[Orgmode] Re: [Babel] Difficult to follow code execution in HTML exported file

[Sébastien Vauban]

Hi Eric,

"Eric Schulte" wrote:
> Sébastien Vauban <address@hidden> writes:
>> #+TITLE:     Tables don't have their name exported
>>
>> Chunks of code are exported to HTML with their parameters, such as table
>> names. But *tables aren't exported with their name*.
>>
>> * Playing with data and code
>>
>> Here is one table:
>>
>> #+tblname: numbers-1
>> |        1 |
>> |        2 |
>> |       45 |
>> |     test |
>> | 3.141592 |
>>
>> Another one is here:
>>
>> #+tblname: numbers-2
>> |       21 |
>> |       22 |
>> |      245 |
>> |    test2 |
>> | 23.14159 |
>>
>> When applying the following chunk of code to some data (find who is
>> =numbers-1=!):
>>
>> #+srcname: add-type
>> #+begin_src emacs-lisp :var data=numbers-1 :exports both
>> (mapcar
>>  (lambda (line)
>>    (let ((number (car line)))
>>      (list number (type-of number))))
>>  data)
>> #+end_src
>>
>> I get the following results:
>>
>> #+results: add-type
>> #+BEGIN_RESULT
>> |        1 | integer |
>> |        2 | integer |
>> |       45 | integer |
>> |     test | string  |
>> | 3.141592 | float   |
>> #+END_RESULT
>
> The title of your email mentions code execution, but the body seems to
> focus on export of table names.  I'll reply to the latter and my
> apologies if I've missed something related to the former.

You're right that there is *not necessarily* execution per se, though the
function is well *executed* and outputs results in the example I gave.
And, imagine you read that page on Worg, you can't completely follow the
execution chain: the code refers to some data that is invisible in HTML. You
currently can't output that information...

My titles aren't orthogonal: if I'm clear, it's because table names are not
exported that it's difficult to understand how documented code has produced
the displayed results. If not yet done, put your mind in "literate
programming" documentation style, and "reproducible research". It really is
about exporting both data, code and results...


> Table names have existed in Org-mode since before the existence of
> active code blocks, and I don't think they have ever been exported, so
> the export of table names would be a Org-mode wide feature request.

OK.


> I imagine that such a change would meet with some resistance, at least I
> know I would not want all of my table names exported by default.

You name it... "by default", meaning such a behavior should be "switchable":
on or off.


> Is there a reason you don't just add the table name manually? e.g.
>
> Numbers-1
> #+tblname: numbers-1
> |        1 |
> |        2 |
> |       45 |
> |     test |
> | 3.141592 |

I'm not really enthousiast about solutions that would be manual.

On the contrary, even if table names were always exported, the name can easily
disappear from your documents with just a (very) little bit of CSS or LaTeX
code. In CSS, just apply "display: none" on the DIV, and you're done. I don't
really understand the resistance you're talking about, then.

To sum up, if the info is there, it's really easy to remove it (even
automatically!). If it's not there, it's quite a tedious task to add it
(manually)...

Do you understand the need I'm trying to express?

Best regards,
  Seb

-- 
Sébastien Vauban