Category : BASIC Source Code
Archive   : QBFAQR01.ZIP
Filename : MBF2NUM.QRM

Ä Area: R-QBasic ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ
Msg#: 13834 Date: 07-30-92 03:33
From: Nick Dyer Read: Yes Replied: No
To: Robert Laszko Mark:
Subj: Numeric conversion help
ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ

RL>
RL>I'm writing a routine to display the number of active records in a
RL>given PCBoard message base file. According to PCBoard's file format
RL>this number is contained in bytes 8 through 11, stored as a Basic S
RL>Precision Real Number. Problem is .... I don't know how to convert
RL>these four bytes into a usable, decimal equivalant.

I think you will find that the following will help if you use Basic

If the 4 byte Basic Real is stored in a string say A$ then,

Q&=CVSMBF(A$)

will place in variable Q& the converted value.

RL>
RL>So what now, coach?? ..... What I'm NOT looking for is the name of
RL>C or Pascal function to do the conversion (but if that's all you ha
RL>I'll take it 🙂 ). I'm using a higher-level language to manipulate

Michael Stewart showed me this technique:

Sub MsFloatToLong (MsFloat As String, L1 As Long)

I1& = Asc(Mid$(MsFloat$, 1))
I2& = Asc(Mid$(MsFloat$, 2))
I3& = Asc(Mid$(MsFloat$, 3))
I4& = Asc(Mid$(MsFloat$, 4))


L1& = ((I1& + (I2& * 256) + (I3& * 65536)) Or &H800000) / (2 ^ (24 -
(I4& - &H80)))


End Sub

Calling the above sub passing the variables containing the four byte basic
real string and a receiving long integer variable will carry out the
conversion necessary.

An unsung hero did this in C, (Author Unknown)

Unsigned long mbf_to_int (m1,m2,m3,exp)

unsigned int m1,m2,m3,exp;

{
return (((m1 + ((unsigned long) m2 << 8) + ((unsigned long) m3 <<
16)) | 0x800000L) >> (24 - (exp - 0x80)));
}


I hope all this helps you. Good Luck!!


Regards
Nick Dyer [email protected]

-!-
þ Orator V1.0 #1þ Mom looted LA & all I got was this lousy tagline.


PCRelay:ALMAC -> #1598 RelayNet (tm)
4.11 ALMAC PC Board BBS +44 (0)324-665371