Re: [Bug] \~ can overwrite previous text.

[Alejandro Colomar]

Hi Branden,

On 3/7/23 12:20, G. Branden Robinson wrote:
> Hi Alex,
> 
> At 2023-03-07T02:15:09+0100, Alejandro Colomar wrote:
>> With the above patch, my bug is fixed, but the following test fails:
>>
>>     XFAIL: tmac/tests/e_footnotes-work-with-columns.sh
>>
>> which is about <https://savannah.gnu.org/bugs/?58736>.
> 
> Keep in mind that an XFAIL is an _expected_ failure.

Ahhhh.  Now it makes sense.

>  So it should have
> been XFAILing before you applied your patch.  If it's still XFAILing
> now, nothing has changed.

I didn't test that. :|
Should have/Will do.

Then the only read FAIL I have is one related to fonts that I suspect
has nothing to do with these patches.  Anyway, I'll test without them
to check.  From the two proposed patches, which one makes more sense
to you?  I don't think there's any valid reason to want negative space
at all, so I think returning early is the best thing, but maybe I'm
missing some obscure groff(1) stuff (although nothing that the
regression tests would miss, it seems).

> 
> When an XFAIL unexpectedly passes, it is marked in red as an XPASS and
> the "make check" target exits with a nonzero status.
> 
> https://www.gnu.org/software/automake/manual/html_node/Scripts_002dbased-Testsuites.html
> 
> I wouldn't _expect_ what you're changing to have any effect on that
> me(7) macro package test; it is concerned with vertical placement of
> things, and the code you're experimenting on concerns adjustment of a
> single output line.

Yep, that's what made me believe that my patch was correct and maybe
something else was broken, but if that test is expected to break, then
the right thing seems to be broken :)

> 
> That said, groff regularly overturns my expectations, especially when I
> fiddle with deep internals.  :)> 
> Regards,
> Branden

Cheers,

Alex


-- 
<http://www.alejandro-colomar.es/>
GPG key fingerprint: A9348594CE31283A826FBDD8D57633D441E25BB5

OpenPGP_signature
Description: OpenPGP digital signature