Escaping again!

Escaping again!

[Rustom Mody]

[-- Attachment #1: Type: text/plain, Size: 473 bytes --]

I have some pseudo-haskell in slides I am preparing with org and ox-reveal

- \(+\):: Int → Int → Int
- \(-\):: Int → Int →Int
- \(\leq\):: Int → Int → Bool
- \(=\):: Int → Int → Bool

Those '::' are haskell for has type
However putting a space before the '::' makes it into a definition list!

Ive also tried making the \( \) enclose the whole line -- no use!

So how to escape the :: and tell org: "Please dont make this a definition
list?"

[-- Attachment #2: Type: text/html, Size: 635 bytes --]

Re: Escaping again!

[Bastien]

Hi Rustom,

Rustom Mody <rustompmody@gmail.com> writes:

> - \(+\):: Int → Int → Int
> - \(-\):: Int → Int →Int
> - \(\leq\):: Int → Int → Bool
> - \(=\):: Int → Int → Bool

1. \(+\) :: Int → Int → Int
2. \(-\) :: Int → Int →Int
3. \(\leq\) :: Int → Int → Bool
4. \(=\) :: Int → Int → Bool

would do -- but this is not entirely satisfactory.

For now description lists accept both "-" and "+".  I'm all for
allowing only "-" so that we could use 

+ \(+\) :: Int → Int → Int
+ \(-\) :: Int → Int →Int
+ \(\leq\) :: Int → Int → Bool
+ \(=\) :: Int → Int → Bool

in your example.

What do you and others think?

-- 
 Bastien

Re: Escaping again!

[Rustom Mody]

[-- Attachment #1: Type: text/plain, Size: 2483 bytes --]

Hi Bastien

On Fri, May 30, 2014 at 7:30 PM, Bastien <bzg@gnu.org> wrote:

> Hi Rustom,
>
> Rustom Mody <rustompmody@gmail.com> writes:
>
> > - \(+\):: Int → Int → Int
> > - \(-\):: Int → Int →Int
> > - \(\leq\):: Int → Int → Bool
> > - \(=\):: Int → Int → Bool
>
> 1. \(+\) :: Int → Int → Int
> 2. \(-\) :: Int → Int →Int
> 3. \(\leq\) :: Int → Int → Bool
> 4. \(=\) :: Int → Int → Bool
>
> would do -- but this is not entirely satisfactory.
>
> For now description lists accept both "-" and "+".  I'm all for
> allowing only "-" so that we could use
>
> + \(+\) :: Int → Int → Int
> + \(-\) :: Int → Int →Int
> + \(\leq\) :: Int → Int → Bool
> + \(=\) :: Int → Int → Bool
>
> in your example.
>
> What do you and others think?
>
>
If you are asking me about a proposed change, I thank you for the
consideration :-)

Please dont take the following too seriously -- I am just spoilt by the
fact that I am an old programmer.

Here is a command I ran on the org sources and its output:

$ grep -r '\\\\\\\\\\\\\\\\\\\\' .
./contrib/lisp/ox-groff.el:          "\\(\\\\\\\\\\)?[ \t]*\n" ".br\n"
output)))


./lisp/ox-latex.el:            "\\(\\\\\\\\\\)?[ \t]*\n" " \\\\\\\\\n"
output)))
./lisp/ox-latex.el:              "\\(\\\\\\\\\\)?[ \t]*\n" " \\\\\\\\\n"
contents)))
./lisp/ox-odt.el:            "\\(\\\\\\\\\\)?[ \t]*\n" "<text:line-break/>"
output t)))
./lisp/ox-man.el:      (setq output (replace-regexp-in-string
"\\(\\\\\\\\\\)?[ \t]*\n" ".br\n"
./lisp/ox-html.el:         "\\(\\\\\\\\\\)?[ \t]*\n"
./lisp/ox-html.el:           "\\(\\\\\\\\\\)?[ \t]*\n"
./lisp/ox-texinfo.el:            "\\(\\\\\\\\\\)?[ \t]*\n" " @*\n" output)))
./lisp/ox-texinfo.el:             "\\(\\\\\\\\\\)?[ \t]*\n" " \\\\\\\\\n"
contents)))
./lisp/org.el:    (looking-at "\\\\\\\\\\($\\|[^\\\\]\\)")))

So sometimes we need that much ESCAPE-ing.

I realise that the general support for escaping in org is much harder than
in a programming language like lisp because there are so many different
contexts and different entities to escape.

However I would also like to humbly submit that if one doesn't have
systematic general escaping, there will always be legitimate uses that will
not be addressable.

Anyways… if you are doing this just for me (!) very kind of you!
For now I am getting along using '꞉' (Unicode 0xA789).

Regards,
Rusi

-- 
http://blog.languager.org

[-- Attachment #2: Type: text/html, Size: 3706 bytes --]

Re: Escaping again!

[Loyall, David]

>[...] if one doesn't have systematic general escaping, there
> will always be legitimate uses that will not be addressable.

+1

As a lowly user, I have often wished for a hypothetical function called org-escapify-region.  (And of course the reverse function.)

I've never even looked for one, though, because it is my understanding that org does not have general-case escaping, and without that, no such function could be complete (and stay complete).

I understand that it won't be easy, since every character I can see on my keyboard already has a purpose (or three).

To review: my requested new function is conceptually simple.  If the implementation can't be simple due to the lack of general-case escaping, then maybe this is an argument in favor of a deep change to org, but, that's a conversation I don't know how to participate in.  So good luck and thanks for your consideration, and thanks for such a useful piece of software. :)

Cheers,
--Dave

Re: Escaping again!

[Bastien]

Hi Rustom,

Rustom Mody <rustompmody@gmail.com> writes:

> Anyways… if you are doing this just for me (!) very kind of you!

Well, I will make a separate thread asking if other users would be
fine with the change -- but glad you found the solution above.

-- 
 Bastien

Re: Escaping again!

[Rustom Mody]

[-- Attachment #1: Type: text/plain, Size: 588 bytes --]

David Loyall wrote:

> >[...] if one doesn't have systematic general escaping, there
> > will always be legitimate uses that will not be addressable.

> +1
>
> As a lowly user, I have often wished for a hypothetical function called org-escapify-region.  (And of
> course the reverse function.)

Just another instance that I had to deal with today.
When using reveal and ox-reveal one needs to set
#+REVEAL_ROOT: "path/to/reveal.js"

I wanted to 'comment out' that line.
After trying ':' and single-quote neither of which worked, I finally got it
by making it into:

#+REVEAL_ROOT_XX: ...

[-- Attachment #2: Type: text/html, Size: 910 bytes --]

Re: Escaping again!

[Nick Dokos]

Rustom Mody <rustompmody@gmail.com> writes:

> David Loyall wrote:
>
>> >[...] if one doesn't have systematic general escaping, there
>> > will always be legitimate uses that will not be addressable.
>
>> +1
>>
>> As a lowly user, I have often wished for a hypothetical function called org-escapify-region.  (And of
>> course the reverse function.)
>
> Just another instance that I had to deal with today.
> When using reveal and ox-reveal one needs to set
> #+REVEAL_ROOT: "path/to/reveal.js"
>
> I wanted to 'comment out' that line.
> After trying ':' and single-quote neither of which worked, I finally got it by making it into:
>
> #+REVEAL_ROOT_XX: ...

I usually just do

# +FOO

adding a space.

Nick

Re: Escaping again!

[Bastien]

Nick Dokos <ndokos@gmail.com> writes:

> I usually just do
>
> # +FOO

Well, I do

# #+FOO so that uncommenting makes this right again.

-- 
 Bastien

Re: Escaping again!

[Nicolas Goaziou]

Hello,

Bastien <bzg@gnu.org> writes:

> For now description lists accept both "-" and "+".  I'm all for
> allowing only "-" so that we could use 
>
> + \(+\) :: Int → Int → Int
> + \(-\) :: Int → Int →Int
> + \(\leq\) :: Int → Int → Bool
> + \(=\) :: Int → Int → Bool
>
> in your example.
>
> What do you and others think?

I have no objection, with appropriate documentation, obviously. 

Be careful, though, as some functions need to be updated accordingly:
`org-list-struct', `org-element--list-struct',
`org-at-item-description-p', `org-list-item-body-column',
`org-at-item-timer-p', `org-list-insert-item'... The latter should
probably use `org-at-item-description-p'.


Regards,

-- 
Nicolas Goaziou

Re: Escaping again!

[Bastien]

Hi Nicolas and Rustom,

Nicolas Goaziou <n.goaziou@gmail.com> writes:

> I have no objection, with appropriate documentation, obviously. 
>
> Be careful, though, as some functions need to be updated accordingly:
> `org-list-struct', `org-element--list-struct',
> `org-at-item-description-p', `org-list-item-body-column',
> `org-at-item-timer-p', `org-list-insert-item'... The latter should
> probably use `org-at-item-description-p'.

Quite a lot of work for a minor useful restriction.

Let's put that aside for 8.4.

But if nobody disagree, I think this is a good restriction.

-- 
 Bastien

Re: Escaping again!

[Gregor Zattler]

Hi Bastien,
* Bastien <bzg@gnu.org> [30. May. 2014]:
[...]
> For now description lists accept both "-" and "+".  I'm all for
> allowing only "-" so that we could use 
[...]
> in your example.
> 
> What do you and others think?

I'd like to be able to use a mix of "-" or "+" in list items as to
indicate positive respective negative aspects of something:

* should org-mode support a mix of "+" and "-" as the first char in plain lists? 
  + would be a great to indicate positive and negative aspects of
    something
  - difficult to implement


Ciao, Gregor
-- 
 -... --- .-. . -.. ..--.. ...-.-

Re: Escaping again!

[Nicolas Goaziou]

Hello,

Gregor Zattler <telegraph@gmx.net> writes:

> I'd like to be able to use a mix of "-" or "+" in list items as to
> indicate positive respective negative aspects of something:
>
> * should org-mode support a mix of "+" and "-" as the first char in plain lists? 
>   + would be a great to indicate positive and negative aspects of
>     something
>   - difficult to implement

This was suggested a couple of years ago, and rejected. This would be
confusing. For example, it would not be always possible to "fix" mixed
bullets if mixing was not intended.

You can use two lists to separate positive and negative aspects of
something. IMO, this is also clearer.

Note that this is orthogonal to the issue at hand, so it belongs to
another thread.


Regards,

-- 
Nicolas Goaziou