Re: [O] Syntax of Org Babel ":results" header argument

2013-03-18 Thread Sebastien Vauban
Hello all,

Andreas Leha wrote:
> t...@tsdye.com (Thomas S. Dye) writes:
>> Eric Schulte  writes:
>>> Bastien  writes:
 "Sebastien Vauban" writes:

> As there was no reaction to this, I'd like to bump it up. At least, to 
> either
> have a discussion on this, or a clearly stated "no go".

 I tend to agree with Jay here and I prefer the minimalist syntax,
 although values for the :results parameter are heterogeneous.
>>>
>>> +1 for maintaining the existing concise syntax.
>>
>> +1 for stable syntax.
>
> +1 one more vote

I've got the authoritative answers I was expecting. It was just a RFI. Case is
closed!

Best regards,
  Seb

-- 
Sebastien Vauban




Re: [O] Syntax of Org Babel ":results" header argument

2013-03-18 Thread Andreas Leha
t...@tsdye.com (Thomas S. Dye) writes:

> Eric Schulte  writes:
>
>> Bastien  writes:
>>
>>> Hi Sébastien,
>>>
>>> "Sebastien Vauban"
>>>  writes:
>>>
 As there was no reaction to this, I'd like to bump it up. At least, to 
 either
 have a discussion on this, or a clearly stated "no go".
>>>
>>> I tend to agree with Jay here and I prefer the minimalist syntax,
>>> although values for the :results parameter are heterogeneous.
>>>
>>
>> +1 for maintaining the existing concise syntax.
>
> +1 for stable syntax.
>

+1 one more vote

Andreas




Re: [O] Syntax of Org Babel ":results" header argument

2013-03-17 Thread Thomas S. Dye


Eric Schulte  writes:

> Bastien  writes:
>
>> Hi Sébastien,
>>
>> "Sebastien Vauban"
>>  writes:
>>
>>> As there was no reaction to this, I'd like to bump it up. At least, to 
>>> either
>>> have a discussion on this, or a clearly stated "no go".
>>
>> I tend to agree with Jay here and I prefer the minimalist syntax,
>> although values for the :results parameter are heterogeneous.
>>
>
> +1 for maintaining the existing concise syntax.

+1 for stable syntax.

Tom

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




Re: [O] Syntax of Org Babel ":results" header argument

2013-03-17 Thread Eric Schulte


Bastien  writes:

> Hi Sébastien,
>
> "Sebastien Vauban"
>  writes:
>
>> As there was no reaction to this, I'd like to bump it up. At least, to either
>> have a discussion on this, or a clearly stated "no go".
>
> I tend to agree with Jay here and I prefer the minimalist syntax,
> although values for the :results parameter are heterogeneous.
>

+1 for maintaining the existing concise syntax.

-- 
Eric Schulte
http://cs.unm.edu/~eschulte




Re: [O] Syntax of Org Babel ":results" header argument

2013-03-17 Thread Bastien


Hi Sébastien,

"Sebastien Vauban"
 writes:

> As there was no reaction to this, I'd like to bump it up. At least, to either
> have a discussion on this, or a clearly stated "no go".

I tend to agree with Jay here and I prefer the minimalist syntax,
although values for the :results parameter are heterogeneous.

Also, I think the change you proposal can be backward-compatible,
and in that case, making it after 8.0 is fine.

In any case, I'll let Eric and Nicolas advice have priority over
mine here.

Best,

-- 
 Bastien




Re: [O] Syntax of Org Babel ":results" header argument

2013-03-15 Thread Jay Kerns
Greetings,

On Fri, Mar 15, 2013 at 1:30 PM, shripad sinari
 wrote:

[snip]

I am happy with the current behavior, but
that isn't to say it couldn't possibly be improved or that I
disagree with Sebastien's suggestions.  It did take me quite some
time to figure out what the heck was going on with it all, but
now that I know, I like it a lot.

Few comments:

- :results graphics makes the list even longer, yes?  :-) I'm not
  sure that every language supports it and I don't believe it's
  currently in the manual. And I don't think it directly affects
  what Sebastien is asking about.

- Out of the myriad combinations, I only use a few regularly.

- I hope that a change, if any, would be mindful of
  brevity. Particularly with inline src blocks, it is much
  shorter to write

  : SRC_foo[:results value scalar raw append]{blah}

  than it is to write

  : SRC_foo[:results_collect value :results_type scalar
:results_handling append :results_wrapper raw]{blah}

  (or however it would be decided to handle that).

-- 
Jay



Re: [O] Syntax of Org Babel ":results" header argument

2013-03-15 Thread shripad sinari
I think Sebastian's suggestions are very nice and would be very helpful.

Shripad
Tucson, AZ


On Fri, Mar 15, 2013 at 4:41 AM, Eric S Fraga  wrote:

> Sebastien Vauban  writes:
>
> > Hello,
> >
> > As there was no reaction to this, I'd like to bump it up. At least, to
> either
> > have a discussion on this, or a clearly stated "no go".
>
> Okay, if nobody else answers, here I go!  I am happy with the current
> approach as it's easier for me to see, at a glance, what is intended to
> happen with the results and a single word key is easier to type.
>
> However, I would not be distraught if a change were made along the lines
> you propose.
>
> --
> : Eric S Fraga, GnuPG: 0xC89193D8FFFCF67D
> : in Emacs 24.3.50.1 and Org release_8.0-pre-72-gc66641
>
>
>


Re: [O] Syntax of Org Babel ":results" header argument

2013-03-15 Thread Eric S Fraga
Sebastien Vauban  writes:

> Hello,
>
> As there was no reaction to this, I'd like to bump it up. At least, to either
> have a discussion on this, or a clearly stated "no go".

Okay, if nobody else answers, here I go!  I am happy with the current
approach as it's easier for me to see, at a glance, what is intended to
happen with the results and a single word key is easier to type.

However, I would not be distraught if a change were made along the lines
you propose.

-- 
: Eric S Fraga, GnuPG: 0xC89193D8FFFCF67D
: in Emacs 24.3.50.1 and Org release_8.0-pre-72-gc66641




Re: [O] Syntax of Org Babel ":results" header argument

2013-03-15 Thread Sebastien Vauban
Hello,

As there was no reaction to this, I'd like to bump it up. At least, to either
have a discussion on this, or a clearly stated "no go".

Best regards,
Seb

"Sebastien Vauban" wrote:
> Before Org 8 is out, I'm willing to put light on some last syntax which I find
> counter-intuitive and not along the lines of the rest: it concerns the
> `results' parameter.
>
> Let's sum up first the list of all parameters:
>
> 1. Collection
>
>- :results value
>- :results output
>
> 2. Type of results (when :results is set to `value'):
>
>- Result types
>
>  + :results vector
>  + :results scalar
>  + :results list
>  + :results file
>
>- Result wrappers
>
>  + :results raw
>  + :results drawer
>  + :results org (removed, right?)
>  + :results html
>  + :results code
>  + :results latex
>  + :results pp
>
> 3. Handling
>
>- :results replace
>- :results silent
>- :results none
>- :results append
>- :results prepend
>
> As you see (by the shown structure), the different values answer different
> questions:
>
> - How the results should be collected from the source code block?
> - How they will be inserted into the Org mode buffer?
> - How to interpret/wrap the results?
> - How the results should be handled?
>
> And answering many of these questions at the same time means giving a
> *multi-value* to the parameter, such as:
>
>   :results list append
>
> Wouldn't it make more sense (and be more easily parsed by the machine and be
> cleaner and less error-prone for us, poor humans) if `results' would be split
> in different parameters for the different questions they answer, each of those
> parameters getting at most one value?
>
> Something along the lines of:
>
>   :results_type file :results_insertion append
>
>   (those names may be ugly, it just for the purpose of explaining my idea).
>
> I know that it's the ultimate moment to discuss such a change, would there be
> consensus, before Org 8 is out.

-- 
Sebastien Vauban




[O] Syntax of Org Babel results

2013-03-10 Thread Sebastien Vauban
Hello,

Before Org 8 is out, I'm willing to put light on some last syntax which I find
counter-intuitive and not along the lines of the rest: it concerns the
`results' parameter.

Let's sum up first the list of all parameters:

1. Collection

   - :results value
   - :results output

2. Type of results (when :results is set to `value'):

   - Result types

 + :results vector
 + :results scalar
 + :results list
 + :results file

   - Result wrappers

 + :results raw
 + :results drawer
 + :results org (removed, right?)
 + :results html
 + :results code
 + :results latex
 + :results pp

3. Handling

   - :results replace
   - :results silent
   - :results none
   - :results append
   - :results prepend

As you see (by the shown structure), the different values answer different
questions:

- How the results should be collected from the source code block?
- How they will be inserted into the Org mode buffer?
- How to interpret/wrap the results?
- How the results should be handled?

And answering many of these questions at the same time means giving a
*multi-value* to the parameter, such as:

  :results list append

Wouldn't it make more sense (and be more easily parsed by the machine and be
cleaner and less error-prone for us, poor humans) if `results' would be split
in different parameters for the different questions they answer, each of those
parameters getting at most one value?

Something along the lines of:

  :results_type file :results_insertion append

  (those names may be ugly, it just for the purpose of explaining my idea).

I know that it's the ultimate moment to discuss such a change, would there be
consensus, before Org 8 is out.

Best regards,
  Seb

-- 
Sebastien Vauban