This reverts commit c870111546.
The constification of a pointer to a pointer caused unexpected issues,
and xorg-devel was unable to come up with a clean, safe, reasonable way
to handle them, so we're chalking this up for now as yet another mistake
in the Xlib API definition we'll be living with.
See https://bugs.freedesktop.org/show_bug.cgi?id=32098 for details.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Fixes numbering of actual chapters to match their filenames and to
make the table/figure numbering match the references in the text.
Reported-by: jelmd on freenode irc
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Updates code & docs for XInternAtoms.
The single atom name argument to XInternAtom was already const char *
in the code, but not the docs, so updated it in the docs too.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net>
When checking for the OS/2 platform, $target_os is used. However, unless
building a cross compiler, the $host* strings contain the platform
details for the build system. See:
http://www.gnu.org/software/automake/manual/automake.html#Cross_002dCompilation
$host_os is already being used to determine the transport and threading
options.
Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
The OS/2 platform requires some utility functions as well as having a
non-32 bit wchar_t. Fix the configure check so that it doesn't also
affect the nios2 cpu, which wouldn't influence these operating system
issues.
Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
Tested-by: Frans Meulenbroeks <fransmeulenbroeks@gmail.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Currenlty the xorg.css file is copied in each location
where a DocBook/XML file resides. This produces about
70 copies in the $(docdir) install tree.
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
fop is used to generate the pdf and ps formats of the documentation.
This can significantly slow down the build, especially when creating all
the compose key charts. Since few people probably want the full set of
doc formats, set the default to 'no'.
The default parameter for XORG_WITH_FOP is only available in recent
macros. Users generating configure from older macros will just get
'auto' as the default.
Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
Reviewed-by: Gaetan Nadon <memsize@videotron.ca>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Add C.UTF-8 locale as an alias for en_US.UTF-8
Add C.ASCII locale as an alias for C
(C.UTF-8 is the default locale for cygwin. It also exists in debian,
although I don't think it's intended for use by humans.)
Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk>
Acked-by: Julien Cristau <jcristau@debian.org>
Reviewed-by: James Cloos <cloos@jhcloos.com>
Make expects prerequisites to be files with valid timestamps, and
directories are treated as always being out of date. Thus, any targets
depending on directories will always be rebuilt.
Instead, the doc rules are changed to always create the target's leading
directory. This should prevent the documentation from being rebuilt when
"make install" is run.
Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
This addresses a build failure which can result from <X11/Xlocale.h> and
<xlocale.h> being included in the same code since they both used the same
_XLOCALE_H_ protection.
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
Adds compose-chart.pl to generate DocBook/XML documents listing compose
keys, and Makefile rules to generate HTML & PDF output from them if xmlto
is present.
https://bugs.freedesktop.org/show_bug.cgi?id=19379
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
Reviewed-by: James Cloos <cloos@jhcloos.com>
Tested-by: Gaetan Nadon <memsize@videotron.ca>
On a 4 core CPU with gmake -j 16 the nls subdir builds in half the time,
plus this simplifies the next set of changes.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Mikhail Gusarov <dottedmag@dottedmag.net>
Reviewed-by: James Cloos <cloos@jhcloos.com>
Tested-by: Gaetan Nadon <memsize@videotron.ca>
The man page suffix is the same for all libraries on a given
platform and is not configurable. It should have been removed
in commit 09edc6de6.
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
Inspired by a pattern in NoMachine's NX. Consistently zeroed buffers
compress better with ssh and friends. Note that you'll need to rebuild
all your protocol libraries to take advantage of this.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Libtool’s is now sufficiently fast that DOLT is no longer
worth the bother, even on those few systems where is works.
This reverts commit 3e9afd501e
and part of commit d31e644c65.
It conflicted with commit f6a4fd0c76
which moved dolt from configure.ac to m4/dolt.m4.
And it addresses: http://bugs.freedesktop.org/show_bug.cgi?id=28188
Signed-off-by: James Cloos <cloos@jhcloos.com>
This option was added in commit 6e752ea120 with no explanation.
The section number is provoded by XORG_MANPAGE_SECTIONS
There is no case where libX11 should be different than other libs
The option was also used to disable building of the man pages,
which build in 14 secs. No indication this is required.
If there is a requirement from system builders to disable building
of man pages, it could be done consistently for all modules.
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
The X Test Suite's XDisplayString test checks the invariant
XDisplayString(XOpenDisplay(str)) == str. The Xlib XOpenDisplay violates this
invariant by expanding str to the canonical form "host:display.scrn" (unless
HAVE_LAUNCHD is set and it starts with "/tmp/launch"). E.g., this expands ":1"
to ":1.0":
400|26 1 1 19:26:41|IC Start
200|26 1 19:26:41|TP Start
520|26 1 00032625 1 1|VSW5TESTSUITE PURPOSE 1
520|26 1 00032625 1 2|Assertion XDisplayString-1.(A)
520|26 1 00032625 1 3|A call to XDisplayString returns the string that was used
520|26 1 00032625 1 4|as the argument to the XOpenDisplay call that returned the
520|26 1 00032625 1 5|value used as the display argument.
520|26 1 00032625 1 6|METH: Open a connection using XOpenDisplay.
520|26 1 00032625 1 7|METH: Obtain the display string using XDisplayString.
520|26 1 00032625 1 8|METH: Verify that the value of the string is the parameter used in XOpenDisplay.
520|26 1 00032625 1 9|METH: Close the display using XCloseDisplay.
520|26 1 00032625 1 10|REPORT: XDisplayString() returned ":1.0" instead of ":1".
220|26 1 1 19:26:41|FAIL
410|26 1 1 19:26:41|IC End
Fix this by deleting all of the code to construct the canonical path and just
stashing a copy of the original display_name in dpy->display_name.
Signed-off-by: Aaron Plattner <aplattner@nvidia.com>
Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
Store the shadow files in git as any other man page.
Move man pages to man dir and use the common makefile
Local fix in CVS for bug 5628 is not required
as the problem has been fixed in
util-macros d9062e4077ebfd0985baf8418f3d0f111b9ddbba
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
There were no special symbols needing cpp.
Everything can be handled by the default MAN_SUBSTS in util-macros.
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
_XReply would always call dequeue_pending_request on errors. When it
got an error for the current request, it would call dequeue, then break
out of the loop; then, if it had an error in the event queue, it would
compare it with the sequence number of the now-freed pending request.
_XReply already stored that sequence number in dpy->last_request_read
before freeing it, so look at that instead.
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=29412
Signed-off-by: Jamey Sharp <jamey@minilop.net>
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Since gnaborretni is primarily used in LTR locales, the ? part of the
sequence reflects the position of the bulge in the ?. When scanning an
inverted interrobang left-to-right, you see the bulge first, so the
compose sequence ?! is used; upright interrobang shows the bulge
last, so it's !?.
This is a change in behaviour, but I was unable to come up with anything
better. ??!! or !!?? would match the patterns for ¿ and ¡, but they
would delay evaluation of them; "¿Qué?" wouldn't show the ¿ until you hit
Q. Likewise ?!?! and !?!? would delay showing the interrobang itself.
~!? and ~?! were considered but are arguably less intuitive.
Reviewed-by: James Cloos <cloos@jhcloos.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
This move fixes a Java class cast exception in the glossary.
The problem was introduced in commit
26f4f0d508
It may not happen on all versions of the doc toolchain.
There is no reason why indexterm cannot appear in glossdef,
this is a workaround to an implementation problem found by
trial and error.
Signed-off-by: Gaetan Nadon <memsize@videotron.ca>
Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Unicode is even more pointlessly awesome! Add a binding for Compose-\o/
to U+1F64C PERSON RAISING BOTH HANDS IN CELEBRATION.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
If we get input in the style of 0xdeadbeef, just return that exact
keysym. Introduces a dependency on strtoul, which I'm told is OK on all
the systems we care about.
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
