Main Content


(To be removed) Read logged records from disk to MATLAB workspace

Support for the OPC Data Access (DA) standard will be removed in a future release. Instead use OPC Unified Architecture (UA). See Compatibility Considerations.


S = OPCREAD('LogFileName')
S = opcread('LogFileName','PropertyName','PropertyValue',...)
TSCell = opcread('LogFileName','DataType','timeseries')
[I,V,Q,TS,ET] = opcread('LogFileName','DataType',DType,...)


S = OPCREAD('LogFileName') returns all available records from the OPC log file named LogFileName. If no extension is specified as part of LogFileName, then .olf is used.

S is an NRec-by-1 structure array, where NRec is the number of records returned. S contains the fields 'LocalEventTime' and 'Items'. LocalEventTime is a date vector corresponding to the local event time for that record. Items is an NItems-by-1 structure array containing the fields show below.

Field Name



The fully qualified item ID, as a character vector.


The data value. The data type is dependent on the original Item's DataType property.


The data quality, as a character vector.


The time the value was changed, as a date vector.

S = opcread('LogFileName','PropertyName','PropertyValue',...) limits the data read from the specified OPC log file based on the properties and values provided. Valid property names and property values are defined in the table below.

Property Name

Property Value


Specify the required records as [startRec endRec]. If no records fall within those bounds, opcread returns empty outputs.


Specify the date range for records as [startDt endDt]. The dates must be in MATLAB® date number format. If no records fall within those bounds, opcread returns empty outputs.


Specify the required item IDs as a character vector, string, or array. If no records match the required ItemIDs, OPCREAD returns empty outputs.

TSCell = opcread('LogFileName','DataType','timeseries') assigns the data received from the OPC log file to a cell array of time series objects. TSCell contains as many time series objects as there are items in the group, with the name of each time series object set to the item ID. The quality value stored in the time series object is offset from the quality value returned by the OPC server by 128. The quality displayed by each is the same. Because each record logged might not contain information for every item, the time series objects have only as many data points as there are records containing information about that particular item ID.

[I,V,Q,TS,ET] = opcread('LogFileName','DataType',DType,...) assigns the data retrieved from the OPC log file to separate arrays. Valid data types for DType are 'double', 'single', 'int8', 'int16', 'int32', 'uint8', 'uint16', 'uint32', 'logical', 'currency', 'date', and 'cell'.

I is a 1-by-NItem cell array of item names.

V is an NRec-by-NItem array of values with the data type specified. If a data type of 'cell' is specified, V is a cell array containing data in the returned data type for each item. Otherwise, V is a numeric array of the specified data type.


DType must be set to 'cell' when retrieving records containing character vectors or arrays of values.

Q is an NRec-by-NItem array of quality character vectors for each value in V.

TS is an NRec-by-NItem array of MATLAB date numbers representing the time when the relevant value and quality were stored on the OPC server.

ET is an NRec-by-1 array of MATLAB date numbers, corresponding to the local event time for each record.

Each record logged may not contain information for every item returned, since data for that item may not have changed from the previous update. When data is returned as a numeric matrix, the missing item columns for that record are filled as follows.


The corresponding value entry is set to the previous value of that item, or to NaN if there is no previous value.


The corresponding quality entry is set to 'Repeat'.


The corresponding time stamp entry is set to the first valid time stamp for that record.


Configure and start a logging task. Wait for the task to complete.

da = opcda('localhost','Matrikon.OPC.Simulation');
grp = addgroup(da,'ExOPCREAD');
itm1 = additem(grp,'Triangle Waves.Real8');
itm2 = additem(grp,'Saw-Toothed Waves.Int2');
grp.LoggingMode = 'disk';
grp.RecordsToAcquire = 30;
grp.LogFileName = 'ExOPCREAD.olf';

Retrieve the first two records into a structure:

s = opcread('ExOPCREAD.olf','Records',[1, 2]);

Retrieve all the data and plot it with a legend:

[itmID,val,qual,tStamp] = opcread('ExOPCREAD.olf', ...
datetick x keeplimits

Version History

Introduced before R2006a

expand all