Re: mdocml: Churn-ish check-in getting mdoc_parseln() and man_parseln() to

[Kristaps Dzonsons <kristaps@bsd.lv>]

Ingo,

> This must not be const.

Registers should be written only by libroff (they're roff constructs).
It's up to libmdoc and libman if they want to cue the front-ends by
looking at both their internal state and the registers.  This will be
apparent by the next commit that implements the cue for SYNOPSIS formatting.

> .Sh SYNOPSIS implies .nr nS 1,
> any other .Sh implies .nr nS 0.
> 
> Apart from that, it might be possible to get your approach working.
> The decisive point will be implementing the frontend cues.

That'll be in the next patch.  I'm committing a big fat clean-up right
now that stuffs regset into struct mdoc/man/roff, because it must reach
deep into the system and I'm tired of changing function prototypes.

> The nice thing about my patch is that the backend and the frontend
> can use identical roff_reg_(get|set)* functions, so some functions
> can be shared between mdoc_validate and mdoc_term.

True, but I don't want the front-ends looking at raw registers.

In general, the compilers should tell the front-ends exactly what to do
with flags and cached data: I want dumb front-ends in terms of semantics.

> With your approach, you must be careful to not implement the same
> thing twice, once in the backend using the roff register data
> structure, then the same logic in the frontend using a different
> cues data structure.
> 
> So, i shall not merge this yet, but wait how it goes.  :)

Thanks for the comments!  Keep an eye peeled.

Kristaps
--
 To unsubscribe send an email to tech+unsubscribe@mdocml.bsd.lv