octal values in double/float arrays

There is an issue when octal values are used in double or float arrays, but they can be used in Integer arrays without any problems.

`double[][] vCIFHa = {{0.54,-0x1.83e3a5cp26},{-0x1.013a519cp30,014371144027}};`

And is there a reason why arrays for shorts are not supported, but they seem to work for all other standard data types?