Bug #2776
event time bug from gammalib 1.4.0 to 1.5.2
| Status: | Closed | Start date: | 12/14/2018 | |
|---|---|---|---|---|
| Priority: | Normal | Due date: | ||
| Assigned To: |  Knödlseder Jürgen | % Done: | 0% | |
| Category: | - | |||
| Target version: | - | |||
| Duration: |
Description
Event times (GCTAEventAtom::time()) are bugged in gammalib 1.5.2. They were correct in 1.4.0 .
The code to reproduce this is:
from gammalib import GObservations, GCTAObservation
obs = GObservations()
ob = GCTAObservation()
ob.load('VR50990.fits')
ob.id('VR50990')
obs.append(ob)
for ob in obs :
for ev in ob.events() :
print(ev.time().utc())
On gammalib 1.4.0, when I read in the Veritas datafile VR50990.fits, it prints:
2010-04-14T10:14:47 2010-04-14T10:14:48 2010-04-14T10:14:48 2010-04-14T10:14:50 2010-04-14T10:15:00 2010-04-14T10:15:01 2010-04-14T10:15:04
which are the correct times for those events.
When I take the same file, and run that script on gammalib 1.5.2, the same events are at times:
2003-12-15T22:14:49 2003-12-15T22:14:50 2003-12-15T22:14:50 2003-12-15T22:14:52 2003-12-15T22:15:02 2003-12-15T22:15:03 2003-12-15T22:15:06
which are a few years before VERITAS was built .
Recurrence
No recurrence.
Related issues
History
#1
Updated by Knödlseder Jürgen over 5 years ago
Does the FITS file contain the reference time key words?
#2
Updated by Kelley-Hoskins Nathan over 5 years ago
- File VR81417.chunk3.clean.fits added
The EVENTS table has the time-related header keywords:
MJDREFI 53856 MJDREFF 0.0 TIMEUNIT s TIMESYS TT TIMEREF local
And events have columns like:
OBS_ID EVENT_ID TIME 81417 247243 313216622.2168968 81417 247649 313216623.7063196 81417 248645 313216627.4837822 81417 248931 313216628.5783541 81417 249323 313216630.0244226
I’ve added the fits file these came from, which should have events on 2016-04-03, from 4:21-4:51 am UTC.
#3
Updated by Knödlseder Jürgen over 5 years ago
- Status changed from New to Feedback
- Assigned To set to Knödlseder Jürgen
Running your example file using the current code I get:
>>> import gammalib
>>> obs=gammalib.GCTAObservation('VR81417.chunk3.clean.fits')
>>> print(obs)
=== GCTAObservation ===
Name ......................:
Identifier ................: 81417
Instrument ................: VERITAS
Event file ................: VR81417.chunk3.clean.fits
Event type ................: EventList
Statistic .................: cstat
Ontime ....................: 480 s
Livetime ..................: 439.374101047894 s
Deadtime correction .......: 0.915362710516446
User energy range .........: undefined
=== GCTAPointing ===
Pointing direction ........: (RA,Dec)=(83.6342417597327,21.5144715979806)
Response function .........: undefined
=== GCTAEventList ===
Number of events ..........: 262 (disposed in "VR81417.chunk3.clean.fits")
Time interval .............: 57481.1812731481 - 57481.2021180556 days
Energy interval ...........: 0.02 - 500 TeV
Region of interest ........: RA=83.634242, DEC=21.514472 [0,0] Radius=2.25 deg
=== GSkyRegions ===
Number of regions .........: 0
>>> print(obs.gti())
=== GGti ===
Number of intervals .......: 1
Ontime ....................: 1801.00000101328 sec
Elapsed time ..............: 1801.00000101328 sec
MJD range .................: 57481.1812731481 - 57481.2021180556 s (TT)
UTC range .................: 2016-04-03T04:19:54 - 2016-04-03T04:49:55 s (TT)
Reference MJD .............: 53856
>>> for ev in obs.events():
... print(ev.time().utc())
...
2016-04-03T04:35:54
2016-04-03T04:35:56
2016-04-03T04:35:59
2016-04-03T04:36:00
2016-04-03T04:36:02
2016-04-03T04:36:03
2016-04-03T04:36:03
2016-04-03T04:36:03
2016-04-03T04:36:05
2016-04-03T04:36:05
Everything seems okay here. I get exactly the same result with ctools/gammalib version 1.5.2.
#4
Updated by Knödlseder Jürgen over 5 years ago
Interesting: when running the code as you did I get the following with gammalib 1.5.2:
>>> from gammalib import GObservations, GCTAObservation
>>> obs = GObservations()
>>> ob = GCTAObservation()
>>> ob.load('VR81417.chunk3.clean.fits')
>>> ob.id('VR50990')
>>> obs.append(ob)
<gammalib.cta.GCTAObservation; proxy of <Swig Object of type 'GCTAObservation *' at 0x10b925840> >
>>> for ob in obs :
... for ev in ob.events() :
... print(ev.time().utc())
...
2009-12-04T16:35:56
2009-12-04T16:35:58
2009-12-04T16:36:01
2009-12-04T16:36:02
2009-12-04T16:36:04
Appending a CTA observation to the container seems to change the time reference (after appending it is
51544.5 while before it was 53856):>>> print(obs[0].gti())
=== GGti ===
Number of intervals .......: 1
Ontime ....................: 1801.00000101328 sec
Elapsed time ..............: 1801.00000101328 sec
MJD range .................: 57481.1812731481 - 57481.2021180556 s (TT)
UTC range .................: 2016-04-03T04:19:54 - 2016-04-03T04:49:55 s (TT)
Reference MJD .............: 51544.5
This is however fixed in the
devel version of the code which gives>>> from gammalib import GObservations, GCTAObservation
>>> obs = GObservations()
>>> ob = GCTAObservation()
>>> ob.load('VR81417.chunk3.clean.fits')
>>> ob.id('VR50990')
>>> obs.append(ob)
<gammalib.cta.GCTAObservation; proxy of <Swig Object of type 'GCTAObservation *' at 0x10b925840> >
>>> for ob in obs :
... for ev in ob.events() :
... print(ev.time().utc())
...
2016-04-03T04:35:54
2016-04-03T04:35:56
2016-04-03T04:35:59
2016-04-03T04:36:00
2016-04-03T04:36:02
2016-04-03T04:36:03
2016-04-03T04:36:03
2016-04-03T04:36:03
2016-04-03T04:36:05
2016-04-03T04:36:05
>>> print(obs[0].gti())
=== GGti ===
Number of intervals .......: 1
Ontime ....................: 1801.00000101328 sec
Elapsed time ..............: 1801.00000101328 sec
MJD range .................: 57481.1812731481 - 57481.2021180556 s (TT)
UTC range .................: 2016-04-03T04:19:54 - 2016-04-03T04:49:55 s (TT)
Reference MJD .............: 53856
I guess this relates to #2054. Can you check with the
devel version of the code?#5
Updated by Knödlseder Jürgen over 5 years ago
- Related to Change request #2054: MJD reference date in the output file of the ctbin task added
#6
Updated by Kelley-Hoskins Nathan over 5 years ago
I’m trying to build it, but getting some python-related error:
$ ./configure --prefix=$PWD LDFLAGS=-L/gpfs/pace2/project/pc1/nkelleyh3/software/cfitsio/lib CPPFLAGS=-I/gpfs/pace2/project/pc1/nkelleyh3/software/cfitsio/include/ --without-mwl --without-lat --without-com
$ make -j 4
...
/nv/hp11/nkelleyh3/data/software/miniconda3//bin/python setup.py build_ext
Python extension module building:
Link cfitsio into Python module.
Link readline into Python module.
Include CTA instrument interface in gammalib Python module.
running build_ext
building '_app' extension
creating build
creating build/temp.linux-x86_64-3.6
creating build/temp.linux-x86_64-3.6/gammalib
gcc -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I../include -I/gpfs/pace2/project/pc1/nkelleyh3/softw
are/cfitsio/include/ -I../inst/cta/include -I/nv/hp11/nkelleyh3/data/software/miniconda3/include/python3.6m -c gammalib/app_wrap.
cpp -o build/temp.linux-x86_64-3.6/gammalib/app_wrap.o -fopenmp
cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++
gammalib/app_wrap.cpp: In function ‘swig_module_info* SWIG_Python_GetModule()’:
gammalib/app_wrap.cpp:2452:51: error: ‘PyCObject_Import’ was not declared in this scope
(char*)"type_pointer" SWIG_TYPE_TABLE_NAME);
^
gammalib/app_wrap.cpp: In function ‘void SWIG_Python_SetModule(swig_module_info*)’:
gammalib/app_wrap.cpp:2521:92: error: ‘PyCObject_FromVoidPtr’ was not declared in this scope
PyObject *pointer = PyCObject_FromVoidPtr((void *) swig_module, SWIG_Python_DestroyModule);
^
gammalib/app_wrap.cpp:2512:22: warning: unused variable ‘swig_empty_runtime_method_table’ [-Wunused-variable]
static PyMethodDef swig_empty_runtime_method_table[] = { {NULL, NULL, 0, NULL} };/* Sentinel */
^
gammalib/app_wrap.cpp: In function ‘swig_type_info* SWIG_Python_TypeQuery(const char*)’:
gammalib/app_wrap.cpp:2544:60: error: ‘PyCObject_AsVoidPtr’ was not declared in this scope
descriptor = (swig_type_info *) PyCObject_AsVoidPtr(obj);
^
gammalib/app_wrap.cpp:2549:51: error: ‘PyCObject_FromVoidPtr’ was not declared in this scope
obj = PyCObject_FromVoidPtr(descriptor, NULL);
^
gammalib/app_wrap.cpp: In function ‘PyObject* _wrap_new_GLog__SWIG_1(PyObject*, PyObject*)’:
gammalib/app_wrap.cpp:7460:28: error: ‘PyString_Check’ was not declared in this scope
if (PyString_Check(obj0)) {
^
...
gammalib/app_wrap.cpp:12075:106: error: ‘PyString_AsString’ was not declared in this scope
temp2 = GFilename(std::string(PyString_AsString(PyUnicode_AsEncodedString(obj1, "utf-8", "Error ~"))));
^
gammalib/app_wrap.cpp: In function ‘PyObject* _wrap_GApplicationPars_load__SWIG_1(PyObject*, PyObject*)’:
gammalib/app_wrap.cpp:12142:28: error: ‘PyString_Check’ was not declared in this scope
if (PyString_Check(obj1)) {
^
gammalib/app_wrap.cpp:12143:59: error: ‘PyString_AsString’ was not declared in this scope
temp2 = GFilename(std::string(PyString_AsString(obj1)));
^
gammalib/app_wrap.cpp:12147:106: error: ‘PyString_AsString’ was not declared in this scope
temp2 = GFilename(std::string(PyString_AsString(PyUnicode_AsEncodedString(obj1, "utf-8", "Error ~"))));
^
gammalib/app_wrap.cpp: In function ‘PyObject* _wrap_GApplicationPars_load(PyObject*, PyObject*)’:
gammalib/app_wrap.cpp:12218:35: error: ‘PyString_Check’ was not declared in this scope
if (PyString_Check(argv[1]) ||
^
gammalib/app_wrap.cpp:12242:35: error: ‘PyString_Check’ was not declared in this scope
if (PyString_Check(argv[1]) ||
^
gammalib/app_wrap.cpp: In function ‘PyObject* _wrap_GApplicationPars_save(PyObject*, PyObject*)’:
gammalib/app_wrap.cpp:12290:28: error: ‘PyString_Check’ was not declared in this scope
if (PyString_Check(obj1)) {
^
gammalib/app_wrap.cpp:12291:59: error: ‘PyString_AsString’ was not declared in this scope
temp2 = GFilename(std::string(PyString_AsString(obj1)));
^
gammalib/app_wrap.cpp:12295:106: error: ‘PyString_AsString’ was not declared in this scope
temp2 = GFilename(std::string(PyString_AsString(PyUnicode_AsEncodedString(obj1, "utf-8", "Error ~"))));
^
gammalib/app_wrap.cpp: In function ‘PyObject* _wrap_GApplicationPar_filename__SWIG_0(PyObject*, PyObject*)’:
gammalib/app_wrap.cpp:13579:28: error: ‘PyString_Check’ was not declared in this scope
if (PyString_Check(obj1)) {
^
gammalib/app_wrap.cpp:13580:59: error: ‘PyString_AsString’ was not declared in this scope
temp2 = GFilename(std::string(PyString_AsString(obj1)));
^
gammalib/app_wrap.cpp:13584:106: error: ‘PyString_AsString’ was not declared in this scope
temp2 = GFilename(std::string(PyString_AsString(PyUnicode_AsEncodedString(obj1, "utf-8", "Error ~"))));
^
gammalib/app_wrap.cpp: In function ‘PyObject* _wrap_GApplicationPar_filename(PyObject*, PyObject*)’:
gammalib/app_wrap.cpp:14277:35: error: ‘PyString_Check’ was not declared in this scope
if (PyString_Check(argv[1]) ||
^
error: command 'gcc' failed with exit status 1
make[3]: *** [build] Error 1
make[3]: Leaving directory `/gpfs/pace2/project/pc1/nkelleyh3/software/gammalib-timebug/pyext'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/gpfs/pace2/project/pc1/nkelleyh3/software/gammalib-timebug/pyext'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/gpfs/pace2/project/pc1/nkelleyh3/software/gammalib-timebug'
make: *** [all] Error 2
My system:
$ uname -a Linux iw-h31-20.pace.gatech.edu 2.6.32-573.12.1.el6.x86_64 #1 SMP Mon Nov 23 12:55:32 EST 2015 x86_64 x86_64 x86_64 GNU/Linux
My python is through anaconda:
$ python --version Python 3.6.7 :: Anaconda, Inc.
#7
Updated by Knödlseder Jürgen over 5 years ago
Compiling against Anaconda Python is not a good idea. It may or may not work. Just use the Vanilla Python on your machine.
#8
Updated by Kelley-Hoskins Nathan over 5 years ago
Ok, I rebuilt it with regular python, and printed out the event times using devel, they come out correct now.
#9
Updated by Knödlseder Jürgen about 5 years ago
- Status changed from Feedback to Closed