On 18 Aug, GOTO Kentaro wrote:
> Hi, 
> 
> # Don't call me Mr. Goto please ;-)
> 
> In message "[ruby-talk:00683] Re: Summary of discussion about RD (Re: Documenta tion about RD?)"
>     on 99/08/18, clemens.hintze / alcatel.de <clemens.hintze / alcatel.de> writes:
> 

[...]

>>Also used for email addresses?
> 
> Yes. ((<URL:mailto:gotoken / notwork.org>)) is also legal URL. 

I mean something like: ((<c.hintze / gmx.net>)) :-)

> 
>>> Tosh suggested to use "(x ... x)" instead of "((x ...x))". Tosh think 
>>> "(x ... x)" is more simple and more naturally embeded in plain text.
>>> But Mr.Goto is afraid that "(x ... x)" is so simple that it may cause
>>> some trouble.
>>
>>I think you are right, but Goto also :-) The `((x..x))' syntax seems to
>>be more error prone!

Arrrghhh! Forgot my dictionary. I meant something like "not so easy to
mistakenly used". I mean *NOT* error prone!!! Sorry! Forgive me!
Urrrgghhs! :-]

> 
> Shall we elect? :-)

No! Not necessary! Simply use `((x...x))'! Do not need to chose ;-)))

> 
>>May I suggest another thing? If there is a text like 
>>
>>	`(( This is text ))' or `(( This is text))'
>>
>>you could treat that *not* as inline-element, and remove *one* leading
>>whitespace! I.e. it would become `((This is text ))' or `((This is
>>text))' respectively!
> 
> Hmm, What do you ame by this feature? But whitespece treating should
> be decide. Japanese guys maybe not sensitive this problem :-)

I can think a situation, where I want to express just that: ((*no*)).
Not meant inline but verbose.

If I would use the `(('((*no*))'))' construct, it would look a little
bit, erhm clumsy?!?

Furthermore I assume that text set within `(('...'))' would be set into
another font? I would like to be able to set it using the current font.

Perhaps we could also use it as: \((*no*))? Perhaps it would be enough
that there is a `\' in front of the inline element to prevent it from
taken into consideration as inline-element? It may be a better solution
than the `ignore blanks' one.

What do you think?

[...]

> 
> Well, we often use embedding document as `lines killer' in daily hack. 
> But current RD does not deal with such a hackish comment, all embedding
> document will be printed. Though it maybe right (pedantic) manner, 
> I feel too rigit. For example
> 
> =begin ignore
> Ahhhhhhh!! why doesn't callcc work as I intend?
> 
> p $ccccc
> p $aaaaa
> =end ignore

Ohhh! That makes sense, I think. I sometimes use that too!

May I make another proposal? Do it as now, take all into consideration,
but perhaps allow rd2html and family to do chose what to take into
consideration.

Somthing with following syntax:

	rd2xxx -use <kind>[,<kind>]... <file>

and

	rd2xxx -usenot <kind>[,<kind>]... <file>

The first example would only take multiline comments into
consideration, if they begin with any of `=begin <kind>'. All other
comments would be ignored!

The second example would take all into consideration *except* all these
ones, that begin with any of `=begin <kind>'.

For example:

	rd2xxx -use rd MyScript.rb 

That would only takes multiline comments into consideration that begin
with `=begin RD' or `=begin rd'.

	rd2xxx -use rd,html MyScript.rb

would take only these into consideration that begins with 
`=begin {rd|RD}' or `=begin {HTML|html}'.

Perhaps e.g. `rd2html' could set the `-use html' flag on any case, if
another `-use' is used.

	rd2xxx -usenot ...

Would works vice versa.

With this solution we also would have the following solved...

>>> Mr.Goto also suggested HTML-embeding like Perl. i.e. use following type of 
>>> "=begin ... =end" to embed HTML (or Markup Language).
>>>  =begin html
>>>  <br>Figure 1.<br>
>>>  =end html
>>> It is suggested in [ruby-list:16162].
>>
>>Perhaps simply insert them into the corresponding result?
> 
> I think so. I just intended it like `#ifdef .. #end' without
> `#else'. But I agree it causes to be too complicated and, probably,
> makes difficult to read the document.

I don't think so. It would power up the tool, I think! I could write a
large `.rd' document, that could be used to produce README, tutorial,
reference and even an online helptext by putting it simply into
different `=begin <kind>' sections! :-))))

Don't forget, I could write the doc into a `.`rd' file, not necessarily
into the script. Perhaps only a part of the doc would be contained into
the script, neh? :-)

> 
>>> Sorry of my poor English...
>>
>>Nothing to forgive. Mine is also not better ;-))))
> 
> Maybe, we are developing a kind of Pidgin language :-)

I am developing such language, every day I try to use it ;-))))))

> 
> -- gotoken

\cle