be a very capable
package overall, but i decided to stick with Mutt after a number of
hours wrestling with Gnus (and its user manual, which i often found more
confusing than enlightening) to get even the basic functionality i
wanted. Later, and in contrast, i got basic functionality within about
an hour of installing mu4e; and i now have little incentive to try Gnus
again as an MUA. i'm more than willing to gradually and continually
tweak a system to better meet my needs, but i'm not likely to be able to
get to that point when even basic setup is such a struggle 

But to keep this discussion more on topic :-), this issue has informed
how i've designed org-vcard - i want it to be as easy to use out of the
box as possible, including making it as easy as possible for users to
take advantage of its underlying flexibility[1]. This has taken more
work than it might otherwise have done, but i think it's the right
approach to take.

[1] i (and many others, it seems) found GNOME 3 problematic in this
regard. i don't use it myself, but have in the past supported users who
did, and i wasn't thrilled to find GNOME 3 had taken away GUI access to
various configuration options important to my users, rather than
e.g. hiding them in an Advanced section with HERE BE DRAGONS!!!
warnings when people try to access those options.

 I am mainly talking about bbdb3 now, since I can't remember the
 variable names in bbdb2. For phone numbers one can use free form style
 by calling bbdb-insert-field with a prefix or change the variable
 bbdb-phone-style

 [snip]

 As for postal codes, either turn the checking off by setting
 bbdb-check-postcode to nil or change the variable
 bbdb-legal-postcodes

*nod*

Still, i'm surprised there's no default support for Australian-style
phone numbers and postcodes; are there really that few Australians
making use of BBDB?

In any case, if some people are happy with a BBDB-based solution to
managing their contacts, that's fine with me! It's just that i want an
Org-based system for myself. :-)


Alexis.



Re: [O] ANN: org-vcard. Export/import vCards. Backwards-compatible with org-contacts.el.

2014-08-06 Thread Alexis

Gour writes:

 what do you think about BBDB-v3? Many people like it, but I must admit
 I haven't take closer look at it?

i haven't tried using BBDB-v3, only BBDB-v2, several years ago. i found
the latter, hm, 'clunky'. (Similar to how, until the advent of mu4e, i
found no Emacs-based MUA with maildir support which i found as usable as
Mutt.) And iirc, part of the problem might have been lack of (full)
support for Australian phone numbers and/or postcodes, which at the time
i really didn't want to wrestle with.

In any case, org-contacts.el has nowwhetted my appetite for an Org-based
contacts solution - given the slogan Your life in plain text, contacts
management certainly seems to me to fall within Org's remit. :-)

 However, I wonder whether it's flexible enough to define one's own
 properties of format of one's contact data and what about syncing?

i've just had a quick scan through the BBDB-v3 source, and
superficially, it looks like the properties are hard-coded.

Regarding syncing, the BBDB page on SourceForge suggests that existing
support for this might be rather limited - i only see mention of
PalmPilot syncing. i would certainly be interested to know if anyone's
created CardDAV support for BBDB - given that Google Contacts can be
accessed via CardDAV, and given (what i imagine to be) the large number
of people syncing their Android phone's contacts with Google Contacts,
such support might well be in demand!

 I wish you all the best hoping org-mode users will find decent solution
 for handling contacts soon.

Thank you! i'm hoping my ongoing work with org-vcard might eventually
contribute towards this. :-)


Alexis.



[O] Creating a new contacts style for org-vcard [was: Re: ANN: org-vcard. ...]

2014-08-06 Thread Alexis
 (cdr (assoc (downcase fieldtype) 
fengshu-style-properties))
 (org-get-heading t t)
  (setq end-vcard t
(setq content (concat content
  (org-vcard-export-line END:VCARD  t)))
(setq output (concat output content)))
  (if search-result
  (re-search-backward \\s *:FIELDTYPE:\\s *name)))
(org-vcard-write-to-destination output destination)
(cond
 ((string= buffer destination)
  (message Exported contacts data to *org-vcard-export* buffer.))
 ((string= file destination)
  (message Exported contacts data to file.)))


(defun org-vcard-import-to-fengshu (source destination)
  ;; 
)
--- END ---

Save the file.

The org-vcard-import-to-fengshu function is deliberately left as a stub;
i'm just wanting to give an overall idea of how one can create new
org-vcard style functionality. Most of the definition of
org-vcard-export-from-fengshu comes from org-vcard-export-from-tree -
i've marked the section where the former differs substantially from the
latter.

(e) Run the command `org-vcard-reload-styles`.

(f) Open your example content in a buffer. You'll need to change the
FIELDTYPEs from cells-folder and emails-folder to cell and
email, respectively, due to how org-vcard's property-mapping system
works. You'll also need to make sure that there's a newline after the
fifth instance of address@hidden; my code doesn't address the case
where there's no such final newline.

(g) Either run the `org-vcard-export` command with source buffer,
destination buffer, style fengshu, language en and version 4.0;
or enable org-vcard-mode and select the corresponding items from the
Org-vCard menu.

(h) Hopefully, the result should now be available in the
*org-vcard-export* buffer. :-)

Please let me know if you have any questions!


Alexis.



Re: [O] ANN: org-vcard. Export/import vCards. Backwards-compatible with org-contacts.el.

2014-08-06 Thread Alexis

Eric Abrahamsen writes:

 If by properties you mean arbitrary key-value data, BBDB does indeed
 support that -- properties are known as fields, and xfields are
 user-designated fields. Labels and values can be arbitrarily
 designated by the user, and with a bit of coding you can format them
 in unusual ways, or have them do things.

Ah okay - thanks for clearing that up!


Alexis.



Re: [O] MobileOrg

2014-08-06 Thread Alexis

Subhan Michael Tindall writes:

 There’s also a few areas I’d like to do or see some enhancements on
 the mobile end, for example handling display or edit of properties is
 very limited

*nod* i started to address this issue with this pull request:

https://github.com/matburt/mobileorg-android/pull/434

which got merged  But given the lack of commits and releases more
generally, i've been reticent to work on this further.


Alexis.



Re: [O] ANN: org-vcard. Export/import vCards. Backwards-compatible with org-contacts.el.

2014-08-03 Thread Alexis

Feng Shu writes:

 Is it possible support this style?

In general, i'd prefer not to implement support /myself/ for everyone's
preferred style; the idea is that people implement their own style,
including:

* the relevant mapping structure which maps Org properties to vCard
properties/types;

* the org-vcard-export-from-[new style] function; and

* the org-vcard-import-to-[new style] function.

And, preferably, new tests to accompany the new style. :-)

Then, if a new style turns out to be particularly popular, i'll consider
including it in org-vcard by default.

Having said all that, as a proof-of-concept, i'll try to create a
mapping and export function for the style you describe, for vCard 4.0,
and will post the relevant code here. :-)


Alexis.



Re: [O] ANN: org-vcard. Export/import vCards. Backwards-compatible with org-contacts.el.

2014-08-01 Thread Alexis

Gour wrote:

 Excuse me for dumb question, but is this package meant to be something
 like org-contacts NG (something which we would really like to have)?

i don't think that's a dumb question at all!

My original motivation for developing org-vcard was to create part of a
pipeline for synchronising contacts between Org and my smartphone. My
intended plan for this pipeline is:

1. Create org-vcard to at least enable easy export and import of vCard
files between my smartphone and Org.

2. Using org-vcard as a library, create org-carddav (which i hope to
start working on shortly) in order to be able to synchronise contacts
stored in Org with arbitrary CardDAV servers.

3. Using org-carddav, set up a way of synchronising my Org-based
contacts with my smartphone-based contacts.

i'm taking this approach because:

(a) The MobileOrg release process currently appears to be stalled. A few
pull requests i've submitted to the project, one of which was intended
as the first step in making MobileOrg org-contacts.el-aware, have been
accepted, but there has been no new release including these
changes/fixes.

(b) (i) i feel org-contacts.el, as it stands, is too inflexible. It's only
aware of a relatively small number of properties, and extending this
requires directly messing around in the org-contacts.el code. i feel
that this has made it difficult to expand the list of properties
available, as it basically requires everyone to agree on exactly what
properties to add, and how.

(b) (ii) Further to (b) (i), i feel people should be able to define
their own properties, in their preferred language, without having to
write code for this.

(b) (iii) More generally, previous discussions on this topic on this
list have convinced me that it's folly to expect everyone to agree on a
single style for contacts in Org, and that what is needed is a system
that can easily accommodate the development of new contact styles,
potentially allowing an ecosystem/marketplace of styles to develop, with
particularly popular styles being considered for inclusion by default in
the system.

Given all the above, yes, i would like to see org-vcard become the basis
for an Org contacts NG system. (Which, to answer Feng Shu's question,
would mean that the 'tree' style would be available by default in such a
system.) Whether to actually take this approach, however, is something
i'll let the community decide. :-)


Alexis.



[O] ANN: org-vcard. Export/import vCards. Backwards-compatible with org-contacts.el.

2014-07-31 Thread Alexis

Hi all,

i'm pleased to announce the initial release of org-vcard, a package for
Org-based export and import of vCards:

* Backwards-compatible with org-contacts.el.

  org-vcard comes with a built-in contacts style called 'flat', which
  adheres to org-contacts' method of structuring contacts and contact
  information. It not only supports the properties specified in
  org-contacts.el, but many other properties as well.

* Basic support for vCard 4.0, 3.0 and 2.1.

  org-vcard is working towards full compliance with the vCard 4.0 (RFC
  6350), 3.0 (RFC 2426 and RFC 4770) and 2.1 specifications.

* New contacts style: 'tree'.

  org-vcard introduces a new style for Org contacts, called 'tree'.

* Highly customisable, via Emacs' customize interface.

  Modify existing contact styles; change the labels used to map contact
  details in org-mode to various vCard properties/types, or add new
  ones. Create completely new contact styles by plugging in your own
  code to handle export and import.

Further details are available on the GitHub page:

https://github.com/flexibeast/org-vcard

The package can be installed from MELPA, or downloaded and installed
manually.

Hope people find it useful!


Alexis.



[O] org-contacts to vCard

2014-07-05 Thread Alexis

Hi all,

i recently posted about connecting `org-contacts` with MobileOrg:

(1) https://lists.gnu.org/archive/html/emacs-orgmode/2014-05/msg00971.html

Part of the resulting thread was about fleshing out an 'official' spec
for the org-contacts format. Another part was about using DAV as part of
a putative syncing process.

This has led me to start working on an `org-carddav` package[*], à la
the existing `org-caldav` package (regardless of whether a DAV-based
approach ends up being used for syncing between `org-contacts` and
MobileOrg); this in turn has led me to start working on an `org-vcard`
package. Which brings me back to the issue of the org-contacts format.

The most recent post on this topic that i can find is:

(2) https://lists.gnu.org/archive/html/emacs-orgmode/2014-06/msg00140.html

Taking into consideration people's comments on this matter in thread
(1), i'd like to propose the following:

--

Define a property `ORG_CONTACTS_STYLE`. The default value is 'flat', and
represents the way org-contacts currently works:

,* People
,** Alexis
,:PROPERTIES:
,:CELL:  999 999
,:PHONE: 00   
,:EMAIL: addres...@example.com
,:END:

A second value `ORG_CONTACTS_STYLE` can have is 'tree':

,* People
,** Alexis
,:PROPERTIES:
,:KIND: individual 
,:FIELDTYPE: name
,:END:
,*** Landline
,:PROPERTIES:
,:FIELDTYPE: landline
,:END:
, 00  
,*** Mobile
,:PROPERTIES:
,:FIELDTYPE: mobile
,:END:
,  999 999
,*** Email 1
,:PROPERTIES:
,:PREFERRED: 
,:FIELDTYPE: email
,:END:
, addres...@example.com
,*** Email 2
,:PROPERTIES:
,:FIELDTYPE: email
,:END:
, addres...@example.com

For each value of ORG_CONTACTS_STYLE, define a default mapping:

,#+ORG_CONTACTS_STYLE: flat

,+--+-+
,| PROPERTIES field | vCard field |
,+--+-+
,| [ heading with 'KIND' property ] | FN  |
,| CELL | TEL;CELL|
,| MOBILE   | TEL;CELL|
,| LANDLINE | TEL;HOME|
,| PHONE| TEL |
,| EMAIL| EMAIL   |
,| ...  | ... |

,#+ORG_CONTACTS_STYLE: tree

,+--+-+
,| FIELDTYPE field  | vCard field |
,+--+-+
,| name | FN  |
,| cell | TEL;CELL|
,| mobile   | TEL;CELL|
,| landline | TEL;HOME|
,| phone| TEL |
,| email| EMAIL   |
,| ...  | ... |


Users would be able to change and extend the above mappings as works
best for them. And of course, there's potential to define entirely new
mappings.

`org-vcard` would then be able to export org-contacts to vCard depending
on the value of ORG_CONTACTS_STYLE. And since the default value of this
property would be 'flat', backwards-compatibility would be preserved.

--

The reason i'm proposing a new contacts style of 'tree' is to create one
that more takes advantage of Org being an outliner. Rather than having a
huge flat property list with no ability to show/hide relevant groups, it
seems to me to make sense to be able to do something like:

,* People
,** Alexis
,*** Work
, Phone
, Email
,*** Home
, Phone
, Email
,* Organisations
,** Acme
,*** Coyote Services
, Anvils
,* Phone
,* Email
, Rope
,*** Snake Oils

and be able to show/hide the relevant sections as required.

Further, i believe the mappings approach avoids having to make fixed and
final decisions which everyone has to agree on - different people will
have different needs in different use-cases. And inserting new contact
data could be facilitated via an interactive function which does a
`completing-read` using the values of the first column in the
currently-active mapping table, then creating the relevant PROPERTIES
drawer with the appropriate value(s).

Finally: yes, i am personally willing to do all the coding to support
this proposal. :-)

What do people think?


Alexis.

--

[*] Though i'm already encountering the same possible issue with the
`url-dav` package that i've found when trying to use `org-caldav` with
a DAViCal server:

https://github.com/dengste/org-caldav/issues/45

Is anyone successfully using `org-caldav` (or `url-dav`) with a DAViCal
server?



Re: [O] babel and computing a number of months

2014-06-02 Thread Alexis

Alan Schmitt writes:

 I need to work with dates for some code/scripts I'm writing in a
 document making heavy usage of source blocks and babel evaluation. The
 good news is that I have access to many programming languages, so the
 bad news is I don't know which one to choose. The problem I want to
 solve is the following: I want to compute the number of months between
 march 1st, 2014, and the beginning of the current month (so right now
 it's 3, but on may 31st it was 2). Is there an easy way to do it in a
 babel supported language?

How about Perl with either DateTime::Moonpig:

https://metacpan.org/pod/DateTime::Moonpig

or Time::Piece:

https://metacpan.org/pod/Time::Piece


Alexis.



Re: [O] Hiding checked items in a checkbox list (FR?)

2014-06-01 Thread Alexis

Sergio Pokrovskij writes:

 Yes (actually I have about 80 items in my Springpad list; if the full
 list consists of mere 10 items, it fits the screen of my smartphone,
 and there is no real need to hide the checked items).

*nod* Fair enough; i was just making up some numbers in order to be
concrete.

   Al 2. Initially mark all of them as 'unneeded' by checking all
   Al items with a single command.

 Possibly so.  Actually all the performed items get checked as I buy
 them, and those which for some reason I could not buy should preserve
 their open (unchecked, unimplemented) status.  So normally I need not
 perform this step (provided that what I have checked at the store gets
 synchronized back to the Big Brother); but in some infrequent cases it
 is nice to have such a command (though I guess it can be easily done
 with an M-% in emacs).

*nod*

   Al 3. Uncheck (say) 3 items that are 'needed'.
   Al 4. Run org-mobile-push.

 Do I have to do it manually every time (more than once)?  Aren't such
 files synchronized automatically?

It depends on how you've set up MobileOrg. i sync via ssh, and thus sync
manually via org-mobile-push and org-mobile-pull from within Emacs. If
you use e.g. Dropbox, things might be different - i can't comment on
that.

   Al 5. MobileOrg displays only the 3 'needed' items; the other 7
   Al checked (i.e. unneeded) items are hidden.

 Yes (in fact, normally there are more of them, a dozen at least).

See above - at this stage, for the purposes of understanding the core of
the flow you're wanting, the exact numbers don't matter so much. But in
terms of potential UI access to such flows, i'll certainly keep in mind
that lists might include many dozens of items.

   Al 6. As items are purchased and checked off the list, they can
   Aleither:
   Al(a) remain visible; or
   Al(b) become hidden upon being checked.

   Al Is that correct?

 Yes.  Normally (b) is sufficient for me, because normally I have just
 a screenful of interesting items.  But I can imagine that people who
 need longer shopping lists (more that one screen of them) would prefer
 to have a MobileOrg command to hide the newly-checked items as well.

*nod* Fair enough. i'll check the MobileOrg issues page:

https://github.com/matburt/mobileorg-android/issues

for any equivalent and/or similar proposals, and submit a new issue if
there aren't any. Either way, i'll see what i can do about adding this
functionality. :-) Note, though, that i'm not the lead developer, so i
don't have any control over if/when any such functionality will be (a)
included in the master branch and (b) included in a release.


Alexis.



Re: [O] Hiding checked items in a checkbox list (FR?)

2014-05-31 Thread Alexis

Sergio Pokrovskij writes:

 The next problem is that MobileOrg should respect the preliminary
 hiding done before the visit to the shop.  I do not request that it
 hide the checked (= bought) items as well (actually I'd prefer it to
 leave them checked and visible; but also its hiding-on-the-fly is
 quite acceptable).

Sorry, but i'm not sure i follow. Do you mean something like the
following example workflow?

1. Have list of (say) 10 grocery items in org-mode.
2. Initially mark all of them as 'unneeded' by checking all items with a
   single command.
3. Uncheck (say) 3 items that are 'needed'.
4. Run org-mobile-push.
5. MobileOrg displays only the 3 'needed' items; the other 7 checked
   (i.e. unneeded) items are hidden.
6. As items are purchased and checked off the list, they can either:
   (a) remain visible; or
   (b) become hidden upon being checked.

Is that correct?


Alexis.



Re: [O] org-contacts development

2014-05-25 Thread Alexis

Daimrod daim...@gmail.com writes:

 Sure, but then how would we define the map? with a fixed set of rules?
 with a user customizable map (e.g. '((MOBILE - TELL;CELL)
 (EMAIL_WORK - EMAIL;TYPE=work)))?

More the latter. My initial (again, brainstorming!) thinking has an
user-modifiable data structure along the lines of the following:

+---+++
| elisp var |org property| ~ vCard equivalent |
+---+++
| org-contacts-email-work-property  |EMAIL_WORK  |EMAIL;TYPE=work |
| org-contacts-mobile-work-property |MOBILE_WORK |TEL;CELL;TYPE=work  |
+---+++

English-language properties for the org property column could be
provided by default, but users who wanted to use non-English property
names could do so by modifying the relevant entries.

 I'm not saying that it's impossible nor that we shouldn't do it but that
 we need to think about it first. :)

Oh 100% agreed! That's why i've started this discussion here, so as to
get input, feedback, and hopefully at least rough consensus, on how to
approach all this. :-)

 Well, if people are willing to help, I'll see if I can come up with a
 proposal.

:-) Sounds good!


Alexis.



Re: [O] org-contacts development

2014-05-24 Thread Alexis

Rasmus ras...@gmx.us writes:

 You could also put yourself a slightly different challenge: map
 between the org-contacts-format and an carddav server that you
 establish as a backend on your phone.  E.g. google contacts is a
 carddav server, I think.

 I use DAVdroid for syncing my calendar and my contacts with my
 OwnCloud and it very, very well!  I sync my agenda with my OwnCloud
 using org-caldav, which works great except when you have multiple time
 stamps per heading.  For contacts I tried Asynk to sync my bbdb
 contacts with OwnCloud, but I was using too many fields unknown to it,
 and my contacts got messed up.

 Mayhaps you already thought of this approach.  I think it is more
 future-proof.

i wasn't aware of CardDAV - thanks for pointing it out! My thoughts of
how to deal with org-contacts data have indeed been revolving around
vCard though, so this approach looks interesting indeed. :-)


Alexis.



Re: [O] org-contacts development

2014-05-24 Thread Alexis

Bastien b...@gnu.org writes:

 Improving org-contacts.el and making it part of Org core are two
 separate issues -- not everything that is part of Org core gets a
 lot of attention, and some contributed packages do.

Oh okay, fair enough!

 I'd say: go ahead with whatever conventions you want to enforce
 and the community will step up -- or just applaud :)

Heh, okay. :-)

 PS: Actually, being part of Org core sometimes slow down things,
 because substantial contributions then need to come from people
 who assign their copyright to the Free Software Foundation.

*nod* Good point  i'd be happy to assign copyright to the FSF,
myself, but i know that doing so is an issue for many people.


Alexis.



Re: [O] org-contacts development

2014-05-24 Thread Alexis

Alexander Baier alexander.ba...@mailbox.org writes:

 About the spec, might it make sense to use a standard already in
 existence as a starting point or even a major inspiration?  Something
 like vCard for example?

Yeah, i like the idea of building on top of vCard, given how widely it's
used and supported, and that it doesn't seem to be a horrible format
 Those who feel otherwise are going to need to speak up soon or
forever hold their peace. :-)


Alexis.



Re: [O] org-contacts development

2014-05-24 Thread Alexis

Xebar Saram zelt...@gmail.com writes:

 This sounds awesome, i would love syncing org contacts with my android
 phone

:-) It's good to know i'm not the only one wanting this functionality -
gives me more incentive to actually implement it!


Alexis.



Re: [O] org-contact Documentation

2014-05-24 Thread Alexis

On Fri, 23 May 2014 15:17:04 +0200, Esben Stien wrote:

 Sync'ing with ldap would also be great, but I guess that's a bit
 premature.

Well, if org-contacts becomes based in vCard, as i'd like, i guess it
might be possible to use LDAP vCard services 


Alexis.



Re: [O] Extending org-contacts with properties: naming properties

2014-05-24 Thread Alexis

Hi Esben,

The way that org-contacts currently works is that contact details are
grouped together in the same PROPERTIES drawer, e.g.

* Alexis
:PROPERTIES:
:EMAIL:  ale...@example.com
:PHONE:  -
:END:

and that's what i've assumed in my MobileOrg code for parsing
org-contacts data. i imagine that moving away from such a data layout
would require substantial changes to the org-contacts.el code, which i
have no intention of doing - particularly given the work that would be
needed to ensure backwards-compatibility with people's existing data.

For reference, at the moment, in my own org-contacts file, i've set up
grouping via a headings hierarchy like:

* Lists
** Org-mode
:PROPERTIES:
:EMAIL:  emacs-orgmode@gnu.org
:END:
* People
** Mum and Dad
:PROPERTIES:
:PHONE:  -
:END:
* Organisations
** Free Software Foundation
:PROPERTIES:
:EMAIL:  i...@fsf.org
:END:
*** A. Person
:PROPERTIES:
:EMAIL:  a.per...@fsf.org
:END:


Alexis.



Re: [O] org-contacts development

2014-05-24 Thread Alexis

Bastien b...@gnu.org writes:

 Alexis flexibe...@gmail.com writes:

 *nod* Good point  i'd be happy to assign copyright to the FSF,
 myself, but i know that doing so is an issue for many people.

 Here you go!

 http://orgmode.org/cgit.cgi/org-mode.git/plain/request-assign-future.txt

 :)

Heh, thank you! i'll definitely make sure to complete that form and send
it in. :-)


Alexis.



Re: [O] org-contacts development

2014-05-24 Thread Alexis

Alexander Baier alexander.ba...@mailbox.org writes:

 For what it's worth, I would love to help out in any way I can.  I use
 org-contacts myself and know elisp well enough to also do some
 implementing.  But I wouldn't mind writing some documentation, either.

Thank you! At this point i'm studying RFC 6350 (vCard 4) and the
org-contacts.el source, trying to work out how to bring more of the
former to the latter with a minimal amount of work and fuss. :-)

 I, however, don't know much about Android programming or the syncing
 aspect of this task.

*nod* i'm not sure how much work is involved in the syncing aspect, and
so my first priority is to at least be able to export individual
contacts from MobileOrg to an individual vCard file that can be imported
into the Android Contacts system, and vice versa. There might already be
a FOSS CardDAV library for Android out there that could be used for
syncing, but i've not researched that yet.


Alexis.



Re: [O] org-contacts development

2014-05-24 Thread Alexis

Rasmus ras...@gmx.us writes:

 Check here:

 https://f-droid.org/repository/browse/?fdfilter=carddavfdpage=1page_id=0

 DAVdroid works very well.

Oh okay, i didn't realise DAVdroid was FOSS.

DAVdroid makes use of the ez-vcard library, which is BSD-licensed, and
certainly seems like it might be useful. However, i'm not sure to what
extent the sync-specific functionality of DAVdroid can be separated from
the rest of the app; superficially, it seems like it's fairly tightly
integrated. But if nothing else, it does show how it can be done! Thank
you. :-)


Alexis.



Re: [O] org-contacts development

2014-05-24 Thread Alexis

Daimrod daim...@gmail.com writes:

 So, as you said, we would need to define and document a specification
 for org-contacts. And we need to be clear from the beginning about
 what it can do and what it can not do. For example, it is unlikely
 that org-contacts will be a 1:1 mapping with the vCard format in its
 current form because vCard properties can have values.

 e.g.
 PROP1;PROP2=VAL:FOO BAR
   ^

Agreed. But it seem to me we could at least map e.g. TEL;CELL: (which
is how my Android phone vCard-exports a mobile number) to
e.g. :MOBILE: or :PHONE_CELL:. And we could map things like
e.g. EMAIL;TYPE=work: to e.g. :EMAIL_WORK:. This is just
brainstorming, however. :-)

 An approach would be to keep using properties whenever it's possible
 to keep the format simple when possible and use subtree instead of
 properties when necessary.

 e.g.
 #+BEGIN_SRC org
   ,* Contact Name
   ,** PHONE
  :PROPERTIES:
  :TYPE: WORK
  :END:
   - num1
   - num2
 #+END_SRC

 But then we would need a special property to indicate that a subtree
 is a contact and ignore subtree without this property (just like it
 does with the EMAIL property ATM).

 #+BEGIN_SRC org
   ,* Contact Name
 :PROPERTIES:
 :TYPE: org-contacts
 :END:

   ,** PHONE
  :PROPERTIES:
  :TYPE: WORK
  :END:
   - num1
   - num2
 #+END_SRC

 And of course it would be nice if we could keep as much compatibility
 with the current format :)

Well, to me this looks broadly similar to what Esben has proposed:

https://lists.gnu.org/archive/html/emacs-orgmode/2014-05/msg01055.html

Although i like the idea of such a format in principle, my concern (as i
noted in my reply to Esben) is that this might require a substantial
modification of org-contacts.el, both to accommodate this format and to
ensure backwards-compatibility. If this is indeed the case, and someone
else is willing to commit to being the lead on undertaking that work,
then i'll certainly support that work and write relevant MobileOrg code
to work with both the new and old formats. Otherwise, from my
perspective of wanting to simply add some more fields and be able to
transfer contact details between MobileOrg and my phone's Contacts
system, it's more than i'm willing to take on at this point.


Alexis.



[O] org-contacts development

2014-05-22 Thread Alexis

Hi all,

i use org-contacts as my primary system for contact
management. Consequently, i'd love to be able to make use of my
org-contacts data on my Android phone.

To that end, i recently wrote some code for MobileOrg-Android which adds
basic PROPERTIES drawer support:

https://github.com/matburt/mobileorg-android/pull/434

What i'd now like to do is to add support for transferring data back and
forth between my org-contacts file and the Contacts store on my
phone. The challenge is the mapping between these two systems.

For example, org-contacts provides only one EMAIL property, which can
contain multiple addresses separated by spaces, whereas Android's
ContactsContract.CommonDataKinds.Email class is able to distinguish
between different addresses for different purposes:

http://developer.android.com/reference/android/provider/ContactsContract.CommonDataKinds.Email.html

What would be useful would be an 'official', fleshed-out spec for
org-contacts data, which handles a greater range of contact-related
info. At the moment, for example, my org-contacts file makes use of the
properties:

#+PROPERTY: LANDLINE
#+PROPERTY: MOBILE
#+PROPERTY: POST
#+PROPERTY: RESIDENCE

A search of this list's archives for references to 'org-contacts':

https://lists.gnu.org/archive/cgi-bin/namazu.cgi?query=org-contactssubmit=Search!idxname=emacs-orgmodemax=20result=normalsort=date%3Alate

suggests that org-contacts is something people are using heavily enough
that they're writing code, ad-hoc, to provide functionality they
require, e.g.

http://lists.gnu.org/archive/html/emacs-orgmode/2013-11/msg00869.html

i'm wondering if it might now be appropriate for org-contacts to become
part of org-mode proper, rather than simply a contrib, to reduce
unnecessary duplication of efforts. i suspect that, at the moment, a
number of people interested in making use of org-contacts might be
reluctant to do so (or do so more heavily) because it's not an
'official' part of org-mode. Yet contacts management seems to me to be
functionality well within org-mode's remit.

Fleshing out an extended spec for org-contacts data could be part of the
process of making org-contacts a first-class citizen of org-mode, and
provide a more solid foundation on which people can build (and share)
the org-contacts functionality they're after. And in my own particular
case, it would greatly facilitate my work in org-contacts / Android
Contacts integration. :-)

What do people think?


Alexis.



[O] Possible bug? Links using 'file+Emacs:' don't work. (Was: Re: Unable to get file link search to work)

2014-02-23 Thread Alexis

Christian Moe m...@christianmoe.com writes:

 Just to say, I reproduce it, so I think it's a bug. Note, the below
 only happens for me when using file+emacs: to force opening in
 emacs. The links work with regular file: links, or just starting the
 link with ./ Since my .org files open with emacs anyway, I normally
 wouldn't add +emacs. But the manual suggests I could.

Hi Christian,

Thanks for taking the time to look into this!

i was using 'file+emacs:' because when i just used 'file:', the .org
file would be opened using 'less', not Emacs. As a result of your email,
i checked out the documentation for org-open-at-point; it said that
providing a prefix argument would force opening in Emacs. So i removed
the '+emacs' from my example link, and then tried opening it via
C-u C-c C-o - and it worked!

i checked the value of org-file-apps, to see why org-open-at-point was
using 'less', rather than Emacs, to open .org files. It turns out there
was a typo in how i was setting the value of org-file-apps, causing it
to have a nonsense value - and i guess this caused org-mode to fall back
to using 'less' as a default. Setting the value of org-file-apps to a
correct value - including (org . emacs) as part of it - resulted in
'file:' links being opened in Emacs, rather than with 'less'.

However, 'file+emacs:' links still exhibit the incorrect behaviour i
described in my original email, and which you experienced also. So it
looks like there is indeed a bug here - or, the 'file+emacs:' example
should be removed from section 4.3 of the Org manual.

Thanks again!


Alexis.



Re: [O] Possible bug? Links using 'file+Emacs:' don't work.

2014-02-23 Thread Alexis

Nicolas Goaziou n.goaz...@gmail.com writes:

 Alexis flexibe...@gmail.com writes:

 However, 'file+emacs:' links still exhibit the incorrect behaviour i
 described in my original email, and which you experienced also. So it
 looks like there is indeed a bug here - or, the 'file+emacs:' example
 should be removed from section 4.3 of the Org manual.

 This should be fixed in master branch.

 Thank you for reporting it.

Hi Nicolas,

Have just tested your fix - works for me. :-) Thanks for fixing this so
quickly!


Alexis.



Re: [O] Unable to get file link search to work

2014-02-22 Thread Alexis

Bump. Any suggestions as to whether the below is me doing something
wrong (more likely) or a bug (less likely)?

Thanks!


Alexis flexibe...@gmail.com writes:

 Hi all,

 i'm running org-mode 8.2.5h in Emacs 24.3.

 Say i have file a.org:

   * Main
   ** Group
   *** Firstname Lastname

 My reading of

 http://orgmode.org/manual/External-links.html#External-links

 and

 http://orgmode.org/manual/Search-options.html#Search-options

 is that i should be able to, in b.org - which resides in the same
 directory as a.org - have something like:

   [[file+emacs:a.org::*Firstname]]

 such that when i move point over that link, and enter C-c C-o
 (i.e. org-open-at-point), i will be taken to a buffer containing a.org,
 with point on the 'Firstname Lastname' heading. Instead, what happens is
 that a new buffer is created, attached to a new file called, literally,
 'a.org::*Firstname'.

 Similarly, if in b.org i have something like:

   [[file+emacs:a.org::2]]

 i am not taken to line 2 of a.org, but instead to a new buffer attached
 to a new file literally called 'a.org::2'.

 i have observed this behaviour with
 org-link-search-must-match-exact-headline set to either 'query-to-create
 or to nil.

 What am i doing wrong?


 Alexis.




[O] Unable to get file link search to work

2014-02-21 Thread Alexis

Hi all,

i'm running org-mode 8.2.5h in Emacs 24.3.

Say i have file a.org:

  * Main
  ** Group
  *** Firstname Lastname

My reading of

http://orgmode.org/manual/External-links.html#External-links

and

http://orgmode.org/manual/Search-options.html#Search-options

is that i should be able to, in b.org - which resides in the same
directory as a.org - have something like:

  [[file+emacs:a.org::*Firstname]]

such that when i move point over that link, and enter C-c C-o
(i.e. org-open-at-point), i will be taken to a buffer containing a.org,
with point on the 'Firstname Lastname' heading. Instead, what happens is
that a new buffer is created, attached to a new file called, literally,
'a.org::*Firstname'.

Similarly, if in b.org i have something like:

  [[file+emacs:a.org::2]]

i am not taken to line 2 of a.org, but instead to a new buffer attached
to a new file literally called 'a.org::2'.

i have observed this behaviour with
org-link-search-must-match-exact-headline set to either 'query-to-create
or to nil.

What am i doing wrong?


Alexis.



[O] Bad formatting in documentation

2013-05-24 Thread Alexis Praga
Hi,

I just stumbled upon a bad formatting on the website (
http://orgmode.org/worg/org-faq.html ).
In Why doesn't C-c a call the agenda? Why don't some org keybindings
work?, we can see

#+BEGIN_SRC emacs-lisp ;; The following lines are always needed. Choose
your own keys. (add-to-list 'auto-mode-alist '(\\.org\\' . org-mode))
(global-set-key \C-cl 'org-store-link) (global-set-key \C-ca
'org-agenda) (global-set-key \C-cb 'org-iswitchb) #+END_SRC emacs-lisp



-- 

   Alexis Praga
___

Ph.D Student
CERFACS
alexis.pr...@gmail.com


[O] the right way to build OMPL export and import

2013-04-24 Thread Alexis Gallagher
Hi,

I would love to be able to export org documents to opal, so that I can read 
them with the various commercial outlining apps on platforms without emacs -- 
e.g, iOS. The ideal thing would be if I could import OPML as well.

Is anyone working on this already?

If not, does anyone have any pointers on the right way to go about this, so 
that the work would go smoothly and be acceptable upstream? This seems like a 
good time to ask given the recent consolidation of export facilities around the 
internal parser org-element.el.

I presume any OPML exporter should be based on that, correct? Is any one of the 
existing exporters a particularly clean example to work from?

For OPML import, is org-element-interpret-data the best starting point?

Alexis