GNU bug report logs -
#7955
configure does not thoroughly test for ImageMagick includes
Previous Next
Reported by: Glenn Morris <rgm <at> gnu.org>
Date: Tue, 1 Feb 2011 20:31:02 UTC
Severity: minor
Found in version 24.0.50
Fixed in version 24.1
Done: Glenn Morris <rgm <at> gnu.org>
Bug is archived. No further changes may be made.
To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 7955 in the body.
You can then email your comments to 7955 AT debbugs.gnu.org in the normal way.
Toggle the display of automated, internal messages from the tracker.
Report forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Tue, 01 Feb 2011 20:31:02 GMT)
Full text and
rfc822 format available.
Message #3 received at submit <at> debbugs.gnu.org (full text, mbox):
Package: emacs
Version: 24.0.50
Severity: minor
I tried to build the current trunk on a RHEL 4.8 system.
This has ImageMagick 6.0.7 installed.
configure decided that ImageMagick was usable:
checking for Wand... yes
checking IMAGEMAGICK_CFLAGS...
checking IMAGEMAGICK_LIBS... -lWand -lMagick
checking for MagickExportImagePixels... no
[...]
Does Emacs use imagemagick? yes
However, the build failed because of this:
image.c:7455:29: warning: wand/MagickWand.h: No such file or directory
The system has /usr/include/wand/magick_wand.h, but it does not
have /usr/include/wand/MagickWand.h. Looks like the former is now
deprecated. The latter does exist on RHEL 5.5, with ImageMagick 6.2.8.
Obviously I can just re-configure using --without-imagemagick, but
Emacs should do this for me. Either configure should test for some
required minimum version of ImageMagick (I was not able to find out in
what version wand/MagickWand.h was added); or image.c should fall back
to wand/magick_wand.h if it can.
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Wed, 02 Feb 2011 02:41:01 GMT)
Full text and
rfc822 format available.
Message #6 received at 7955 <at> debbugs.gnu.org (full text, mbox):
Joakim, mind taking a look at this? Thanks.
Glenn Morris <rgm <at> gnu.org> writes:
> Package: emacs
> Version: 24.0.50
> Severity: minor
>
> I tried to build the current trunk on a RHEL 4.8 system.
> This has ImageMagick 6.0.7 installed.
> configure decided that ImageMagick was usable:
>
> checking for Wand... yes
> checking IMAGEMAGICK_CFLAGS...
> checking IMAGEMAGICK_LIBS... -lWand -lMagick
> checking for MagickExportImagePixels... no
> [...]
> Does Emacs use imagemagick? yes
>
> However, the build failed because of this:
>
> image.c:7455:29: warning: wand/MagickWand.h: No such file or directory
>
> The system has /usr/include/wand/magick_wand.h, but it does not
> have /usr/include/wand/MagickWand.h. Looks like the former is now
> deprecated. The latter does exist on RHEL 5.5, with ImageMagick 6.2.8.
>
> Obviously I can just re-configure using --without-imagemagick, but
> Emacs should do this for me. Either configure should test for some
> required minimum version of ImageMagick (I was not able to find out in
> what version wand/MagickWand.h was added); or image.c should fall back
> to wand/magick_wand.h if it can.
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Wed, 02 Feb 2011 09:23:02 GMT)
Full text and
rfc822 format available.
Message #9 received at 7955 <at> debbugs.gnu.org (full text, mbox):
Chong Yidong <cyd <at> stupidchicken.com> writes:
> Joakim, mind taking a look at this? Thanks.
I'll ask the ImageMagick maintainers.
If that fails, I can try to add a specific configure check for the include file
we need.
If I could get a pointer to a virtualization image for rhel 4.8 that
would be splendid too.
>
> Glenn Morris <rgm <at> gnu.org> writes:
>
>> Package: emacs
>> Version: 24.0.50
>> Severity: minor
>>
>> I tried to build the current trunk on a RHEL 4.8 system.
>> This has ImageMagick 6.0.7 installed.
>> configure decided that ImageMagick was usable:
>>
>> checking for Wand... yes
>> checking IMAGEMAGICK_CFLAGS...
>> checking IMAGEMAGICK_LIBS... -lWand -lMagick
>> checking for MagickExportImagePixels... no
>> [...]
>> Does Emacs use imagemagick? yes
>>
>> However, the build failed because of this:
>>
>> image.c:7455:29: warning: wand/MagickWand.h: No such file or directory
>>
>> The system has /usr/include/wand/magick_wand.h, but it does not
>> have /usr/include/wand/MagickWand.h. Looks like the former is now
>> deprecated. The latter does exist on RHEL 5.5, with ImageMagick 6.2.8.
>>
>> Obviously I can just re-configure using --without-imagemagick, but
>> Emacs should do this for me. Either configure should test for some
>> required minimum version of ImageMagick (I was not able to find out in
>> what version wand/MagickWand.h was added); or image.c should fall back
>> to wand/magick_wand.h if it can.
--
Joakim Verona
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Wed, 02 Feb 2011 19:24:02 GMT)
Full text and
rfc822 format available.
Message #12 received at 7955 <at> debbugs.gnu.org (full text, mbox):
joakim <at> verona.se wrote:
> If I could get a pointer to a virtualization image for rhel 4.8 that
> would be splendid too.
I don't think you can get one without paying Red Hat cash money;
but Scientific Linux:
http://www.scientificlinux.org/
http://www.scientificlinux.org/distributions/4x/48/
or CentOS
http://www.centos.org/
http://mirror.centos.org/centos/4/isos/
ought to be functionally equivalent in every way.
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Thu, 03 Feb 2011 08:09:02 GMT)
Full text and
rfc822 format available.
Message #15 received at 7955 <at> debbugs.gnu.org (full text, mbox):
http://www.imagemagick.org/discourse-server/viewtopic.php?f=2&t=17985
basically:
,----
| The magick_wand.h header file was deprecated after ImageMagick version 6. Its still included in the distribution but probably should not be used since it will be removed in version 7 of ImageMagick (due in 3-4 years). It should not be a problem. Change your configure.ac file to check for MagickWand.h then use a conditional compile, something like:
|
| #if defined (HAVE_MAGICKWAND_H)
| #include <wand/MagickWand.h>
| #else
| #include <wand/magick_wand.h>
| #endif /* HAVE_MAGICKWAND_H */
`----
--
Joakim Verona
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Tue, 08 Feb 2011 19:12:02 GMT)
Full text and
rfc822 format available.
Message #18 received at 7955 <at> debbugs.gnu.org (full text, mbox):
OK, so can you implement that then?
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Fri, 11 Feb 2011 23:30:03 GMT)
Full text and
rfc822 format available.
Message #21 received at 7955 <at> debbugs.gnu.org (full text, mbox):
I have a patch attempt now. I was going to test it and push it to the
imagemagick branch but experienced some infuriating setbacks.
If you could help me test and commit that would be spectacular.
=== modified file 'src/image.c'
--- src/image.c 2011-02-01 19:46:21 +0000
+++ src/image.c 2011-02-11 23:23:42 +0000
@@ -7452,7 +7452,12 @@
/* The GIF library also defines DrawRectangle, but its never used in Emacs.
Therefore rename the function so it doesnt collide with ImageMagick. */
#define DrawRectangle DrawRectangleGif
+#if defined (HAVE_WAND_MAGICKWAND_H)
#include <wand/MagickWand.h>
+#else
+#include <wand/magick_wand.h>
+#endif /* HAVE_MAGICKWAND_H */
+
/* imagemagick_load_image is a helper function for imagemagick_load,
which does the actual loading given contents and size, apart from
=== modified file 'configure.in'
--- configure.in 2011-01-31 23:54:50 +0000
+++ configure.in 2011-02-11 23:20:24 +0000
@@ -1832,6 +1832,7 @@
CFLAGS="$CFLAGS $IMAGEMAGICK_CFLAGS"
LIBS="$IMAGEMAGICK_LIBS $LIBS"
AC_CHECK_FUNCS(MagickExportImagePixels)
+ AC_CHECK_HEADERS(wand/MagickWand.h)
fi
fi
fi
--
Joakim Verona
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Sat, 12 Feb 2011 01:03:02 GMT)
Full text and
rfc822 format available.
Message #24 received at 7955 <at> debbugs.gnu.org (full text, mbox):
Seems fine on RHEL5. On RHEL4, build still fails.
configure says:
checking IMAGEMAGICK_CFLAGS...
checking IMAGEMAGICK_LIBS... -lWand -lMagick
checking for MagickExportImagePixels... no
checking wand/MagickWand.h usability... no
checking wand/MagickWand.h presence... no
checking for wand/MagickWand.h... no
Does Emacs use imagemagick? yes
make says:
image.c: In function `imagemagick_load_image':
image.c:7521: warning: implicit declaration of function `MagickWandGenesis'
image.c:7525: warning: implicit declaration of function `MagickSetResolution'
image.c:7532: warning: implicit declaration of function `MagickPingImageBlob'
image.c:7560: warning: implicit declaration of function `AcquireExceptionInfo'
image.c:7560: warning: assignment makes pointer from integer without a cast
image.c:7725: warning: implicit declaration of function `PixelGetMagickColor'
image.c:7809: warning: implicit declaration of function `MagickWandTerminus'
image.c: In function `Fimagemagick_types':
image.c:7899: error: too many arguments to function `GetMagickList'
make[1]: *** [image.o] Error 1
/usr/include/magick/magick.h says:
**GetMagickList(const char *,unsigned long *),
Looks like it may be best for configure to check for a minimum
imagemagick version.
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Sat, 12 Feb 2011 22:31:01 GMT)
Full text and
rfc822 format available.
Message #27 received at 7955 <at> debbugs.gnu.org (full text, mbox):
Glenn Morris <rgm <at> gnu.org> writes:
> Seems fine on RHEL5. On RHEL4, build still fails.
>
> configure says:
>
> checking IMAGEMAGICK_CFLAGS...
> checking IMAGEMAGICK_LIBS... -lWand -lMagick
> checking for MagickExportImagePixels... no
> checking wand/MagickWand.h usability... no
> checking wand/MagickWand.h presence... no
> checking for wand/MagickWand.h... no
> Does Emacs use imagemagick? yes
ImageMagick-6.0 is almost 7 years old. Let's just set HAVE_IMAGEMAGICK
back to "no" if AC_CHECK_HEADER doesn't find MagickWand.h, and not try
to handle the magick_wand.h case.
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Sat, 12 Feb 2011 23:29:02 GMT)
Full text and
rfc822 format available.
Message #30 received at 7955 <at> debbugs.gnu.org (full text, mbox):
Chong Yidong wrote:
> ImageMagick-6.0 is almost 7 years old. Let's just set HAVE_IMAGEMAGICK
> back to "no" if AC_CHECK_HEADER doesn't find MagickWand.h, and not try
> to handle the magick_wand.h case.
I'm fine with not supporting old versions, but AFAIK we have no idea if
an ImageMagick with MagickWand.h also satisfies all the other
assumptions Emacs has made about the API. So, again, I think it is
better if configure tests for the minimum version of ImageMagick that is
known to work.
As a data point, 6.2.8, which is over 4 years old, seems fine.
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Sun, 13 Feb 2011 00:20:03 GMT)
Full text and
rfc822 format available.
Message #33 received at 7955 <at> debbugs.gnu.org (full text, mbox):
Chong Yidong <cyd <at> stupidchicken.com> writes:
> Glenn Morris <rgm <at> gnu.org> writes:
>
>> Seems fine on RHEL5. On RHEL4, build still fails.
>>
>> configure says:
>>
>> checking IMAGEMAGICK_CFLAGS...
>> checking IMAGEMAGICK_LIBS... -lWand -lMagick
>> checking for MagickExportImagePixels... no
>> checking wand/MagickWand.h usability... no
>> checking wand/MagickWand.h presence... no
>> checking for wand/MagickWand.h... no
>> Does Emacs use imagemagick? yes
>
> ImageMagick-6.0 is almost 7 years old. Let's just set HAVE_IMAGEMAGICK
> back to "no" if AC_CHECK_HEADER doesn't find MagickWand.h, and not try
> to handle the magick_wand.h case.
I managed to push to the imagemagick branch. it checks for IM version
now. Im still not sure which version to require yet though so I just
picked one.
--
Joakim Verona
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Mon, 14 Feb 2011 20:05:02 GMT)
Full text and
rfc822 format available.
Message #36 received at 7955 <at> debbugs.gnu.org (full text, mbox):
joakim <at> verona.se wrote:
> I managed to push to the imagemagick branch.
Why are you still using that branch?
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Mon, 14 Feb 2011 20:29:02 GMT)
Full text and
rfc822 format available.
Message #39 received at 7955 <at> debbugs.gnu.org (full text, mbox):
Glenn Morris <rgm <at> gnu.org> writes:
> joakim <at> verona.se wrote:
>
>> I managed to push to the imagemagick branch.
>
> Why are you still using that branch?
I thought it was suitable to work on a branch since I find the autoconf
machinery to be brittle and I'm not even able to test the patch properly
atm. The branch has been updated from trunk so I thought it was ok.
Why is it a concern?
--
Joakim Verona
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Tue, 15 Feb 2011 02:43:01 GMT)
Full text and
rfc822 format available.
Message #42 received at 7955 <at> debbugs.gnu.org (full text, mbox):
joakim <at> verona.se wrote:
> I thought it was suitable to work on a branch since I find the autoconf
> machinery to be brittle and I'm not even able to test the patch properly
> atm. The branch has been updated from trunk so I thought it was ok.
>
> Why is it a concern?
I'm not concerned, just seems like it makes more work to continue using
a branch after the feature in question has been merged to the trunk.
There are several uses of a ">=" test with PKG_CHECK_MODULES in
configure already, so I think you can feel pretty confident about adding
another.
Information forwarded
to
owner <at> debbugs.gnu.org, bug-gnu-emacs <at> gnu.org
:
bug#7955
; Package
emacs
.
(Thu, 07 Apr 2011 21:01:01 GMT)
Full text and
rfc822 format available.
Message #45 received at 7955 <at> debbugs.gnu.org (full text, mbox):
Please can we have this fix installed in the trunk and close this report.
Reply sent
to
Glenn Morris <rgm <at> gnu.org>
:
You have taken responsibility.
(Tue, 12 Apr 2011 03:57:02 GMT)
Full text and
rfc822 format available.
Notification sent
to
Glenn Morris <rgm <at> gnu.org>
:
bug acknowledged by developer.
(Tue, 12 Apr 2011 03:57:02 GMT)
Full text and
rfc822 format available.
Message #50 received at 7955-done <at> debbugs.gnu.org (full text, mbox):
Version: 24.1
Done.
bug archived.
Request was from
Debbugs Internal Request <help-debbugs <at> gnu.org>
to
internal_control <at> debbugs.gnu.org
.
(Tue, 10 May 2011 11:24:04 GMT)
Full text and
rfc822 format available.
This bug report was last modified 14 years and 102 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.