save in h5 format

CMakeLists.txt

@@ -17,3 +17,15 @@ add_subdirectory(tests)
 
 add_executable(example example.cpp)
 target_link_libraries(example PUBLIC binFileParser)
+
+include(FetchContent)
+# Import FetchContent module:
+FetchContent_Declare(
+    highfive
+    URL
+    https://github.com/BlueBrain/HighFive/archive/refs/tags/v2.7.1.zip
+)
+FetchContent_MakeAvailable(highfive)
+
+add_executable(example_h5 example_h5.cpp)
+target_link_libraries(example_h5 PUBLIC binFileParser HighFive)
\ No newline at end of file

example_h5.cpp

+#include <string>
+#include <iostream>
+#include <chrono>
+#include <highfive/H5DataSet.hpp>
+#include <highfive/H5DataSpace.hpp>
+#include <highfive/H5File.hpp>
+
+#include "BinFileParser.h" 
+
+
+void print_execution_time()
+{
+    static auto startTime = std::chrono::high_resolution_clock::now();
+    auto endTime = std::chrono::high_resolution_clock::now();
+    std::cout << std::chrono::duration_cast<std::chrono::milliseconds>(endTime - startTime).count() << "ms" << std::endl;
+    startTime = endTime;
+}
+
+
+int main(int argc, char** argv)
+{
+    std::string fpath("/media/ming/DATA/projects/python_parser_for_DT5550W_bin_file/test_data/test.data");
+    DT5550WBinFile binFileHandle(fpath, true);
+    // Depending on available memory size, chunkSize can be made smaller/larger
+    const int chunkSize = 100000;
+    // Max number of events to process
+    const int maxN = 1000000;
+    int numberOfPulsesRead = 0;
+    // suppose the events you want to export are stored in a vector called `newEvents`
+    std::vector<CITIROCEvent> newEvents;
+    // Read the file in chunks of 100k events until end of file (EOF)
+    while (numberOfPulsesRead < maxN && binFileHandle.isGood())
+    {
+        auto newChunk = binFileHandle.readNextNEvents(chunkSize);
+        numberOfPulsesRead += newChunk.size();
+        std::cout << "Read " << newChunk.size() << "events" << '\n';
+        newEvents.insert(newEvents.end(), newChunk.begin(), newChunk.end()); // save the events we just read
+    }
+    print_execution_time();
+
+    // save data to txt files
+    std::ofstream outfile_LG("timestamps.txt");
+    if(!outfile_LG){// file couldn't be opened
+        std::cerr << "Error: timestamps.txt could not be opened" << std::endl;
+        exit(1);
+    }
+    for (int i=0; i<newEvents.size();i++)
+    {
+        for (int j=0; j<32; j++)
+        {
+            outfile_LG << newEvents[i].chargeLG[j] << ' '; // separated by space
+        }
+        outfile_LG << '\n';
+    }
+    outfile_LG.close();
+    print_execution_time();
+    
+    std::vector<std::vector<uint16_t>> LG_charges(newEvents.size(), std::vector<uint16_t>(32, 0));
+    std::vector<uint64_t> timestamps(newEvents.size(), 0);
+    for (int i=0; i<newEvents.size();i++)
+    {
+        LG_charges[i] = newEvents[i].chargeLG;
+    }
+    for (int i=0; i<newEvents.size();i++)
+    {
+        timestamps[i] = newEvents[i].RunEventTimecode;
+    }
+    const std::string FILE_NAME("newEvents.h5");
+    try {
+        // Create a new file using the default property lists.
+        HighFive::File file("newEvents.h5", HighFive::File::ReadWrite | HighFive::File::Create | HighFive::File::Overwrite);
+
+        // Create the dataset
+        HighFive::DataSet LG_dataset = file.createDataSet<uint16_t>("LGcharges", HighFive::DataSpace::From(LG_charges));
+        // write it
+        LG_dataset.write(LG_charges);
+
+        HighFive::DataSet tt_dataset = file.createDataSet<uint16_t>("timestamps", HighFive::DataSpace::From(timestamps));
+        tt_dataset.write(timestamps);
+
+    } catch (HighFive::Exception& err) {
+        // catch and print any HDF5 error
+        std::cerr << err.what() << std::endl;
+    }
+    print_execution_time();
+
+    return 0;
+}
\ No newline at end of file

read_h5.py

+'''
+Description: 
+Author: Ming Fang
+Date: 1969-12-31 18:00:00
+LastEditors: Ming Fang
+LastEditTime: 2023-08-15 16:00:04
+'''
+import numpy as np
+import h5py
+
+
+hf = h5py.File('newEvents.h5', 'r')
+LGcharges = hf.get('LGcharges')[()]
+print(LGcharges.shape)
+print(LGcharges[0])
+timestamps = hf.get('timestamps')[()]
+print(timestamps.shape)
+print(timestamps[0])