fcc-edm master not compatible with podio master (operator<< conflict)

[cbernet]

Hello, I've been trying to compile the master version of podio (664b301775979) with the latest version of fcc-edm (22b5abc) on ubuntu 16.04 with gcc version 5.4.0.

Podio compiles and works fine, but the fcc-edm compilation fails with the error messages shown below (*).

The problem is that the ostream operators that are now added by podio to the components are conflicting with the ones in https://github.com/HEP-FCC/fcc-edm/blob/master/utilities/ParticleUtils.h.
And indeed, if I remove the operators from ParticleUtils, fcc-edm compiles and works.

That sounds like a valid complaint from the compiler, and I don't understand how it could work at all at some point. Would anybody have a suggestion to make it work? I'd like to avoid removing the operators in ParticleUtils.

Cheers!

Colin

(*)

/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc: In function ‘std::ostream& operator<<(std::ostream&, const fcc::Particle&)’:
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:105:29: error: call of overloaded ‘operator<<(std::ostream&, const fcc::BareParticle&)’ is ambiguous
   operator<<(out, ptc.core());
                             ^
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:91:15: note: candidate: std::ostream& operator<<(std::ostream&, const fcc::BareParticle&)
 std::ostream& operator<<(std::ostream& out, const fcc::BareParticle& ptc) {
               ^
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:103:15: note: candidate: std::ostream& operator<<(std::ostream&, const fcc::Particle&)
 std::ostream& operator<<(std::ostream& out, const fcc::Particle& ptc) {
               ^
In file included from /home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:2:0:
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.h:50:15: note: candidate: std::ostream& operator<<(std::ostream&, const fcc::MCParticle&)
 std::ostream& operator<<(std::ostream& out, const fcc::MCParticle& ptc);
               ^
In file included from /home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:10:0:
/home/cbernet/FCC/fcc-edm/datamodel/datamodel/MCParticle.h:136:15: note: candidate: std::ostream& fcc::operator<<(std::ostream&, const fcc::ConstMCParticle&)
 std::ostream& operator<<( std::ostream& o,const ConstMCParticle& value );
               ^
In file included from /home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:9:0:
/home/cbernet/FCC/fcc-edm/datamodel/datamodel/Particle.h:137:15: note: candidate: std::ostream& fcc::operator<<(std::ostream&, const fcc::ConstParticle&)
 std::ostream& operator<<( std::ostream& o,const ConstParticle& value );
               ^
In file included from /home/cbernet/FCC/fcc-edm/datamodel/datamodel/Particle.h:3:0,
                 from /home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:9:
/home/cbernet/FCC/fcc-edm/datamodel/datamodel/BareParticle.h:22:22: note: candidate: std::ostream& fcc::operator<<(std::ostream&, const fcc::BareParticle&)
 inline std::ostream& operator<<( std::ostream& o,const fcc::BareParticle& value ){ 
                      ^
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc: In function ‘std::ostream& operator<<(std::ostream&, const fcc::MCParticle&)’:
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:112:29: error: call of overloaded ‘operator<<(std::ostream&, const fcc::BareParticle&)’ is ambiguous
   operator<<(out, ptc.core());
                             ^
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:91:15: note: candidate: std::ostream& operator<<(std::ostream&, const fcc::BareParticle&)
 std::ostream& operator<<(std::ostream& out, const fcc::BareParticle& ptc) {
               ^
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:103:15: note: candidate: std::ostream& operator<<(std::ostream&, const fcc::Particle&)
 std::ostream& operator<<(std::ostream& out, const fcc::Particle& ptc) {
               ^
/home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:109:15: note: candidate: std::ostream& operator<<(std::ostream&, const fcc::MCParticle&)
 std::ostream& operator<<(std::ostream& out, const fcc::MCParticle& ptc) {
               ^
In file included from /home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:10:0:
/home/cbernet/FCC/fcc-edm/datamodel/datamodel/MCParticle.h:136:15: note: candidate: std::ostream& fcc::operator<<(std::ostream&, const fcc::ConstMCParticle&)
 std::ostream& operator<<( std::ostream& o,const ConstMCParticle& value );
               ^
In file included from /home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:9:0:
/home/cbernet/FCC/fcc-edm/datamodel/datamodel/Particle.h:137:15: note: candidate: std::ostream& fcc::operator<<(std::ostream&, const fcc::ConstParticle&)
 std::ostream& operator<<( std::ostream& o,const ConstParticle& value );
               ^
In file included from /home/cbernet/FCC/fcc-edm/datamodel/datamodel/Particle.h:3:0,
                 from /home/cbernet/FCC/fcc-edm/utilities/ParticleUtils.cc:9:
/home/cbernet/FCC/fcc-edm/datamodel/datamodel/BareParticle.h:22:22: note: candidate: std::ostream& fcc::operator<<(std::ostream&, const fcc::BareParticle&)
 inline std::ostream& operator<<( std::ostream& o,const fcc::BareParticle& value ){ 
                      ^
utilities/CMakeFiles/utilities.dir/build.make:134: recipe for target 'utilities/CMakeFiles/utilities.dir/ParticleUtils.cc.o' failed
make[2]: *** [utilities/CMakeFiles/utilities.dir/ParticleUtils.cc.o] Error 1
make[2]: *** Waiting for unfinished jobs....
CMakeFiles/Makefile2:1115: recipe for target 'utilities/CMakeFiles/utilities.dir/all' failed
make[1]: *** [utilities/CMakeFiles/utilities.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
[ 86%] Built target datamodel-dictgen

[vvolkl]

Hello Colin!

the operators in podio already seem to give decent output to me. Maybe we can improve them so the ones in ParticleUtils can be removed after all?
Cheers,
Valentin

[cbernet]

Hi Valentin! The user wants physic specific data, and this data might depend on the collider. For example for a particle in h h collision, we want to print pt eta, while in ee collision we want E theta. Generic podio operators will not do the job. The podio operators are only good for debugging software, not for physics. Cheers, Colin

…

Le 10 janv. 2018 à 14:13, vvolkl ***@***.***> a écrit : Hello Colin! the operators in podio already seem to give decent output to me. Maybe we can improve them so the ones in ParticleUtils can be removed after all? Cheers, Valentin — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#52 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AD8ku0Z1P0xiPBuHrUzhndX7OldMkNajks5tJLdlgaJpZM4RIfz4>.

[vvolkl]

Hi Colin,

I see what you mean! and probably you need the operator << to have convenient debugging in python using print? But strictly speaking, we have the same problem with the code that is currently in ParticleUtils, no? It should be usuable by ee/hh/eh and there is no way to make it specific on the collider. The simplest solution I can think of is to have a print function in a ee / hh namespace ( without the convenience of <<, of course :/)

[cbernet]

We could make the operator in ParticleUtils generic enough by making it print pt, e, eta, theta, phi, … I think that would be a good way forward, but I don’t know how the operator clash can be solved. Le 10 janv. 2018 à 15:18, vvolkl <[email protected]> a écrit :

…

Hi Colin, I see what you mean! and probably you need the operator << to have convenient debugging in python using print? But strictly speaking, we have the same problem with the code that is currently in ParticleUtils, no? It should be usuable by ee/hh/eh and there is no way to make it specific on the collider. The simplest solution I can think of is to have a print function in a ee / hh namespace ( without the convenience of <<, of course :/) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#52 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AD8ku9I-hQzihfLnzCNb07BV_8JAx9viks5tJMalgaJpZM4RIfz4>.