Question on nested-blocks syntax

Question on nested-blocks syntax

[Albert Krewinkel]

Hi list,

I was wondering about the current block syntax, specifically about
nested blocks.  The current way to get nested blocks requires escaping
of the nested block commands, e.g.

#+BEGIN_EXAMPLE
  ,#+BEGIN_EXAMPLE
    Meta-example
  ,#+END_EXAMPLE
#+END_EXAMPLE

My personal feeling is that it would be nicer to allow nesting blocks
simply by indenting them:

#+BEGIN_EXAMPLE
  #+BEGIN_EXAMPLE
    Indented meta-example
  #+END_EXAMPLE
#+END_EXAMPLE

I'd be very interested to learn if this (IMHO nicer) second option could
be supported or if there are æsthetical or technical reasons for not
supporting that syntax?

I'm the author of pandoc's org-mode reader and we initially supported
the latter syntax as it felt cleaner.  Since we are striving for maximum
comparability with Emacs as the reference implementation, this specific
feature has since been removed from the reader.

Is there any chance that indentation-based nesting will become legit
org-mode syntax?

Thanks,

Albert


-- 
Albert Krewinkel
GPG: 8eed e3e2 e8c5 6f18 81fe  e836 388d c0b2 1f63 1124

Re: Question on nested-blocks syntax

[Nicolas Goaziou]

Hello,

Albert Krewinkel <albert@zeitkraut.de> writes:

> Is there any chance that indentation-based nesting will become legit
> org-mode syntax?

I don't think so. It would introduce unnecessary complexity.

Regards,

-- 
Nicolas Goaziou

Re: Question on nested-blocks syntax

[Albert Krewinkel]

Nicolas Goaziou <mail@nicolasgoaziou.fr> writes:

> Albert Krewinkel <albert@zeitkraut.de> writes:
>
>> Is there any chance that indentation-based nesting will become legit
>> org-mode syntax?
>
> I don't think so. It would introduce unnecessary complexity.

I guess that makes sense.  Too bad, though.

Thanks for the quick reply,

-- 
Albert Krewinkel
GPG: 8eed e3e2 e8c5 6f18 81fe  e836 388d c0b2 1f63 1124

Re: Question on nested-blocks syntax

[Nicolas Goaziou]

Albert Krewinkel <albert+orgmode@zeitkraut.de> writes:

> Nicolas Goaziou <mail@nicolasgoaziou.fr> writes:

>> I don't think so. It would introduce unnecessary complexity.
>
> I guess that makes sense.  Too bad, though.

My answer was terse. Let me elaborate a bit.

There are a few issues that come to mind when considering this. 

First, you cannot use regexps to parse an example block anymore. You
have to read a line after another until you find the end of the block.
This is more complex.

Also, it defeats the "-i" flag, which allows an user to ignore
indentation in an example block.

In any case, as long as fontification is not based on top of the parser,
I think we should refrain from introducing too deep changes in the
syntax without a serious motivation (e.g., speed concerns, or ambiguous
cases) behind it. I sympathize with aesthetics concerns, Org syntax,
particularly blocks, is "massive", but I think this is not a sufficient
motivation here.

Regards,

Re: Question on nested-blocks syntax

[Rasmus]

Nicolas Goaziou <mail@nicolasgoaziou.fr> writes:

> I sympathize with aesthetics concerns, Org syntax, particularly blocks,
> is "massive", but I think this is not a sufficient motivation here.

I wonder what you mean by "massive" here?  Do you mean the fact that there
is an explicit beginning and end?

Aside: I think a big getting rid of the FORTRAN-70 "touch" (capitalized
keywords and delimiters) would make Org must more ascetically pleasing.
(We’d just need to change default usage of templates and block
capitalization).

Rasmus

-- 
With monopolies the cake is a lie!

Re: Question on nested-blocks syntax

[Nicolas Goaziou]

Rasmus <rasmus@gmx.us> writes:

> Nicolas Goaziou <mail@nicolasgoaziou.fr> writes:
>
>> I sympathize with aesthetics concerns, Org syntax, particularly blocks,
>> is "massive", but I think this is not a sufficient motivation here.
>
> I wonder what you mean by "massive" here?  Do you mean the fact that there
> is an explicit beginning and end?

I mean that, visually, it can be a bit invasive, at times.

> Aside: I think a big getting rid of the FORTRAN-70 "touch" (capitalized
> keywords and delimiters) would make Org must more ascetically pleasing.
> (We’d just need to change default usage of templates and block
> capitalization).

I guess we could do that. 

However, I think it is already easy to write every keyword as lower
case. Moreover, there is no lower case for "#".


Regards,

Re: Question on nested-blocks syntax

[Rasmus]

Nicolas Goaziou <mail@nicolasgoaziou.fr> writes:

>> Aside: I think a big getting rid of the FORTRAN-70 "touch" (capitalized
>> keywords and delimiters) would make Org must more ascetically pleasing.
>> (We’d just need to change default usage of templates and block
>> capitalization).
>
> I guess we could do that. 
>
> However, I think it is already easy to write every keyword as lower 
> case.

As said, templates, and blocks inserted via org-structure-template-alist
are uppercased (as least the default).

> Moreover, there is no lower case for "#".

Neither is there a uppercase...


-- 
evidence suggests Snowden used a powerful tool called monospaced fonts