[6.38] Allow retrieving dataset top-level field names

tree/dataframe/inc/ROOT/RDF/RInterface.hxx

@@ -1523,7 +1523,17 @@ public:
       // RemoveDuplicates should preserve ordering of the columns: it might be meaningful.
       RDFInternal::RemoveDuplicates(columnNames);
 
-      auto selectedColumns = RDFInternal::ConvertRegexToColumns(columnNames, columnNameRegexp, "Snapshot");
+      std::vector<std::string> selectedColumns;
+      try {
+         selectedColumns = RDFInternal::ConvertRegexToColumns(columnNames, columnNameRegexp, "Snapshot");
+      }
+      catch (const std::runtime_error &e){
+         // No columns were found, try again but consider all input data source columns
+         if (auto ds = GetDataSource())
+            selectedColumns = RDFInternal::ConvertRegexToColumns(ds->GetColumnNames(), columnNameRegexp, "Snapshot");
+         else
+            throw e;
+      }
 
       if (RDFInternal::GetDataSourceLabel(*this) == "RNTupleDS") {
          RDFInternal::RemoveRNTupleSubFields(selectedColumns);

tree/dataframe/inc/ROOT/RDF/RInterfaceBase.hxx

@@ -216,6 +216,7 @@ public:
    RInterfaceBase(RDFDetail::RLoopManager &lm, const RDFInternal::RColumnRegister &colRegister);
 
    ColumnNames_t GetColumnNames();
+   ColumnNames_t GetDatasetTopLevelFieldNames();
 
    std::string GetColumnType(std::string_view column);
 

tree/dataframe/inc/ROOT/RNTupleDS.hxx

@@ -126,6 +126,7 @@ class RNTupleDS final : public ROOT::RDF::RDataSource {
    std::unordered_map<ROOT::DescriptorId_t, std::string> fFieldId2QualifiedName;
    std::vector<std::string> fColumnNames;
    std::vector<std::string> fColumnTypes;
+   std::vector<std::string> fTopLevelFieldNames;
    /// List of column readers returned by GetColumnReaders() organized by slot. Used to reconnect readers
    /// to new page sources when the files in the chain change.
    std::vector<std::vector<ROOT::Internal::RDF::RNTupleColumnReader *>> fActiveColumnReaders;
@@ -222,6 +223,7 @@ public:
    void SetNSlots(unsigned int nSlots) final;
    std::size_t GetNFiles() const final { return fFileNames.empty() ? 1 : fFileNames.size(); }
    const std::vector<std::string> &GetColumnNames() const final { return fColumnNames; }
+   const std::vector<std::string> &GetTopLevelFieldNames() const final { return fTopLevelFieldNames; }
    bool HasColumn(std::string_view colName) const final;
    std::string GetTypeName(std::string_view colName) const final;
    std::vector<std::pair<ULong64_t, ULong64_t>> GetEntryRanges() final;

tree/dataframe/src/RInterfaceBase.cxx

@@ -100,6 +100,28 @@ ROOT::RDF::ColumnNames_t ROOT::RDF::RInterfaceBase::GetColumnNames()
    return ret;
 }
 
+/////////////////////////////////////////////////////////////////////////////
+/// \brief Retrieve the names of top-level field names
+///
+/// For data sources that support hierarchical dataset schemas, such as TTree
+/// or RNTuple, this function will retrieve the names of top-level fields. For
+/// example, if the schema contains a user class with a data member, only
+/// the name of the top-level field containing the user class object would be
+/// reported, but not the name of the data member sub-field.
+///
+/// For all other data sources, returns the list of all available dataset columns.
+ROOT::RDF::ColumnNames_t ROOT::RDF::RInterfaceBase::GetDatasetTopLevelFieldNames()
+{
+   ROOT::RDF::ColumnNames_t ret;
+   if (auto ds = GetDataSource()) {
+      ret = ROOT::Internal::RDF::GetTopLevelFieldNames(*ds);
+      // Sorting to be consistent with GetColumnNames
+      std::sort(ret.begin(), ret.end());
+   }
+
+   return ret;
+}
+
 /////////////////////////////////////////////////////////////////////////////
 /// \brief Return the type of a given column as a string.
 /// \return the type of the required column.

tree/dataframe/src/RNTupleDS.cxx

@@ -393,6 +393,12 @@ ROOT::RDF::RNTupleDS::RNTupleDS(std::unique_ptr<ROOT::Internal::RPageSource> pag
 
    AddField(fPrincipalDescriptor, "", fPrincipalDescriptor.GetFieldZeroId(),
             std::vector<ROOT::RDF::RNTupleDS::RFieldInfo>());
+
+   auto topLevelFields = fPrincipalDescriptor.GetTopLevelFields();
+   const auto nTopLevelFields = std::distance(topLevelFields.begin(), topLevelFields.end());
+   fTopLevelFieldNames.reserve(nTopLevelFields);
+   for (const auto &field : topLevelFields)
+      fTopLevelFieldNames.push_back(field.GetFieldName());
 }
 
 namespace {

tree/dataframe/test/datasource_ntuple.cxx

@@ -828,3 +828,15 @@ TEST(RNTupleDS, Int8)
    std::vector<std::int8_t> expected{0, 1, 2, 3, 4};
    EXPECT_EQ(expected, df.Take<std::int8_t>("x").GetValue());
 }
+
+TEST_F(RNTupleDSTest, GetTopLevelFieldNames)
+{
+   ROOT::RDataFrame df{fNtplName, fFileName};
+
+   EXPECT_VEC_EQ(
+      df.GetDatasetTopLevelFieldNames(),
+      std::vector<std::string>{"VecElectron", "electron", "energy", "jets", "nElectron", "nnlo", "pt", "rvec", "tag"});
+   EXPECT_VEC_EQ(df.GetColumnNames(),
+                 std::vector<std::string>{"VecElectron", "VecElectron.pt", "electron", "electron.pt", "energy", "jets",
+                                          "nElectron", "nnlo", "pt", "rvec", "tag"});
+}

tree/dataframe/test/datasource_tree.cxx

@@ -57,6 +57,15 @@ TEST(RTTreeDS, BranchWithNestedSameName)
    expect_vec_eq(branchNames, expectedBranchNames);
 }
 
+TEST(RTTreeDS, GetDatasetTopLevelFieldNames)
+{
+   InputTreeRAII dataset{};
+
+   ROOT::RDataFrame df{dataset.fTreeName, dataset.fFileName};
+   auto branchNames = df.GetDatasetTopLevelFieldNames();
+   expect_vec_eq(branchNames, std::vector<std::string>{"toplevel"});
+}
+
 #ifdef R__USE_IMT
 struct Dataset20164RAIII {
    const char *fTreeName{"tree_20164"};