#37643 closed update (fixed)
root @5.34.03_1: update to 5.34.04
Reported by: | cjones051073 (Chris Jones) | Owned by: | macports-tickets@… |
---|---|---|---|
Priority: | Normal | Milestone: | |
Component: | ports | Version: | |
Keywords: | haspatch maintainer | Cc: | mattiafrancescomoro@…, g5pw (Aljaž Srebrnič) |
Port: | root |
Description
- Update ROOT to the latest 5.34.04 release
- Add --disable-cocoa to the default configure argument list, to make sure it is disabled unless the cocoa variant is explicitly enabled.
- Minor comment change...
Attachments (3)
Change History (25)
Changed 12 years ago by cjones051073 (Chris Jones)
comment:1 Changed 12 years ago by larryv (Lawrence Velázquez)
Cc: | mattiafrancescomoro@… added |
---|---|
Summary: | Update ROOT to 5.34.04 → root @5.34.03_1: update to 5.34.04 |
Version: | 2.1.2 |
comment:2 Changed 12 years ago by g5pw (Aljaž Srebrnič)
comment:3 follow-up: 5 Changed 12 years ago by cjones051073 (Chris Jones)
your log file appears to be corrupted. Please do a clean rebuild ('sudo port clean root' first) and if it still files post the log file again.
comment:4 follow-up: 6 Changed 12 years ago by cjones051073 (Chris Jones)
also, please isolate the minimum set of variants that cause the failure. Do not just build with them all ...
comment:5 follow-up: 8 Changed 12 years ago by g5pw (Aljaž Srebrnič)
Replying to jonesc@…:
your log file appears to be corrupted. Please do a clean rebuild ('sudo port clean root' first) and if it still files post the log file again.
It is a clean logfile, I'm pretty sure i cleaned beforehand, and it looks like all phases are there, why do you say it's corrupted?
comment:6 Changed 12 years ago by g5pw (Aljaž Srebrnič)
Replying to jonesc@…:
also, please isolate the minimum set of variants that cause the failure. Do not just build with them all ...
I'm trying to isolate the problematic variant now. On a macbook, it does take some time... :)
comment:8 Changed 12 years ago by cjones051073 (Chris Jones)
Replying to g5pw@…:
Replying to jonesc@…:
your log file appears to be corrupted. Please do a clean rebuild ('sudo port clean root' first) and if it still files post the log file again.
It is a clean logfile, I'm pretty sure i cleaned beforehand, and it looks like all phases are there, why do you say it's corrupted?
no it is not. The last part I can read is
:info:build bin/rmkdepend -R -fcint/reflex/src/Base.d -Y -w 1000 -- -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONF
everything after that is unreadable. Or is it just me ... ?
Chris
comment:9 Changed 12 years ago by cjones051073 (Chris Jones)
just to clarify, when I say your log is not clean, I mean I simply cannot read it at all. It contains unreadable characters. not that you did not perform a clean port build.
Chris
comment:10 follow-up: 12 Changed 12 years ago by cjones051073 (Chris Jones)
just to clarify more. If I just download your log and cat it, I get
<snip> :info:build clang++ -O2 -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFIG -pthread -o cint/reflex/src/Any.o -c /opt/local/var/macports/build/_opt_local_var_macports_sources_svn.macports.org_dports_science_root/root/work/root/cint/reflex/src/Any.cxx :info:build bin/rmkdepend -R -fcint/reflex/src/Array.d -Y -w 1000 -- -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFIG -pthread -D__cplusplus -- /opt/local/var/macports/build/_opt_local_var_macports_sources_svn.macports.org_dports_science_root/root/work/root/cint/reflex/src/Array.cxx :info:build bin/rmkdepend -R -fcint/reflex/src/Base.d -Y -w 1000 -- -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONF
that is it. everything after that is unreadable for some reason....
So it tells me nothing as is...
cheers Chris
comment:11 Changed 12 years ago by cjones051073 (Chris Jones)
Also, I have tried numerous combinations of variants, wth OSX 10.8 and Xcode 4.5.2 and they all work just fine for me ....
comment:12 follow-up: 13 Changed 12 years ago by g5pw (Aljaž Srebrnič)
Replying to jonesc@…:
just to clarify more. If I just download your log and cat it, I get
<snip> :info:build clang++ -O2 -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFIG -pthread -o cint/reflex/src/Any.o -c /opt/local/var/macports/build/_opt_local_var_macports_sources_svn.macports.org_dports_science_root/root/work/root/cint/reflex/src/Any.cxx :info:build bin/rmkdepend -R -fcint/reflex/src/Array.d -Y -w 1000 -- -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFIG -pthread -D__cplusplus -- /opt/local/var/macports/build/_opt_local_var_macports_sources_svn.macports.org_dports_science_root/root/work/root/cint/reflex/src/Array.cxx :info:build bin/rmkdepend -R -fcint/reflex/src/Base.d -Y -w 1000 -- -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFthat is it. everything after that is unreadable for some reason....
So it tells me nothing as is...
cheers Chris
Huh, that's strange. I'll try to rebuild and re-post the log. However, root with +avahi+clang31+cocoa+fftw3+fitsio+graphviz+gsl+ldap+minuit2+mysql55+opengl+postgresql92+python32+roofit+ruby+soversion+ssl+tmva+xml builds cleanly, so I suspect it's either pythia or qt_mac that causes the error. Will rebuild with just +pythia and post logs soon!
comment:13 Changed 12 years ago by g5pw (Aljaž Srebrnič)
Replying to g5pw@…:
Replying to jonesc@…:
just to clarify more. If I just download your log and cat it, I get
<snip> :info:build clang++ -O2 -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFIG -pthread -o cint/reflex/src/Any.o -c /opt/local/var/macports/build/_opt_local_var_macports_sources_svn.macports.org_dports_science_root/root/work/root/cint/reflex/src/Any.cxx :info:build bin/rmkdepend -R -fcint/reflex/src/Array.d -Y -w 1000 -- -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFIG -pthread -D__cplusplus -- /opt/local/var/macports/build/_opt_local_var_macports_sources_svn.macports.org_dports_science_root/root/work/root/cint/reflex/src/Array.cxx :info:build bin/rmkdepend -R -fcint/reflex/src/Base.d -Y -w 1000 -- -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFthat is it. everything after that is unreadable for some reason....
Huh, that's strange. I'll try to rebuild and re-post the log. However, root with +avahi+clang31+cocoa+fftw3+fitsio+graphviz+gsl+ldap+minuit2+mysql55+opengl+postgresql92+python32+roofit+ruby+soversion+ssl+tmva+xml builds cleanly, so I suspect it's either pythia or qt_mac that causes the error. Will rebuild with just +pythia and post logs soon!
comment:14 follow-up: 15 Changed 12 years ago by cjones051073 (Chris Jones)
The problem with your log might be the encoding, or you have created a binary file not a text file somehow... gedit gives me when I try and open it
gedit has not been able to detect the character coding. Please check that you are not trying to open a binary file. Select a character coding from the menu and try again.
Please make sure it is a proper text file.
Chris
Changed 12 years ago by g5pw (Aljaž Srebrnič)
Failed build with default variants+pythia
comment:15 Changed 12 years ago by g5pw (Aljaž Srebrnič)
Replying to jonesc@…:
The problem with your log might be the encoding, or you have created a binary file not a text file somehow... gedit gives me when I try and open it
gedit has not been able to detect the character coding. Please check that you are not trying to open a binary file. Select a character coding from the menu and try again.Please make sure it is a proper text file.
Beats me, I haven't touched it! Maybe it got corrupted during upload? Anyway, I reuploaded the failed build log, this time with only +pythia variant enabled.
comment:16 follow-up: 17 Changed 12 years ago by cjones051073 (Chris Jones)
did you manage to reproduce the build failure and create a new log ?
comment:17 Changed 12 years ago by g5pw (Aljaž Srebrnič)
Replying to jonesc@…:
did you manage to reproduce the build failure and create a new log ?
Yes, I overwrote the attached main.log I uploaded before (the corrupted one).
comment:18 Changed 12 years ago by cjones051073 (Chris Jones)
Hmmm. default variants + pythia builds fine for me. See the attached reference log.
I wonder if something is fishy in your pythia port installation. Can you please check it, reinstall it from scratch. This is the only thing I can think of looking at your errors. The failures are quite odd....
cheers Chris
Changed 12 years ago by cjones051073 (Chris Jones)
comment:19 Changed 12 years ago by cjones051073 (Chris Jones)
Also please confirm the default variants + pythia build fine with the correct released version, 5.34.03. Please post a build log for that version.
Chris
comment:20 follow-up: 21 Changed 12 years ago by cjones051073 (Chris Jones)
If you still have failures after checking your pythia port, please check what is in your version of
/opt/local/include/Analysis.h
as the first error you get is
:info:build /opt/local/include/Analysis.h:38:22: error: unknown type name 'Event'; did you mean 'event'?
for me, this file starts with
#ifndef Pythia8_Analysis_H #define Pythia8_Analysis_H #include "Basics.h" #include "Event.h" #include "PythiaStdlib.h"
and the compile command is
:info:build clang++ -O2 -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFIG -pthread -I/opt/local/include -o montecarlo/pythia8/src/TPythia8.o -c /opt/local/var/macports/build/_opt_local_var_macports_sources_svn.macports.org_dports_science_root/root/work/root/montecarlo/pythia8/src/TPythia8.cxx
so the include for Event.h should be found via the -I/opt/local/include which for me does include Event.h .... Why your build is failing is a bit odd...
Chris
comment:21 Changed 12 years ago by g5pw (Aljaž Srebrnič)
Resolution: | → fixed |
---|---|
Status: | new → closed |
Replying to jonesc@…:
If you still have failures after checking your pythia port, please check what is in your version of
/opt/local/include/Analysis.hfor me, this file starts with
#ifndef Pythia8_Analysis_H #define Pythia8_Analysis_H #include "Basics.h" #include "Event.h" #include "PythiaStdlib.h"
Yap, same here!
and the compile command is
:info:build clang++ -O2 -m64 -pipe -Wshadow -W -Wall -Woverloaded-virtual -fsigned-char -fno-common -Iinclude -DR__HAVE_CONFIG -pthread -I/opt/local/include -o montecarlo/pythia8/src/TPythia8.o -c /opt/local/var/macports/build/_opt_local_var_macports_sources_svn.macports.org_dports_science_root/root/work/root/montecarlo/pythia8/src/TPythia8.cxxso the include for Event.h should be found via the -I/opt/local/include which for me does include Event.h .... Why your build is failing is a bit odd...
Chris
Hmm.. that's quite suspicious. I have a similar failure with another port I was making, which seemed not to include headers which were in fact #include'd. very strange!
Since this looks like a problem on my machine, I commited your diff to the repo in r101697. I'll investigate some more and eventually reopen the ticket if I find something out.
comment:22 Changed 12 years ago by cjones051073 (Chris Jones)
Great thanks ! I agree this sounds like some oddity with your machine, hope you find it ;)
I'm attaching a failed build log, with all variants selected. Error seems to be in the pythia variant.