ctools

I updated it from my fork, and I have different problems when I switch over to devel instead of 1520-dm-profiles (I dont know why I didn’t mention it earlier). I think I’ve just been messing up the merges and rebases or something, since git seems to think everything’s intact:

$ git remote -v
origin    https://cta-gitlab.irap.omp.eu/nkelhos/gammalib.git (fetch)
origin    https://cta-gitlab.irap.omp.eu/nkelhos/gammalib.git (push)

$ git fetch origin

$ git status
On branch devel
Your branch is up-to-date with 'origin/devel'.
Untracked files:
  (use "git add <file>..." to include in what will be committed)

    bin/
    include/gammalib/
    lib/
    share/

nothing added to commit but untracked files present (use "git add" to track)

$ git branch -a
  1520-dm-profiles
* devel
  remotes/origin/1111-implement-GSkyRegionRing
  remotes/origin/1113-implement-altaz-pointing
  remotes/origin/1121-implement-GSkyRegionMap
  remotes/origin/1327-VERITAS-Support
  remotes/origin/1432-fix_psf_king_zero_params
  remotes/origin/1508-support-config-file-analysis
  remotes/origin/1520-dm-profiles
  remotes/origin/1876-correct-containment-radius
  remotes/origin/HEAD -> origin/devel
  remotes/origin/devel
  remotes/origin/integration
  remotes/origin/master
  remotes/origin/release
  remotes/origin/spi

I’m thinking its not a good practice for me to keep a branch out for this long (months). I’m not sure what to do. Can we merge nkelhos/gammalib/1520-dm-profiles into gammalib/gammalib/1520-dm-profiles, then I can delete my fork nkelhos/gammalib and refork gammalib/gammalib again?

For the moment, I’ve switched to just using gammalib/gammalib, which works without problems, so I think something went wrong with my rebasing/merging of nkelhos/gammalib. I’m not even sure there were any changes in nkelhos/gammalib/1520-dm-profiles, so I may just throw it away and start with a fresh fork of gammalib/gammalib.

Alright, something’s still not working right. I do:

$ git clone https://cta-gitlab.irap.omp.eu/ctools/ctools.git .
$ ./autogen.sh
$ make distclean
$ ./configure --prefix=$PWD LDFLAGS=-L$FITSSYS/lib/  CPPFLAGS=-I$FITSSYS
$ make -j 4
...
ctool.cpp: In member function ‘GEbounds ctool::create_ebounds()’:
ctool.cpp:545:31: error: ‘extname_ebounds’ is not a member of ‘gammalib’
             if (file.contains(gammalib::extname_ebounds)) {
                               ^
ctool.cpp:554:36: error: ‘extname_energies’ is not a member of ‘gammalib’
             else if (file.contains(gammalib::extname_energies)) {
                                    ^
ctool.cpp:556:25: error: ‘class GEbounds’ has no member named ‘set’
                 ebounds.set(GEnergies(ebinfile));
                         ^
ctool.cpp:561:35: error: ‘extname_ebounds’ is not a member of ‘gammalib’
                                   gammalib::extname_ebounds+"\", " 
                                   ^
ctool.cpp:563:35: error: ‘extname_energies’ is not a member of ‘gammalib’
                                   gammalib::extname_energies+"\" found in FITS " 
                                   ^
ctool.cpp:594:25: error: ‘class GEbounds’ has no member named ‘set’
                 ebounds.set(GEnergies(ebinfile));
                         ^
make[3]: *** [ctool.lo] Error 1

But I still run into the same problems. The gammalib I’m using is gammalib/nkelhos/1520-dm-profiles.

The problem is that your GammaLib needs an update. The things that are missing are things that are new in GammaLib, here specifically

gammalib::extname_ebounds

Did those changes get merged to gammalib/gammalib/1520-dm-profiles? I just made a clean fork (deleted my old fork) of 1520-dm-profiles today, and I still have the same missing ebounds error.

No, I have not touched this branch since a long time. You have to rebase the branch on devel.

Ah, alright. its sometimes hard to keep track of what code is where. I’ve rebased gammalib, but ctools is now erroring out about a missing GCTAEventList::has_phase() function.

Making all in ctphase
make[3]: Entering directory `/nv/pc1/nkelleyh3/software/ctools/src/ctphase'
/bin/sh ../../libtool --tag=CXX   --mode=compile g++ -DHAVE_CONFIG_H -I. -I../.. -I../../src/support  -I/nv/hp11/nkelleyh3/data/software/cfitsio -I/nv/hp11/nkelleyh3/data/software/gammalib-nkelhos/include/gammalib  -g -O2 -fopenmp -MT ctphase.lo -MD -MP -MF .deps/ctphase.Tpo -c -o ctphase.lo ctphase.cpp
g++ -DHAVE_CONFIG_H -I. -I../.. -I../../src/support -I/nv/hp11/nkelleyh3/data/software/cfitsio -I/nv/hp11/nkelleyh3/data/software/gammalib-nkelhos/include/gammalib -I/nv/hp11/nkelleyh3/data/software/cfitsio -I/nv/hp11/nkelleyh3/data/software/gammalib-nkelhos/include/gammalib  -g -O2 -fopenmp -MT ctphase-main.o -MD -MP -MF .deps/ctphase-main.Tpo -c -o ctphase-main.o `test -f 'main.cpp' || echo './'`main.cpp
libtool: compile:  g++ -DHAVE_CONFIG_H -I. -I../.. -I../../src/support -I/nv/hp11/nkelleyh3/data/software/cfitsio -I/nv/hp11/nkelleyh3/data/software/gammalib-nkelhos/include/gammalib -g -O2 -fopenmp -MT ctphase.lo -MD -MP -MF .deps/ctphase.Tpo -c ctphase.cpp  -fPIC -DPIC -o .libs/ctphase.o
ctphase.cpp: In member function ‘void ctphase::phase_events(GCTAObservation*)’:
ctphase.cpp:561:13: error: ‘class GCTAEventList’ has no member named ‘has_phase’
     events->has_phase(true);
             ^
make[3]: *** [ctphase.lo] Error 1

Any ideas what pack of code has this has_phase() function?

I merged in the has_phase() method yesterday, so I guess you need again to rebase gammalib on devel.