Re: makeinfo --docbook should include a newline in @docbook

2014-03-30 Thread Aharon Robbins
Hi Karl.

> Current makeinfo when processing @docbook ... @end docbook leaves off
> the trailing newline.
>
> I can't tell you how much hassle that the trailing newline before @end is.
> In some cases it is absolutely neeeded.  In other cases it absolutely
> needs to be ignored.  I see no way to get it right.  (It's even worse
> with @macro.)

OK. I didn't appreciate the depth of the problem.

> It seems that, in practice, newlines are not ignored at the end of
> @if... conditionals, but are ignored at the end of raw conditionals.

I guess so.  But for my uses of @docbook, I need the newline.

>  would be a true no-op, if that's what you want.
>
> Another approach is apparently to use an empty comment:
> @docbook
> stuff where you want the newline...
> @c
> @end docbook

OK, I'll try some of these kinds of things and see what I end up with.

>   peformance, while
>
> peformance -> performance 

Thanks! (This may have been a manual typo, I'll check.)

Thanks for the feedback.

Arnold



Re: makeinfo --docbook should include a newline in @docbook

2014-03-28 Thread Karl Berry
Current makeinfo when processing @docbook ... @end docbook leaves off
the trailing newline.

I can't tell you how much hassle that the trailing newline before @end is.
In some cases it is absolutely neeeded.  In other cases it absolutely
needs to be ignored.  I see no way to get it right.  (It's even worse
with @macro.)

This is one of the main reasons why we invented the
@inline... conditionals, which do not ignore trailing whitespace.  See
the Inline Conditionals node.

It seems that, in practice, newlines are not ignored at the end of
@if... conditionals, but are ignored at the end of raw conditionals.

so that the result doesn't run together 

 would be a true no-op, if that's what you want.

Another approach is apparently to use an empty comment:
@docbook
stuff where you want the newline...
@c
@end docbook

I'm sure there are other ways.  But I don't think trying to alter
newline-keeping/removing behavior again would be fruitful.

peformance, while

peformance -> performance 
:)

sorry,
karl