Binary Array

[dragonguy]

All this time iv'e been using the Byte Array, whereas there has been the Binary Array all along which is far superior, it can save other things apart from bytes and has compression methods, Iv'e now switched from Byte Array to Binary Array but I have a few questions about it.

1) Are it's Bytes Signed or Unsigned? -127 to 128 or 0 to 255?

2) How big are Floats, Integers & Strings byte-wise? I need to know so I don't accidently overwrite Bytes needed for other stuff.
Don't know what the other datas are, lol.

3) Whats a Work-Space?

[James]

1) There are seperate expressions for signed and unsigned data types

2) Float - 4 bytes, integer - 4 bytes, strings - each letter is one byte

3) Each workspace is a seperate array. You can store data and load and save files all from within one workspace, and then change to another one and have a completely clean slate to work from while your original workspace is still held to switch back to. You can also save all of the workspaces to a single file, and load that file back in.

[dragonguy]

1) There are seperate expressions for signed and unsigned data types

Explain.

Edit: I understand how to get a Signed/Unsigned byte from the Array using it's expressions but how do I write them?

[Looki]

Wikipedia.

[Nifflas]

Edit: I understand how to get a Signed/Unsigned byte from the Array using it's expressions but how do I write them?

This may this may sound weird, but no matter if you need the byte to be signed or unsigned, you just use the "set byte" action and input the desired value.

This is possible simply because the binary data of the range 0-128 is identical in both signed and unsigned bytes, only the -127 to -1 range and 129-255 range differs. Input a value between 0 and 255 for unsigned, and input a value between -127 to 128 for signed.

[dragonguy]

Ok that makes sense.

[Dynasoft]

When you load a byte to a larger number (demonstrating with 10-bit, MMF's numbers are actually 32-bit which is huge to write in binary):

       byte         =       (decimal)       =      MMF-size number
00000000 - 01111111 =    0 - 127 (both)     = 00 00000000 - 00 01111111
10000000 - 11111111 = -128 - -1  (signed)   = 11 10000000 - 11 11111111
10000000 - 11111111 =  128 - 255 (unsigned) = 00 10000000 - 00 11111111

So as you can see it makes a difference on loading the number whether it's signed or unsigned. For a signed number the highest value bit needs to be extended through the extra bits for the value to stay negative. For an unsigned number the extra bits HAVE to be 0.
For saving you're coming from a larger number, like this (demonstrating with 10-bit):

     MMF-size number      = (decimal)  =       byte
00 00000000 - 00 01111111 =    0 - 127 = 00000000 - 01111111
11 10000000 - 11 11111111 = -128 - -1  = 10000000 - 11111111
00 10000000 - 00 11111111 =  128 - 255 = 10000000 - 11111111

As you can see the last 8 bits of your larger number are identical for numbers that would go into a signed (-128 - -1) or unsigned (128 - 255) byte, so you don't need to say which it is, because it doesn't matter.
You only need to worry about whether it's a signed or unsigned number when converting to a larger number, like byte->int.

[dragonguy]

Ok that doesn't make sense, especially seeing as bytes are 8-bit.

[Dynasoft]

I've tried to clarify the confusing bit.
The byte is only 8 bits, but all numbers in MMF are 32-bit, so it has to translate from one to the other when loading and saving bytes. I used a 10-bit number in my example because writing out a full 32-bit number in binary is huge.

[James]

I've tried to clarify the confusing bit.

Pun intended?