Main Content

readNext

Read next available data item from each stream of ADTF DAT file selection

    Description

    example

    data = readNext(streamReader) reads the next available data item in each stream of the adtfStreamReader object streamReader across. data is a structure containing nested structures, arrays based on the contents of the files. By default, readNext initially reads from the first data item in each stream of the selection. It then reads subsequent data items during successive calls, up to the last available item across all streams. To reset the reading operation back to the first data item, use the reset object function of the adtfStreamReader object.

    When reading from a selection that contains multiple streams, if some streams reach the end of selection criteria, then this function reads only from the remaining data streams.

    Examples

    collapse all

    This example shows how to extract and visualize a video stream, stored in an ADTF DAT file. It also shows how to write the video stream into a video file.

    Download the sample video DAT file.

    downloadURL  = 'https://ssd.mathworks.com/supportfiles/driving/data/sample_can_video_dat.zip';
    dataFolder   = fullfile(tempdir, 'adtf-video', filesep); 
    options      = weboptions('Timeout', Inf);
    zipFileName  = [dataFolder, 'sample_can_video_dat.zip'];
    folderExists = exist(dataFolder, 'dir');
    
    % Create a folder in a temporary directory to save the downloaded file.
    if ~folderExists  
        mkdir(dataFolder); 
        disp('Downloading sample_can_video_dat.zip (9.74 MB)...') 
        websave(zipFileName, downloadURL, options); 
        
        % Extract contents of the downloaded file.
        disp('Extracting sample_can_video_dat.zip...') 
        unzip(zipFileName, dataFolder); 
    end

    Create the ADTF File Reader object.

    datFileName = fullfile(dataFolder,"sample_can_video.dat");
    fileReader  = adtfFileReader(datFileName) 
    fileReader = 
                  DataFileName: "C:\Users\latriwal\AppData\Local\Temp\adtf-video\sample_can_video.dat"
           DescriptionFileName: ""
               PluginDirectory: ""
                   StreamCount: 2
                    StreamInfo: 
    
        StreamIndex    StreamName      StreamType       StartTime     EndTime      ItemCount
        ___________    __________    _______________    _________    __________    _________
    
             1         {'rawcan'}    {'UNSUPPORTED'}      0 sec      14.805 sec       743   
             2         {'video' }    {'adtf/image' }      0 sec      14.799 sec       149   
    
    
    

    From the StreamInfo property, note that the index of the video stream is 2. Use the select function of the adtfFileReader object, to select the video stream for reading. The returned adtfStreamReader object has all the information about the selection.

    streamReader = select(fileReader,2) 
    streamReader = 
      adtfStreamReader with properties:
    
               DataFileName: "C:\Users\latriwal\AppData\Local\Temp\adtf-video\sample_can_video.dat"
        DescriptionFileName: ""
            PluginDirectory: ""
                StreamIndex: 2
                 StartIndex: 1
                   EndIndex: 149
         CurrentIndexOffset: 0
                  StartTime: [0×0 duration]
                    EndTime: [0×0 duration]
                  DataCount: 149
    
    

    Note that the value of CurrentIndexOffset is 0. This signifies that the next readNext call will return the first item.

    Preview the first image frame from the stream.

    firstFrame = readNext(streamReader);
    imshow(firstFrame.Data.Item)

    Before creating a video, use the reset function to start reading from the first frame. This resets the value of CurrentIndexOffset to 0.

    reset(streamReader);
    fprintf("CurrentIndexOffset = %d\n",streamReader.CurrentIndexOffset)
    CurrentIndexOffset = 0
    

    Create a VideoWriter object that you can use to write image frames to a video file. Specify a frame rate of 1 frame per second.

    videoWriterObj = VideoWriter("example_video.avi"); 
    videoWriterObj.FrameRate = 1; 
    open(videoWriterObj); 

    Using the streamReader object, iterate over the data items in the selection one-by-one. The hasNext function determines if there is an item left to read as we are incrementally reading the file. readNext returns the data item which is basically a structure containing the data and the associated timestamp. In every iteration, extract the image frame and write it to the video file.

    while streamReader.hasNext()
        streamData = streamReader.readNext(); 
        imageFrame = streamData.Data.Item; 
        frame      = im2frame(streamData.Data.Item, gray); 
        writeVideo(videoWriterObj, frame); 
    end 

    Alternatively, you can read all the image frames at once, using the read function, and iterate over it later.

    allData = read(streamReader)
    allData = struct with fields:
        StreamIndex: 2
               Data: [149×1 struct]
    
    

    Close the connection with the video file.

    close(videoWriterObj);
    close all

    Visualize the output file example_video.avi using Video Viewer.

    implay("example_video.avi") 

    Input Arguments

    collapse all

    Stream reader, specified as an adtfStreamReader object.

    Output Arguments

    collapse all

    Next available data item from each stream of the selection, returned as an N-by-1 structure array. N is the number of streams in the selection. Each element in data is a structure that has these fields:

    • StreamIndex — Index of the stream the data comes from.

    • Data — Structure containing the next data item from the stream, and its timestamp in microseconds.

    Version History

    Introduced in R2022a