Dear Alex,
>On Tue, Nov 3, 2009 at 1:20 AM, Axel Naumann <Axel.Naumann_at_cern.ch> wrote:
> another "before we look at your example"-question is: do you build a dictionary for the slot class? And if so, is > that a genreflex dictionary or a rootcint one? That would help us understand where the issue could come from.
I used genreflex to generate the dictionary, and I should have built the dictionary for the slot class. The following is part of requirements file which will generate the dictionary:
apply_pattern install_more_includes more="EventDisplayBase"
apply_pattern linker_library library=EventDisplayBase
apply_pattern ld_library_path
apply_pattern reflex_dictionary dictionary=EventDisplayBase \
selectionfile=${EventDisplayBase_root}/dict/dict.xml \
headerfiles=${EventDisplayBase_root}/dict/dict.h
macro EventDisplayBaseDict_use_linkopts $(use_linkopts)
You can untar the tarball to check the fmwk/EventDisplayBase/dict/dict.{h,xml} for the dictionary classes.
Thanks,
Xinchun
Cheers, Axel.
Bertrand Bellenot wrote on 11/02/2009 06:13 PM:
Hi,
I don't have LCGCMT nor Xerces-C, but I'll try to take a look as soon as
possible, and I'll let you know.
In the meanwhile, you can check that the slots really exist by looking
in the code (they should be public methods)
Cheers, Bertrand.
-----Original Message-----
From: owner-roottalk_at_root.cern.ch [mailto:owner-roottalk_at_root.cern.ch]
On Behalf Of Xinchun Tian
Sent: Monday, November 02, 2009 6:03 PM
To: roottalk_at_root.cern.ch
Cc: tianxc_at_fnal.gov
Subject: [ROOT] signal/slot communication problems
Dear ROOTers,
I am in the process of converting SRT managed FMWK project
(http://enrico1.physics.indiana.edu/fmwk/wiki/index.php?n=FMWK.FMWK?from
=Main.HomePage) to a CMT project. I used LCGCMT as the glue package and
the policy is basically from GaudiPolicy with slightly modification. The
conversion is almost done, but get problems on EventDisplayBase package
which is using signal/slot communication mechanism.
If you happen to using LCGCMT, and have ROOT/Xerces-C installed, it is
quite straightforward to build EventDisplayBase and reproduce the error.
The tarball is attached.
shell> cd fmwk/EventDisplayBase/cmt
shell> cmt config
shell> source setup.sh
shell> cmt br cmt config
shell> cmt br make
shell> test_evdb.exe
include path:
-I/Users/tianxc/nova/nova-trunk/../external/ROOT/5.22.00c/i386-darwin-gc
c40-dbg/root/include
include path:
-I/Users/tianxc/nova/nova-trunk/../external/ROOT/5.22.00c/i386-darwin-gc
c40-dbg/root/include
Error in <TQObject::CheckConnectArgs>: slot HandleMenu(int) does not
exist Error in <TQObject::CheckConnectArgs>: slot HandleMenu(int) does
not exist Error in <TQObject::CheckConnectArgs>: slot HandleMenu(int)
does not exist Error in <TQObject::CheckConnectArgs>: slot
EditConfig(int) does not exist Error in <TQObject::CheckConnectArgs>:
slot HandleMenu(int) does not exist Error in
<TQObject::CheckConnectArgs>: slot HandleMenu(int) does not exist Error
in <TQObject::CheckConnectArgs>: slot PrevEvt() does not exist Error in
<TQObject::CheckConnectArgs>: slot NextEvt() does not exist Error in
<TQObject::CheckConnectArgs>: slot ReloadEvt() does not exist Error in
<TQObject::CheckConnectArgs>: slot FileList() does not exist Error in
<TQObject::CheckConnectArgs>: slot Print() does not exist Error in
<TQObject::CheckConnectArgs>: slot GoTo() does not exist Error in
<TQObject::CheckConnectArgs>: slot GoTo() does not exist Error in
<TQObject::CheckConnectArgs>: slot GoTo() does not exist Error in
<TQObject::CheckConnectArgs>: for signal/slot consistency checking need
to specify class name as argument to RQ_OBJECT macro Error in
<TQObject::CheckConnectArgs>: for signal/slot consistency checking need
to specify class name as argument to RQ_OBJECT macro Error in
<TQObject::CheckConnectArgs>: slot CloseWindow() does not exist
test_evdb.exe [0]
The SRT version of EventDisplayBase works well. The ROOT version is
5.22.00c, the systems are Mac OS 10.5.8 and Sci. Linux Fermi 4.4.
Any help? Thanks in advance.
Regards,
Xinchun
Received on Mon Nov 02 2009 - 18:35:54 CET
This archive was generated by hypermail 2.2.0 : Mon Nov 02 2009 - 23:50:02 CET