NDFileNetCDF¶
| author: | Mark Rivers, University of Chicago |
|---|
Contents
Overview¶
NDFileNetCDF inherits from NDPluginFile. This plugin saves data in the netCDF file format, which is a portable self-describing binary file format supported by UniData at UCAR (University Corporation for Atmospheric Research). There are netCDF libraries for C, C++, Fortran, and Java. Other languages, including Matlab and IDL have built-in support for netCDF. There are also add-on interfaces available for Python, Ruby and other languages.
The netCDF plugin supports all NDArray data types and any number of array dimensions. It also has full support for NDArray attributes. It will write all attributes associated with the NDArray to the file. If multiple arrays are written to a single netCDF file (stream or capture mode) then each attribute will be an array, with the attribute value for each NDArray in the file being stored. Note that the number and data types of attributes must not be changed while file capture or file streaming are in progress because that would change the structure of the attribute array. Also the colorMode attribute must not be changed while capture or streaming is in progress, because that would change the structure of the NDArray data.
The netCDF classic file format does not handle 64-bit integer data, so cannot actually store NDArrays with types NDInt64 or NDUInt64, or NDAttributes of type NDAttrInt64 or NDAttrUInt64. This plugin works around this restriction by casting the data to float64, so the netCDF library thinks it is writing float64 (double) data, and marks the netCDF data in in the file as double. However the netCDF global attribute “datatype” (enum NDDataType_t) will correctly indicate the actual datatype. File readers will need to cast the data to the actual datatype after reading the data with the netCDF library functions.
The NDFileNetCDF class documentation describes this class in detail.
Configuration¶
The NDFileNetCDF plugin is created with the NDFileNetCDFConfigure
command, either from C/C++ or from the EPICS IOC shell.
NDFileNetCDFConfigure (const char *portName, int queueSize, int blockingCallbacks,
const char *NDArrayPort, int NDArrayAddr, size_t maxMemory,
int priority, int stackSize)
For details on the meaning of the parameters to this function refer to the detailed documentation on the NDFileNetCDFConfigure function in the NDFileNetCDF.cpp documentation and in the documentation for the constructor for the NDFileNetCDF class.
The following is the header contents of a netCDF file produced by this plugin. This information was produced with the following command:
ncdump -h test_netCDF_68.nc
netcdf test_netCDF_68 {
dimensions:
numArrays = UNLIMITED ; // (10 currently)
dim0 = 240 ;
dim1 = 320 ;
dim2 = 1 ;
attrStringSize = 256 ;
variables:
int uniqueId(numArrays) ;
double timeStamp(numArrays) ;
float array_data(numArrays, dim0, dim1, dim2) ;
int Attr_colorMode(numArrays) ;
double Attr_AcquireTime(numArrays) ;
double Attr_RingCurrent(numArrays) ;
int Attr_ImageCounter(numArrays) ;
char Attr_CameraModel(numArrays, attrStringSize) ;
int Attr_BinX(numArrays) ;
int Attr_BinY(numArrays) ;
double Attr_AttrTimeStamp(numArrays) ;
double Attr_ROI0Mean(numArrays) ;
double Attr_ROI1Mean(numArrays) ;
char Attr_FilePath(numArrays, attrStringSize) ;
char Attr_FileName(numArrays, attrStringSize) ;
// global attributes:
:dataType = 6 ;
:NDNetCDFFileVersion = 3. ;
:numArrayDims = 3 ;
:dimSize = 1, 320, 240 ;
:dimOffset = 0, 0, 0 ;
:dimBinning = 1, 2, 2 ;
:dimReverse = 0, 0, 0 ;
:Attr_colorMode_DataType = "Int32" ;
:Attr_colorMode_Description = "Color mode" ;
:Attr_colorMode_Source = ;
:Attr_colorMode_SourceType = "Driver" ;
:Attr_AcquireTime_DataType = "Float64" ;
:Attr_AcquireTime_Description = "Camera acquire time" ;
:Attr_AcquireTime_Source = "13SIM1:cam1:AcquireTime" ;
:Attr_AcquireTime_SourceType = "EPICS_PV" ;
:Attr_RingCurrent_DataType = "Float64" ;
:Attr_RingCurrent_Description = "Storage ring current" ;
:Attr_RingCurrent_Source = "S:SRcurrentAI" ;
:Attr_RingCurrent_SourceType = "EPICS_PV" ;
:Attr_ImageCounter_DataType = "Int32" ;
:Attr_ImageCounter_Description = "Image counter" ;
:Attr_ImageCounter_Source = "ARRAY_COUNTER" ;
:Attr_ImageCounter_SourceType = "Param" ;
:Attr_CameraModel_DataType = "String" ;
:Attr_CameraModel_Description = "Camera model" ;
:Attr_CameraModel_Source = "MODEL" ;
:Attr_CameraModel_SourceType = "Param" ;
:Attr_BinX_DataType = "Int32" ;
:Attr_BinX_Description = "X binning" ;
:Attr_BinX_Source = "13SIM1:ROI1:0:BinX_RBV" ;
:Attr_BinX_SourceType = "EPICS_PV" ;
:Attr_BinY_DataType = "Int32" ;
:Attr_BinY_Description = "Y binning" ;
:Attr_BinY_Source = "13SIM1:ROI1:0:BinY_RBV" ;
:Attr_BinY_SourceType = "EPICS_PV" ;
:Attr_AttrTimeStamp_DataType = "Float64" ;
:Attr_AttrTimeStamp_Description = "Time stamp" ;
:Attr_AttrTimeStamp_Source = "TIME_STAMP" ;
:Attr_AttrTimeStamp_SourceType = "Param" ;
:Attr_ROI0Mean_DataType = "Float64" ;
:Attr_ROI0Mean_Description = "Mean value ROI 0" ;
:Attr_ROI0Mean_Source = "MEAN_VALUE" ;
:Attr_ROI0Mean_SourceType = "Param" ;
:Attr_ROI1Mean_DataType = "Float64" ;
:Attr_ROI1Mean_Description = "Mean value ROI 0" ;
:Attr_ROI1Mean_Source = "MEAN_VALUE" ;
:Attr_ROI1Mean_SourceType = "Param" ;
:Attr_FilePath_DataType = "String" ;
:Attr_FilePath_Description = "File path" ;
:Attr_FilePath_Source = "13SIM1:netCDF1:FilePath_RBV" ;
:Attr_FilePath_SourceType = "EPICS_PV" ;
:Attr_FileName_DataType = "String" ;
:Attr_FileName_Description = "File name" ;
:Attr_FileName_Source = "13SIM1:netCDF1:FileName_RBV" ;
:Attr_FileName_SourceType = "EPICS_PV" ;
}
ncdump is one of a number of very useful command line utilities that come with the netCDF package. The -h option to ncdump means to dump only the header information, not the variable data. This is an explanation of this output:
- dimensions: numArrays is the number of arrays in the file. It will be 1 for files collected in Single mode, and is normally >1 for files collected in Capture or Stream mode. For each array dim0 is the slowest varying dimension, dim1 the next slowest, etc. attrStringSize is the maximum string length for string attributes.
- variables: There are 15 variables in this netCDF file. uniqueId is the unique ID number of each array. timeStamp is the timestamp in seconds for each array. array_data is the array data. Its data type depends on the data type of the NDArray data passed in the callbacks. It dimensions are [numArrays, dim0, dim1, …dimN]. This notation is in the Fortran syntax where the slowest varying dimension comes first in the list. The remaining variables all have the prefix Attr_ and are the attributes for the arrays. Each can have its own data type, and all have the numArrays elements.
- global attributes. dataType is the NDDataType_t enum value for the array data type. numArrayDims is the number of dimensions in each array. array_data has 1 more dimension than this, numArrays, because it contains all of the array callback data. dimSize is an array[numArrayDims] containing the size of each dimension, with the fastest varying dimension first. dimOffset, dimBinning, and dimReverse are the values of the offset, binning and reverse fields in the NDDimension_t structure for each dimension. The remaining netCDF global attributes all have the prefix Attr_, and describe the NDArray attribute values. For each NDArray attribute there is information on the data type, a description, source string and source type.
There is an IDL function, read_nd_netcdf that can be used to read the netCDF files created by this plugin. This routine is contained in the CARS IDL detector package. This function is also contained in the areaDetector distribution in the Viewers/IDL directory.
There is a plugin for the popular ImageJ program that can be used to read the netCDF files created by this plugin. This ImageJ plugin can be downloaded here. This plugin is also contained in the areaDetector distribution in the Viewers/ImageJ/EPICS_areaDetector directory.
