try - beginning of try block in try-catch control instruction
catch - beginning of catch block in try-catch control instruction
The try - catch control instruction can be used to manage codes that could possibly generate errors.
When a try - catch control instruction is executed, normally only the statements between the try and catch keywords are executed. However, if an error occurs during execution of any of these statements, the error is recorded, the remaining statements up to the catch keyword are skipped and the statements between the catch and end keywords are executed using the default error handling mode (see: errcatch .
The recorded error can be retreived using the lasterror function.
The catch statements as well as the catch keyword can be omitted if no alternative statements are given.
Note that one can also use the execstr function with 'errcatch' argument for error handling. This can be particularily useful for handling syntactical errors.
file_path=TMPDIR+'/wrong' try u=mopen(file_path,'r') x=mget(10,'c',u) catch disp(['file '+file_path+ 'cannot be read', 'using default values for x']) x=1:10 end [error_message,error_number]=lasterror(%t)
error , execstr , if , lasterror , errcatch ,
Serge Steer, INRIA