Import Geant4 9.6.0 source tree

This commit is contained in:
Gabriele Cosmo
2016-06-09 17:01:34 +02:00
parent b1eb5424d2
commit e2d2f9810a
10384 changed files with 698580 additions and 628834 deletions
@@ -17,6 +17,312 @@ committal in the CVS repository !
* Reverse chronological order (last date on top), please *
----------------------------------------------------------
09 November 12: V.Ivant (emutils-V09-05-56)
08 November 12: V.Ivant (emutils-V09-05-55)
- G4VEnergyLossProcess - fixed non-reproducibility for ions
08 November 12: V.Ivant (emutils-V09-05-54)
- G4VMultipleScattering - improved logic of computation of PostStep
safety and true step length, added more accurate protection
against case when true length exceed physics step limit
06 November 12: V.Ivant (emutils-V09-05-53)
- G4VMultipleScattering - fixed non-reproducibility for ions, added
protection - do not sample scattering if range exceed true step
length less than minGeom parameter (1e-6 mm)
03 November 12: V.Ivant (emutils-V09-05-52)
- G4VEnergyLossProcess, G4VEmProcess, G4VMultipleScattering - fixed
printout at initialisation
02 November 12: V.Ivant (emutils-V09-05-51)
- G4VMultipleScattering - ion multiple scattering is defined
in the class constructor
- G4VMscModel - build table for all particles except case when
multiple scattering process is shared with G4GenericIon
26 October 12: V.Ivant (emutils-V09-05-50)
- G4VMultipleScattering - return version 48
- G4VEmProcess - changed initialisation for ions
25 October 12: V.Ivant (emutils-V09-05-49)
- G4VMultipleScattering - moved relocation AlongStep and do not
call msc PostStep
21 October 12: V.Ivant (emutils-V09-05-48)
- G4EmBiasingManager - cleanup the code; do not modify ParticleChange
when apply RussianRoulette (monor CPU saving); moved
RussianRoulette method to inline; added range cut option
- G4VEmProcess, G4VEnergyLossProcess, G4EmProcessOption - add
corresponding modifiction needed for biasing interface
20 October 12: V.Ivant (emutils-V09-05-47)
- G4VMultipleScattering - fixed typo in previous tag;
compute true path length even is msc model is inactive
20 October 12: V.Ivant (emutils-V09-05-46)
- G4VMultipleScattering - critical fix for stopping particles
(ATLAS report on small steps of low-energy ions)
19 October 12: V.Ivant (emutils-V09-05-45)
- G4VEnergyLossProcess, G4VMultipleScattering - attempt to optimise
run time methods called at each step
- G4EmCalculator - added ComputeGammaAttenuationLength method
- added warning to obsolete methods and classes
18 October 12: V.Ivant (emutils-V09-05-44)
- G4VEnergyLossProcess - define flag isIon for all nucleus except
deuteron, triton and special DNA ions
- G4VEmProcess, G4EmModelManager - minor reduce of number of operations
at each step
- G4VMscModel - minor optimisation of inline methods
05 October 12: V.Ivant (emutils-V09-05-43)
- G4VMultipleScatteringProcess - check that true step length proposed
by msc model do not exceed physical true length
01 October 12: V.Ivant (emutils-V09-05-42)
- G4VAtomDeexcitation - updated initialisation for the case
when de-excitation is not used
27 September 12: V.Ivant (emutils-V09-05-41)
- G4VMscModel - added GetDEDX method
21 September 12: V.Ivant (emutils-V09-05-40)
- G4VAtomDeexcitation - removed forgotten debug printout
- G4VMultipleScatteringProcess - minor cleanup
20 September 12: V.Ivant (emutils-V09-05-39)
- G4VAtomDeexcitation - fixed logic of activation of de-excitation
per region and for all geometry
20 September 12: V.Ivant (emutils-V09-05-38)
- G4VAtomDeexcitation - added computation of sum of energies of
de-excitation gamma and electron produced AlongStep;
use logical OR instead of AND for de-exitation flags
- G4VEnergyLossProcess - fixed energy balance for the case of de-excitation
AlongStep
17 September 12: V.Ivant (emutils-V09-05-37)
- G4VAtomDeexcitation - fixed de-excitation AlongStep by substitution
which limited de-ecitation from ion ionisation
- G4VEnergyLossProcess - take into account modifications for
G4VAtomDeexcitation in order to keep energy balance
06 August 12: V.Ivant (emutils-V09-05-36)
- G4VMultipleScatteringProcess - added call to ReLocate PostStep
06 August 12: V.Ivant (emutils-V09-05-35)
- G4VMultipleScatteringProcess - change final position PostStep to
bypass bug in parallel transportation
28 August 12: V.Ivant (emutils-V09-05-34)
- G4VMultipleScatteringProcess - improved debug printout
24 August 12: V.Ivant (emutils-V09-05-33)
- G4VMultipleScatteringProcess, G4VMscModel - sampling scattering
is moved from PostStep to AlongStep
- G4VEmProcess - fixed minor memory leak at destruction
14 August 12: V.Ivant (emutils-V09-05-32)
- G4VEnergyLossProcess - fixed non-reproducibility by extention
of initialisation in StertTracking method for ions
- G4VEmProcess - removed over initialisation in StartTracking
- G4EmModelManager - improve printout on models
09 August 12: V.Ivant (emutils-V09-05-31)
- G4VMultipleScattering - allowed to change position and direction
AlongStep
01 August 12: V.Ivant (emutils-V09-05-30)
- G4EnergyLossMessenger - Allow to apply bising UI commands only
in G4_Idle state
30 July 12: V.Ivant (emutils-V09-05-29)
- G4VEnergyLossProcess, G4VEmProcess, G4VMultipleScattering -
use the same list of particles for printout at initialisation
in the case of verboseLevel=1
20 July 12: V.Ivant (emutils-V09-05-28)
- G4VEnergyLossProcess - similar fix as for G4VEmProcess
- G4VEmAngularGenerator - cleanup comments, removed obsolete methods
19 July 12: V.Ivant (emutils-V09-05-27)
- G4VEmProcess - fixed problem introduced in previous tag making
possible infinite loop in DNA tracking of 6 eV electron
17 July 12: V.Ivant (emutils-V09-05-26)
- G4VEnergyLossProcess, G4VEmProcess - updated tracking parameters
and flags in PostStepGetPhysicalInteractionLength
- G4EmBiasingManager - fixed bug in bremsstrahlung splitting, which
practically forbid splitting
13 July 12: V.Ivant (emutils-V09-05-25)
- G4VEnergyLossProcess, G4LossTableManager - removed unused method
InitialiseMSC
13 July 12: V.Ivant (emutils-V09-05-24)
- G4EmBiasingManager - do not using dynamic weight for Russian roulette
12 July 12: V.Ivant (emutils-V09-05-23)
- G4VEnergyLossProcess, G4VMultipleScattering, G4VEmModel - added
implementation of new method StartTracking
- G4VMscModel - changed initialisation of the new track using
StartTracking method
- G4VEmAngularDistribution - add temporary dummy implementation of virtual
sampling methods - will be removed when all generators will migrated
to the new signature
10 July 12: G.Cosmo (emutils-V09-05-22)
- Explicitly use inclusion of headers for system of units and physical
constants, in plan to remove implicit inclusion from globals.hh.
05 July 12: V.Ivant (emutils-V09-05-21)
- G4EmCorrections, G4VEmModel, G4VMscModel - fixed shadowing
04 July 12: V.Ivant (emutils-V09-05-20)
- G4VMscModel - method Value() is moved to G4VEmModel
- G4VEmProcess - added extra option allowing to build two different tables
for low-energy and high-energy using different interpolation method,
for that added optinal extra energy limit and SetMinKinEnergyPrim(G4double)
method, improved initialisation of tables and initialisation printout
- G4EmTableType - added type fIsCrossSectionPrim
- G4EmModelManager - added possibility to build high energy table of
cross sections
- G4VEmAngularDistribution - cleanup user interface
08 June 12: D. Sawkey
- G4VEmProcess fix for first part of bug 1314. Force if step==1 (see 30May12)
08 June 12: V.Ivant (emutils-V09-05-19)
- Fixed shadowing
06 June 12: V.Ivant (emutils-V09-05-18)
- G4VMscModel - trying optimise inline methods
- G4VMultipleScattering - trying optimise initialisation and reduce
number of checks in run time
- G4LossTableManager - moved inline methods to the source
- G4EmBiasingmanager - add check on number of secondaries PostStep,
if it is not 1 then disable both spliting and Russian roulette,
this does not affect ordinary delta-electron and bremssrahlung
secondary biasing (splitting and Russian roulette);
Dynamic weight approach (weight is function of secondary energy)
is applied for Russian roulette if biasing limit energy is set
below 1 GeV, alternatively costant factor is used
05 June 12: V.Ivant (emutils-V09-05-17)
- G4VMscModel - allow build transport x-section table for d,t,He3,He4
01 June 12: D. Sawkey (emutils-V09-05-16)
- G4EnergyLossMessenger: Put back in guidance for high energy limit for
bremsplitting, russian roulette
31 May 12: D. Sawkey
G4EmBiasingManager:Put back in high energy limit for Russian roulette,
bremsplitting
31 May 12: V.Ivant (emutils-V09-05-15)
30 May 12: D. Sawkey
- G4VEnergyLossProcess bug 1314 fix: for forced biasing, now force if step==1
30 May 12: D. Sawkey
- G4VEmProcess, G4VEnergyLossProcess. for forced interaction, cross-
section biasing, set weight consistently
29 May 12: D. Sawkey
- G4EmBiasingManager, G4VEmProcess, G4VEnergyLossProcess,
G4EmProcessOptions, G4EnergyLossMessenger: Change brem splitting to
EGS-style with unique secondaries. Only brem split particles with
weight > 1/N (N=brem splitting factor), and only Russian Roulette
with weight < 1. Energy limit no longer used.
28 May 12: V.Ivant (emutils-V09-05-14)
- G4EmCalculator - fixed FindRegion
18 May 12: V.Ivant (emutils-V09-05-13)
- G4EmModelManager - improved printout of model energy limits
- G4VEmModel - added inline methods HighEnergyActivationLimit(),
LowEnergyActivationLimit(), ForceBuildTableFlag(), ForceBuildTable()
- G4VMscModel - updated GetRange() and GetEnergy() methods allowing to
use the same model class for many particles
09 May 12: V.Ivant (emutils-V09-05-12)
- G4ElectronIonPair - use inverse FanoFactor to simplify formula
09 May 12: V.Ivant (emutils-V09-05-11)
- G4VMultipleScatetring - improved printout at initialisation
- G4ElectronIonPair - cleanup implementation of sampling of number of ions
26 April 12: V.Ivant (emutils-V09-05-10)
- G4EmCalculator - fixed GetMeanFreePath for msc processes (bug introduced
in previous tags)
20 April 12: V.Ivant (emutils-V09-05-09)
- G4LossTableManager - added methods MinKinEnergy, MaxkinEnergy
- G4EmModelManager - extended printout if cross section table is built for
the model
- G4EmCalculator - improved model selection for msc processes
- G4VMultipleScatetring - removed transport cross section table,
simplified AlongStepGetPhysicalInteractionLength
method, added protection against zero
kinetic energy into PostStepDoIt,
general cleanup of the code
- G4VMscModel - added building of transport cross section table and new
method GetTransportCrossSection to be used in run time
- G4LossTableBuilder - added method BuildTableForModel
19 April 12: V.Ivant (emutils-V09-05-08)
- G4EmCorrections : A.Ribon fixed reproducibility violation due to the cashing
of IonHighOrderCorrections with a vector depending on the ion charge.
The correction depends not only on the ion charge but also on its
mass, therefore the cashing based on the ion charge only depends on
which isotope is encountered first, which depends on the history,
and violated the reproducibility. The fix consists of using a map
for cashing, with the key given by the PDG encoding of the ion,
which is unique for each isotope.
- G4VEmProcess - use ProposeWeight instead of ProposeParentWeight
to fix weights of secondary particle for gamma processes
- G4LossTableManager - minor correction of number of bins per decade
15 April 12: V.Ivant (emutils-V09-05-07)
- G4LossTableBuilder - added method BuildTableForModel allowing to build
G4PhysicsTable for a model
- G4VEmModel - added virtual method Value, by default computes cross section
- G4VMscModel - added method GetTransportMeanFreePath
19 March 12: V.Ivant (emutils-V09-05-06)
- G4EmCalculator - fixed compillation warning at gcc 4.6
15 March 12: V.Ivant (emutils-V09-05-05)
- G4LossTableManager - added a check and protection against double
registration of the same process; this addition allows to use
the same EM energy loss process for different particle types
- G4EmCalculator - modifed method to access a process taking into account
that the same process may used by different particle type;
added new methods ActiveForParticle, FindEnLossProcess,
FindDiscreteProcess, FindMscProcess
5 March 12: V.Ivant (emutils-V09-05-04)
- G4VEnergyLossProcess fixed crash due to apply G4hIonisation for ions;
- G4VEnergyLossProcess, G4VEmProcess, G4LossTableBuilder - delete old
G4PhysicsVectors if new vectors are built between runs
- G4VAtomDeexcitation - use safe method to extract integer Z
29 February 12: V.Ivant (emutils-V09-05-03)
- G4VEnergyLossProcess fixed computation of delta-electron cross section
for ions and very small cuts (about 1 um), reported by Sergio Ibarmia
21 February 12: V.Ivant (emutils-V09-05-02)
- G4VEmAngularDistribution added new method Initialise
21 February 12: V.Ivant (emutils-V09-05-01)
- G4VEmAngularDistribution added two new methods SampleCosinePolarAngle
and SampleDirection
7 February 12: V.Ivant (emutils-V09-05-00)
- G4VAtomDeexcitation - add activation of deexcitation for the world
if deexcitation is activated but G4Regions for deexcitation
are not defined
10 November 11: V.Ivant (emutils-V09-04-51)
- G4LossTableManager, G4EmModelManager - fixed defects reported by
Coverity - private assignment operator and copy constructor added