[libvirt] [PATCH] docs: Drop unused rule for internals/%.html.tmp target
We're using the %.html.tmp for all html files now so drop the unused one and rather make sure the needed directory exists. This fixes build failures as described in https://www.redhat.com/archives/libvir-list/2015-August/msg00603.html --- configure.ac | 1 + docs/Makefile.am | 9 --------- 2 files changed, 1 insertion(+), 9 deletions(-) diff --git a/configure.ac b/configure.ac index 9066ed6..749dbb7 100644 --- a/configure.ac +++ b/configure.ac @@ -2774,6 +2774,7 @@ AC_DEFINE_UNQUOTED([isbase64],[libvirt_gl_isbase64],[Hack to avoid symbol clash] AC_DEFINE_UNQUOTED([base64_encode],[libvirt_gl_base64_encode],[Hack to avoid symbol clash]) AC_DEFINE_UNQUOTED([base64_encode_alloc],[libvirt_gl_base64_encode_alloc],[Hack to avoid symbol clash]) +AC_CONFIG_COMMANDS([mkdir], [$MKDIR_P docs/internals]) AC_CONFIG_FILES([run], [chmod +x,-w run]) AC_CONFIG_FILES([\ diff --git a/docs/Makefile.am b/docs/Makefile.am index daf37b6..563baca 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -218,15 +218,6 @@ $(srcdir)/hvsupport.html.in: $(srcdir)/hvsupport.pl $(api_DATA) \ %.png: %.fig convert -rotate 90 $< $@ -internals/%.html.tmp: internals/%.html.in subsite.xsl page.xsl sitemap.html.in - @if [ -x $(XSLTPROC) ] ; then \ - echo "Generating $@"; \ - $(MKDIR_P) internals; \ - name=`echo $@ | sed -e 's/.tmp//'`; \ - $(XSLTPROC) --stringparam pagename $$name --nonet \ - $(top_srcdir)/docs/subsite.xsl $< > $@ \ - || { rm $@ && exit 1; }; fi - %.html.tmp: %.html.in site.xsl page.xsl sitemap.html.in $(acl_generated) @if [ -x $(XSLTPROC) ] ; then \ echo "Generating $@"; \ -- 2.1.4
On 19.08.2015 14:40, Guido Günther wrote:
We're using the %.html.tmp for all html files now so drop the unused one and rather make sure the needed directory exists.
This fixes build failures as described in
https://www.redhat.com/archives/libvir-list/2015-August/msg00603.html --- configure.ac | 1 + docs/Makefile.am | 9 --------- 2 files changed, 1 insertion(+), 9 deletions(-)
diff --git a/configure.ac b/configure.ac index 9066ed6..749dbb7 100644 --- a/configure.ac +++ b/configure.ac @@ -2774,6 +2774,7 @@ AC_DEFINE_UNQUOTED([isbase64],[libvirt_gl_isbase64],[Hack to avoid symbol clash] AC_DEFINE_UNQUOTED([base64_encode],[libvirt_gl_base64_encode],[Hack to avoid symbol clash]) AC_DEFINE_UNQUOTED([base64_encode_alloc],[libvirt_gl_base64_encode_alloc],[Hack to avoid symbol clash])
+AC_CONFIG_COMMANDS([mkdir], [$MKDIR_P docs/internals]) AC_CONFIG_FILES([run], [chmod +x,-w run]) AC_CONFIG_FILES([\ diff --git a/docs/Makefile.am b/docs/Makefile.am index daf37b6..563baca 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -218,15 +218,6 @@ $(srcdir)/hvsupport.html.in: $(srcdir)/hvsupport.pl $(api_DATA) \ %.png: %.fig convert -rotate 90 $< $@
-internals/%.html.tmp: internals/%.html.in subsite.xsl page.xsl sitemap.html.in - @if [ -x $(XSLTPROC) ] ; then \ - echo "Generating $@"; \ - $(MKDIR_P) internals; \ - name=`echo $@ | sed -e 's/.tmp//'`; \ - $(XSLTPROC) --stringparam pagename $$name --nonet \ - $(top_srcdir)/docs/subsite.xsl $< > $@ \ - || { rm $@ && exit 1; }; fi - %.html.tmp: %.html.in site.xsl page.xsl sitemap.html.in $(acl_generated) @if [ -x $(XSLTPROC) ] ; then \ echo "Generating $@"; \
ACK Michal
On Thu, Aug 20, 2015 at 03:12:36AM +0200, Michal Privoznik wrote:
On 19.08.2015 14:40, Guido Günther wrote:
We're using the %.html.tmp for all html files now so drop the unused one and rather make sure the needed directory exists.
This fixes build failures as described in
https://www.redhat.com/archives/libvir-list/2015-August/msg00603.html --- configure.ac | 1 + docs/Makefile.am | 9 --------- 2 files changed, 1 insertion(+), 9 deletions(-)
diff --git a/configure.ac b/configure.ac index 9066ed6..749dbb7 100644 --- a/configure.ac +++ b/configure.ac @@ -2774,6 +2774,7 @@ AC_DEFINE_UNQUOTED([isbase64],[libvirt_gl_isbase64],[Hack to avoid symbol clash] AC_DEFINE_UNQUOTED([base64_encode],[libvirt_gl_base64_encode],[Hack to avoid symbol clash]) AC_DEFINE_UNQUOTED([base64_encode_alloc],[libvirt_gl_base64_encode_alloc],[Hack to avoid symbol clash])
+AC_CONFIG_COMMANDS([mkdir], [$MKDIR_P docs/internals]) AC_CONFIG_FILES([run], [chmod +x,-w run]) AC_CONFIG_FILES([\ diff --git a/docs/Makefile.am b/docs/Makefile.am index daf37b6..563baca 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -218,15 +218,6 @@ $(srcdir)/hvsupport.html.in: $(srcdir)/hvsupport.pl $(api_DATA) \ %.png: %.fig convert -rotate 90 $< $@
-internals/%.html.tmp: internals/%.html.in subsite.xsl page.xsl sitemap.html.in - @if [ -x $(XSLTPROC) ] ; then \ - echo "Generating $@"; \ - $(MKDIR_P) internals; \ - name=`echo $@ | sed -e 's/.tmp//'`; \ - $(XSLTPROC) --stringparam pagename $$name --nonet \ - $(top_srcdir)/docs/subsite.xsl $< > $@ \ - || { rm $@ && exit 1; }; fi - %.html.tmp: %.html.in site.xsl page.xsl sitemap.html.in $(acl_generated) @if [ -x $(XSLTPROC) ] ; then \ echo "Generating $@"; \
ACK
Pushed. Thanks, -- Guido
On Thu, Aug 20, 2015 at 10:50:54AM +0200, Guido Günther wrote:
On Thu, Aug 20, 2015 at 03:12:36AM +0200, Michal Privoznik wrote:
On 19.08.2015 14:40, Guido Günther wrote:
We're using the %.html.tmp for all html files now so drop the unused one and rather make sure the needed directory exists.
This fixes build failures as described in
https://www.redhat.com/archives/libvir-list/2015-August/msg00603.html ---
I'm glad it fixed your problem, but now it doesn't build on old autoconf/automake. And it also generates the pages using different schema. What was the issue for you (why you originally needed this patch)? Couldn't that be fixed differently and this one reverted?
configure.ac | 1 + docs/Makefile.am | 9 --------- 2 files changed, 1 insertion(+), 9 deletions(-)
diff --git a/configure.ac b/configure.ac index 9066ed6..749dbb7 100644 --- a/configure.ac +++ b/configure.ac @@ -2774,6 +2774,7 @@ AC_DEFINE_UNQUOTED([isbase64],[libvirt_gl_isbase64],[Hack to avoid symbol clash] AC_DEFINE_UNQUOTED([base64_encode],[libvirt_gl_base64_encode],[Hack to avoid symbol clash]) AC_DEFINE_UNQUOTED([base64_encode_alloc],[libvirt_gl_base64_encode_alloc],[Hack to avoid symbol clash])
+AC_CONFIG_COMMANDS([mkdir], [$MKDIR_P docs/internals]) AC_CONFIG_FILES([run], [chmod +x,-w run]) AC_CONFIG_FILES([\ diff --git a/docs/Makefile.am b/docs/Makefile.am index daf37b6..563baca 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -218,15 +218,6 @@ $(srcdir)/hvsupport.html.in: $(srcdir)/hvsupport.pl $(api_DATA) \ %.png: %.fig convert -rotate 90 $< $@
-internals/%.html.tmp: internals/%.html.in subsite.xsl page.xsl sitemap.html.in - @if [ -x $(XSLTPROC) ] ; then \ - echo "Generating $@"; \ - $(MKDIR_P) internals; \ - name=`echo $@ | sed -e 's/.tmp//'`; \ - $(XSLTPROC) --stringparam pagename $$name --nonet \ - $(top_srcdir)/docs/subsite.xsl $< > $@ \ - || { rm $@ && exit 1; }; fi - %.html.tmp: %.html.in site.xsl page.xsl sitemap.html.in $(acl_generated) @if [ -x $(XSLTPROC) ] ; then \ echo "Generating $@"; \
ACK
Pushed. Thanks, -- Guido
-- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
On Mon, Aug 31, 2015 at 05:22:23PM +0200, Martin Kletzander wrote:
On Thu, Aug 20, 2015 at 10:50:54AM +0200, Guido Günther wrote:
On Thu, Aug 20, 2015 at 03:12:36AM +0200, Michal Privoznik wrote:
On 19.08.2015 14:40, Guido Günther wrote:
We're using the %.html.tmp for all html files now so drop the unused one and rather make sure the needed directory exists.
This fixes build failures as described in
https://www.redhat.com/archives/libvir-list/2015-August/msg00603.html ---
I'm glad it fixed your problem, but now it doesn't build on old autoconf/automake. And it also generates the pages using different schema. What was the issue for you (why you originally needed this patch)? Couldn't that be fixed differently and this one reverted?
I had some off list discussion about Centos 5 and it seems the MKDIR_P in configure.ac just isn't expanded correctly, that could be worked around, I guess. The reason I proposed this change was that from 1.2.17 on the build failed with: make[4]: Entering directory '/tmp/buildd/libvirt-1.2.17~rc1/debian/build/docs' missing XHTML1 DTD cat: internals/locking.html.tmp: No such file or directory Makefile:2385: recipe for target 'internals/locking.html' failed make[4]: *** [internals/locking.html] Error 1 that was caused by locking.html.tmp using the %.html.tmp: instead of the internals/%.html.tmp: rule. So the rule removal was more of a cleanup. Creating the directory via configure.ac was already sufficient to "fix" the problem. See: https://www.redhat.com/archives/libvir-list/2015-June/msg01603.html for some more details. Cheers, -- Guido
On 31.08.2015 18:11, Guido Günther wrote:
On Mon, Aug 31, 2015 at 05:22:23PM +0200, Martin Kletzander wrote:
On Thu, Aug 20, 2015 at 10:50:54AM +0200, Guido Günther wrote:
On Thu, Aug 20, 2015 at 03:12:36AM +0200, Michal Privoznik wrote:
On 19.08.2015 14:40, Guido Günther wrote:
We're using the %.html.tmp for all html files now so drop the unused one and rather make sure the needed directory exists.
This fixes build failures as described in
https://www.redhat.com/archives/libvir-list/2015-August/msg00603.html ---
I'm glad it fixed your problem, but now it doesn't build on old autoconf/automake. And it also generates the pages using different schema. What was the issue for you (why you originally needed this patch)? Couldn't that be fixed differently and this one reverted?
I had some off list discussion about Centos 5 and it seems the MKDIR_P in configure.ac just isn't expanded correctly, that could be worked around, I guess.
The reason I proposed this change was that from 1.2.17 on the build failed with:
make[4]: Entering directory '/tmp/buildd/libvirt-1.2.17~rc1/debian/build/docs' missing XHTML1 DTD cat: internals/locking.html.tmp: No such file or directory Makefile:2385: recipe for target 'internals/locking.html' failed make[4]: *** [internals/locking.html] Error 1
that was caused by locking.html.tmp using the %.html.tmp: instead of the internals/%.html.tmp: rule.
So the rule removal was more of a cleanup. Creating the directory via configure.ac was already sufficient to "fix" the problem.
Guido, I think the original problem you're seeing is just an ordering problem. If you revert you patch, and apply this one: diff --git a/docs/Makefile.am b/docs/Makefile.am index 563baca..529f695 100644 --- a/docs/Makefile.am +++ b/docs/Makefile.am @@ -162,7 +162,7 @@ EXTRA_DIST= \ sitemap.html.in aclperms.htmlinc \ todo.pl hvsupport.pl todo.cfg-example -acl_generated = aclperms.htmlinc +acl_generated = $(srcdir)/aclperms.htmlinc $(srcdir)/aclperms.htmlinc: $(top_srcdir)/src/access/viraccessperm.h \ $(srcdir)/genaclperms.pl Makefile.am @@ -174,7 +174,7 @@ MAINTAINERCLEANFILES = \ $(addprefix $(srcdir)/,$(devhelphtml)) \ $(addprefix $(srcdir)/,$(internals_html)) \ $(addprefix $(srcdir)/,$(dot_php)) \ - $(srcdir)/hvsupport.html.in $(srcdir)/aclperms.htmlinc + $(srcdir)/hvsupport.html.in $(acl_generated) all-am: web are you able to build libvirt on debian? Can you test it please? Michal
On Tue, Sep 01, 2015 at 03:27:05PM +0200, Michal Privoznik wrote: [..snip..]
Guido, I think the original problem you're seeing is just an ordering problem. If you revert you patch, and apply this one:
This gives the same error. Feel free to revert my change and I keep the patch in Debian until we tracked down the root cause and don't interfere with the 1.2.19 release. Cheers, -- Guido
On Wed, Sep 02, 2015 at 09:27:00AM +0200, Guido Günther wrote:
On Tue, Sep 01, 2015 at 03:27:05PM +0200, Michal Privoznik wrote: [..snip..]
Guido, I think the original problem you're seeing is just an ordering problem. If you revert you patch, and apply this one:
This gives the same error. Feel free to revert my change and I keep the patch in Debian until we tracked down the root cause and don't interfere with the 1.2.19 release.
The release was done without the revert, but newer Debian is more important then super-old CentOS. The problem is that we still have broken subsites (the internal pages). I tried reproducing your problem on our debian installation and I couldn't. I have no idea how the versions go, etc. But I really badly want to fix this. What are the versions and steps to reproduce for the original problem you were trying to fix? I'm starting fresh installation to finally fix this once and for all. Martin
Cheers, -- Guido
-- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
Hi, On Mon, Sep 07, 2015 at 03:26:26PM +0200, Martin Kletzander wrote:
On Wed, Sep 02, 2015 at 09:27:00AM +0200, Guido Günther wrote:
On Tue, Sep 01, 2015 at 03:27:05PM +0200, Michal Privoznik wrote: [..snip..]
Guido, I think the original problem you're seeing is just an ordering problem. If you revert you patch, and apply this one:
This gives the same error. Feel free to revert my change and I keep the patch in Debian until we tracked down the root cause and don't interfere with the 1.2.19 release.
The release was done without the revert, but newer Debian is more important then super-old CentOS. The problem is that we still have broken subsites (the internal pages). I tried reproducing your problem on our debian installation and I couldn't. I have no idea how the versions go, etc. But I really badly want to fix this. What are the versions and steps to reproduce for the original problem you were trying to fix? I'm starting fresh installation to finally fix this once and for all.
I'm doing a: git revert e5470dd0e0a5be02cbf18a882cfc676b39d1c1a5 mkdir build && cd build ../autogen.sh && make -j8 distcheck (note the distcheck) and it then fails with: make[4]: Entering directory '/var/scratch/src/libvirt/upstream/libvirt/build/libvirt-1.2.20/_build/docs' Generating internals/rpc.html.tmp Generating internals/locking.html.tmp Generating internals/oomtesting.html.tmp Generating internals/command.html.tmp /bin/bash: line 3: internals/rpc.html.tmp: No such file or directory /bin/bash: line 3: internals/locking.html.tmp: No such file or directory rm: /bin/bash: line 3: internals/oomtesting.html.tmp: No such file or directory rm: cannot remove 'internals/rpc.html.tmp'/bin/bash: line 3: internals/command.html.tmp: No such file or directory : No such file or directorycannot remove 'internals/locking.html.tmp': No such file or directory Makefile:2366: recipe for target 'internals/locking.html.tmp' failed make[4]: *** [internals/locking.html.tmp] Error 1 make[4]: *** Waiting for unfinished jobs.... Makefile:2366: recipe for target 'internals/rpc.html.tmp' failed make[4]: *** [internals/rpc.html.tmp] Error 1 rm: cannot remove 'internals/oomtesting.html.tmp': No such file or directory rm: cannot remove 'internals/command.html.tmp': No such file or directory Makefile:2366: recipe for target 'internals/oomtesting.html.tmp' failed Cheers, -- Guido
On Mon, Sep 07, 2015 at 07:15:26PM +0200, Guido Günther wrote:
Hi, On Mon, Sep 07, 2015 at 03:26:26PM +0200, Martin Kletzander wrote:
On Wed, Sep 02, 2015 at 09:27:00AM +0200, Guido Günther wrote:
On Tue, Sep 01, 2015 at 03:27:05PM +0200, Michal Privoznik wrote: [..snip..]
Guido, I think the original problem you're seeing is just an ordering problem. If you revert you patch, and apply this one:
This gives the same error. Feel free to revert my change and I keep the patch in Debian until we tracked down the root cause and don't interfere with the 1.2.19 release.
The release was done without the revert, but newer Debian is more important then super-old CentOS. The problem is that we still have broken subsites (the internal pages). I tried reproducing your problem on our debian installation and I couldn't. I have no idea how the versions go, etc. But I really badly want to fix this. What are the versions and steps to reproduce for the original problem you were trying to fix? I'm starting fresh installation to finally fix this once and for all.
I'm doing a:
git revert e5470dd0e0a5be02cbf18a882cfc676b39d1c1a5 mkdir build && cd build ../autogen.sh && make -j8 distcheck
(note the distcheck) and it then fails with:
make[4]: Entering directory '/var/scratch/src/libvirt/upstream/libvirt/build/libvirt-1.2.20/_build/docs' Generating internals/rpc.html.tmp Generating internals/locking.html.tmp Generating internals/oomtesting.html.tmp Generating internals/command.html.tmp /bin/bash: line 3: internals/rpc.html.tmp: No such file or directory /bin/bash: line 3: internals/locking.html.tmp: No such file or directory rm: /bin/bash: line 3: internals/oomtesting.html.tmp: No such file or directory rm: cannot remove 'internals/rpc.html.tmp'/bin/bash: line 3: internals/command.html.tmp: No such file or directory : No such file or directorycannot remove 'internals/locking.html.tmp': No such file or directory
Makefile:2366: recipe for target 'internals/locking.html.tmp' failed make[4]: *** [internals/locking.html.tmp] Error 1 make[4]: *** Waiting for unfinished jobs.... Makefile:2366: recipe for target 'internals/rpc.html.tmp' failed make[4]: *** [internals/rpc.html.tmp] Error 1 rm: cannot remove 'internals/oomtesting.html.tmp': No such file or directory rm: cannot remove 'internals/command.html.tmp': No such file or directory Makefile:2366: recipe for target 'internals/oomtesting.html.tmp' failed
No message like that for me. I completes with 2 test failures (namely qemuxml2argvtest and virsh-uriprecedence), but no problem with any html files being generated. And I tried this on both stable and unstable debian installations. Would you mind also trying one patch I proposed upstream? I know it will seem a bit far-fetched, desperate times call for desperate measures, I guess. I'm talking about this one: https://www.redhat.com/archives/libvir-list/2015-September/msg00194.html Thanks, Martin
Cheers, -- Guido
On Tue, Sep 08, 2015 at 09:58:12AM +0200, Martin Kletzander wrote: [..snip..]
No message like that for me. I completes with 2 test failures (namely qemuxml2argvtest and virsh-uriprecedence), but no problem with any
That's caused by: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=781232 if the maintainer doesn't act before that it will be fixed with my libxml2 NMU: https://ftp-master.debian.org/deferred.html
html files being generated. And I tried this on both stable and unstable debian installations.
Would you mind also trying one patch I proposed upstream? I know it will seem a bit far-fetched, desperate times call for desperate measures, I guess. I'm talking about this one:
https://www.redhat.com/archives/libvir-list/2015-September/msg00194.html
Nope, same error. I'll try to have another look later this week. Cheers, -- Guido
On Tue, Sep 08, 2015 at 08:30:19PM +0200, Guido Günther wrote:
On Tue, Sep 08, 2015 at 09:58:12AM +0200, Martin Kletzander wrote: [..snip..]
No message like that for me. I completes with 2 test failures (namely qemuxml2argvtest and virsh-uriprecedence), but no problem with any
That's caused by:
Oh, and I thought that would be fixed by 8f17d0eaae7ee. So it looks like the libxml2 version there is somewhere between 8eb55d782a2b and beb7281055db.
if the maintainer doesn't act before that it will be fixed with my libxml2 NMU:
OK, cool, thanks for the info.
html files being generated. And I tried this on both stable and unstable debian installations.
Would you mind also trying one patch I proposed upstream? I know it will seem a bit far-fetched, desperate times call for desperate measures, I guess. I'm talking about this one:
https://www.redhat.com/archives/libvir-list/2015-September/msg00194.html
Nope, same error. I'll try to have another look later this week.
Thanks, I think we really need to revert your fix in the meantime, though, as the internal documentation pages are broken: https://libvirt.org/internals/locking.html
Cheers, -- Guido
Hi, On Wed, Sep 09, 2015 at 09:12:03AM +0200, Martin Kletzander wrote:
On Tue, Sep 08, 2015 at 08:30:19PM +0200, Guido Günther wrote:
On Tue, Sep 08, 2015 at 09:58:12AM +0200, Martin Kletzander wrote: [..snip..]
No message like that for me. I completes with 2 test failures (namely qemuxml2argvtest and virsh-uriprecedence), but no problem with any
That's caused by:
Oh, and I thought that would be fixed by 8f17d0eaae7ee. So it looks like the libxml2 version there is somewhere between 8eb55d782a2b and beb7281055db.
if the maintainer doesn't act before that it will be fixed with my libxml2 NMU:
OK, cool, thanks for the info.
html files being generated. And I tried this on both stable and unstable debian installations.
Would you mind also trying one patch I proposed upstream? I know it will seem a bit far-fetched, desperate times call for desperate measures, I guess. I'm talking about this one:
https://www.redhat.com/archives/libvir-list/2015-September/msg00194.html
Nope, same error. I'll try to have another look later this week.
Thanks, I think we really need to revert your fix in the meantime, though, as the internal documentation pages are broken:
Sure. Ack from me to revert this. Cheers, -- Guido
participants (3)
-
Guido Günther -
Martin Kletzander -
Michal Privoznik