Commit 0b175024f77df4e3eb9588307b47affc735882d2
1 parent
eb902dfd
Create bugfix release 1.7.1
Showing
5 changed files
with
37 additions
and
200 deletions
ChangeLog
NEWS
| 1 | -New Features and Important Changes in GammaLib 1.7.0 | |
| 1 | +New Features and Important Changes in GammaLib 1.7.1 | |
| 2 | 2 | |
| 3 | -2 July 2020 | |
| 3 | +20 July 2020 | |
| 4 | 4 | |
| 5 | 5 | |
| 6 | 6 | 1. Introduction |
| ... | ... | @@ -12,80 +12,25 @@ made since the last release of GammaLib. |
| 12 | 12 | 2. Public interface modifications |
| 13 | 13 | --------------------------------- |
| 14 | 14 | The following classes have been added: |
| 15 | -- GResponseCache | |
| 16 | -- GModelSpatialRadialRing | |
| 17 | -- GModelSpectralTable | |
| 18 | -- GSPIEventBin | |
| 19 | -- GSPIEventCube | |
| 20 | -- GSPIInstDir | |
| 21 | -- GSPIModelDataSpace | |
| 22 | -- GSPIObservation | |
| 23 | -- GSPIResponse | |
| 15 | +- none | |
| 24 | 16 | |
| 25 | 17 | The following classes have been removed: |
| 26 | -- GCTAResponseCache | |
| 18 | +- none | |
| 27 | 19 | |
| 28 | 20 | The following classes have been renamed: |
| 29 | 21 | - none |
| 30 | 22 | |
| 31 | 23 | The following methods have been added: |
| 32 | -- GTime.datetime() - Datetime conversion in Python interface | |
| 33 | -- GTime.julian_epoch() - Returns Julian epoch for time | |
| 34 | -- GGti::overlap(GTime&, GTime&) - Compute overlap with time interval | |
| 35 | -- GInstDir::hash() - Returns a cached value | |
| 36 | -- GObservations::npred(string&, string&) - Compute predicted number of events | |
| 37 | -- GObservations::npred(GModel&, string&) - Compute predicted number of events | |
| 38 | -- GObservations::remove_response_cache() - Remove response cache for all models | |
| 39 | -- GObservations::remove_response_cache(string&) - Remove response cache for model | |
| 40 | -- GObservation::remove_response_cache(string&) - Remove response cache for model | |
| 41 | -- GSkyDir::precess() - Precess sky position to epoch | |
| 42 | -- GSkyDir::sun() - Set sky position of Sun | |
| 43 | -- GSkyDir::moon() - Set sky position of Moon | |
| 44 | -- GSkyDir::rotate() - Rotate sky direction (radians variant) | |
| 45 | -- GSkyDir::GSkyDir(GVector&) - Sky direction vector constructor | |
| 46 | -- GSkyMap::extract() - Extract section of sky map | |
| 47 | -- GSkyMap::region_circle() - Get bounding region circle of sky map | |
| 48 | -- GSkyRegionCircle::operator==() - Equality operator | |
| 49 | -- GSkyRegionCircle::operator!=() - Non-equality operator | |
| 50 | -- GEbounds::set() - Set energy boundaries for different methods | |
| 51 | -- GEnergies::set() - Set energies for different methods | |
| 52 | -- GModelPar::has_factor_min() - Test if parameter has factor minimum | |
| 53 | -- GModelPar::has_factor_max() - Test if parameter has factor maximum | |
| 54 | -- GModelPar::remove_factor_min - Remove factor minimum | |
| 55 | -- GModelPar::remove_factor_max - Remove factor maximum | |
| 56 | -- GModelSpatial::flux() - Extract flux from sky region | |
| 57 | -- GModelSpatialDiffuseCube::mc_cone() - Return MC simulation cone | |
| 58 | -- GModelSpatialDiffuseMap::mc_cone() - Return MC simulation cone | |
| 59 | -- GCTAObservation::remove_response_cache(string&) - Remove response cache for model | |
| 60 | -- GCTAResponse::remove_response_cache(string&) - Remove response cache for model | |
| 61 | -- GCTAResponseCache::remove(string&) - Remove response cache for model | |
| 62 | -- GCOMObservation::drm(GModels&) - Compute model cube | |
| 63 | -- GCOMObservation::remove_response_cache(string&) - Remove response cache for model | |
| 64 | -- GCOMTim::save(GFilename&, bool&) - Save TIM into FITS file | |
| 65 | -- GCOMTim::write(GFitsBinTable&) - Write TIM into FITS binary table | |
| 66 | -- GCOMDri::cone_content() - Get DRI content within event cone | |
| 24 | +- none | |
| 67 | 25 | |
| 68 | 26 | The following methods have been removed: |
| 69 | -- GEbounds::set_lin() | |
| 70 | -- GEbounds::set_log() | |
| 71 | -- GEnergies::set_lin() - method is now private | |
| 72 | -- GEnergies::set_log() - method is now private | |
| 73 | -- GModelSpatialDiffuseCube::maps() | |
| 74 | -- GModelSpatialDiffuseCube::pixels() | |
| 27 | +- none | |
| 75 | 28 | |
| 76 | 29 | The following methods have been renamed: |
| 77 | -- GModelSpatialDiffuseCube::set_mc_cone() to GModelSpatialDiffuseCube::mc_cone() | |
| 78 | -- GModelSpatialDiffuseMap::set_mc_cone() to GModelSpatialDiffuseMap::mc_cone() | |
| 79 | -- GResponse::irf(GEvent&, GSource&, GObservation&) to GResponse::irf_spatial() | |
| 80 | -- GLATResponse::irf(GLATEventAtom&, GSource&, GObservation&) to GLATResponse::irf_spatial_atom() | |
| 81 | -- GLATResponse::irf(GLATEventBin&, GSource&, GObservation&) to GLATResponse::irf_spatial_bin() | |
| 30 | +- none | |
| 82 | 31 | |
| 83 | 32 | The arguments for the following methods have been changed: |
| 84 | -- GTime::utc(int&) - Add optional precision value | |
| 85 | -- GSkyDir::posang(GSkyDir&, std::string&) - Add optional coordinate system string | |
| 86 | -- GSkyDir::posang_dir(GSkyDir&, std::string&) - Add optional coordinate system string | |
| 87 | -- GEbounds::GEbounds(int&, GEnergy&, GEnergy&, std::string&, double&) - Replace bool by string and double | |
| 88 | -- GEnergies::GEnergies(int&, GEnergy&, GEnergy&, std::string&, double&) - Replace bool by string and double | |
| 33 | +- none | |
| 89 | 34 | |
| 90 | 35 | The return value of the following methods has been changed: |
| 91 | 36 | - none |
| ... | ... | @@ -93,12 +38,7 @@ The return value of the following methods has been changed: |
| 93 | 38 | |
| 94 | 39 | 3. Configuration |
| 95 | 40 | ----------------- |
| 96 | -The order of the Python module imports was changed since there were some issues | |
| 97 | -with type casting with the present order. This is some side effect of how SWIG | |
| 98 | -handles type casting among multiple modules. | |
| 99 | - | |
| 100 | -The example binaries as well as the cpp source code file and Python scripts are | |
| 101 | -now installed in the share/examples folder (#2971). | |
| 41 | +None | |
| 102 | 42 | |
| 103 | 43 | |
| 104 | 44 | 4. Application module |
| ... | ... | @@ -123,28 +63,7 @@ None |
| 123 | 63 | |
| 124 | 64 | 8. Model module |
| 125 | 65 | --------------- |
| 126 | -The GModelSpatialRadialRing class was added that implements a radial ring model | |
| 127 | -with uniform source intensity (#3253). | |
| 128 | - | |
| 129 | -The set_mc_cone() methods of the GModelSpatialDiffuseCube and GModelSpatialDiffuseMap | |
| 130 | -class were renamed to mc_cone() and mc_cone() getter methods were implemented | |
| 131 | -for both classes. This solves a problem with pickling both classes (#3179). | |
| 132 | - | |
| 133 | -Set the GModelSpatialDiffuseCube::m_region member using the m_cube::region_circle() | |
| 134 | -method and implemented the GModelSpatialDiffuseCube::contains() method (#3145). | |
| 135 | - | |
| 136 | -The method GModelSpatial::flux() was added that extracts the flux by integrating | |
| 137 | -over a specific sky region. For the moment only circular sky regions are | |
| 138 | -implemented (#2961). | |
| 139 | - | |
| 140 | -The GModelSpatialDiffuseCube::maps() and GModelSpatialDiffuseCube::pixels() | |
| 141 | -methods were removed since the same information can be accessed through the | |
| 142 | -GModelSpatialDiffuseCube::cube().nmaps() and cube().npix() methods. Also, the | |
| 143 | -GModelSpatialDiffuseCube::cube() method now fetches the cubes so that the | |
| 144 | -information is always reliable (#1761). | |
| 145 | - | |
| 146 | -The GModelSpectralTable class was added that implements a spectral table model. | |
| 147 | -This allows fitting of arbitrary spectral models (#599). | |
| 66 | +None | |
| 148 | 67 | |
| 149 | 68 | |
| 150 | 69 | 9. Numerics module |
| ... | ... | @@ -154,95 +73,17 @@ None |
| 154 | 73 | |
| 155 | 74 | 10. Observation module |
| 156 | 75 | ---------------------- |
| 157 | -The GResponseCache class was added as a copy of GCTAResponseCache to implement | |
| 158 | -a generic response cache for response computation. The GResponseCache class now | |
| 159 | -implements a three-level response cache, with the first level being the name, | |
| 160 | -the second being a hash value of the true and measured energies, and the third | |
| 161 | -level a hash value of the instrument direction. The hash values are implemented | |
| 162 | -as uint64_t 64 Bit unsigned integers. The former CTA specific response cache was | |
| 163 | -removed (#3202). | |
| 164 | - | |
| 165 | -An GInstDir::hash() method was added that returns a unique 64 Bit unsigned | |
| 166 | -integer value for every class instance that can be used for response caching | |
| 167 | -(#3202). | |
| 168 | - | |
| 169 | -Add GObservations::npred() methods that compute the number of predicted events | |
| 170 | -for a model with a given name or for a given reference to a model. Optionally | |
| 171 | -an instrument string can be passed so that the number of predicted events are | |
| 172 | -only computed for that instrument (#3062). | |
| 173 | - | |
| 174 | -Implement remove_response_cache() methods for GObservations and GObservation | |
| 175 | -classes that serve as a hook to remove all models or one model from a response | |
| 176 | -cache. The actual cache deallocation needs to be implemented at the level of the | |
| 177 | -derived classes (#2973). | |
| 178 | - | |
| 179 | -Add GGti::overlap() method to compute the temporal overlap in seconds with | |
| 180 | -a specified time interval. | |
| 181 | - | |
| 182 | -Avoid rounding of seconds in case that digits after the comma are requested in | |
| 183 | -GTime::utc() method (#2917). | |
| 184 | - | |
| 185 | -Add GTime.datetime() method to the Python interface to convert times from and | |
| 186 | -to datetime.datetime objects. Add an optional precision argument to the GTime.utc() | |
| 187 | -to set the number of digits after the comma for the seconds (#2909). | |
| 188 | - | |
| 189 | -The interface of the GEbounds and GEnergies constructors that sets a given number | |
| 190 | -of energy boundaries or values was changed. The bool argument that toggled between | |
| 191 | -linear and logarithmic spacing was replaced by a string that can take the values | |
| 192 | -"LIN", "LOG" or "POW". For the "POW" method an additional optional argument was added | |
| 193 | -to specify the power-law index. The GEbounds::set() and GEnergies::set() methods | |
| 194 | -were added with the same method arguments, and the former set_lin() and set_log() | |
| 195 | -methods were removed from the public interface (#2881). | |
| 196 | - | |
| 197 | -Add GTime.julian_epoch() methods that return the Julian epoch for a given time | |
| 198 | -(#1854). | |
| 76 | +None | |
| 199 | 77 | |
| 200 | 78 | |
| 201 | 79 | 11. Optimization module |
| 202 | 80 | ----------------------- |
| 203 | -Fixed a bug in the handling of parameter boundaries for negative scale factors in | |
| 204 | -GOptimizerPar. Before the minimum and maximum were inverted, now they are kept and | |
| 205 | -correctly applied (#3072). | |
| 81 | +None | |
| 206 | 82 | |
| 207 | 83 | |
| 208 | 84 | 12. Sky module |
| 209 | 85 | -------------- |
| 210 | -Added equality and non-equality operators to the GSkyRegionCircle class. | |
| 211 | - | |
| 212 | -The position angle is now computed by GSkyDir::posang() and GSkyDir::posang_deg() | |
| 213 | -for the coordinate system that is specified by the coordsys argument. By default | |
| 214 | -this argument is set to "CEL", hence by default the methods return the position | |
| 215 | -angle with respect to celestial coordinates (#3199). | |
| 216 | - | |
| 217 | -Added GSkyMap::region_circle() method to compute a bounding region circle that | |
| 218 | -encloses all sky map pixels (#3145). | |
| 219 | - | |
| 220 | -The GSkyMap operators used the sky map value access operator that bi-linearly | |
| 221 | -interpolates sky map values even for maps with identical definition. This could | |
| 222 | -lead to rounding errors. For maps with identical definition the operators now | |
| 223 | -directly act on the pixel values, avoid thus the bi-linear interpolation. This | |
| 224 | -has also the side effect to speed up the computation for maps with identical | |
| 225 | -definition (#3126). | |
| 226 | - | |
| 227 | -Exclude the upper sky map boundary in GSkyMap::contains() method. Originally the | |
| 228 | -boundary was excluded, but this was changed in commit 0ecb6a07 when the | |
| 229 | -GSkyMap::flux() method was added since the flux method evaluates intensities at | |
| 230 | -the upper pixel boundaries. Now the GSkyMap::flux() method evaluates intensities | |
| 231 | -just below the upper pixel boundaries (#3012). | |
| 232 | - | |
| 233 | -Two GSkyMap::extract() methods were added. The first method extracts a patch of | |
| 234 | -pixels, defined by the start and stop pixel indices in X and Y direction. The second | |
| 235 | -method extracts all patch of pixels that is defined by inclusion sky regions (#2990). | |
| 236 | - | |
| 237 | -A GSkyDir::moon() method was added that sets the sky direction to the position of | |
| 238 | -the Moon for a given date and a specific epoch (#1855). | |
| 239 | - | |
| 240 | -A GSkyDir::sun() method was added that sets the sky direction to the position of | |
| 241 | -the Sun for a given date and a specific epoch. To support the method, a | |
| 242 | -GSkyDir::precess() method was added that allows precessing a sky direction to | |
| 243 | -a specific epoch (#1854). | |
| 244 | - | |
| 245 | -The GSkyProjection::coordsys() method now returns CEL instead of EQU (#1829). | |
| 86 | +None | |
| 246 | 87 | |
| 247 | 88 | |
| 248 | 89 | 13. Support module |
| ... | ... | @@ -282,38 +123,14 @@ None |
| 282 | 123 | |
| 283 | 124 | 20. CTA interface |
| 284 | 125 | ----------------- |
| 285 | -Move the GCTAResponseCache class from the CTA module to the observation module | |
| 286 | -so that response caching becomes available for all instruments (#3202). | |
| 287 | - | |
| 288 | -Remove obsolete ndirs(), nerecos() and netrues() methods for GCTAResponseCache | |
| 289 | -class (#2893). | |
| 290 | - | |
| 291 | -Implement remove_response_cache() methods for GCTAObservation and | |
| 292 | -GCTAReponse, allowing de-allocation of models in the response cache (#2973). | |
| 293 | - | |
| 294 | -The GCTAModelAeffBackground, GCTAModelIrfBackground and GCTAModelBackground | |
| 295 | -classes were adapted so that the npred() method now also works for RoI centres | |
| 296 | -that differ from the pointing direction. This now allows using the corresponding | |
| 297 | -models in an unbinned analysis where the RoI centres differ from the pointing | |
| 298 | -directions (#2695). | |
| 126 | +None | |
| 299 | 127 | |
| 300 | 128 | |
| 301 | 129 | 21. COMPTEL interface |
| 302 | 130 | --------------------- |
| 303 | -Add the method GCOMObservation::drm(GModels&) that computes a DRM model cube | |
| 304 | -using the models that are in a model container (#2973). | |
| 305 | - | |
| 306 | -Implement GCOMObservation::remove_response_cache() allowing de-allocation of | |
| 307 | -models in the response cache (#2973). | |
| 308 | - | |
| 309 | -Add GCOMTim::write() and GCOMTim::save() methods for writing and saving TIM | |
| 310 | -datasets (#2973). | |
| 311 | - | |
| 312 | -Correctly write model attributes in GCOMModelDRBFitting class. | |
| 131 | +None | |
| 313 | 132 | |
| 314 | 133 | |
| 315 | 134 | 22. INTEGRAL/SPI interface |
| 316 | 135 | -------------------------- |
| 317 | -Add instrument module for INTEGRAL/SPI to analyse and specifically fit data from | |
| 318 | -the SPI telescope aboard ESA's INTEGRAL mission (#733, #734, #735, #736, #740, | |
| 319 | -#742, #745, #746). | |
| 136 | +None | ... | ... |
README.md
doc/source/admin/release_history/1.7.1.rst
0 → 100644