bug-texinfo
[Top][All Lists]
Advanced

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

Re: texinfo-6.3.90 pretest


From: Gavin Smith
Subject: Re: texinfo-6.3.90 pretest
Date: Sat, 29 Apr 2017 11:25:44 +0100
User-agent: Mutt/1.5.23 (2014-03-12)

On Sat, Apr 29, 2017 at 12:55:21PM +0300, Eli Zaretskii wrote:
> I'm not following: what is followed immediately by the end-of-line on
> my system?
> 
> The text in tp/tests/coverage/formatting-cr.texi around line 25 is
> this:
> 
>  21 @include coverage_macro-cr.texi^M
>  22 ^M
>  23 @copying^M
>  24 In copying^M
>  25 @mymacro{}^M
>  26 @end copying^M
>  27 ^M
>  28 @titlepage^M
> 
> In each of these lines, the ^M character is immediately followed by a
> newline character.  So where's the @^M command here?
> 
> Do you perhaps mean line 49 of coverage_macro-cr.texi?  It has this:
> 
>   @@ followed by a new line^M
>   @^M

Yes. I suspect on your system the CR's are being stripped but on the
system where the reference test results were generated they weren't.

> 
> > >   +formatting-cr.texi:25: warning: accent command `@'' must not be 
> > >   followed by new line (possibly involving @mymacro)
> > 
> > In the input, @' was at the end of a line. In the reference results,
> > I suspect that a CR character appeared first, otherwise this error 
> > message would have been given.
> 
> Where's all that?  coverage_macro-cr.texi has this on line 639:
> 
>   @@' @'^M
> 
> Is this what you are alluding to?

Yes.

> 
> > Do you know if the tests still pass with Texinfo 6.3 on your system?
> 
> You mean, whether running the Texinfo 6.3 test suite still works, with
> the code base of that version?  Or do you mean something else?

I meant running with the code base of that version. That would confirm
whether the problem was introduced since then.

> Anyway, what does this test try to verify?  A ^M followed by a newline
> on Windows is in general indistinguishable from a newline, so is the
> issue being tested here relevant to Windows?

The test was introduced in SVN revision 5345 on 2013-08-27. I think
it is supposed to test that the CR characters are being ignored, but
they aren't in the macro definition, so this looks like a bug.




reply via email to

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