Significant restructure of parsing

kdlong · Jan 19, 2020 · 7daba24 · 7daba24
1 parent 9bd87c7
commit 7daba24
Show file tree

Hide file tree

Showing 15 changed files with 231 additions and 255 deletions.
diff --git a/GeneratorInterface/Core/interface/LHEWeightHelper.h b/GeneratorInterface/Core/interface/LHEWeightHelper.h
@@ -19,23 +19,13 @@
 namespace gen {
     class LHEWeightHelper : public WeightHelper {
     public:
-        LHEWeightHelper();
-
-        //// possibly add more versions of this functions for different inputs
-        void parseLHEFile(std::string filename);
-        void parseWeightGroupsFromHeader(std::vector<std::string> lheHeader);
+        LHEWeightHelper() : WeightHelper() {};
+        void setHeaderLines(std::vector<std::string> headerLines);
+        void parseWeights();
+        void buildGroups();
+        std::unique_ptr<WeightGroupInfo> buildGroup(const ParsedWeight& weight);
     private:
-        std::map<std::string, std::string> weightAttributeMapFromHeaderLine(std::string line);
-        void loadAttributeNames(std::string baseName, std::vector<std::string> altNames ={});
-        std::map<std::string, std::string> getAttributeMap(std::string);
-        std::string sanitizeText(std::string);
-        bool isAWeight(std::string);
-
-        std::regex weightGroupStart_;
-        std::regex weightGroupEnd_;
-        std::regex weightContent_;
-
-        std::map<std::string, std::string> nameConversionMap_;
+        std::vector<std::string> headerLines_;
     };
 }
 

diff --git a/GeneratorInterface/Core/interface/WeightHelper.h b/GeneratorInterface/Core/interface/WeightHelper.h
@@ -6,6 +6,10 @@
 #include "SimDataFormats/GeneratorProducts/interface/WeightGroupInfo.h"
 #include "SimDataFormats/GeneratorProducts/interface/PdfWeightGroupInfo.h"
 #include "SimDataFormats/GeneratorProducts/interface/WeightsInfo.h"
+#include "SimDataFormats/GeneratorProducts/interface/UnknownWeightGroupInfo.h"
+#include "SimDataFormats/GeneratorProducts/interface/PdfWeightGroupInfo.h"
+#include "SimDataFormats/GeneratorProducts/interface/ScaleWeightGroupInfo.h"
+#include "SimDataFormats/GeneratorProducts/interface/MEParamWeightGroupInfo.h"
 #include <boost/algorithm/string.hpp>
 
 namespace gen {
@@ -15,6 +19,14 @@ namespace gen {
         PdfUncertaintyType uncertaintyType; 
     };
 
+    struct ParsedWeight {
+        std::string id;
+        size_t index;
+        std::string groupname;
+        std::string content;
+        std::unordered_map<std::string, std::string> attributes;
+    };
+
     class WeightHelper {
     public:
         WeightHelper();
@@ -23,17 +35,29 @@ namespace gen {
         }
         std::unique_ptr<GenWeightProduct> weightProduct(std::vector<gen::WeightsInfo>, float w0);
         std::unique_ptr<GenWeightProduct> weightProduct(std::vector<double>, float w0);
-        void setGroupInfo();
-        bool currentGroupIsScale();
-        bool currentGroupIsMEParam();
-        bool currentGroupIsPdf();
+        void setModel(std::string model) { model_ = model; }
         int addWeightToProduct(std::unique_ptr<GenWeightProduct>& product, double weight, std::string name, int weightNum, int groupIndex);
         int findContainingWeightGroup(std::string wgtId, int weightIndex, int previousGroupIndex);
     protected:
+        std::string model_;
+        std::vector<ParsedWeight> parsedWeights_;
         const std::vector<PdfSetInfo> pdfSetsInfo;
         std::map<std::string, std::string> currWeightAttributeMap_;
         std::map<std::string, std::string> currGroupAttributeMap_;
         edm::OwnVector<gen::WeightGroupInfo> weightGroups_;
+        bool isScaleWeightGroup(const ParsedWeight& weight);
+        bool isMEParamWeightGroup(const ParsedWeight& weight);
+        bool isPdfWeightGroup(const ParsedWeight& weight);
+        void updateScaleInfo(const ParsedWeight& weight);
+        void updatePdfInfo(const ParsedWeight& weight);
+        std::string searchAttributes(const std::string& label, const ParsedWeight& weight);
+
+        // Possible names for the same thing
+        const std::unordered_map<std::string, std::vector<std::string>> attributeNames_ = {
+            {"muf",  {"muR", "MUR", "muf","facscfact"}},
+            {"mur",  {"muF", "MUF", "mur","renscfact"}},
+            {"pdf", {"PDF", "PDF set", "lhapdf", "pdf", "pdf set", "pdfset"}}
+        };
     };
 }
 

diff --git a/GeneratorInterface/Core/plugins/GenWeightProductProducer.cc b/GeneratorInterface/Core/plugins/GenWeightProductProducer.cc
@@ -23,6 +23,7 @@
 #include "GeneratorInterface/Core/interface/GenWeightHelper.h"
 
 #include "FWCore/ServiceRegistry/interface/Service.h"
+#include <boost/algorithm/string.hpp>
 
 class GenWeightProductProducer : public edm::one::EDProducer<edm::BeginLuminosityBlockProducer> {
 public:
@@ -34,6 +35,7 @@ class GenWeightProductProducer : public edm::one::EDProducer<edm::BeginLuminosit
   gen::GenWeightHelper weightHelper_;
   edm::EDGetTokenT<GenLumiInfoHeader> genLumiInfoToken_;
   edm::EDGetTokenT<GenEventInfoProduct> genEventToken_;
+  const edm::EDGetTokenT<GenLumiInfoHeader> genLumiInfoHeadTag_;
 
   void produce(edm::Event&, const edm::EventSetup&) override;
   void beginLuminosityBlockProduce(edm::LuminosityBlock& lb, edm::EventSetup const& c) override;
@@ -44,7 +46,8 @@ class GenWeightProductProducer : public edm::one::EDProducer<edm::BeginLuminosit
 //
 GenWeightProductProducer::GenWeightProductProducer(const edm::ParameterSet& iConfig) :
     genLumiInfoToken_(consumes<GenLumiInfoHeader, edm::InLumi>(iConfig.getParameter<edm::InputTag>("genInfo"))),
-    genEventToken_(consumes<GenEventInfoProduct>(iConfig.getParameter<edm::InputTag>("genInfo")))
+    genEventToken_(consumes<GenEventInfoProduct>(iConfig.getParameter<edm::InputTag>("genInfo"))),
+    genLumiInfoHeadTag_(mayConsume<GenLumiInfoHeader, edm::InLumi>(iConfig.getParameter<edm::InputTag>("genLumiInfoHeader")))
 {
   produces<GenWeightProduct>();
   produces<GenWeightInfoProduct, edm::Transition::BeginLuminosityBlock>();
@@ -69,6 +72,14 @@ GenWeightProductProducer::produce(edm::Event& iEvent, const edm::EventSetup& iSe
 
 void
 GenWeightProductProducer::beginLuminosityBlockProduce(edm::LuminosityBlock& iLumi, edm::EventSetup const& iSetup) {
+    edm::Handle<GenLumiInfoHeader> genLumiInfoHead;
+    iLumi.getByToken(genLumiInfoHeadTag_, genLumiInfoHead);
+    if (genLumiInfoHead.isValid()) {
+        std::string label = genLumiInfoHead->configDescription();
+        boost::replace_all(label,"-","_");
+        weightHelper_.setModel(label);
+    }
+
     if (weightNames_.size() == 0) {
         edm::Handle<GenLumiInfoHeader> genLumiInfoHandle;
         iLumi.getByToken(genLumiInfoToken_, genLumiInfoHandle);

diff --git a/GeneratorInterface/Core/plugins/LHEWeightProductProducer.cc b/GeneratorInterface/Core/plugins/LHEWeightProductProducer.cc
@@ -46,9 +46,7 @@ class LHEWeightProductProducer : public edm::one::EDProducer<edm::BeginLuminosit
 
 };
 
-//
-// constructors and destructor
-//
+// TODO: Accept a vector of strings (source, externalLHEProducer) exit if neither are found
 LHEWeightProductProducer::LHEWeightProductProducer(const edm::ParameterSet& iConfig) :
     lheLabel_(iConfig.getParameter<std::string>("lheSourceLabel")),
     lheRunInfoToken_(consumes<LHERunInfoProduct, edm::InRun>(lheLabel_)),
@@ -90,7 +88,7 @@ LHEWeightProductProducer::beginRun(edm::Run const& run, edm::EventSetup const& e
             headerWeightInfo = *iter;
     }
 
-	weightHelper_.parseWeightGroupsFromHeader(headerWeightInfo.lines());
+	//weightHelper_.parseWeightGroupsFromHeader(headerWeightInfo.lines());
 }
 
 void

diff --git a/GeneratorInterface/Core/src/GenWeightHelper.cc b/GeneratorInterface/Core/src/GenWeightHelper.cc
@@ -17,7 +17,9 @@ namespace gen {
         for (std::string weightName : weightNames) {
             if(weightName.find("LHE") != std::string::npos) {
                 // Parse as usual, this is the SUSY workflow
-                continue;
+                // std::vector<std::string> info;
+                // boost::split(info, weightName, boost::is_any_of(","));
+                weightGroups_.push_back(new gen::UnknownWeightGroupInfo("inGen"));
             }
             // Working on the not-so-nice assumption that all non-LHE gen weights are PS weights
             else if (weightGroups_.size() == 0) {