`smoothing_enabled' undeclared

`smoothing_enabled' undeclared

[Herbert Euler]

Hello,

It seems HEAD sync was incorrect again.  When compiling
src/w32term.c on Windows:

gcc -I. -DWIN32_LEAN_AND_MEAN -D_WIN32_WINNT=0x0500 -D_X86_=1 -c  
-mno-cygwin -m
cpu=i686 -O2  -Di386 -D_CRTAPI1=_cdecl  -O2 -Demacs=1 -DWINDOWSNT -DDOS_NT 
-DHAV
E_CONFIG_H -I../nt/inc -D_UCHAR_T -DHAVE_NTGUI=1 -DUSE_CRT_DLL=1 
-DPURESIZE=5000
000 -o oo-spd/i386/w32term.o w32term.c
w32term.c: In function `w32_initialize':
w32term.c:6551: `smoothing_enabled' undeclared (first use in this function)
w32term.c:6551: (Each undeclared identifier is reported only once
w32term.c:6551: for each function it appears in.)
w32term.c:6553: `smoothing_type' undeclared (first use in this function)
make[2]: *** [oo-spd/i386/w32term.o] Error 1
make[2]: Leaving directory `E:/emacs-cvs/src'
make[1]: *** [bootstrap-temacs] Error 2
make[1]: Leaving directory `E:/emacs-cvs/src'
make: *** [bootstrap-gmake] Error 2

Regards,
Guanpeng Xu

_________________________________________________________________
Don't just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/

Re: `smoothing_enabled' undeclared

[Eli Zaretskii]

> From: "Herbert Euler" <herberteuler@hotmail.com>
> Bcc: 
> Date: Fri, 23 Jun 2006 11:20:50 +0800
> 
> It seems HEAD sync was incorrect again.  When compiling
> src/w32term.c on Windows:
> 
> gcc -I. -DWIN32_LEAN_AND_MEAN -D_WIN32_WINNT=0x0500 -D_X86_=1 -c  
> -mno-cygwin -m
> cpu=i686 -O2  -Di386 -D_CRTAPI1=_cdecl  -O2 -Demacs=1 -DWINDOWSNT -DDOS_NT 
> -DHAV
> E_CONFIG_H -I../nt/inc -D_UCHAR_T -DHAVE_NTGUI=1 -DUSE_CRT_DLL=1 
> -DPURESIZE=5000
> 000 -o oo-spd/i386/w32term.o w32term.c
> w32term.c: In function `w32_initialize':
> w32term.c:6551: `smoothing_enabled' undeclared (first use in this function)
> w32term.c:6551: (Each undeclared identifier is reported only once
> w32term.c:6551: for each function it appears in.)
> w32term.c:6553: `smoothing_type' undeclared (first use in this function)
> make[2]: *** [oo-spd/i386/w32term.o] Error 1

I cannot reproduce this: w32term.c compiles for me just fine.  And the
error message looks very strange anyway, since w32term.c has only 6474
lines (and was last modified on 5th June, so if there were a problem,
lots of people would have complained by now).

Can you see what is going on around the lines that GCC complains
about?

Re: `smoothing_enabled' undeclared

[Herbert Euler]

>From: Eli Zaretskii <eliz@gnu.org>
>Reply-To: Eli Zaretskii <eliz@gnu.org>
>To: "Herbert Euler" <herberteuler@hotmail.com>
>CC: emacs-devel@gnu.org
>Subject: Re: `smoothing_enabled' undeclared
>Date: Fri, 23 Jun 2006 13:10:54 +0300
>
> > From: "Herbert Euler" <herberteuler@hotmail.com>
> > Bcc:
> > Date: Fri, 23 Jun 2006 11:20:50 +0800
> >
> > It seems HEAD sync was incorrect again.  When compiling
> > src/w32term.c on Windows:
> >
> > gcc -I. -DWIN32_LEAN_AND_MEAN -D_WIN32_WINNT=0x0500 -D_X86_=1 -c
> > -mno-cygwin -m
> > cpu=i686 -O2  -Di386 -D_CRTAPI1=_cdecl  -O2 -Demacs=1 -DWINDOWSNT 
>-DDOS_NT
> > -DHAV
> > E_CONFIG_H -I../nt/inc -D_UCHAR_T -DHAVE_NTGUI=1 -DUSE_CRT_DLL=1
> > -DPURESIZE=5000
> > 000 -o oo-spd/i386/w32term.o w32term.c
> > w32term.c: In function `w32_initialize':
> > w32term.c:6551: `smoothing_enabled' undeclared (first use in this 
>function)
> > w32term.c:6551: (Each undeclared identifier is reported only once
> > w32term.c:6551: for each function it appears in.)
> > w32term.c:6553: `smoothing_type' undeclared (first use in this function)
> > make[2]: *** [oo-spd/i386/w32term.o] Error 1
>
>I cannot reproduce this: w32term.c compiles for me just fine.  And the
>error message looks very strange anyway, since w32term.c has only 6474
>lines (and was last modified on 5th June, so if there were a problem,
>lots of people would have complained by now).
>
>Can you see what is going on around the lines that GCC complains
>about?

The lines around there are:

    /* Determine if Cleartype is in use.  Used to enable a hack in
       the char metric calculations which adds extra pixels to
       compensate for the "sub-pixels" that are not counted by the
       system APIs. */
    cleartype_active =
      SystemParametersInfo (SPI_GETFONTSMOOTHING, 0, &smoothing_enabled, 0)
      && smoothing_enabled
      && SystemParametersInfo (SPI_GETFONTSMOOTHINGTYPE, 0, &smoothing_type, 
0)
      && smoothing_type == FE_FONTSMOOTHINGCLEARTYPE;

I "grep"ed in src directory, the result is:

$ pwd
/cygdrive/e/emacs-cvs/src

$ grep -nR smoothing *
w32term.c:6551:      SystemParametersInfo (SPI_GETFONTSMOOTHING, 0, 
&smoothing_enabled, 0)
w32term.c:6552:      && smoothing_enabled
w32term.c:6553:      && SystemParametersInfo (SPI_GETFONTSMOOTHINGTYPE, 0, 
&smoothing_type, 0)
w32term.c:6554:      && smoothing_type == FE_FONTSMOOTHINGCLEARTYPE;

By the way, it's the Unicode 2 branch.

Regards,
Guanpeng Xu

_________________________________________________________________
Don't just search. Find. Check out the new MSN Search! 
http://search.msn.com/

Re: `smoothing_enabled' undeclared

[Nick Roberts]

 > By the way, it's the Unicode 2 branch.

Not exactly `By the way'.  You stated at the start of your message that the
problem was in HEAD.  This is pretty important preliminary information.  Also,
if you do find a bug in the Unicode 2 branch, could you please try to find
out if it's present in HEAD (Emacs 22.0.50) too.

Thanks,

Nick

Re: `smoothing_enabled' undeclared

[Stefan Monnier]

>> By the way, it's the Unicode 2 branch.
> Not exactly `By the way'.  You stated at the start of your message that
> the problem was in HEAD.  This is pretty important
> preliminary information.  Also, if you do find a bug in the Unicode
> 2 branch, could you please try to find out if it's present in HEAD (Emacs
> 22.0.50) too.

Every branch has a "head".  The main branch is called "the trunk".

But you're right: since CVS has been so good at mixing up "HEAD" and "trunk"
(especially since there is/was no way to refer explicitly to the trunk on
cvs's command line), people should be extra careful when using "HEAD" to try
and avoid any confusion.



        Stefan

Re: `smoothing_enabled' undeclared

[Nick Roberts]

Stefan Monnier writes:
 > >> By the way, it's the Unicode 2 branch.
 > > Not exactly `By the way'.  You stated at the start of your message that
 > > the problem was in HEAD.  This is pretty important
 > > preliminary information.  Also, if you do find a bug in the Unicode
 > > 2 branch, could you please try to find out if it's present in HEAD (Emacs
 > > 22.0.50) too.
 > 
 > Every branch has a "head".  The main branch is called "the trunk".

HEAD is a tag.  Is it a tag for the head of the trunk or the head of the
branch which the working directory is in?


-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: `smoothing_enabled' undeclared

[Miles Bader]

Nick Roberts <nickrob@snap.net.nz> writes:
>  > Every branch has a "head".  The main branch is called "the trunk".
>
> HEAD is a tag.  Is it a tag for the head of the trunk or the head of the
> branch which the working directory is in?

The CVS docs are maddeningly vague about this...

I thought it was the latter, but I just did a bit of testing, and the
result of using -rHEAD is different depending on which command you use.

E.g., if, in a working directory where everything has the sticky tag
"emacs-unicode-2" (i.e., that's the current branch), then if I test on
the file "src/xfaces.c", which is different in the trunk and on the
branch, the following commands yield these results:

   * "cvs diff -rHEAD src/xfaces.c" produces no output -- so presumably
     it's diffing against the latest revision of _the current branch_.

   * "cvs update -rHEAD src/xfaces.c" _changes_ the sticky tag from the
     file to be "HEAD", and updates it to be the latest revision on _the trunk_.

I don't know if I'm missing something, but this seems like pretty dumb
behavior... no wonder people are confused about what HEAD means.

[I suppose the reason it is this way is that they simply didn't
special-case HEAD in places where they really should have, so the result
is probably internally consistent but confusing for users...]

So I think that you shouldn't use "-rHEAD" with any command that would
set the sticky tag when given a real branch tag name, but it should be
OK when used with commands like diff.

-Miles
-- 
`Cars give people wonderful freedom and increase their opportunities.
 But they also destroy the environment, to an extent so drastic that
 they kill all social life' (from _A Pattern Language_)

Re: `smoothing_enabled' undeclared

[Bob Rogers]

   From: Miles Bader <miles@gnu.org>
   Date: Mon, 03 Jul 2006 08:33:24 +0900

   Nick Roberts <nickrob@snap.net.nz> writes:
   >  > Every branch has a "head".  The main branch is called "the trunk".
   >
   > HEAD is a tag.  Is it a tag for the head of the trunk or the head of the
   > branch which the working directory is in?

   The CVS docs are maddeningly vague about this...

   I thought it was the latter, but I just did a bit of testing, and the
   result of using -rHEAD is different depending on which command you use.

   . . .

      * "cvs update -rHEAD src/xfaces.c" _changes_ the sticky tag from the
	file to be "HEAD", and updates it to be the latest revision on
	_the trunk_.

It's even worse than that.  If you edit src/xfaces.c, you'll find you
can't commit the change; you must first do "cvs update -A src/xfaces.c"
to get it fully back onto the trunk.  If you look at emacs/CVS/Entries,
you'll notice that "trunk" is denoted by the absence of a tag (field
between the last two slashes), not an explicit "HEAD".

   IMHO, "-rHEAD" is strictly to be avoided.

					-- Bob Rogers
					   http://rgrjr.dyndns.org/

Re: `smoothing_enabled' undeclared

[Stefan Monnier]

>> > Every branch has a "head".  The main branch is called "the trunk".
>> HEAD is a tag.  Is it a tag for the head of the trunk or the head of the
>> branch which the working directory is in?

It's the head of the current branch.

> The CVS docs are maddeningly vague about this...

At best,y es.

> I thought it was the latter, but I just did a bit of testing, and the
> result of using -rHEAD is different depending on which command you use.

All cases where HEAD refers to the trunk are either historical or just
plain bugs.

>    * "cvs update -rHEAD src/xfaces.c" _changes_ the sticky tag from the
>    file to be "HEAD", and updates it to be the latest revision on _the
>    trunk_.

And it puts the src/xfaces.c file in a state that CVS itself doesn't
understand: i.e. it's a bug.  Same thing with any other pseudo-tag
like BASE.

> [I suppose the reason it is this way is that they simply didn't
> special-case HEAD in places where they really should have, so the result
> is probably internally consistent but confusing for users...]

That's about right, although a bit too generous.


        Stefan

Re: `smoothing_enabled' undeclared

[Eli Zaretskii]

> From: "Herbert Euler" <herberteuler@hotmail.com>
> Cc: emacs-devel@gnu.org
> Bcc: 
> Date: Sat, 24 Jun 2006 10:16:02 +0800
> 
> By the way, it's the Unicode 2 branch.

That explains everything.  (You didn't say that in your original
message, only mentioned the "sync with HEAD", which wasn't clear
enough to guess you are using a branch.)

Sorry, I cannot help you here, as I don't track that branch.