groff
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v2] tzfile.5: Fix indentation


From: Alejandro Colomar
Subject: Re: [PATCH v2] tzfile.5: Fix indentation
Date: Sun, 17 Mar 2024 20:31:55 +0100

On Sun, Mar 17, 2024 at 02:24:29PM -0500, G. Branden Robinson wrote:
> [looping in groff mailing list to pitch a terminological reform]
> 
> Hi Alex,

Hi Banden,

> At 2024-03-17T19:56:10+0100, Alejandro Colomar wrote:
> > You already gained the first two spaces from the previous RS, in which
> > you're nested.  Remember that we have
> > 
> >  .RS "\w'  'u"
> >  .IP \(bu "\w'\(bu  'u"
> > 
> > a few lines above.  The second RS, by default, does the same as the
> > previous IP, that is, "\w'\(bu  'u".
> 
> I see that I need to clarify the groff_man(7) page in this department.
> In fact, I suspect Doug McIlroy's term "prevailing indent" was
> tailor-made for expressing this behavior.  But I want to amend it, in
> documentation and the "an.tmac" source file, to "prevailing _inset_",
> because "indentation" is overloaded to also refer to the additional
> spacing applied to (at least some) lines of an `IP`, `TP`, or
> (deprecated) `HP` paragraph.

Hmmm, I was about to say inset, but double-checked groff_man(7) to be
sure of the exact term, and then I got confused even more:

$ man groff_man | grep '   \.\(IP\|RS\) \['
     .RS [insetā€amount]
     .IP [tag [indentation]]


Since RS uses the same amount that IP used before it, it seems they
could be the same thing.  But then RS uses inset and IP uses
indentation.  How do pears be added to apples?  :)

Have a lovely day!
Alex

> 
> The resulting concept is simple.
> 
> prevailing inset = base paragraph inset + sum of relative insets
> 
> Equivalently:
> 
> prevailing inset = value of `BP` register[1] + amounts in `RS` calls[2]
> 
> Regards,
> Branden
> 
> [1] forthcoming in groff 1.24
> 
> [2] for each active inset, if specified; if not, the indentation of the
>     previous paragraph in the (sub)section is used, and if none (as with
>     `P` and its synonyms), then the value of the `IN` register
> 
> I _think_ I've got that right.



-- 
<https://www.alejandro-colomar.es/>
Looking for a remote C programming job at the moment.

Attachment: signature.asc
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]