source: trunk/MagicSoft/Mars/Changelog@ 3203

Please  do  not write behind the end of line tag so that it is  possible
to   send  a   Changelog   file  to  a  line  printer  directly.  Thanks.


                                                 -*-*- END OF LINE -*-*-
 2004/02/16: Markus Gaug

   * mcalib/Makefile, mcalib/CalibLinkDef.h:
     - added MExtractBlindPixel and MExtractedSignalBlindPixel

   * mcalib/MExtractBlindPixel.[h,cc]
     - new signal extractor for the Blind Pixel

   * mcalib/MExtractedSignalBlindPixel.[h,cc]
     - new container for the extracted signal of the BlindPixel

   * mcalib/MCalibrationCalc.[h,cc]
     - does not extract signal from Blind Pixel any more, searches 
       for MExtractedSignalBlindPixel and fills the hists from there
  
   * macros/calibration.C
   * macros/bootcampstandardanalysis.C
   * macros/pedphotcalc.C
     - introduce the blind pixel and PIN diode extraction in the macro

   * manalysis/MPedestalCam.h 
     - changed type of fTotalEntries from ULong_t to UInt_t 
     - fixed documentation of fTotalEntries

   * manalysis/MPedCalcPedRun.h 
     - changed type of fNumSamplesTot from ULong_t to UInt_t 

   * mcalib/MHGausEvents.h
     - changed type of fCurrentSize from ULong_t to UInt_t 
    
   * mcalib/MCalibrationConfig.h
     - removed gkConversionHiLo and gkConversionHiLoErr which is now
       directly in the class MCalibrationPix 



 2004/02/16: Thomas Bretz

   * manalysis/AnalysisLinkDef.h:
     - added missing MCerPhotEvtIter

   * merpp.cc:
     - added possible usage of database

   * star.cc:
     - moved source code into new class MJStar - preliminary

   * manalysis/MCerPhotPix.cc:
     - updated class header
     - fixed Print-output

   * mbase/MAGIC.h:
     - removed definition of MLog

   * mbase/MArgs.h:
     - removed include of MAGIC.h

   * mbase/MLog.h:
     - added extern definition of gLog

   * mbase/MParContainer.[h,cc]:
     - move body of constructors to source file

   * mcalib/MCalibrate.cc, mcalib/MCalibrateData.cc,
     mraw/MRawCrateArray.cc, mraw/MRawCrateData.cc,
     mraw/MRawEvtData.cc, mraw/MRawEvtHeader.cc, mraw/MRawRead.cc,
     mraw/MRawEvtPixelIter.cc, mraw/MRawFileWrite.cc:
     - fixed documentation header

   * mcalib/MMcCalibrationCalc.[h,cc]:
     - fixed documentation header
     - fixed usage of FindCreateObj and FindObject
     - removed data member to count entries (use histogram instead)
     - use TH1 function Integral instead of doing it manually

   * mfilter/MFCosmics.cc:
     - fixed documentation

   * mimage/MHillas.cc, mimage/MImgCleanStd.cc,
     mimage/MNewImagePar.[h,cc]:
     - replaced for-loops by Iterators

   * mraw/MRawFileRead.cc, mraw/MRawSocketRead.cc:
     - fixed documentation header
     - check return value of MRawRunHeader::ReadEvt

   * mraw/mraw/MRawRunHeader.[h,cc]:
     - fixed documentation header
     - added return value to MRawRunHeader::ReadEvt
     - added constant number for max format version - preliminary 
       (should by a static data member)

   * mcalib/MHCalibrationPINDiode.cc:
     - do not delete pointers in the destructor which are already
       deleted in the base class
     - reset histograms in Clear()

   * mcalib/MHCalibrationPixel.cc:
     - after deletion of instances in Clear() initialize pointer to NULL

   * mcalib/MCalibrationCalc.cc, mcalib/MHCalibrationBlindPixel.cc:
     - fixed usage of logging stream output levels

   * manalysis/MPedCalcPedRun.[h,cc]:
     - removed obsolete fSignals

     - removed obsolete fNumPixels

   * manalysis/MPedestalCam.h:
     - added '->' to fArray

   * mfileio/MWriteRootFile.cc:
     - changed some output to make debugging easier

   * mjobs/MJCalibration.cc:
     - fixed a bug in drawing

   * mjobs/MJExtractSignal.cc:
     - removed MPedestalCam from MWriteRootFile output.
       When creating the branch it crashes - reason unknown I'll
       investigate further



 2004/02/16: Markus Gaug

   * mcalib/Makefile, mcalib/CalibLinkDef.h, macros/calibration.C:
     - added MHCalibrationRelTimePix

   * mcalib/MHCalibrationRelTimePix.[h,cc],
     mcalib/MHCalibrationRelTimeCam.[h,cc]:
     - new classes to fit and display the relative arrival times. 
     - filled from MArrivalTime 
     - need: MFillH("MHCalibrationRelTimeCam","MArrivalTime")

   * manalysis/MCalibrationCalc.cc
     - take out all uncommented code referring to the relative arrival time

   * manalysis/MArrivalTime.h 
     - changed function GetNumPixels to GetSize

   * macros/bootcampstandardanalysis.C, macros/pedphotcalc.C
     - included cosmics rejection which was in MCalibrationCalc before



 2004/02/14: Markus Gaug

   * mcalib/Makefile, mcalib/CalibLinkDef.h:
     - added MHCalibrationChargePINDiode

   * mcalib/MHCalibrationChargePINDiode.[h,cc]
     - new class to fill the PIN Diode histograms. Derives from 
       MHCalibrationChargePix and is filled with MFillH from the 
       container MExtractedSignalPINDiode. 
       Will replace MHCalibrationPINDiode entirely soon.

   * mcalib/MHCalibrationChargePix.[h,cc]
     - new base class for the calibration charges histogram pixel. 
       Derives from MH and can be filled with the MH with the MH 
       functions. 

   * mcalib/MExtractPINDiode.[h,cc]
     - new signal extractor for the PIN Diode

   * mcalib/MExtractedSignalPINDiode.[h,cc]
     - new container for the extracted signal of the PIN Diode

   * mcalib/MCalibrationCalc.[h,cc]
     - remove the PIN Diode signal extraction 

   * manalysis/MHPedestalPix.[h,cc]
     - new version of MHPedestalPixel, deriving from MHGausEvents.

   * manalysis/MHPedestalPixel.[h,cc]
     - old version removed, since obsolete (and erroneous)

   * manalysis/MHPedestalCam.[h,cc]
     - histogramming part of MPedestalCam now in own class 

   * manalysis/MPedestalPix.h
     - removed include of MHPedestalPixel

   * macros/calibration.C 
     - changed pedestal displays accordingly

   * mjobs/MJPedestal.h
     - class TCanvas included (did not compile otherwise)

   * mhbase/MH.cc
     - included call to FindGoodLimits(..) in CutEdges()
     - axe from pointer to reference
     - "CutEdges" renamed to "StripZeros"
 
   * manalysis/MHPedestalPixel.cc, mcalib/MHCalibrationPixel.cc,
     mcalib/MHCalibrationBlindPixel.cc:
     - "CutEdges" renamed to "StripZeros"

   * mcalib/MHGausEvent.[h,cc], mcalib/Makefile, mcalib/CalibLinkDef.h:
     - replaced by the improved version: MHGausEvents.[h,cc]



 2004/02/14: Thomas Bretz

   * manalysis/MCerPhotEvt.[h,cc]:
     - added 'Iterator' facility, this will replace some for-loops
       in the near future
     
   * mbase/MTime.[h,cc]:
     - added a more powerfull interface to get and interprete the
       MTime contents as string
     - added a new constructor

   * mreport/MReportTrigger.h:
     - fixed GetPixContent

   * mtools/MCubicCoeff.cc, mtools/MCubicSpline.[h,cc]:
     - many small changes to simple details (like order of includes)
     - some speed improvements
     - many small simplifications
     - changed parts of the code to be more C++ like (eg Iterators
       instead of for-loops)
     - disentangles some if-cases
     - replaced some math.h function by TMath::
     - removed data-member fN (obsolete with iterators)



 2004/02/13: Wolfgang Wittek

   * mcalib/MCalibrateData.[h,cc]
     - new class; 
       like MCalibrate, but in ReInit it fills the MPedPhot container
       using informations from MPedestal, MExtracteSignalCam and 
       MCalibrationCam

   * mcalib/Makefile, mcalib/CalibLinkDef.h
     - added MCalibrateData

   * manalysis/MPedestalWorkaround.[h,cc]
     - create MMcEvt container and fill with theta = 10.0
     - set pedestalRMS = 0. if it is > 10.0

   * manalysis/MSupercuts.[h,cc]
     - change default values of parameters

   * manalysis/MSupercutsCalc.[h,cc]
     - change dNOMLOGSIZE from 4.1 to 5.0

   * mhist/MHBlindPixels.[h,cc]
     - change MPedestalCam to MPedPhotCam

   * mhist/MHSigmaTheta.[h,cc]
     - automatic binning for pixel number doesn't work
       use default binning



 2004/02/13: Markus Gaug

   * mcalib/MCalibrationPix.[h,cc]
     - added variables gkAverageQE, gkAverageQEErr, and fAverageQE, 
       fAverageQEErr
     - gkAverageQE initializes to 0.2 (according to David)
     - gkAverageQEErr initializes to 0.02 (dito)
     - added new member function: SetAverageQE
     - call to GetMeanConversionFFactorMethod returns: 
       fMeanConversionFFactorMethod/fAverageQE. 
     - call to GetErrorConversionFFactorMethod returns: 
       errors of fMeanConversionFFactorMethod and fAverageQEErr added 
       quadratically

   * mcalib/Makefile
     - split into two parts one comment line 


 2004/02/13: Thomas Bretz

   * mbadpixels/MBadPixelsTreat.[h.cc]:
     - added for new bad pixel treatment

   * mbadpixels/Makefile, mbadpixels/BadPixelsLinkDef.h:
     - added MBadPixelsTreat

   * merpp.cc:
     - changed the name of the "DC" tree to "Currents" to make
       MReportRead work correctly with this tree

   * mbase/MTime.[h,cc]:
     - Changed UpDMagicTime to support nanosec instead of millisec
     - exchanged Reset() by Clear() not to clear the yy/mm/dd 
       information away in an eventloop

   * mraw/MRawEvtHeader.[h,cc]:
     - implemented decoding of the time-stamp for raw-data files
       with version>2. Thanks to Marcos.

   * mraw/MRawRead.[h,c]:
     - pipe file version number to MRawEvtHeader::ReadEvt
     - create fake time for file-version<3

   * macros/readMagic.C:
     - changed to use MInteractiveTask

   * mbase/MPrint.cc:
     - fixed debugging level of some output
     
   * mhist/MHCamera.[h,cc]:
     - fixed handling in AddNotify (necessayr due to change of
       inheritance from MCamEvent)
     - replaced cout by gLog
     - added fNotify to list of cleanups

   * mhist/MHEvent.cc, mjobs/MJCalibration.cc,
     mjobs/MJPedestal.cc:
     - fixed usage of AddNotify

   * mfileio/MReadReports.cc:
     - use a different way to determin whether the tree is a valid tree,
       skip invalid trees
       
   * mfileio/MReadTree.cc:
     - added another sanity check at the beginning of PreProcess
     
   * mhbase/MFillH.[h,cc]:
     - added bit kCanSkip which skips MFillH in case the necessary container
       wasn't found.
     - cd into the corresponding canvas before filling the MH



 2004/02/12: Markus Gaug
 
   * mbase/MArray.[h,cc]:
     - changed name of StripZeros

   * mcalib/MHGausEvent.cc:
     - corrected call to StripZeros

   * mcalib/MCalibrationPix.[h,cc]
     - implemented dummy for combined method

   * mcalib/MCalibrate.[h,cc]
     - implemented a dummy for calibration factor 1. (requested by
       Javier Rico)
     - implemented combined method
     - check for valid methods moved to ReInit

   * mcalib/MCalibrationCalc.cc
     - removed initialization of non-used nsamples

   * mcalib/MHCalibrationPixel.[h,cc], mcalib/MCalibrationPix.[h,cc],
     mcalib/MCalibrationCalc.[h,cc]
     macros/calibration.C
     - implemented new histograms: 
       HSinglePheFADCSlices and HPedestalFADCSlices
       They are filled with the FADC slice values of either a s.phe. 
       event or a pedestal event. The decision is taken with the 
       variable: MCalibrationCalc::fBlindPixelSinglePheCut, to be set 
       with a setter 

   * mtools/MFFT.[h,cc]
     - PowerSpectrumDensity of TArrayI implemented



 2004/02/12: Thomas Bretz
 
   * mbase/MArray.[h,cc]:
     - changed argument of CutEdges from pointer to reference

   * mgeom/MGeomPix.h:
     - fixed calculation of area of pixel. It was too big for a 
       factor of 2

   * mjobs/MJCalibration.cc:
     - slight change of name of a MHCamera

   * manalysis/MExtractSignal.cc:
     - slight change to the creation of the satpixels list
     
   * mcalib/MHCalibrationBlindPixel.cc, mcalib/MHCalibrationPixel.cc,
     mcalib/MHGausEvent.cc:
     - corrected call to ProjectArray
     - corrected call to CutEdges
     
   * mfilter/MFCosmics.[h,cc]:
     - small fixes to logging output
     - small simplification to return statement
     - declared CosmicsRejection const

   * mhbase/MH.[h,cc]:
     - changed argument of ProjectArray from pointer to reference
     - added missing calcualtion of minimum
     - removed obsolete SetEntries
     - changed SetDirectory from NULL to gROOT



 2004/02/12: Javier López

   * macros/pointspreadfunction.C
     - added new macro that fits with a 2D gaussian the DC spot for an 
       star. It gives you the RMS of the PSF and the center of the star,
       very useful for misspointing studies.



 2004/02/11: Hendrik Bartko

   * manalysis/MExtractSignal2.cc:
     - introduced a validity check for the set-function of the window 
       size (even number of slices, WindowSize < NumberSlices)

   * manalysis/MArrivalTimeCalc2.[h,cc], manalysis/Makefile, 
     manalysis/AnalysisLinkDef.h:
     - new: introduced the calculation of the arrival time as the signal 
       weighted mean of the time slices in the maximum window



 2004/02/11: Markus Gaug

   * mcalib/MCalibrationCalc.cc:
     - the blind pixel signal extractor introduced one week ago, counted 
       one FADC slice too little, thus the normalization w.r.t. pedestals 
       was wrong. Now corrected.

   * manalysis/MExtractSignal.cc, manalysis/MExtractSignal2.cc:
     - when "WARNING - Lo Gain saturated" occurs, display also the
       corr. pixel numbers.

   * mbase/MArray.[h,cc]:
     - introduced function to cut array edges with zeros. 
       (Thomas, should we keep it there or move it to another class? )

   * mcalib/MCalibrationCalc.[h,cc]:
     - new function SetBlindPixelRange to set start and end of FADC slices 
       used for the blind pixel signal extraction. 
     - remove fEvents which are not necessary any more after the shift 
       to MFCosmics.

   * mcalib/MHCalibration(Blind)Pixel.[h,cc]:
     - show the fourier spectrum now with axes in frequency [Hz] instead 
       of inverted events. 

   * mcalib/MHGausEvent.[h,cc], mcalib/Makefile, mcalib/CalibLinkDef.h:
     - new base class for all kind of events with variables having a 
       Gaussian distribution. Will serve as base class for the 
       existing and new calibration histogram classes. 



 2004/02/11: Abelardo Moralejo

   * mmain/MEventDisplay.cc
     - Switched order of MMcPedestalCopy and MCerPhotCalc in tasklist:
       MMcPedestalCopy must be executed earlier!



 2004/02/10: Raquel de los Reyes

   * mhist/MHPixVsTime.[h,cc]:
     - Added a default draw function and replaced the TGraph object 
       by a pointer to a TGraph object.

   * mreport/MReportTrigger.h:
     - Now it dereives from a MCamEvent class and implemented the
       GetPixContent() and DrawPixelContent() functions.

   * maccros/CCDataCheck.C:
     - An update of the previous macro. New plots: subsystems status and
       macrocells trigger.



 2004/02/10: Markus Gaug

   * mcalib/MCalibrationCam.cc:
     - fixed documentation

   * mhbase/MH.[h,cc]:
     - new function ProjectArray

   * mcalib/MHCalibrationPixel.[h,cc],
     mcalib/MHCalibrationBlindPixel.[h,cc]:
     - use ProjectArray from MH to plot the projection of the fourier 
       spectrum

   * mfilter/MFCosmics.[h,cc], mfilter/Makefile,
     mfilter/FilterLinkDef.h:
     - new filter removing cosmics, the same as in MCalibrationCalc
       where it was removed now. 
       Call: MFCosmics cosmics; 
             MContinue cont(&cosmics); 
             tlist.AddToList(&cont);

   * mcalib/MCalibrationCalc.[h,cc]:
     - removed cosmics rejection from there
 
   * macros/calibration.C, mjobs/MJCalibration.cc:
     - changed cosmics rejection to the filter algorithm



 2004/02/09: Markus Gaug

   * mcalib/MCalibrationBlindPix.[h,cc]:
     - added histogram to display the average in FADC slices of all 
       single and double phe events

   * mcalib/MCalibrationCalc.cc:
     - fill histograms of single-phe events with higain and logain
       sampls

   * mcalib/MCalibrationCam.[h,cc], mcalib/MCalibrationCalc.cc,
     mcalib/MCalibrationConfig.h:
     - take all pixel areas out of MGeomPix::GetA(), remove all previous
       areas 
     - calculate photon fluxes from blind pixel and PIN diode, use 
       area only to fill the conversion factors in MCalibrationPix 


   * mcalib/MJCalibration.cc:
     - display of number of photons for Blindpixel and PIN Diode Method 
       changed to display of photon flux which is the actually 
       calculated quantity



 2004/02/09: Thomas Bretz

   * mbadpixels/*:
     - added

   * Makefile:
     - added MBadPixels

   * manalysis/MGeomApply.cc:
     - fixed class description

   * mjobs/MJCalibration.[h,cc], mjobs/MJExtractSignal.[h,cc]:
     - added bad pixel treatment

   * mjobs/MJPedestal.[h,cc]:
     - reorganized order of functions
     - removed a nonsense scrren output (some BLABABLA or so)

   * mjobs/Makefile:
     - added include for mbadpixels

   * mmain/MBrowser.cc:
     - do not call construtor of TGTReeLBEntry if pic0=0



 2004/02/09: Raquel de los Reyes

   * mcamera/MCameraLids.h, mreport/MReportCamera.h
     - Added the "Get" functions to access the private data 
       members of the class.



 2004/02/09: Markus Gaug

   * MMcCalibrationUpdate.cc, MCalibrate.cc, MCalibrationPix.cc,
     MCalibrationPix.h:
     - replace MCalibrationPix->IsChargeFitValid() by IsChargeValid()
       because we can calibrate with the histogram mean and RMS even 
       if the fit is not OK. Thus, there is a question: IsFitted() and 
       IsChargeValid(), i.e. does the charge make any sense? 

   * MCalibrationCam.cc, MCalibrationConfig.h, MCalibrationCalc.cc:
     - found serious bug in calculation of conversion factors for the 
       outer pixels: 
       GetMeanConversionBlindPixelMethod and 
       GetMeanConversionPINDiodeMethod 
       did not correct for the area of the outer pixel, this did only 
       MCalibrationCam::GetPixelContent which is not used by MCalibrate



 2004/02/06: Markus Gaug

   * mgeom/MGeomCam.h 
     - replaced in Comment of fPixRatio: 
       "ratio between size of pixel idx and pixel 0" by
       "ratio between size of pixel 0 and pixel idx" 

   * manalysis/MPedestalCam.cc
     - catch calls to MHPedestalPixel in case that they are not filled 
       This led previous calls to segmentation violations
     - remove include of MHExtractedSignalPix
 
   * manalysis/MHExtractedSignalPix.[h,cc]
     - is everything in MHPedestalPixel, so removed it again

   * mjobs/MJPedestal.cc
     - put the projections into Display()
     - correct the input of Display (0: pedestal, 1: pedestalerror, 
                                     2: pedRMS, 3: pedRMSerror)

   * manalysis/MHPedestalPixel.cc
     - increased number of bins in for charges plot

   * mtools/MFFT.[h,cc]
     - possibility to have PSD directly from TArray

   * mjobs/MJCalibration.[h,cc]
     - fixed displays
     - replaced MExtractSignal by MExtractSignal2

   * mcalib/MHCalibrationPixel.[h,cc], mcalib/MHCalibrationBlindPixel.[h,cc],
     mcalib/MCalibrationPix.[h,cc], mcalib/MCalibrationBlindPix.[h,cc],
     mcalib/MCalibrationPINDiode.[h,cc], mcalib/MHCalibrationPINDiode.[h,cc]:
     - remove histograms MHChargevsN..., now keep TArrays directly
     - check for oscillations for all pixels (and you will not trust 
       your eyes when you look at the results :-((((

   * mcalib/MCalibrationCalc.[h,cc]
     - remove ArrivalTime -> go to separate class
     - remove BlindPixel and PIN Diode overflow, now handled directly 
       inside the corr. classes. 
     - remove SkipTimeFits -> go the arrival time class

   * mcalib/MCalibrationCam.h
     - new getters for number of photons
     - removed timeslicewidth -> go the corr. time containers

   * mcalib/MCalibrationCam.cc
     - removed times of GetPixelContent
     - added OscillationChecks in GetPixelContent

   * manalysis/MMcCalibrationUpdate.cc
     - remove calpix.SkipTimeFits which is now obsolet

   * macros/calibration.C
     - all previous changes now in this macro.
     - (NB: Thomas, du brauchst in MJCalibration nicht mehr aendern, 
       das ging diesmal in einem Aufwasch)



 2004/02/06: Thomas Bretz

   * mhist/MHPixVsTime.[h,cc]:
     - changed such that you can set the name of the time container now 

   * mraw/MRawRunHeader.[h,cc]:
     - added GetMaxPixId()

   * mraw/MRawEvtData.cc:
     - fixed wrong size calculation for fABflags



 2004/02/06: Oscar Blanch

   * mmc/MFadcDefine.h && mmc/MMcFadcHeader.cxx
     - We cahnged the name (and values) of the global variable
        MFADC_RESPONSE_AMPLITUDE to  MFADC_RESPONSE_INTEGRAL



 2004/02/05: Markus Gaug

   * mcalib/MHCalibrationPixel.[h,cc]
     - put an NDF limit for the fit
     - removed all TStrings and replace them by Form(..)
     - Check for nan's of all returned fit values

   * mcalib/MCalibrationPix.[h,cc], mcalib/MCalibrationCam.cc:
     - reshuffled HiLoGainConversion, calculation of F-Factor method, 
     - calculation of total F-Factor of the readout.

   * mcalib/MCalibrationCam.cc
     - include relative times to display with GetPixelContent

   * mcalib/MCalibrationCalc.[h,cc]
     - flag to skip Hi-Lo Calibration
     - Hi-LO Graph not filled any more after overflow of ChargevsN
     - fill the blind pixel differently from the rest of pixels

   * mcalib/MCalibrate.cc:
     - call to const MCalibrationPix &pix replaced by MCalibrationPix &pix

   * manalysis/MArrivalTimeCalc.cc
     - default step width 100 ps. 
     - look for half maximum (at rising edge)

   * manalysis/MExtractSignal.[h,cc]
     - removed uncommented code (arrival time)
     - set start byte back by one like it was before TB's changes 
       (does not affect the calculation of number of FADC slices)

   * manalysis/MExtractSignal2.[h,cc]
     - removed uncommented code (arrival time)

   * manalysis/MHPedestalPixel.[h,cc]
     - is now filled with total pedestals (not per slice), but results are 
       renormalized to values per slice (this is not to screw up with the 
       pedestal RMS and the derived uncertainties)
     - new function Renorm(nslices)

   * manalysis/MPedestalCam.[h,cc]
     - introduced function Clone() (like proposed by TB)
     - call to MHPedestalPixel operator() not const any more
     - take out lots of uncommented code 
     - found bug in initialization of hists. SetupFill is called in 
       PreProcess, therefore, TObjArray is not yet enlarged there. 
       Now, InitUseHists is called in Fill
     - MHPedestalPixel is now filled with total pedestals (not per slice), 
        but results are 
       renormalized to values per slice (this is not to screw up with the 
       pedestal RMS and the derived uncertainties)

   * manalysis/MPedCalcPedRun.cc
     - added author
     - split calculation of RMS into two steps to make it more clear to 
       the user

   * macros/calibration.C
     - found bug introduced two days ago, namely calculating pedestlas with 
       MExtractSignal2 which is now fixed
     - Skip HiLoConversionCalibration



 2004/02/05: Abelardo Moralejo

   * macros/starmc2.C
     - added 3rd argument, "kFALSE", to call of MWriteRootFile::Add
       for both MMcEvt and MMcRunHeader since these containers are
       not present in real data. Name of macro is no longer adequate if
       we will use it for calibrated real data (to be changed if so).



 2004/02/05: Sebastian Raducci

   * mtools/MCubicSpline.[h,cc]
     - added
     - this class will take the place of the TSpline class to 
       calculate arrival times

   * mtools/MCubicCoeff.[h,cc]
     - added
     - this class is used by MCubicSpline to store coefficients

   * macros/spline.C
     - added
     - test macro. Shows how to use the MCubicSpline class
     - it could be used to test the methods of the Spline class



 2004/02/05: Thomas Bretz

   * mmain/MBrowser.cc:
     - fixed a double deletion of a TGLayoutHints



 2004/02/04: Markus Gaug

   * mmain/MBrowser.cc:
     - corrected TGTReeLBEntry for TGTreeLBEntry



 2004/02/04: Thomas Bretz
 
   * manalysis/MExtractSignal2.cc:
     - don't allow odd numbers for the window size

   * mfilter/MFilterList.cc:
     - corrected a wrong logging message

   * mmain/MBrowser.cc:
     - added '/data/MAGIC' as shortcut in the combo box
     
   * mraw/MRawEvtPixelIter.cc:
     - optimized calculation of MaxIdx



 2004/02/04: Abelardo Moralejo

  * macros/starmc2.C
    - removed unnecessary variables BinsHigh, BinsLow



 2004/02/04: Markus Gaug

  * manalysis/MPedestalCam.[h,cc], manalysis/MPedestalPix.[h,cc],
    manalysis/MPedCalcPedRun.[h,cc], manalysis/MHPedestalPixel.[h,cc],
    macros/calibration.C:
    - histograms are now filled with MFillH as proposed and coded by 
      Thomas Bretz. Some modifications had to be done, however. 
    - Class to hold MHPedestalPixel is still MPedestalCam
    - MPedCalcPedRun does nothing any more (except from what is was 
      doing before)

  * manalysis/MExtractSignal.cc
    - remove uncommented code

  * mcalib/MCalibrationCalc.[h,cc]
    - modified way to handle histogram overflow: Now flags are set and 
      ChargevsN histogram is not tried to be filled subsequently



 2004/02/03: Markus Gaug
 
   * manalysis/MHExtractedSignalPix.[h,cc]
     - added Thomas B. modified version of MHPedestalPixel. Later will 
       remove MHPedestalPixel

   * manalysis/MCalibrationPix.[h,cc], manalysis/MHCalibrationPixel.[h,cc],
     manalysis/MCalibrationCam.cc, macros/calibration.C:
     - now split completely absolute and relative times. Absolute 
       times are not fitted, but their histogram mean and rms are 
       returned.



 2004/02/03: Thomas Bretz

   * mfileio/MReadTree.cc:
     - fixed handling of -1 entries in AddFile: root 3.10/02 crashed

   * manalysis/MExtractSignal.[h,cc]:
     - moved code to extract the signal from Process to
       new member function FindSignal, I left the arrival time 
       extraction as commented-out code in the code just as a 
       demonstration to do this very fast.



 2004/02/02: Markus Gaug

  * manalysis/MPedCalcPedRun.cc
    - corrected formulae in Class description

  * mcalib/MCalibration.[h,cc], mcalib/MHCalibration.[h,cc],
    manalysis/MMcCalibrationUpdate.cc:
    - calculate relative times and absolute times. Fit only 
      the relative ones. 



 2004/02/02: Abelardo Moralejo

   * manalysis/MMcCalibrationUpdate.cc, MMcPedestalCopy.cc
     - removed file type check (MC or data) from PreProcess since now
       run headers are not yet known at that point.

   * mcalib/MMcCalibrationCalc.cc
     - same as above.

   * NEWS
     - updated.



 2004/02/02: Thomas Bretz

   * manalysis/MExtractSignal2.[h,cc]:
     - added new task which takes a given number of slices with the
       highest integral content as signal.

   * manalysis/Makefile, manalysis/AnalysisLinkDef.h:
     - added MExtractSignal2



 2004/01/30: Abelardo Moralejo

   * macros/starmc.C
     - changed default FADC integration range (more adequate for coming
       camera version).

   * macros/mccalibrate.C, starmc2.C
     - added. First macro transforms input files from camera simulation
       into files with pixel data already calibrated in photons. The 
       second reads in an already calibrated data file and produces an 
       output file with Hillas parameters. Both together do the same as
       starmc.C, but it is convenient to have files containing 
       calibrated pixel information.



 2004/01/30: Thomas Bretz
 
   * mbase/MDirIter.h:
     - added GetNumEntries

   * mhist/MHEvent.cc:
     - added a dynamic_cast from MParContainer to MCamEvent to get the
       correct pointer to the correct base-class.

   * mbase/MStatusDisplay.cc:
     - In EventInfo check whether gPad==0
     
   * mfileio/MReadMarsFile.cc:
     - removed reading RunHeaders from PreProcess, this resulted
       in strange behaviour of writing RunHeaders
     - fixed consistency check for RunType accordingly

   * mfilter/MFilterList.cc:
     - do not skip adding filter if only a filter with the same name 
       exists
       
   * mraw/MRawFileRead.cc:
     - added an output message if run header is not accessible
     
   * mraw/MRawRunHeader.[h,cc]:
     - initialize fRunType to new enum kRTNone

   * mfilter/MFGeomag.cc:
     - set fResult to kFALSE at the beginning of Process
     - according to this removed setting fResult=kFALSE before return
     - replaced some float by Float_t
     - added some const-qualifiers
     - replaced (rig<0.5/pr*R*(r2-r1)+r1) by (rig-r1)*pr<rnd
       with rnd = R * (r2-r1)/2 to make sure that we cannot
       devide by 0

   * manalysis/MPedCalcPedRun.cc:
     - added some comments

   * Makefile:
     - added mjobs

   * mjobs/*:
     - added

   * mbase/Makefile, mbase/BaseLinkDef.h:
     - added MRunIter

   * mbase/MRunIter.[h,cc]:
     - added



 2004/01/30: Markus Gaug

   * manalysis/MPedestalCam.h
     - added function IsUsedHists()

   * manalysis/MPedestalCam.cc
     - modified GetPixelContent

   * manalysis/MPedestalCam.cc, manalysis/MPedCalcPedRun.cc:
     - include storage of number of used FADC slices to compare later 
       the pedestal per slice ( and sigma per slice)
     - return relative differences in GetPixelContent

   * mcalib/MCalibrationCalc.[h,cc], mcalib/MHCalibrationBlindPixel.[h,cc]:
     - fBlindPixelId and fPINDiodeId now like in PedestalCam
     - fill Blind Pixel with information about its fitted pedestal if 
       available
     - initialize charges histogram with a bigger range
     - use pedestal information to constrain the pedestal fit ranges

   * manalysis/MHPedestalPixel.cc
     - initialize charges histogram with a bigger range

   * macros/calibration.C
     - SetRange of MExtractSignal to 1,14 (see emails of yesterday)
     - apply new GetPixelContents of MPedestalCam (more calculations 
       and displays of errors, correction for different FADC slice 
       number)

   * manalysis/MExtractSignal.cc
     - not access to MArrivalTime any more. The same information is 
       stored in MRawEvt->GetPixelContent(...)

   * manalysis/MCalibrationCalc.cc
     - retrieve default arrival time from MRawEvt 

   * manalysis/MCalibrationBlindPix.cc, manalysis/MCalibrationPix.cc
     - check for NaN's in fit results



 2004/01/29: Abelardo Moralejo

   * mfileio/MWriteFile.h
     - added missing declaration of ReInit

   * manalysis/MMcCalibrationUpdate.cc
     - added call to MCalibrationCam::SetBlindPixelMethodValid(kTRUE) 
       to adapt to recent changes in MCalibrate.



 2004/01/29: Thomas Bretz

   * manalysis/MArrivalTime.h, manalysis/MCameraData.h,
     manalysis/MCerPhotEvt.h, manalysis/MExtractSignalCam.h,
     manalysis/MPedPhotCam.h, manalysis/MPedestalCam.h,
     mcalib/MCalibrationCam.h, mcamera/MCameraDC.h,
     mraw/MRawEvtData.h, mreflector/MRflEvtData.h:
     - split derivement into MParContainer and MCamEvent

   * manalysis/MExtractSignalCam.[h,cc]:
     - removed obsolete MExtractSignal::ReInit
     - added -> to fArray

   * mbase/MParList.h:
     - added a new bit kIsProcessing
     
   * mbase/MTaskList.cc:
     - fixed the handling of kDoNotReset and the bahaviour of
       SetReadyToSave, such that it is possible to write all
       headers 

   * mfileio/MWriteFile.cc:
     - added ReInit which also calls CheckAndWrite
     
   * mgui/MCamEvent.h:
     - does not derive from MParContainer anymore



 2004/01/29: Markus Gaug

   * mcalib/MCalibrationCalc.[h,cc], mcalib/MCalibrationCam.[h,cc], 
     mcalib/MCalibrationConfig.h
     - put gkCalibrationBlindPixelId and gkCalibrationPINDiodeId 
       directly into the corresponding classes.



 2004/01/29: Nicola Galante

   * mhistmc/MHMcTriggerLvl2.[h,cc]:
     - In method MHMcTriggerLvl2::GetHistByName(const TString name) 
       added string checking to return also 
       fHistCellPseudoSize, fHistCellPseudoSizeNorm
       and fHistCellPseudoSizeEnergy.
     - Added inline functions GetHistCellPseudoSize() and 
       GetHistCellPseudoSizeNorm().



 2004/01/28: Javier Rico

   * mcalib/MCalibrate.[h.cc]
     - include possibility to select the calibration method to be used



 2004/01/28: Markus Gaug

   * mcalib/MCalibrate.[h.cc]
     - corrected Int_t ReInit() to Bool_t ReInit(MParList *pList)

   * mcalib/MCalibrationCalc.[h,cc], mcalib/MCalibrationCam.[h,cc]:
     - included flags to indicate if Blind Pixel Method or PIn Diode 
       method was OK. Up to now, program stopped, if one of the two 
       methods was chosen but not able to being executed.

   * manalysis/MArrivalTimeCalc.cc, manalysis/MExtractSignals.cc,
     manalysis/MGeomApply.cc:
     - put all missing InitSize(npixels) into MGeomApply         

   * manalysis/MPedCalcPedRun.[h,cc], manalysis/MPedestalCam.[h,cc],
     manalysis/MPedestalPix.[h,cc], manalysis/MHPedestalPixel.[h,cc],
     manalysis/Makefile, manalysis/AnalysisLinkDef.h:
     - include possibility to fit the pedestals with Gaussian. 
     - old methods are not affected. 
     - add new method with MPedCalcPedRun::SetUseHists()



 2004/01/28: Abelardo Moralejo

   * mcalib/MMcCalibration.cc
     - fixed wrong variable type in call to TH1F constructor (number of
       bins is now integer), which resulted in a compilation warning.



 2004/01/28: Thomas Bretz

   * manalysis/MPedestalCam.cc:
     - removed unwanted CheckBound from operators. CheckBounds is ONLY
       ment for really old MC files and will be removed in the future
       anyhow.

   * mcalib/MHCalibrationPixel.h:
     - removed obsolete include of MFFT 

   * manalysis/MCerPhotAnal.cc, manalysis/MCerPhotCalc.cc,
     manalysis/MPedestalCam.[h,cc]:
     - removed use of obsolete MPedestalCam::CheckBounds. This was
       a workaround for a 'bug' in very old camera files, which is not
       necessary anymore.

   * manalysis/MCerPhotAnal2.[h,cc]:
     - replaced MPedestalCam by MPedPhotCam
     
   * manalysis/MExtractSignal.cc:
     - some small simplification to the signal extraction
     - removed some obsolete variables
     - replaced for by while
     - removed obsolete casts
     
   * mmain/MEventDisplay.cc:
     - gave names to the filters
     - removed adding MHCamera to fList because it was deleted twice

   * mhist/MHCamera.cc:
     - fixed usage of fPainter in Paint() - title was not updated 
       correctly



 2004/01/28: Sebastian Raducci

   * manalysis/MArrivalTime.[h,cc]:
     - removed all the code relative to the cluster evaluation



 2004/01/27: Markus Gaug
 
   * manalysis/MPedestalPix.cc:
     - initialize default values to 0 instead of -1.

   * manalysis/MPedestalCam.[h,cc]:
     - include CheckBounds, also in operator[]

   * mcalib/MCalibrationCam.[h,cc]:
     - apparently, there was a reference to InitSize(UInt_t i) instead 
       of InitSize(Int_t i), where compiler complained, now changed
 
   * mcalib/MHCalibration*, mcalib/MCalibration*:
     - fixed all Clear()'s, Reset()'s, direct intializers



 2004/01/27: Thomas Bretz

   * mdata/MDataChain.cc:
     - added '%' to supported operators



 2004/01/26: Nicola Galante

   * manalysis/MMcTriggerLvl2.[cc,h]:

     - Defined some static constants to avoid ugly "magic numbers"
       in arrays: gsNCells, gsNTrigPixels,
       gsNPixInCell, gsNLutInCell, gsNPixInLut, fNumPixCell.
     - Added method 
       MMcTriggerLvl2::GetCellCompactPixel(int cell, MGeomCam *fCam)
       which computes compact pixels into a given L2T macrocell.
     - Added method 
       MMcTriggerLvl2::CalcBiggerCellPseudoSize() 
       which computes fCellPseudoSize, the maximum Pseudo Size into L2T 
       macrocells
     - Added method 
       MMcTriggerLvl2::GetCellPseudoSize() const 
       which returns fCellPseudoSize
     - Added method 
       MMcTriggerLvl2::IsPixelInCell(Int_t pixel, Int_t cell),
       which controls whether a pixel belongs to a given L2T cell.
     - Added method 
       MMcTriggerLvl2::GetMaxCell() const 
       which returns fMaxCell, the cell with the maximum 
       fCellPseudoSize.



 2004/01/26: Markus Gaug, Michele Doro

   * manalysis/MArrivalTime.[h,cc], manalysis/MArrivalTimeCalc.[h,cc]:
     - moved calculation of arrival time from MArrivalTime to 
       MArrivalTimeCalc. MArrivalTime does not do any Process-job
       anymore
     - TSpline5 now on stack. 
     - function to set the stepsize

   * mcalib/MHCalibrationBlindPixel.[h,cc]:
     - force mu_{0} in Blind Pixel Fit to be around 0 in fKPoisson4
     - implement combined Polya fit and Michele's back-scattered electron
       fit 

   * mcalib/MHCalibrationPixel.[h,cc], mcalib/MCalibrationPix.cc,
     mcalib/MCalibrationPINDiode.cc:
     - combine FitChargeHiGain and FitChargeLoGain into one function

   * macros/calibration.C
     - improve double gauss fit a bit



 2004/01/26: Thomas Bretz

   * mraw/MRawEvtPixelIter.[h,cc]:
     - removed member function GetNumPixels. For more details see mantis.

   * manalysis/MArrivalTime.[h,cc]:
     - reorganized includes
     - fixed InitSize and ClearArray (now Reset) to fit MParContainer
       definition
     - for the moment removed usage of fPixelChecked - THIS WILL CRASH YOUR
       PROGRAM

   * manalysis/MArrivalTimeCalc.[h,cc]:
     - reorganized includes
     - removed many obsolete code, exspecially copy&paste relicts!
     - fixed usage of InitSize
     - removed CleanArray. Reset() is called automatically before Process()

   * manalysis/MExtractedSignalCam.[h,cc]:
     - added SetLogStream
     - removed obsolete include MGeomCam
     - replaced PrintOut by Print
     - added const-qualifiers to all Getters
     - fixed documentation
     
   * manalysis/MExtractedSignalPix.[h,cc]:
     - replaced PrintOut by Print
     - fixed documentation

   * manalysis/MEventRateCalc.cc, manalysis/MPedCalcPedRun.cc:
     - replaced memset by Reset()



 2004/01/24: Abelardo Moralejo

   * macros/starmcstereo.C
     - Added. Example of how to run the analysis chain for MC files
       containing simulation of stereo systems of 2 telescopes.

   * mcalib/MCalibrate.cc
     - Added check that value of signal from MExtractedSignalPix is not
       the default (error) value. In such a case, the signal in 
       MCerPhotEvt is now set to 0. Before, all pixels got very high
       values in MCerPhotEvt in case the MRawEvtData object for the
       event was missing in the file. This happened in MC files, in 
       which in some circumstances an event is written without the
       pixel information (for instance, when running stereo simulations
       the pixel data is missing for non-triggered telescopes, but the
       event headers are present).



 2004/01/23: Markus Gaug

   * manalysis/MExtractSignal.[h,cc], manalysis/MExtractedSignalPix.[h,cc],
     manalysis/MArrivalTime.[h,cc], mcalib/MCalibrationBlindPix.h,
     mcalib/MHCalibrationBlindPixel.[h,cc], mcalib/MCalibrationPINDiode.h,
     mcalib/MHCalibrationPINDiode.[h,cc], mcalib/MCalibrationPix.[h,cc],
     mcalib/MHCalibrationPixel.[h,cc], mcalib/MCalibrationCalc.[h,cc]:
     - use MArrivalTime for Storage (or calculation) of time
       in MExtractSignal
     - pass from TH1I histos to TH1F for times

   * mcalib/MCalibrationBlindPix.h, mcalib/MHCalibrationBlindPixel.[h,cc],
     mcalib/MCalibrationPix.[h,cc], mcalib/MHCalibrationPixel.[h,cc],
     mcalib/MCalibrationCalc.[h,cc]:
     - clean up  setters 
     - pass to bit-fields for all flags

   * mcalib/MCalibrationCam.cc
     - exclude Blind Pixel from Camera display
     - do not fill &val when return kFALSE

   * macros/calibration.C
     - include functions to exclude the blind pixel, 
       calculation of times and quality checks



 2004/01/23: Abelardo Moralejo

   * mcalib/MCalibrate.cc, manalysis/MExtractSignal.cc
     - Added support (using AddSerialNumber) for stereo MC files.



 2004/01/23: Thomas Bretz

   * mbase/MAstro.cc:
     - Fixed calculation of MJD. In Ymd2Mjd a plus should be a minus.

   * mreport/MReport*.[h,cc]:
     - changed all classes to return Int_t instead of Bool_t in 
       InterpreteBody
     - changed MReportFileRead such, that a wrong line results in
       skipping the line instead of abort reading.

   * manalysis/MExtractSignal.cc:
     - do not print information for each pixel - it's too much.
     
   * mbase/MStatusDisplay.cc:
     - when reading a status display do not display tabs with
       the same name than one which is already existing.
       
   * mcalib/MCalibrate.cc:
     - directly store pointer to pixel when setting saturation flag
     
   * mcamera/MCameraDC.cc:
     - fixed some comments
     
   * mhist/MHCamEvent.cc:
     - fixed a crash when no geometry was available

   * mhist/MHCamera.[h,cc]:
     - removed obsolete member function SetPix

   * mbase/MStatusDisplay.[h,cc]:
     - added EventInfo member function
     - implemented preliminary redirection of GetObjectInfo to
       StatusLine2 -> An own status line for this is missing

   * manalysis/MExtractSignal.[h,cc]:
     - implemented StreamPrimitive
     - changed deafults to be static const data members
     
   * manalysis/MGeomApply.[h,cc]:
     - implemented StreamPrimitive



 2004/01/23: Abelardo Moralejo

   * manalysis/MMcCalibrationUpdate.[h,cc] 
     - Added check to guarantee that the FADC simulation parameters of
       all read files (those used for calibration and those analyzed) 
       are the same.
     - Added missing support for analysis of stereo files.

   * mcalib/MMcCalibrationCalc.cc
     - Removed obsolete FIXME comment.
     - Added missing support for analysis of stereo files.


 2004/01/23: Raquel de los Reyes

   * macros/CCDataCheck.C
     - Added the temperature distribution of the optical links
       and the plots of the LV power supplies.
   * mcamera/MCameraLV.h
     - Added the "Get" functions to access the private data 
       members of the class.


 2004/01/22: Abelardo Moralejo

   * manalysis/MMcCalibrationUpdate.[h,cc]
     - Now ratio of high to low gain is taken from MCalibrationCam if
       it existed previously in the parameter list, instead of being
       read again from the MMcFadcHeader. Removed Setter function for
       fADC2PhInner, no longer necessary. Fixed error regarding the 
       pedestal conversion to photons (did not read conversion factor
       from preexisting MCalibrationCam object).

   * mcalib/MMcCalibrationCalc.cc
     - Changed parameters of the histogram, and also the quantity being 
       histogrammed. Check that input data come from a noiseless camera
       file before proceeding to do the calibration. Introduced lower
       size in cut for calibration. Now rhe calibration constant is not
       calculated from the mean of photons/ADC counts, but from the peak
       of the histogram.

   * macros/starmc.C
     - Introduced new scheme. Now there are two loops over two different
       sets of files. First loop calculates the calibration constants,
       second one does the analysis. Introduced comments. Now the 
       histogram used in the light calibration is written to the output
       file.



 2004/01/22: Thomas Bretz

   * mcalib/MCalibrate.[h,cc], mcalib/MCalibrationCam.[h,cc],
     mcalib/MCalibrationBlindPix.[h,cc], mcalib/MCalibrationCalc.[h,cc],
     mcalib/MCalibrationPix.[h,cc], mcalib/MCalibrationPINDiode.[h,cc],
     mcalib/MCalibrationBlindPixel.[h,cc], 
     mcalib/MHCalibrationPixel.[h,cc], mcalib/MMcCalibrationCalc.[h,cc],
     mcalib/MHCalibrationBlindPixel.[h,cc], 
     mcalib/MHCalibrationPINDiode.[h,cc]:
     - cleaned up includes



 2004/01/22: Raquel de los Reyes

   * macros/CCDataCheck.C
     - macro to check the data from the central control (.rep files).



 2004/01/22: Markus Gaug

   * macros/calibration.C
     - made MStatusDisplay 3 bit smaller

   * manalysis/MExtractSignal.cc
     - put fSignals->SetUsedFADCSlices into PreProcess

   * manalysis/MPedCalcPedRun.cc
     - calculate PedRMS as RMS of the sums, later renormalized to units 
       of RMS/slice (eliminate the wrong effect of odd and even slices)

   * mcalib/MCalibrationCalc.cc
     - if bit SkipBlindPixelFit is set, now number of photons outside 
       plexiglass is not intended to be calculated any more (previous
       bug)
     - exclude excluded pixels in eventloop
        
   * mcalib/MHCalibrationBlindPixel.[h,cc]
     - move functions to pointer again (otherwise, Clone will give 
       strange results)



 2004/01/21: Abelardo Moralejo

   * mcalib/MMcCalibrationCalc.[h,cc]
     - Added. First version of the task intended to calculate the
       conversion factors from ADC counts to photons in the MC. It has
       a different aim than the recently renamed class 
       "MMcCalibrationUpdate" (which takes care that the correct 
       pedestals and conversion factors are supplied to each event, but 
       does not really calculate anything).

   * mcalib/Makefile, CalibLinkDef.h
     - Added new class MMcCalibrationCalc

   * manalysis/MMcCalibrationUpdate.[h,cc]
     - Now the container MCalibrationCam is not created or modified if
       it is found that one such container already exists in the 
       Parameter list (from a previous calibration).



 2004/01/21: Thomas Bretz

   * mraw/MRawRead.cc:
     - restore the fake time for all runs (current time stamp in
       the data seems to be toally nonsense)
 
   * mraw/MRawEvtHeader.cc:
     - removed decoding of real time (current time stamp in the data
       seems to be toally nonsense)



 2004/01/21: Abelardo Moralejo

   * mmc/MMcEvt.hxx
     - Uncommented 6 Getter functions (those regarding the number of 
       photons surviving the different stages of simulation).

   * manalysis/MMcCalibrationUpdate.[h,cc]
     - Added (see below).

   * manalysis/MMcCalibrationCalc.[h,cc]
     - Removed: changed name of this class to MMcCalibrationUpdate, 
       which is more representative of what it really does. Moved 
       the filling of the MCalibrationCam container from Process to 
       ReInit, since it does not change on an event by event basis.
       Simplified PreProcess (MCalibrationCam was searched for twice).
       Updated class description and comments.

   * manalysis/Makefile, AnalysisLinkDef.h
     - Adapted to change above

   * macros/starmc.C
     - Adapted to change above. Changed also task instance name from 
       mccalibcalc to mccalibupdate.



 2004/01/21: Raquel de los Reyes

   * mreport/MReportCC.h, mreport/MReportTrigger.h, mcamera/MCameraHV.h 
     and mcamera/MCameraCalibration.h
     - Added the "Get" functions (e.g GetStatus() for fStatus) to access 
       the private data members of the class.



 2004/01/20: Thomas Hengstebeck

   * mranforest/MRanForestCalc.[h,cc]
     - Added member functions Grow (training of RF) and Fill (reading in
       trained forest from file) which simplify macros. 
       One just needs to call them instead of using MRanForestGrow and 
       MRanForestFill (and the related training and fill loops) in a
       macro.



 2004/01/20: Abelardo moralejo

   * manalysis/MMcCalibrationCalc.cc
     - removed call to MCalibrationPix::SetPedestal, which was not 
       needed.



 2004/01/20: Thomas Bretz

   * Makefile:
     - replaced 'make' by '$(MAKE)'
     
   * Makefile.rules:
     - added Makefile itself to dependencies
     
   * mbase/MTask.cc:
     - added comment about ReInit

   * mhbase/MH.cc:
     - do not clear pad in DrawClone if 'same' option given

   * mhist/MHVsTime.cc:
     - evaluate 'same' option

   * mfileio/MReadCurrents.[h,cc]:
     - obsolete, removed

   * manalysis/MCerPhotEvt.cc, manalysis/MCerPhotPix.cc:
     - added some comments



 2004/01/20: Abelardo moralejo

  * mcalib/MHCalibrationPINDiode.cc
    - Added call to SetDirecory(0) for histograms fHErrCharge, 
      fHPCharge, fHPTime.

  * manalysis/MCerPhotPix.h: 
    - Changed position of declaration of fIsSaturated to be the last 
      private variable. Otherwise, since in the constructor it is 
      initialized in the last place, a warning is displayed upon
      compilation in some systems.



 2004/01/20: Markus Gaug
  * mcalib/Makefile, manalysis/Makefile:
    - include directory ../mtools
 
  * mtools/MFFT.[h,cc]
    - class inherits from TObject now



 2004/01/19: Markus Gaug

  * mcalib/MHCalibrationBlindPixel.[h,cc]
    - fixed a bug in fFitLegend due to which program crashed by calls 
      to DrawClone
    - Modified way to change the fit function

  * mcalib/MHCalibrationPixel.[h,cc]
    - reordered function calls
    - removed SetupFill

  * mcalib/MHCalibrationPINDiode.h
    - reordered function calls

  * mcalib/MCalibrationPix.[h,cc]
    - limits to define fFitValid now as variables in class

  * mcalib/MCalibrationCam.[h,cc]
    - reordered function calls
    - incorporate option to exclude pixels 

  * mcalib/MCalibrationBlindPix.h
    - Modified way to change the fit function

  * mcalib/MCalibrationCalc.[h,cc]
    - Modified way to change the fit function
    - incorporate option to exclude pixels from configuration file

  * macros/calibration.C
    - Modified way to change the fit function
    - incorporate option to exclude pixels from configuration file



 2004/01/19: Javier Rico

   * manalysis/MPedCalcPedRun.cc, manalysis/MPedPhotCalc.cc
     - Added documentation of member functions



 2004/01/19: Abelardo Moralejo

   * manalysis/MCerPhotPix.[h,cc]
     - Added fIsSaturated to flag whether low gain of the pixel is
       saturated. Added corresponding getter and setter functions.
       Updated ClassDef to version 2.

   * mcalib/MCalibrate.cc
     - Set the new saturation flag of MCerPhotPix to kTRUE if at least 
       one low gain FADC slice is saturated.

   * mimage/MNewImagePar.[h,cc]
     - Added new member variable fNumSaturatedPixels to count how many
       pixels were saturated. Updated ClassDef to version 2.



 2004/01/19: Thomas Bretz

   * manalysis/MPedPhotCalc.[cc,h]:
     - fixed order of includes
     - removed obsolete forward declarations
     - removed obsolete data member fNumPixels (stored already twice in
       the two TArrayF)
     - fixed some small bugs in the documentation

   * manalysis/MPedCalcPedRun.cc:
     - updated my eMail address



 2004/01/19: Javier Rico

   * macros/dohtml.C, NEWS
     - include some missing info



 2004/01/19: Abelardo Moralejo

   * manalysis/MExtractedSignalPix.h
     - added GetNumLoGainSaturated()



 2004/01/19: Markus Gaug

   * macros/dohtml.C
     - include calibration.C
        
   * manalysis/MExtractSignal.cc
     - warning of logain saturation now only once per event (out of loop)

   * mcalib/MCalibrationCalc.cc, mcalib/MCalibrationPix.[h,cc]
     - new Flag fExcluded to indicate if pixel has been excluded from 
       further analysis



 2004/01/16: Javier Rico & Pepe Flix

   * manalysis/MPedPhotCalc.[cc,h]
     - added
     - new class for the evaluation of pedestals in units of photons
       using the extracted signal from pedestal runs and calibration
       constants defined in MCalibrate

   * manalysis/AnalysisLinkDef.h,  manalysis/Makefile
     - added entries for MPedPhotCalc

   * macros/pedphotcalc.C
     - added
     - example on how to use MPedPhotCalc task



 2004/01/16: Markus Gaug

   * mcalib/MCalibrationFits.h 
     - removed and incorporated in MHCalibrationBlindPixel.h

   * mcalib/MHCalibrationBlindPixel.[h,cc], mcalib/MHCalibrationConfig.h,
     mcalib/MCalibrationBlindPix.h:
     - incorporate Fit functions
     - fixed a bug due to which DrawClone crashed when class was used 
       in a compiled macro

   * mcalib/MCalibrationCalc.cc, mcalib/MCalibrationPix.cc,
     mcalib/MCalibrationBlindPix.cc, mcalib/MCalibrationPINDiode.cc,
     mcalib/MCalibrationCam.cc, mcalib/MHCalibrationBlindPixel.cc,
     mcalib/MHCalibrationPixel.cc:
     - fixed the warns and errs in fLog's



 2004/01/16: Abelardo Moralejo

   * manalysis/MMcCalibrationCalc.cc
     - TMath::Max() is no longer used because it did not compile in some 
       systems for unkown reasons. I replaced it by a safer (and may be 
       slower) code.
     - Added calpix.SetFitValid();

   * mhist/MHEvent.cc
     - In MHEvent::Fill, fixed: second parameter of SetCamContent must 
       be 1 in case of RMS:
        case kEvtPedestalRMS:
           fHist->SetCamContent(*(MCamEvent*)par, 1);



 2004/01/15: Abelardo Moralejo

   * macros/starmc.C
     - Added. Version of star showing new chain to get Hillas 
       parameters (with Size calibrated in photons) from MC files.



 2004/01/15: Javier Rico
        
   * manalysis/MPedCalcPedRun.[h,cc]
     - optimize the running time
     - add (some) documentation
     - correct treatment for the case of several input files 

   * macros/pedvsevent.C
     - added
     - draw pedestal mean and rms vs event# for input pixel# and run 
       file, and compares them to the global pedestal mean and rms 



 2004/01/15: Raquel de los Reyes

   * mhist/MHCamEvent.cc, mhist/MHCamera.cc
     - changed the MHCamEvent::SetupFill and MHCamera::GetBinError
       functions to compute the estimation of the error of the
       mean. It can be done calling to the function
       MCamEvent::SetBit(MHCamera::kVariance) in the macro



 2004/01/15: Markus Gaug

   * mhist/MHCamera.[h,cc]
     - incorporate Thomas's changes
     - replace void CreateProjection() by TH1D *Projection(const char*)

   * macros/calibration.C
     - replace MHCamera->Draw("proj") by MHCamera->Projection()
     - produce better fits to the projection

   * manalysis/MSimulatedAnnealing.[h,cc], 
     mhist/MHSimulatedAnnealing.[h,cc]:
     - moved to directory mtools
     - MSimulatedAnnealing now inherits from TObject

   * manalysis/Makefile, manalysis/AnalysisLinkDef.h,
     mhist/Makefile, mhist/HistLinkDef.h:
     - removed MSimulatedAnnealing
     - removed MHSimulatedAnnealing

   * mtools/Makefile, mtools/AnalysisLinkDef.h:
     - added MSimulatedAnnealing
     - added MHSimulatedAnnealing



 2004/01/15: Thomas Bretz

   * manalysis/AnalysisLinkDef.h:
     - removed MCurrents

   * mhist/MHCamera.[h,cc]:
     - added member function Projection
     - removed fYproj

   * mreport/MReport*.*:
     - added more comments



 2004/01/14: Markus Gaug
  
   * mtools/MFFT.[h,cc]
     - new class to perform Fast Fourier Transforms

   * mtools/Makefile
   * mtools/ToolsLinkDef.h
     - include MFFT

   * manalysis/MExtractedSignalCam.h
     - removed variables fNumHiGainSamples and fNumLoGainSamples
     - function SetNumUsedFADCSlices replaced by SetUsedFADCSlices

   * manalysis/MExtractSignal.cc
     - updated call to SetUsedFADCSlices



 2004/01/14: Abelardo Moralejo

   * manalysis/MMcCalibrationCalc.cc:
     - set pedestal rms per FADC slice to 0.01 counts in case it is 0.
       This can happen (depending on camera simulation parameters), and
       rms = 0 later resulted in the pixel being eliminated in the 
       image cleaning.



 2004/01/14: Thomas Bretz

   * Makefile:
     - replaced old style Makefile.depend by directory specific
       dependancy files
       
   * Makefile.rules:
     - changed obsolete '.cc.o' style to newer style '%.o: %.c'
     - added rule to make dependency files
     - added rule to remove dependancy files.
     - added rmdep as prerequisite to clean
     - include dependency file

   * mbase/Makefile, mcalib/Makefile, mcamera/Makefile,
     mdata/Makefile, mfileio/Makefile, mfilter/Makefile,
     mgeom/Makefile, mgui/Makefile, mhbase/Makefile, 
     mhist/Makefile, mhistmc/Makefile, mimage/Makefile,
     mmain/Makefile, mmc/Makefile, mmontecarlo/Makefile,
     mpointing/Makefile, mranforest/Makefile, mraw/Makefile,
     mreflector/Makefile, mreport/Makefile, mtemp/Makefile,
     mtools/Makefile:
     - removed clean-rule, use the one in Makefile.rules instead

   * mreport/Makefile, mreport/ReportLinkDef.h:
     - added MReportHelp

   * mreport/MReportFileRead.cc:
     - moved code from MReportHelp to new file
     - removed usage of MReport class completely

   * mfilter/MFGeomag.[h,cc]:
     - fixed a bug which causes MFGeomag to fail if MARSSYS is not set
     - changed the second loop from 0->1151 to 1152->2304
     - removed dummy arrays used for reading
     - output filename in case of 'file not found'
     - removed obsolete data members
     - removed obsolete Init function
     - removed forbidden underscore from member variable
     - changed wrong 0/1 in allocation of fResult into kFALSE/TRUE

   * mbase/MStatusDisplay.[h,cc]:
     - added new member function Open
     - added new menu entry kFileOpen
     - changed UpdatePSheader algorithm (the old one was much too slow)

   * manalysis/MCurrents.[h,cc]:
     - removed (old outdated)

   * manalysis/Makefile, manalysis/AnalysisLinkDef.h:
     - removed MCurrents



 2004/01/14: Wolfgang Wittek

   * macros/ONOFFAnalysis.C:
     - current version

   * mhist/MHSigmaTheta.[h,cc]:
     - replace MPedestalCam by MPedPhotCam

   * manalysis/MPad.[h,cc], manalysus/MSigmabar.[h,cc],
     manalysus/MSigmabarCalc.[h,cc], manalysus/MCT1PadONOFF.[h,cc],
     manalysus/MCT1PadSchweizer.[h,cc], manalysus/MPadding.[h,cc]:
     - replace MPedestalCam by MPedPhotCam

   * manalysis/MPedPhotPix.[h,cc]
     - uncomment SetRms()



 2004/01/13: Markus Gaug
 
   * manalysis/MExtractSignal.h
     - seems that the reduced light intensity of the CT1 Pulser
       results in broader pulses -> Set Default time slices to [3,10]

   * mhist/MHCamera.cc
     - make Stat Window in option 'proj' bigger
     - add comment line at beginning of Draw()

   * mcalib/MHCalibrationBlindPixel.[h.cc]
     - add Function DrawClone()
     - fix bug in calculation of integral (GetIntegral("width"))
     - restrict parameters tighter
     - declare fit as not valid, if less than 100 events 
       in single photo-electron peak

   * mcalib/MCalibrationFits.h, mcalib/MHCalibrationBlindPixel.[h.cc]
     - fit now with fixed normalization

   * mcalib/MHCalibrationPixel.[h.cc]
     - new function IsEmpty()
     - Time fit accepted if ChiSquare less than 30. (instead of 20.)

   * mcalib/MCalibrationPix.[h.cc]
     - F-Factor corrected and added error
     - Histograms not fitted if they are empty
     - rejection criteria of fitted charges now > 5 PedRMS (instead 3)

   * mcalib/MCalibrationCam.cc
     - Square Root of Reduced Sigmas not taken, if they are < 0.
     - Reduced Sigma / Charge squared replaced by reduced sigma / charge

   * mcalib/MCalibrationCalc.cc
     - Blind Pixel is not additionally treated as normal pixel any more
     - Blind Pixel Draw replaced by DrawClone

   * macros/calibration.C
     - took out some unused code
     - some default axis labelling corrected



 2004/01/13: Abelardo Moralejo

   * manalysis/MMcPedestalCopy.cc
     - added check of whether input file is a MC file before executing 
       the PreProcess.



 2004/01/13: Wolfgang Wittek

   * macros/ONOFFAnalysis.C, macros/ONAnalysis.C:
     - change macro to handle also MC events with finite sigmabar

   * manalysis/MPadOn.[h,cc]:
     - deleted, is replaced by MPad

   * manalysis/MPadOnOFF.[h,cc]:
     - deleted, is replaced by MPad

   * manalysis/MPad.[h,cc]:
     - new class, replacing MPadON and MPadONOFF
     - handle also MC events with finite sigmabar

   * manalysis/Makefile, manalysis/AnalysisLinkDef.h:
     - replace MPadON and MPadONOFF by MPad



 2004/01/13: Thomas Bretz

   * mraw/MRawRunHeader.[h,cc]:
     - implemented new member function GetRunTypeStr

   * Changelog.03: 
     - added (contains entries of 2003)

   * Changelog:
     - removed entries of 2003

   * mfilter/MFGeomag.cc:
     - simplified open/close of input files

   * manalysis/MEventRate.[h,cc], manalysis/MEventRateCalc.cc:
     - added some comments
     - added new member function GetError

   * Makefile:
     - updated 'make dox' output for convinience

   * macros/dohtml.C:
     - do not print Info's
     - added mcamera
     - added mpointing
     - added mreport

   * mreport/MReportCurrents.cc, mreport/MReportDAQ.cc:
     - removed some obsolete comments (these comments confused THtml)

   * manalysis/MCameraData.[h,cc], mfileio/MCT1ReadAscii.[h,cc],
     mfileio/MCT1ReadPreProc.[h,cc], mfilter/MFCT1SelBasic.cc,
     mfilter/MFSelBasic.cc, mhist/MHCamera.cc, mhist/MHEvent.[h,cc],
     mimage/MImgCleanStd.[h,cc], mimage/MImgCleanTGB.[h,cc],
     mmain/MAnalysis.cc, mmain/MEventDisplay.cc:
     - replaced MPedestalCam by MPedPhotCam
     - removed some obsolete includes

   * mbase/BaseLinkDef.h, mbase/Makefile:
     - added MLogPlugin
     - added MLogHtml

   * mbase/MLog.[h,cc]:
     - added use of plugins

   * mbase/MLogPlugin.[h,cc], mbase/MLogHtml.[h,cc]:
     - added

   * mbase/MStatusDisplay.[h,cc]:
     - added a Save-As dialog



 2004/01/12: Thomas Bretz

   * macros/rootlogon.C:
     - added mcamera to include path
     - added mpointing to include path
     - added mreport to include path
     
   * mbase/MLog.cc:
     - updated my eMail address
     
   * mbase/MParContainer.h:
     - added Read member function
     
   * mbase/MStatusDisplay.[h,cc]:
     - changed size of ps-paper
     - added a user defined bottom line to the ps-files
     - added name (currently unused)
     - added title (used as possible bottom line in ps-file)
     
   * mraw/MRawRunHeader.cc:
     - added some comments

   * status.cc:
     - set name of input as bottom line of status display



 2004/01/12: Abelardo Moralejo

   * mimage/MStereoPar.cc
     - fixed mistake: now fSourceY = scale1*hillas1.GetMeanY() + ...
       instead of         fSourceY = scale2*hillas1.GetMeanY() + ...
       Bug had no effect unless scale1 != scale2 (-> telescopes of 
       different size).



 2004/01/12: Markus Gaug

   * manalysis/MExtractSignal.[h,cc]
     - take out some uncommented code 
     - replace logainshift by two independent variables: 
       logainfirst, logainlast
     - because in most data, maximum occurs in slice 4, shift default 
       from slices 4-9 to slices 3-8.

   * manalysis/MExtractedSignalPix.[h,cc]
     - took out some uncommented code
     - new initializers: instead of -1. now 99999.9



 2004/01/09: Markus Gaug

   * mcalib/MCalibrationCam.[h,cc]
     - take out 1-dim histos because we can use 
       the projections in MHCamera (s.b.)

   * mcalib/MHCalibrationBlindPixel.h
     - make the fit by default quiet

   * mcalib/MHCalibrationPixel.[h,cc]
   * mcalib/MCalibrationPix.[h,cc]
     - Ranges of time slices stored independently for HiGain and LoGain
       in order to allow consecutive checks

   * mhist/MHCamera.[h.cc]
     - add the possibility to call a Draw("proj") which will 
       draw a Y-Projection of the histogram

   * macros/calibration.C
     - apply the drawing of the projections and fit them

   * mcalib/MCalibrationCalc.cc
     - catch the possibility of a calibration file, 
       unintendedly being a pedestal or cosmic file. 
       Program now exits with an error before doing the fits

   * mcalib/MCalibrationPix.[h,cc]
     - fixed a bug in calculation of fFitValid. 
       before, a call to IsFitValid() always returned false



 2004/01/09: Javier L¢pez

   * macros/bootcampstandardanalysis.C
     - added new macro that holds the main scheleton discussed in the 
       december bootcamp in Wuerzburg. It runs first over a pedestal
       run, then over a calibration run, again over the pedestal run
       to compute de pedestal in number of photons and finally over a
       data run to calibrate the data.



 2004/01/06: Sebastian Raducci
   
   * manalysis/MArrivalTime.[h,cc]
     - added new method to calculate photon arrival times. It uses the
       TSpline5 Root Class to interpolate the Fadc Slices. The time is
       the abscissa value of the absolute maximum of the interpolation
     - added new method to find clusters with similar arrival times.
       For now it's a preliminary version. It simply search for adiacent
       pixels having the same arrival time (color).

   * manalysis/MArrivalTimeCalc.[h,cc]
     - modified the Process() method according to the changes in the
       Arrival Time Cointainer

   * mmain/MEventDisplay.cc
     - added new tab in the camera event display to show the arrival
       time. This arrival time is calculated with the spline method
     - the old Arrival Time pad now it's called Max Slice Idx, because
       it's only the index of the slice with the max content.
   
   * mhist/MHEvent.[h,cc]
     - added new fType in te enumeration to handle the arrival time
       histos