Recommended LoKi::Hybrid Functors

This lists the filters recommended for use in the HLT and the selections. See DaVinciTutorial4 for a hands-on tutorial. A longer, but not necessarily up-to-date, list can be found at LoKiParticleFunctions. It also contains examples on how to use these functors in C++ code. As neither are guaranteed to be a complete list, you can always see the DOxygen for the LoKi::Cuts namespace here.

This page refers to DaVinci v22r1.

Particle Functors

Some mnemonic rules:

• The functors with the prefix BPV deal with "the best primary vertex". The best primary vertex is extracted from the desktop-tool using the method IPhysDesktop::relatedVertex
• The functors with the suffix DV get some even-data (e.g. the list of all vertices) through desktop tool
• The functors with the suffix TES get the event data from the Transient Event Store
• The VD as a part of the functor name means that the functor deals with "vertex distance"
• The VDCHI2 as a part of the functor name means that the functor deals with "vertex -distance"
• The SIGN as a part of the functor name means that functor return the value of or distance artificially signed according to some criteria.
• The IP stands for "impact parameter" and MIP means "the minimal (with reswpect to some set of vertices) value of impact parameter"
• Practically all "vertex functors" ( the functors which accept const LHCb::VertexBase* as arguments) have their names starting from V, the most evident exception if the functor PRIMARY
• All indices, e.g. for CHILDCUT or CHILDFUN functors, starts from 1. 0 always means the particle itself.
• For many Particle functors there is an equivalent Particle array functor starting with an A. A notable exception is ADAMASS for ADMASS.

ALL: All

Takes all particles. This is required if one wants to apply no cut.

FilterDesktop.Filter.Code = "ALL" 

ABSID: Absolute value of PID.

Returns the absolute value of the PID. The following lines are equivalent:

FilterDesktop.Filter.Code = "ABSID==211" 
FilterDesktop.Filter.Code = "ABSID=='pi+' " 
FilterDesktop.Filter.Code = "ABSID=='pi-' " 

Note the last line! The comparison (ABSID=='pi-') takes the absolute value on both sides. This avoid having to remember that the has a positive pid (211) while the has a negative pid (-13).

The pid can also be used without the absolute value. See ID.

ACHILD: Compute a function on a particle in an array

The functor ACHILD acts as "meta-functor", it gets the daughter particle number "N" (the second parameter of constructor), apply the functor (the first parameter of contructor) and return the result.

e.g. ACHILD(PT,1) return the transverse momentum of the first particle in the combination, ACHILD(PT,2) return the transverse momentum of the second particle in the combination.

For example:

XXX.CombinationCut = "ACHILD(PT,1)*ACHILD(PT,2)>1500000" 

See the function AMAXCHILD or AMINCHILD, to get the daughter particle by name.

ADMASS: The absolute mass difference to the reference value

Calculates the absolute difference between the measured mass and the PDG reference value. It takes the pid of the reference particle as argument.

FilterDesktop.Filter.Code = "ADMASS('KS0')<50*MeV"

The mass difference can also be used without the absolute value. See DMASS.

ADPDGM and ADPDGMASS: The absolute mass difference to the nominal mass

Returns the difference with the PDG value of a particle mass. Useful to compute the difference with the measured mass. This calls the ParticlePropertySvc at very call and is therefore slow. Use ADMASS('B0') rather than ADPDGM if you know the PID.

AMAXCHILD or AMINCHILD: cut an array and return the max/min value of a functor.

Similar to MINTREE and MAXTREE, however can be performed on a particle array.

e.g. the minimal PT for all negative kaons: AMINCHILD(PT,"K-"==ID)

XXX.CombinationCut = "AMINCHILD(PT,"K-"==ID)*ACHILD(PT,"K+"==ID)>1500000" 

This is computationally expensive, using ACHILD is much quicker if possible.

BPVDIRA: Direction angle

Computes the cosine of the angle between the momentum of the particle and the direction fo flight from the best PV to the decay vertex.

CombineParticles.MotherCut = "BPVDIRA()>0.9999" 

BPVIPCHI2(): IP on related PV

Computes the -IP on the related PV.

CombineParticles.MotherCut = "BPVIPCHI2()<25" 

TODO : So far it needs the ().

BPVLTFITCHI2: the of the proper lifetime fit.

The functor evaluates the of the proper lifetime fit of the particle using ILifetimeFitter tool. This is probably the best measure of the consistency of the hypothesis that the particle originates from the given primary vertex. It is also probably the best "pointing" variable.

CombineParticles.MotherCut = "BPVLTFITCHI2()<16" 

The related primary vertex is extracted from the desktop, the fitter itself is extracted using DVAlgorithm::lifetimeFitter("nick-name"). Please note that it is completely different functor from BPVLTCHI2!

BPVLTIME: the proper lifetime of the particle

The functor evaluates the proper lifetime of the particle using ILifetimeFitter tool. Unfortunately due to very sad conventions adopted for LHCb, the proper time is measured in ns instead of the natural units: .

CombineParticles.MotherCut = "BPVLTIME()>1.5" 

The related primary vertex is extracted from the desktop, the fitter itself is extracted using DVAlgorithm::lifetimeFitter("nick-name").

BPVLTCHI2: the -significance of the proper lifetime of the particle

The functor evaluates the -significance of the proper lifetime of the particle using ILifetimeFitter tool.

CombineParticles.MotherCut = "BPVLTCHI2()>9" 

The related primary vertex is extracted from the desktop, the fitter itself is extracted using DVAlgorithm::lifetimeFitter("nick-name"). Please note that it is completely different functor from BPVLTFITCHI2!

BPVLTSIGNCHI2: the signed -significance of the proper lifetime of the particle

The functor evaluates the signed -significance of the proper lifetime of the particle using ILifetimeFitter tool.

CombineParticles.MotherCut = "BPVLTSIGNCHI2()>-4" 

The related primary vertex is extracted from the desktop, the fitter itself is extracted using DVAlgorithm::lifetimeFitter("nick-name").

BPVVDCHI2: -separation from related PV

Computes the -distance from the related PV: (Pos(bestPV) - Pos(EndVertex))^T ( cov_(bestPV) + cov(EndVertex) )^-1 (Pos(bestPV) - Pos(EndVertex))

CombineParticles.MotherCut = "BPVVDCHI2()>100" 

BPVVDZ: -distance from the end vertex of the particle and the related primary vertex.

The functor computes the -distance from the end vertex of the particle and the related primary vertex.

CombineParticles.MotherCut = "0<BPVVDZ" 

The concept and the name come from Sean Brisbane. The functor is available starting from Phys/LoKiPhys version >= v7r2.

BPVVDR: -distance from the end vertex of the particle and the related primary vertex.

The functor computes the -distance(cylindrical) from the end vertex of the particle and the related primary vertex.

CombineParticles.MotherCut = "0.1<BPVVDR" 

The concept and the name come from Sean Brisbane. The functor is available starting from Phys/LoKiPhys version >= v7r2.

BPVVDRHO: -distance from the end vertex of the particle and the related primary vertex.

The functor computes the -distance(cylindrical) from the end vertex of the particle and the related primary vertex.

CombineParticles.MotherCut = "0.1<BPVVDRHO" 

The concept and the name come from Sean Brisbane. The functor is available starting from Phys/LoKiPhys version >= v7r2.

CHI2PDGM and CHI2PDGMASS: The mass difference to the nominal mass in units of chi2

Returns the difference with the PDG value of a particle mass. This calls the ParticlePropertySvc at very call and is therefore slow. Use CHI2M('B0') rather than CHI2PDGM if you know the PID.

CHI2M and CHI2MASS: The mass difference to some reference mass in units of chi2

Returns the difference with the PDG value of a particle mass. It takes an argument like ADMASS

CombineParticles.MotherCut = "CHI2M('B0')<5" 

CHILDCUT: Applies a cut to a given child

FilterDesktop.Filter.Code = "CHILDCUT ( MIPCHI2DV ( PRIMARY ) > 1 , 1 )" 

In this example one applies an IP cut on the first daughter of the input particle. This requires to know which is the first, second, etc daughter. Can be useful when (N)INTREE won't work. Like here for the slow pion in a where searching for a pion in the tree would also return the daughters of the . Use the safer INTREE and NINTREE instead.

TODO : Is there a way of applying a cut to the daughters only, without navigating the whole tree? From next version NINGENERATION and INGENERATION will do this.

DECTREE: Check if a particle satisfies a certain decay descriptor

FilterDesktop.Filter.Code = "DECTREE ('B0 -> (D- -> K+ pi- pi-) pi+')" 

The example selects the decay of B0 to D- pi+ followed by D- to K+ pi- pi-. This can be useful for example to select one or more decay channels from a stripping line that contains many decay channels

DMASS: The mass difference to the reference value

Calculates the difference between the measured mass and the PDG reference value. It takes the pid of the reference particle as argument. In most reasonable cases one will use the absolute mass difference (see ADMASS) or the mass itself (see MM).

DPDGM and DPDGMASS: The mass difference to the nominal mass

Returns the difference with the PDG value of a particle mass. Useful to compute the difference with the measured mass. This calls the ParticlePropertySvc at very call and is therefore slow. Use DMASS('B0') rather than DPDGM if you know the PID.

ID: Particle ID

Like ABSID but without the absolute value.

INGENERATION: "in generation"

The predicate which checks the existence of the particle satisfying the requirements in the decay tree at the given depth

FilterDesktop.Code = "INGENERATION ( ( 'mu+'==ABSID)  & ( PT > 2 * GeV ) , 2 ) " 

Requires the presence of at least one granddaughter , which the transverse momentum . The generations are defined as: 0 - the particle itself, 1 - the direct children, 2 - grandchildren , etc. The concept and the name come from Patrick Koppenburg. The functor is available starting from Phys/LoKiPhys version >= v7r2.

INTREE: In tree

Requires there is a particle in the decay tree satisfying the requirements.

FilterDesktop.Filter.Code = "INTREE( (ID=='J/psi(1S)') & (BPVVDCHI2>25) ) " 

Requires there is a in the tree more than away from the best vertex.

MAXTREE: Maximum value in the decay tree

Returns the maximum value of some functor in the decay tree. Useful to apply cuts on all particles, or to extract some value from the tree.

CombineParticles.MotherCut = "M-MAXTREE('D0'==ABSID,M)<165.5" 

This example looks for all in the decay, returns the largest mass (which is the mass of the as there's only one) and computes the mass difference with the mother.

MINTREE: Minimal value in the decay tree

Takes a functor as agrument and returns its minimal value searching through the decay tree.

FilterDesktop.Filter.Code = "MINTREE(ABSID=='K+',PT)>1400*MeV" 

This would look for all particles kaons in the decay tree and find their and return the minimum. The cut then requires that all kaon descendents have a .

MINVDDV: minimum distance between the particle's end-vertex and any other vertex

The functor evaluates the minimal 3D-distance between the particle's end-vertex and the vertices form "list". The list of vertices is extracted from the desktop:

CombineParticles.MotherCut = "MINVDDV(PRIMARY)>1" 

Requires that the particle's end-vertex has at least one millimeter separation with respect to any primary vertex.

MINVDSOURCE: minimum distance between the particle's end-vertex and any other vertex

The functor evaluates the minimal 3D-distance between the particle's end-vertex and the vertices form "list". The list of vertices is extracted from the "source":

CombineParticles.MotherCut = "MINVDSOURCE( VSOURCEDV ( PRIMARY ) )>1" 

Requires that the particle's end-vertex has at least one millimeter separation with respect to any primary vertex.

MINVDTES: minimum distance between the particle's end-vertex and any other vertex

The functor evaluates the minimal 3D-distance between the particle's end-vertex and the vertices form "list". The list of vertices is extracted from the TES:

CombineParticles.MotherCut = "MINVDTES('SomeLocationInTES',PRIMARY)>1" 

Requires that the particle's end-vertex has at least one millimeter separation with respect to any primary vertex.

MIPCHI2DV: Minimum IP-

Returns the minimum distance of a particles's trajectory to any set of vertices. Most useful in conjunction with the PRIMARY argument that looks at PVs.

FilterDesktop.Filter.Code = "MIPCHI2DV(PRIMARY)>9"
FilterDesktop.Filter.Code = "MIPCHI2DV()>9"  ## ditto 

This requires that the particle is away from any PV.

MIPDV: Minimum IP

Returns the minimum distance of a particles's trajectory to any set of vertices. Most useful in conjunction with the PRIMARY argument that looks at PVs.

FilterDesktop.Filter.Code = "MIPDV(PRIMARY)>0.15" 

This requires that the particle is 0.15 mm away from any PV.

M: Mass

Returns . Uses the function LHCb::Particle::momentum().M(). For many cases it should be practically the same as MM.

MM: Measured Mass

Returns the measured mass of the particle. The function LHCb::Particle::measuredMass() is used for evaluation.

HltSharedDiMuon.CombinationCuts = "AM<11*GeV"

This selects dimuons up to a mass of 11 GeV.

NINGENERATION: "Number of particles in generation"

The function which counts the particles satisfying the requirements in the decay tree at the given depth:

FilterDesktop.Filter.Code = " 2 == NINGENERATION ( ( 'mu+'==ABSID)  & ( PT > 2 * GeV ) , 2 ) " 

Requires the presence of exactly two granddaughter with the transverse momentum . The generations are defined as: 0 - the particle itself, 1 - the direct children, 2 - grandchildren , etc. The concept and the name come from Patrick Koppenburg. The functor is available starting from Phys/LoKiPhys version >= v7r2.

NINTREE: Number of particles in tree

Returns the number of particles in a tree satisfying some selection criteria.

FilterDesktop.Filter.Code = "2 == NINTREE( (ABSID=='e-') & (PT>1*GeV))" 

Requires that there are two electrons in the tree with .

NMASS: The nominal mass of a particle

Returns the PDG value of a particle mass. Useful to compute the difference with the measured mass. This calls the ParticlePropertySvc at very call and is therefore slow. Use ADMASS('B0') rather than abs(MM-NMASS) if you know the PID.

P: Momentum

Gets the momentum of the particle.

FilterDesktop.Filter.Code = "P>2*GeV"

Gets particles with .

PIDe, PIDmu, PIDK, PIDp, PIDpi : PID

Gets the combined delta-log-likelihood for the given hypothesis (wrt the pion), :

FilterDesktop.Filter.Code = "PIDe-PIDpi>6" 

PT: Transverse momentum

Gets the transverse momentum of the particle. Note that this is evaluated at the first measurement of the particle, which might not be where you want it for pions in decays.

FilterDesktop.Filter.Code = "PT>1*GeV"

Gets particles with .

TRCHI2: the track fit

Gets the of the track fit.

FilterDesktop.Filter.Code = "TRCHI2<1000" 

TRCHI2DOF: per degree of freedom of the track fit

Gets the per degree of freedom of the track fit.

FilterDesktop.Filter.Code = "TRCHI2DOF<20" 

TRPCHI2: -probability the track fit

Gets the -probability of the track fit.

FilterDesktop.Filter.Code = "TRPCHI2>0.01" 

VDMIN: Distance of two vertices

TODO : Not yet supported.

VFASPF: Vertex Function as Particle Function.

Allows to apply vertex functors to the particle's endVertex().

CombineParticles.MotherCut = "VFASPF(VCHI2/VDOF)<10" 

Applies a cut to the vertex of the particle.

Particle Array Functors

Applied to a set of particles, typically the daughters

AHASCHILD: Checks if there is a particle in the array that matches the requirements

CombineParticles.CombinationCut = "AHASCHILD((ABSID=='pi+') & (PT>1*GeV))"

Requires there's at least one pion with a Pt larger than 1 GeV in the combination. This is useful when one requires at least one particle in the decay to meet a requirement.

ADAMASS: The absolute mass difference to the PDG reference value

Calculates the absolute difference between the measured mass and the PDG reference value. It takes the pid of the reference particle as argument.

CombineParticles.CombinationCut = "ADAMASS('KS0')<50*MeV"

AM: Mass of the combination

Returns .

AMAXCHILD, AMINCHILD: Maximal, minimal value in array.

CombineParticles.CombinationCut = "1*GeV<AMINCHILD(MINTREE('pi-'==ABSID,PT)"

In this example one checks that the minimal PT of pions in the decay tree of the particles is in excess of 1 GeV.

AMAXDOCA, AMINDOCA: Distance of closest approach cut

CombineParticles.CombinationCut = "(AMAXDOCA('') < 0.08*mm)"

Checks that the maximum distance of closest approach between all possible pairs of particles is less than 80mum. Equal results can be otained with less CPU time using

CombineParticles.CombinationCut = "ACUTDOCA(0.08*mm,'')"

AMAXDOCACHI2, AMINDOCACHI2: Distance of closest approach significance cut

CombineParticles.CombinationCut = "(AMAXDOCACHI2('') < 10)"

Checks that the maximum distance of closest approach - chi2 between all possible pairs of particles is less than 10. Equal results can be otained with less CPU time using

CombineParticles.CombinationCut = "ACUTDOCACHI2(10,'')"

This cut is particularly useful to gain speed on vertex fits. As a rule of thumb, any vertex chi2 cut can be preceded by a DOCACHI2 cut at the same (total, not reduced!) chi2.

ANUM: Number of particles in the array meeting a requirement

CombineParticles.CombinationCut = "1<ANUM( ('pi+'==ABSID) & (MIPCHI2DV(PRIMARY)>25))"

AP: Sum momentum of the array

Gets the momentum of the array.

CombineParticles.CombinationCut = "AP>1*GeV"

Gets particles with .

APT: Sum transverse momentum of the array

Gets the transverse momentum of the array (that's the of the sum 4-vector, not the sum of the ). Note that this is evaluated at the first measurement of the particles, which might not be where you want it for pions in decays.

CombineParticles.CombinationCut = "APT>1*GeV"

Gets particles with .

Vertex functors

Vertex functors are accessed using VFASPF.

VX, VY, VZ: Vertex x, y or z-position

CombineParticles.MotherCut = "VFASPF(VZ)>-100 & VFASPF(VZ)>100" 

VCHI2 : Vertex

CombineParticles.MotherCut = "VFASPF(VCHI2/VDOF)<10" 

VDOF : Vertex fit number of degrees of freedom

CombineParticles.MotherCut = "VFASPF(VCHI2/VDOF)<10" 

A two-track vertex has one degree of freedom. A three-track vertex has three degrees of freedom.

VMINVDDV: minimum distance between the vertex and any other vertex

The functor evaluates the minimal 3D-distance between the vertex and the vertices from "list". The list of vertices is extracted from the desktop:

CombineParticles.MotherCut = "VFASFP(VMINVDDV(PRIMARY))>1" 

Requires that the particle's end-vertex has at least one millimeter separation with respect to any primary vertex.

VMINVDSOURCE: minimum distance between the vertex and any other vertex

The functor evaluates the minimal 3D-distance between the vertex and the vertices form "list". The list of vertices is extracted from the "source":

CombineParticles.MotherCut = "VFASPF( VMINVDSOURCE( VSOURCEDV ( PRIMARY ) ) ) >1" 

Requires that the particle's end-vertex has at least one millimeter separation with respect to any primary vertex.

VMINVDTES: minimum distance between the vertex and any other vertex

The functor evaluates the minimal 3D-distance between the vertex and the vertices form "list". The list of vertices is extracted from the TES:

CombineParticles.MotherCut = "VFASPF( VMINVDTES ( 'SomeLocationInTES' , PRIMARY ) ) >1" 

Requires that the particle's end-vertex has at least one millimeter separation with respect to any primary vertex.

VPCHI2: -probability

The functor evaluates the -probability of the vertex, taking into account the number of degrees of freedom. The GSL routine gsl_cdf_chisq_Q functon is used for evaluation.

CombineParticles.MotherCut = "VFASPF(VPCHI2)>1*perCent" 

The functor is available starting from Phys/LoKiPhys version >= v7r2

Units

Presently units are not supported easily. This will change in the next release. To avoid confusion between M (mass) and m (meter) for instance a limited set will be defined. Vanya suggests:

• MeV, GeV
• mm, millimeter, cm , centimeter, meter
• perCent
• ns, fs, nanosecond, femtosecond, ps, picoseconds

These units are available starting from Phys/LoKiPhys version >= v7r2

Syntax

Cut are combined using the bit-wide & and | operators, not boolean operators. This also requires that cuts are well protected using parentheses. Example:

FilterDesktop.Filter.Code = "( ((ABSID=='pi+') & (PT>400*MeV)) | 
                               ((ABSID=='KS0') & (PT>1*GeV)))"

Cuts can be split over several lines.

Types

Some allowed symbols are instances, therefore they do not require any parentheses, e.g. P, PT etc... Some symbols are types, e.g. MIPDV, and they require to be instantiated. The parentheses means invocation of constructor. E.g. MIPDV(PRIMARY) is just invocation of the constructor of type MIPDV with parameter PRIMARY (the predicate to select the vertices). the simplest way to check if the symbol is a type or instance is through inspection using the interactive python:

[lxplus097] ~/cmtuser/DaVinci_v19r12/Phys/DaVinci/v19r12/cmt % python
Python 2.5 (r25:51908, Oct 18 2007, 16:04:48)
[GCC 3.4.6 20060404 (Red Hat 3.4.6-8)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>>from LoKiPhys.decorators import *
>>>from LoKiArrayFunctors.decorators import *
>>>P
P
>>>MIPDV
<class 'LoKiPhys.functions.LoKi::Particles::MinImpParDV'>

Use of CombineParticles.

CombineParticles is a LoKI::Algo, and a DVAlgorithm that combines the input particles according to the decay descriptor. There are three cuts applied :

DaughtersCuts :

On the incoming daughter particles. It is a map of "particle" : "cuts"

HltSelB2DstarMu.DaughtersCuts = { 
// PT cut for muons
"mu-" :  "PT>400*MeV", 
// IP_chi2 cut for D0 in D*+ 
"D*(2010)+'" : "MAXTREE('D0'==ABSID,BPVVDCHI2) >6.25 "
}  

CombinationCut

Once a combination has been made according to the decay descriptor but before the vertex fit. This cut is applied to the set of daughters. Therefore only Particle array functors can be used. They all start with an A.

MotherCut

Applied the vertex fit on the mother. All the cuts that require the position of the vertex must be applied in MotherCut, while the others can be applied earlier (saving CPU). Note that for long lived particles like Ks it pays off to apply a loose mass cut in CombinationCut and a harder in MotherCut. The reason is that the vertex fit does a propagation of the momenta through the detector. You thus get the momentum at the Ks vertex, while in CombinationCut it's just the sum of the momenta of the daughters at their first measurement.

A simple example:

HltShared.Members += { "CombineParticles/HltSharedPhi2KK" } 

HltSharedPhi2KK.PhysDesktop.InputLocations = { "Phys/HltNoPIDsKaons" }

HltSharedPhi2KK.DecayDescriptor = "phi(1020) -> K+ K-"

// (IP > 2 sigma^2)
HltSharedPhi2KK.DaughtersCuts = { "K+" : "MIPCHI2DV(PRIMARY)>4" }    

// mass window +/- 300 MeV
HltSharedPhi2KK.CombinationCut = "ADAMASS('phi(1020)')<50" 
 
// chi2/ degrees of freedom < 25 
HltSharedPhi2KK.MotherCut = "VFASPF(VCHI2/VDOF)<25" 

The decay descriptor can be overwritten by DecayDescriptors which allows to reconstruct several decays in one go.

HltShared.Members += { "CombineParticles/HltSharedDstarWithD02KPi" } 
HltSharedDstarWithD02KPi.PhysDesktop.InputLocations = { "Phys/HltSharedD02KPi", "Phys/HltSharedSlowPions" }

// also wrong-sign ones
HltSharedDstarWithD02KPi.DecayDescriptors = {"[D*(2010)+ -> pi+ D0]cc", "[D*(2010)+ -> pi+ D~0]cc" } 

// mass window +/- 50 MeV, PT> 1.25 GeV
HltSharedDstarWithD02KPi.CombinationCut = "(ADAMASS('D*(2010)+')<50) & (APT>1250)"  

// Chi2/nDoF < 25 .  m(D*-D0) = 145.5 MeV, +2
HltSharedDstarWithD02KPi.MotherCut = "(VFASPF(VCHI2/VDOF)<25) & (M-MAXTREE('D0'==ABSID,M)<165.5)"        

Mother Cut, Daughter Cut, Combination Cut

Each cut, which is applicable to mother is also applicable to daughter (and vice versa). Moreover the cut, used for daughter particle, being equipped with the proper CHILD,CHILDFUN,CHILDCUT,MINTREE,MAXTREE,INTREE,NINTREE meta- function, gives IDENTICAL result being applied for the mother particle.

From pure C++ point of view, both DaughterCuts and MotherCuts are objects which get as argument const LHCb::Particle*. But CombinationCut accepts as an argument LoKi::Range_<LHCb::Particle::ConstVector>".

Therefore is requires the DIFFERENT objects. CombinationCut is applied for the combination of "good" daughter particles, and only for the combinations, which satisfy CombinationCut, the effective mother particle is created through teh vertex fit procedure, and MotherCut is applied for this effective mother particle.

Help

Please do not send mails directly to the authors, but send the questions and requests through the following mailing lists:

lhcb-davinci@cern.ch, lhcb-loki@cern.ch, lhcb-bender@cern.ch

In this case more colleagues can profit from solutions to the problems and it will allow a bit more wide distribution of useful information, tricks, recipes, experience and the solutions. Also it will simplify the monitoring of the progress with the implementation of the missing functionality. As alternative to the mailing lists the following actions are also accepted:

1. One can submit the question to LoKi FAQ.
2. One can submit new task or request new feature through Savannah LoKi portal
3. One can submit new task or request new feature through Savannah Bender portal

Examples

Many working examples can be found in the Hlt/HltSelections package from version v6r0 released with DaVinci v19r11.

Feature requests

Feature requests should be sent to the same mailing lists as above. For every new feature a suggested name should be provided. This drastically simplifies the life of developers and results in a MUCH HIGHER PRIORITY for the implementation.

Change of names

Although it is very simple to define aliases for these functors (and many of them are actually available) their use is not encouraged. This is because we would like to define a minimum vocabulary everyone would have to be able to understand. Since this is rather new, suggestions for changes of names are accepted but should be agreed by the community. Please send requests to the mailing lists above. It is suggested to freeze this in the next version.

Suggestions for the names of the filters in CombineParticles are also welcome.

Few related tools/algorithms.

A few simple tools can be used for debugging and/or validation/testing the various LoKi::Hybrid construction.

LoKi::Hybrid::PrintTool

The most primitive tool, it implements the abstract interface IPrintDecay, and it allows to print some information about the decay tree. It is convinient to use the high-level algorithm PrintDecayTree to drive the tool:

// append the algorithm to the list of top level algorithms:
ApplicationMgr.TopAlg += { "PrintDecayTree/Printer"} ;

// declare the actual type/name of the IPrintDecay tool:
Printer.PrintDecayTool = "LoKi::Hybrid::PrintTool/PRINT" ;

// declare the variables to be printed by the tool, each "variable" is just LoKi-functor:
Printer.PRINT.Variables = {  "ID"  , "P/GeV"  ,  "PT/GeV"  , "M/GeV"   ,  "KEY"   } ;

With this configuration one obtains the following printout:

Printer.PRINT             INFO Print the decay tree:  ( B0 ->  mu+  mu-  K+  K-  )
 ---------------------------------------------------------------------------------------------------------
 |#         Decay        |       ID      |    (P/1000)   |   (PT/1000)   |       M       |      KEY      |
 ---------------------------------------------------------------------------------------------------------
 |0 |--> B0              |      511      |     262.81    |     6.1192    |     5099.1    |       5       |
 |1    |--> mu+          |      -13      |     56.877    |     1.6668    |     105.66    |       0       |
 |1    |--> mu-          |       13      |     128.8     |     3.7141    |     105.66    |       1       |
 |1    |--> K+           |      321      |     56.877    |     1.6668    |     493.68    |       2       |
 |1    |--> K-           |      -321     |     20.261    |    0.63129    |     493.68    |       5       |
 ---------------------------------------------------------------------------------------------------------

For the default configuration, all columns are printed using 14.5g "scientific" format and the column title is constructed from the functor itself. However both the format and the table header could be modified, e.g. if one adds into the previous configuration the explicit specification of Format and Header:

// Format: print only 5th (KEY) and 1st (ID) variable from the list of "Variables": 
Printer.PRINT.Format = "| %5% %|6t|| %1% %|14t||"

// specify explicitly the table header: column names
Printer.PRINT.Header = "| Key | PdgID |"

In this case the output table looks like:

Printer.PRINT                         INFO Print the decay tree:  ( B0 ->  mu+  mu-  K+  K-  )
 ---------------------------------------
 |#         Decay        | Key | PdgID |
 ---------------------------------------
 |0 |--> B0              | 5   | 511   |
 |1    |--> mu+          | 0   | -13   |
 |1    |--> mu-          | 1   | 13    |
 |1    |--> K+           | 2   | 321   |
 |1    |--> K-           | 5   | -321  |
 ---------------------------------------

For more details see the Doxygen documentation here.

LoKi::Hybrid::PlotTool

The simple tool, which implements the abstract interface IPlotTool and allows to fill some histograms for the selected particles.

For more details see the Doxygen documentation here.

LoKi::Hybrid::TupleTool

The simple tool, which implements the abstract interface here and allows to fill N-tuple with useful information for the selected particles.

The tool is a specialization of general "tuple tools", the concept developed by Jeremie Borel. The usage is failrly trivial:

ApplicationMgr.TopAlg += { "DecayTreeTuple/MyDecayTreeTuple" } ;

// specify location of input data in TES 
MyDecayTreeTuple.PhysDesktop.InputLocations = {"Phys/HltSelBd2MuMuKstar"} ;

// use various tools, developed by Jeremie
MyDecayTreeTuple.ToolList = {  
      "TupleToolTrigger"
      , "TupleToolMCTruth"
      , "TupleToolMCBackgroundInfo"
      , "TupleToolGeometry"
      , "TupleToolKinematic"
      , "TupleToolPropertime"
      , "TupleToolPrimaries"
     , "TupleToolEventInfo"
     // use also LoKi-based Tuple Tool!!
     , "LoKi::Hybrid::TupleTool/LoKiTool" /// < THIS LINE 
};

// configure LoKi tuple tool:
MyDecayTreeTuple.LoKiTool.Variables = { 
  "mass" : "M/GeV" ,
  "p"       : "P/GeV" ,
  "pt"      : "PT/GeV" 
};

For this example, the tool adds three columns into N-Tuple, defines the names of the column to be "head"+"mass", "head"+"p" and "head"+"pt". The content of the colums will be filled with tresult of evaluation of the following LoKi functors: M/GeV, P/GeV and PT/GeV.

For more details see the Doxygen documentation here.

Direct manipulatuon with factories

Sometime for debugging purposes the direct manipulation with factories could be useful. Also it could have some sense for validation of functors and comparison with some other calculations.

Since the factory is AlgTool in Gaudi sense, first one needs to acquire the factory

#include "LoKi/IHybridFactory.h"

...
const std::string factoryName = "LoKi::Hybrid::Tool/HybridFactory:PUBLIC" ;

LoKi::IHybridFactory* factory = tool<LoKi::IHybridFactory> ( factoryName ) ;

And now one can use the factory for creation of of functors (functions and predicates) :

using namespace LoKi::Types ;
using namespace LoKi::Cuts   ;

// initialize the function with *some* values 

Fun fun = -1.E+10 * ONE ;

// get the code 
const std::string& code = .... ; 

// instantiate the functor:
StatusCode sc = factory->get ( code , fun ) ;

if ( sc.isFailure() ) { .... } // handle the error 

In a similar way one can create predicate (the functor with returns bool):

using namespace LoKi::Types ;
using namespace LoKi::Cuts   ;

// initialize the predicate with *some* values 

Cut cut = NONE ;

// get the code 
const std::string& code = .... ; 

// instantiate the functor:
StatusCode sc = factory->get ( code , cut ) ;

if ( sc.isFailure() ) { .... } // handle the error 

Finally one can operate with fun and cut objects as with regular LoKi functors.

The complete example is available here.

-- PatrickKoppenburg - 28 Mar 2008 -- PatrickKoppenburg - 25 Apr 2008 -- Karol Hennessy - 15 May 2008 -- Vanya Belyaev - 12 Jun 2008 -- PatrickKoppenburg - 18 Aug 2008