Commit db29cc793c6d32af9210d8f41bdf3bf864e705ee
1 parent
24ddea60
Update for version 0.6.1.
Summarize bug fix changes.
Showing
2 changed files
with
31 additions
and
68 deletions
ChangeLog
| 1 | +2012-09-15 Juergen Knoedlseder <jurgen.knodlseder@irap.omp.eu> | ||
| 2 | + | ||
| 3 | + * Bug fix version 0.6.1 released | ||
| 4 | + ================================ | ||
| 5 | + | ||
| 6 | + Do not require gammalib/__init__.py file to be present | ||
| 7 | + Do not install swig files | ||
| 8 | + clang++ compiler takes precedence over g++ compiler | ||
| 9 | + | ||
| 1 | 2012-09-14 Juergen Knoedlseder <jurgen.knodlseder@irap.omp.eu> | 10 | 2012-09-14 Juergen Knoedlseder <jurgen.knodlseder@irap.omp.eu> |
| 2 | 11 | ||
| 3 | * Version 0.6.0 released | 12 | * Version 0.6.0 released |
NEWS
| 1 | -New Features and Important Changes in GammaLib 00-06-00 | 1 | +New Features and Important Changes in GammaLib 00-06-01 |
| 2 | 2 | ||
| 3 | Author(s) : Juergen Knoedlseder | 3 | Author(s) : Juergen Knoedlseder |
| 4 | 4 | ||
| 5 | -September, 14th, 2012 | 5 | +September, 15th, 2012 |
| 6 | 6 | ||
| 7 | 7 | ||
| 8 | 1. Introduction | 8 | 1. Introduction |
| @@ -10,27 +10,27 @@ September, 14th, 2012 | @@ -10,27 +10,27 @@ September, 14th, 2012 | ||
| 10 | This document briefly summarizes module by module the most important | 10 | This document briefly summarizes module by module the most important |
| 11 | changes made since the last release of GammaLib. | 11 | changes made since the last release of GammaLib. |
| 12 | 12 | ||
| 13 | -The code quality was enhanced in this release by resolving errors and | ||
| 14 | -warnings issued by the cppcheck code quality check utility. A few real | ||
| 15 | -errors were detected, the remaining modifications concerned mainly the | ||
| 16 | -coding style and the translation of C-style casts into C++ style casts. | ||
| 17 | - | ||
| 18 | 13 | ||
| 19 | 2. Configuration | 14 | 2. Configuration |
| 20 | ----------------- | 15 | ----------------- |
| 21 | -Added OpenMP support to allow for code parallelization. If available, | ||
| 22 | -OpenMP is enabled by default. OpenMP support may be disabled using | ||
| 23 | -the --disable-openmp flag. | 16 | +The configuration script does no longer check for the presence of the |
| 17 | +gammalib/__init__.py file. This file is generated upon installation | ||
| 18 | +of the Python bindings, and is not shipped with the GammaLib library. | ||
| 19 | +In version 00-06-00, a missing gammalib/__init__.py file was | ||
| 20 | +misinterpreted as a need for swig to build the Python bindings. On | ||
| 21 | +systems without swig the Pyhton bindings were thus not built. | ||
| 22 | + | ||
| 23 | +The swig files are no longer installed in the GammaLib directory. | ||
| 24 | +These files are indeed not needed. | ||
| 24 | 25 | ||
| 25 | -Check for the presence of the Python.h header file and the wrapper | ||
| 26 | -files for Python bindings. | 26 | +The clang++ compiler takes now precedence over the g++ compiler. In |
| 27 | +that way, clang++ will be selected by default on Mac OS X systems | ||
| 28 | +(usually from Mac OS X 10.7 on). | ||
| 27 | 29 | ||
| 28 | 30 | ||
| 29 | 3. Application module | 31 | 3. Application module |
| 30 | --------------------- | 32 | --------------------- |
| 31 | -The GLog class was improved. The printed dates reflect now the time | ||
| 32 | -when a new line was started in a log file. The copy operator has also | ||
| 33 | -been implemented. | 33 | +None |
| 34 | 34 | ||
| 35 | 35 | ||
| 36 | 4. FITS I/O Interface | 36 | 4. FITS I/O Interface |
| @@ -40,28 +40,12 @@ None | @@ -40,28 +40,12 @@ None | ||
| 40 | 40 | ||
| 41 | 5. Linear algebra module | 41 | 5. Linear algebra module |
| 42 | ------------------------ | 42 | ------------------------ |
| 43 | -The matrix classes have been refactored using the actual coding | ||
| 44 | -conventions. | ||
| 45 | - | ||
| 46 | -Two bugs were corrected in GSparseMatrix. First, the non-equal | ||
| 47 | -operator != had a bug that left the major part of the matrix untested. | ||
| 48 | -This had hidden a problem with the fill stack from the unit test. | ||
| 49 | -Once the != operator bug was corrected, the unit test revealed that | ||
| 50 | -the misusage of an internal working array corrupted the fill stack | ||
| 51 | -after a flush. This has been corrected by adding an additional | ||
| 52 | -working array, so that conflicts can no longer arise. | 43 | +None |
| 53 | 44 | ||
| 54 | 45 | ||
| 55 | 6. Model module | 46 | 6. Model module |
| 56 | --------------- | 47 | --------------- |
| 57 | -Added GModelSpatialMap model that allow using skymap for the spatial | ||
| 58 | -component of the model. The Monte Carlo method as well as the function | ||
| 59 | -evaluation methods have been implemented. The function evaluation | ||
| 60 | -methods make use of the GSkymap sky direction access operators that | ||
| 61 | -performs a bi-linear interpolation of the pixels values for a given | ||
| 62 | -direction. Note that the total flux in the sky map is normalized | ||
| 63 | -internally to 1 ph/cm2/s. Negative pixels will automatically be set to | ||
| 64 | -zero. | 48 | +None |
| 65 | 49 | ||
| 66 | 50 | ||
| 67 | 7. Numerics module | 51 | 7. Numerics module |
| @@ -71,10 +55,7 @@ None | @@ -71,10 +55,7 @@ None | ||
| 71 | 55 | ||
| 72 | 8. Observation module | 56 | 8. Observation module |
| 73 | --------------------- | 57 | --------------------- |
| 74 | -The optimization of model parameters makes now use of OpenMP (if | ||
| 75 | -available) for parallelization of the likelihood computation. This | ||
| 76 | -allows the increase of the code execution speed in the presence of | ||
| 77 | -multiple processor cores. | 58 | +None |
| 78 | 59 | ||
| 79 | 60 | ||
| 80 | 9. Optimization module | 61 | 9. Optimization module |
| @@ -84,14 +65,7 @@ None | @@ -84,14 +65,7 @@ None | ||
| 84 | 65 | ||
| 85 | 10. Sky module | 66 | 10. Sky module |
| 86 | -------------- | 67 | -------------- |
| 87 | -Added sky direction access operator to GSkymap class. This operator | ||
| 88 | -returns the intensity in the sky map as function of the sky direction, | ||
| 89 | -performing a bi-linear interpolation between the pixel values of the | ||
| 90 | -map. | ||
| 91 | - | ||
| 92 | -For consistency with other classes, GSkyDir arguments are now passed | ||
| 93 | -by reference in the GSkymap access operators. Also map indices have | ||
| 94 | -been declared by reference. | 68 | +None |
| 95 | 69 | ||
| 96 | 70 | ||
| 97 | 11. Support module | 71 | 11. Support module |
| @@ -106,15 +80,7 @@ None | @@ -106,15 +80,7 @@ None | ||
| 106 | 80 | ||
| 107 | 13. Test module | 81 | 13. Test module |
| 108 | --------------- | 82 | --------------- |
| 109 | -A new module has been introduced that supports unit testing of | ||
| 110 | -GammaLib. The module contains so far three classes: | ||
| 111 | -- GTestSuites is a container class for unit test suites | ||
| 112 | -- GTestSuite implements a specific unit test suite | ||
| 113 | -- GTestCase collects information about a specific test case | ||
| 114 | -GTestSuites outputs the result in a JUnit compliant XML file for test | ||
| 115 | -report analysis. The usual console output is also supported. The | ||
| 116 | -GammaLib unit tests have been rewritten to make use of the new | ||
| 117 | -test suite. | 83 | +None |
| 118 | 84 | ||
| 119 | 85 | ||
| 120 | 14. Multi-wavelength interface | 86 | 14. Multi-wavelength interface |
| @@ -124,22 +90,10 @@ None | @@ -124,22 +90,10 @@ None | ||
| 124 | 90 | ||
| 125 | 15. Fermi-LAT interface | 91 | 15. Fermi-LAT interface |
| 126 | ---------------------- | 92 | ---------------------- |
| 127 | -Added support for Pass 7 data. In particular, P7SOURCE_V6 response | ||
| 128 | -functions are now included in the GammaLib distribution, together | ||
| 129 | -with some test data for the Crab. The LAT response code has also | ||
| 130 | -been refactored to support more cleanly various PSF versions. | ||
| 131 | -Efficiency factors, needed by Pass 7, are now also supported. Note | ||
| 132 | -that we still lack support for the Phi dependence of the IRF, and | ||
| 133 | -also energy dispersion has not yet been implemented. | 93 | +None |
| 134 | 94 | ||
| 135 | 95 | ||
| 136 | 16. CTA interface | 96 | 16. CTA interface |
| 137 | ----------------- | 97 | ----------------- |
| 138 | -Added support for diffuse sources to CTA response. FITS sky maps can | ||
| 139 | -now be used to simulated and to fit CTA event data, both in binned and | ||
| 140 | -unbinned mode. Specifically, the GCTAResponse::irf_diffuse and | ||
| 141 | -GCTAResponse::npred_diffuse methods have been implemented to do the | ||
| 142 | -job. Some optimization has been performed to adjust the precision | ||
| 143 | -of the integrations that need to be performed for the diffuse response | ||
| 144 | -computation. | 98 | +None |
| 145 | 99 |