Re: fix blank line handling in .if

[Kristaps Dzonsons <kristaps@bsd.lv>]

On 05/31/12 12:28, Ingo Schwarze wrote:
> Hi Kristaps,
>
> Kristaps Dzonsons wrote on Thu, May 31, 2012 at 11:49:56AM +0200:
>
>> Good one!  As usual, do just commit and I'll look over the patches
>> in hindsight and see if anything makes me sad.
>
> Thanks, will do so tonight.
>
>> I'm particularly fond of this fix---the trailing space/newline has
>> always been annoying.
>>
>> I anticipate finally (finally!) having some free time by this
>> weekend and next week (projects rolling over), so I'll polish the
>> SQLite fixes (this will initially come with a hit to apropos
>> functionality--only OR searches, for example, but is worthwhile
>> given the massive reduction of code complexity).
>
> OK, so i will try to get that in and tested, too!
> (Or at least to nag... ;-)
>
> One thing i have to do is send a revert patch for the WARNING macro.
> A variable number of arguments for a macro is not portable.
>
>> Given Jesse's mention of many roff macros he'd like to
>> implement---few of them possible without layer violations---I'd like
>> to throw together a concept, as we've discussed before (with fear
>> and awe in our voices), of all macros thrown together and parsed in
>> a common way.
>
> I'm not sure it would be fair with respect to Jesse to do that right
> now; well, he can't get clean earth to till, but at least he should
> have a stable platform.  I don't think you will get that project to
> a stable state until his time is up, so it won't help him.
>  From your perspective, it's probably less work, too, to start that
> after he has moved on to diff(1) - less conflicts, in case he manages
> to produce any patches good enough for commit...

Ingo,

Oh, don't get me wrong---I fully agree.  As mentioned in the last mail, 
a greater priority for me is to clean out the dratted tab-handling code 
in -mdoc column lists.  With this cleaned up, and argument parsing 
unified, there'd be a much clearer basis from which to consider a Grand 
Unification of Macros.

I also want to generalise the [x]html frontend to use macro names as 
classes, which would remove a lot of superfluous code.  But that's an 
entirely different story.

Of course, after such a long caesura, it may take a few hours simply to 
remember what's where in the code-base!  My local tree is a mess of 
sqlite benchmarks and tests.

Best,

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