- Dec 02, 2020
-
-
Stewart Martin-Haugh authored
-
- Dec 01, 2020
-
-
scott snyder authored
-
- Nov 30, 2020
-
-
Nils Erik Krumnack authored
This should address ATLASG-1551, though I didn't perform a complete migration. Doing private sub-sub-tools with the AnaToolHandle in Athena doesn't seem to work correctly anymore. Given that AnaToolHandle is no longer the recommended way of creating sub-tools anymore I'd rather just see if this breaks anything. If it does I'll decide whether to try and fix the AnaToolHandle implementation, or migrate the users to AsgToolConfig instead.
-
Adam Edward Barton authored
-
Frank Winklmeier authored
There is no use-case for deriving from `Monitored::Group`. Remove the virtual `fill` method. This also avoids a warning from clang-tidy about calling a virtual method from the destructor (it would be OK in this case though).
-
- Nov 29, 2020
-
-
Michal Leszczynski authored
-
Michal Leszczynski authored
-
- Nov 27, 2020
-
-
scott snyder authored
Make sure xAOD::EventInfo gets written. Ignore message able missing /Taginfo.
-
Tomasz Bold authored
-
scott snyder authored
Add class ShallowCopyDecorDeps, to help with propagating decoration dependencies across shallow copies.
-
- Nov 26, 2020
-
-
Stewart Martin-Haugh authored
-
Tadej Novak authored
Decorate zero pile-up values in case of no pile-up.
-
Frank Winklmeier authored
-
Frank Winklmeier authored
-
Michal Leszczynski authored
-
Michal Leszczynski authored
-
Vakhtang Tsulaia authored
Partially addresses ATEAM-677
-
When we add a decoration with a decoration handle, we make an alias to the container under the name CONT.DECOR. These aliases are used to express decoration dependencies to the scheduler, and are persistified by AthenaPOOL. The decoration handles assumed that these alias should be there, and in some cases wouldn't work as expected if they were absent. For example, ReadDecorHandle looked up the container via the alias name, while WriteDecorHandle::isAvailable also looked up the container via the alias name. However, the aliases may be missing in some cases. For example, for some EventInfo variables set by the event loop. And aliases may not by available if data are being read by something other than AthenaPOOL. The changes here update the decoration handle classes to allow them to work properly even if the alias is not present. Other than that, the main visible change in behavior is that Read/WriteDecorHandle::key() will now return the key of the container rather than the key of the alias. The alias name may now be retrieved by decorKey(). The behavior of the *DecorHandleKey classes does _not_ change in this regard, and there's probably nothing relying on key() of *DecorHandle. In a bit more detail: ReadDecorHandleKey gets a contHandleKey(), which returns a ReadHandleKey for the container. This is similar to what WriteDecorHandleKey has already been doing, except that the contHandleKey does not get declared to the scheduler. ReadDecorHandle now passes contHandleKey to the base class rather than the ReadDecorHandleKey, and adds decorKey(). We can also clean up isPresent(). WriteDecorHandle no longer has m_contHandle, instead adding decorKey().
-
Vakhtang Tsulaia authored
Partially addresses ATEAM-677
-
- Nov 25, 2020
-
-
Vakhtang Tsulaia authored
Cleanup includes: * Removing legacy scripts * Moving AthenaMP_EventService.py over to AthenaMP/python * Fixing CMakeLists.txt and a few client scripts
-
There should be no need to expose the header of the FileMetaDataTool to other packages, it can be treated as a component and called through its interface. Remove some unneeded headers. Enable checking for thread safety.
-
Attila Krasznahorkay authored
The code was only working correctly with schema evolution when reading standalone objects so far (like xAOD::EventInfo) as it turns out. With the added code it is now also able to deal with std::vector<double> -> std::vector<float> changes on "container types".
-
- Nov 24, 2020
-
-
scott snyder authored
Test automatic evolution of xAOD variables.
-
scott snyder authored
Test automatic schema evolution for xAOD variables.
-
scott snyder authored
Test automatic evolution of xAOD variables.
-
-
- Nov 23, 2020
-
-
scott snyder authored
Ensure that TLS defined in this library actually gets used. Avoids a potential slowdown in accessing TLS seen in simualation. See ATLASSIM-4932.
-
scott snyder authored
Ensure that TLS defined in this library actually gets used. Avoids a potential slowdown in accessing TLS seen in simualation. See ATLASSIM-4932.
-
Debottam Bakshi Gupta authored
-
- Nov 22, 2020
-
-
Michal Leszczynski authored
-
- Nov 20, 2020
-
-
- Nov 19, 2020
-
-
scott snyder authored
Don't crash opening a file that uses some kinds of automatic schema evolution. If the type of a member changes from one class to another (for example, from std::vector<double> to std::vector<float>), then GetExpectedType will give us the on-disk class. To get the class we use in memory, we need to use GetCurrentClass.
-
Tomasz Bold authored
-
-
- Nov 18, 2020
-
-
New TFCSEnergyInterpolationPiecewiseLinear class for piecewise linear interpolation See merge request atlas/athena!38095 (cherry picked from commit c5918933) dd0cab83 New TFCSEnergyInterpolationPiecewiseLinear class for piecewise linear interpolation ffa9240b Made m_linInterpol direct class member 66fd9218 Made logEkin[] and response[] const
-
This has been a long time coming, but one of the reasons for having AsgMessaging as a separate package from AsgTools is to allow xAODRootAccess to depend on it and use the same status codes and messaging as the rest of analysis base. This should take part of the former. I had to add the RECOVERABLE code to the main status code, but since that is also part of the Gaudi StatusCode that seems like a reasonable change/extension. Apart from that this is mostly a global search replace to move all users from TReturnCode to StatusCode, but apart from fixing up a couple of places related to the status code itself I did not try to fix/improve any code in the process. Note that this also revealed a number of tests that failed because they used RETURN_CHECK() incorrectly. This commit doesn't try to address that.
-
- Nov 17, 2020
-
-
-
Debottam Bakshi Gupta authored
-
Peter Onyisi authored
-
Fix LHAPDF reweighting in AnalysisTop See merge request atlas/athena!38269 (cherry picked from commit ea0007cf) 0c65e4b6 Adjust order of execution of event saver in top-xaod.
-