Implementation of TRestDataSet

examples/dataset.rml

@@ -0,0 +1,10 @@
+<?xml version="1.0"?>
+<TRestDataSet name="BabyIAXO">
+    <parameter name="startTime" value="2022/04/28 00:00"/>
+    <parameter name="endTime" value="2022/11/25 13:20"/>
+    <parameter name="filePattern" value="/home/jgalan/rest-framework/source/libraries/axion/examples/full-ray-tracing/BabyIAXO_TrueWolterMicromegasTest*.root"/>
+    <filter metadata="TRestRun::fRunNumber" greaterThan="370" lowerThan="375"/>
+    <filter metadata="TRestRun::fRunTag" contains="Test"/>
+    <observables list="window_transmission:optics_efficiency:axionPhoton_probability:axionPhoton_transmission:boreExitGate_transmission"/>
+    <processObservables list="final"/>
+</TRestDataSet>

source/framework/core/inc/TRestAnalysisTree.h

@@ -266,7 +266,7 @@ class TRestAnalysisTree : public TTree {
     Bool_t EvaluateCuts(const std::string& expression);
     Bool_t EvaluateCut(const std::string& expression);
 
-    TString GetStringWithObservableNames();
+    std::vector<std::string> GetObservableNames();
 
     std::vector<std::string> GetCutObservables(const std::string& cut_str);
 

source/framework/core/inc/TRestDataSet.h

@@ -0,0 +1,103 @@
+/*************************************************************************
+ * This file is part of the REST software framework.                     *
+ *                                                                       *
+ * Copyright (C) 2016 GIFNA/TREX (University of Zaragoza)                *
+ * For more information see https://gifna.unizar.es/trex                 *
+ *                                                                       *
+ * REST is free software: you can redistribute it and/or modify          *
+ * it under the terms of the GNU General Public License as published by  *
+ * the Free Software Foundation, either version 3 of the License, or     *
+ * (at your option) any later version.                                   *
+ *                                                                       *
+ * REST is distributed in the hope that it will be useful,               *
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of        *
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the          *
+ * GNU General Public License for more details.                          *
+ *                                                                       *
+ * You should have a copy of the GNU General Public License along with   *
+ * REST in $REST_PATH/LICENSE.                                           *
+ * If not, see https://www.gnu.org/licenses/.                            *
+ * For the list of contributors see $REST_PATH/CREDITS.                  *
+ *************************************************************************/
+
+#ifndef REST_TRestDataSet
+#define REST_TRestDataSet
+
+#include "ROOT/RDataFrame.hxx"
+#include "TRestMetadata.h"
+#include "TTimeStamp.h"
+
+/// It allows to group a number of runs that satisfy given metadata conditions
+class TRestDataSet : public TRestMetadata {
+   private:
+    /// All the selected runs will have a starting date after fStartTime
+    std::string fStartTime = "2000/01/01";  //<
+
+    /// All the selected runs will have an ending date before fEndTime
+    std::string fEndTime = "3000/12/31";  //<
+
+    /// A glob file pattern that must be satisfied by all files
+    std::string fFilePattern = "";  //<
+
+    /// It contains a list of the observables that will be added to the final tree or exported file
+    std::vector<std::string> fObservablesList;  //<
+
+    /// It contains a list of the process where all observables should be added
+    std::vector<std::string> fProcessObservablesList;  //<
+
+    /// A list of metadata members where filters will be applied
+    std::vector<std::string> fFilterMetadata;  //<
+
+    /// If not empty it will check if the metadata member contains the value
+    std::vector<std::string> fFilterContains;  //<
+
+    /// If the corresponding element is not empty it will check if the metadata member is greater
+    std::vector<Double_t> fFilterGreaterThan;  //<
+
+    /// If the corresponding element is not empty it will check if the metadata member is lower
+    std::vector<Double_t> fFilterLowerThan;  //<
+
+    /// The total integrated run time of selected files
+    Double_t fTotalDuration = 0;  //<
+
+    /// The resulting RDataFrame object after initialization
+    ROOT::RDataFrame fDataSet = 0;  //!
+
+    /// A pointer to the generated tree
+    TTree* fTree = 0;  //!
+
+    /// A list populated by the FileSelection method using the conditions of the dataset
+    std::vector<std::string> fFileSelection;  //!
+
+    void InitFromConfigFile() override;
+
+   protected:
+    virtual std::vector<std::string> FileSelection();
+
+   public:
+    ROOT::RDataFrame GetDataFrame() {
+        if (fTree == nullptr) RESTWarning << "DataFrame has not been yet initialized" << RESTendl;
+        return fDataSet;
+    }
+
+    TTree* GetTree() {
+        if (fTree == nullptr) RESTWarning << "Tree has not been yet initialized" << RESTendl;
+        return fTree;
+    }
+
+    Int_t GetNumberOfColumns() { return fDataSet.GetColumnNames().size(); }
+    Int_t GetNumberOfBranches() { return GetNumberOfColumns(); }
+
+    Double_t GetTotalTimeInSeconds() { return fTotalDuration; }
+
+    void Export(std::string fname);
+
+    void PrintMetadata() override;
+    void Initialize() override;
+    TRestDataSet();
+    TRestDataSet(const char* cfgFileName, std::string name = "");
+    ~TRestDataSet();
+
+    ClassDefOverride(TRestDataSet, 1);
+};
+#endif

source/framework/core/src/TRestAnalysisTree.cxx

@@ -1022,18 +1022,19 @@ Double_t TRestAnalysisTree::GetObservableMinimum(const TString& obsName, Double_
 }
 
 ///////////////////////////////////////////////
-/// \brief It returns a string containing all the observables that exist in the analysis tree.
+/// \brief It returns a vector with strings containing all the observables that exist in
+/// the analysis tree.
 ///
-TString TRestAnalysisTree::GetStringWithObservableNames() {
-    Int_t nEntries = GetEntries();
+std::vector<std::string> TRestAnalysisTree::GetObservableNames() {
+    std::vector<std::string> names;
+
+    // Int_t nEntries = GetEntries();
     auto branches = GetListOfBranches();
-    std::string branchNames = "";
     for (int i = 0; i < branches->GetEntries(); i++) {
-        if (i > 0) branchNames += " ";
-        branchNames += (string)branches->At(i)->GetName();
+        names.push_back((string)branches->At(i)->GetName());
     }
 
-    return (TString)branchNames;
+    return names;
 }
 
 ///////////////////////////////////////////////