[go-nuts] How to represent a Decimal for GoProtobuf ??

On Mon, Oct 27, 2014 at 9:41 AM, Th3x0d3r wrote:

0x400deb851eb851ec

On Monday, October 27, 2014 11:41:25 AM UTC-5, Th3x0d3r wrote:

Hello there !

I'm trying to communicate my go app using goprotobuf with an existing c#
service that uses protobuf-net as serialization, but since there's no
decimal definition on protocol buffers, protobuf-net uses this
representation structure:


message Decimal {
optional uint64 lo = 1; // the first 64 bits of the underlying value
optional uint32 hi = 2; // the last 32 bis of the underlying value
optional uint32 signScale = 3; // the number of decimal digits (bits
1-16), and the sign (bit 0)
}

Go code:

val := math.Float64bits(3.74)
pd.Lo = proto.Uint64(0x400deb851eb851ec) // Using 3.74 Float64bits hexa
representation
pd.Hi = proto.Uint32(0x00000000)
pd.SignScale = proto.Uint32(0x000000020)


But, the C# server side doesn't get the 3.74 :(

Any help ?

Thanks !

On Mon, Oct 27, 2014 at 11:51 AM, Daniel Eloff wrote:

You're trying to stuff a binary floating point number into a decimal
floating point type. That's all wrong.

Decimal in .NET is a decimal floating point number (base 10, not base 2
like binary floats.) It is of the form sign * mantissa / 10^exponent
where mantissa is a 96 bit integer, and exponent is 0-28 (effectively all
negative, so the smallest number is 10^-28.)

So to represent 3.74, you would store 374 in the mantissa, and use 2 for
the exponent. If you need more than 19 digits, you can use math.big.Int
for the mantissa, otherwise uint64 will suffice.

Cheers,
Dan

You received this message because you are subscribed to the Google Groups
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to golang-nuts+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

On Monday, October 27, 2014 2:51:18 PM UTC-4, Daniel Eloff wrote:

You're trying to stuff a binary floating point number into a decimal
floating point type. That's all wrong.

Decimal in .NET is a decimal floating point number (base 10, not base 2
like binary floats.) It is of the form sign * mantissa / 10^exponent
where mantissa is a 96 bit integer, and exponent is 0-28 (effectively all
negative, so the smallest number is 10^-28.)

So to represent 3.74, you would store 374 in the mantissa, and use 2 for
the exponent. If you need more than 19 digits, you can use math.big.Int
for the mantissa, otherwise uint64 will suffice.

Cheers,
Dan

On Mon, Oct 27, 2014 at 12:06 PM, Th3x0d3r wrote:

Thanks ! , can you show some example ??


You received this message because you are subscribed to the Google Groups
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to golang-nuts+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

On Mon, Oct 27, 2014 at 12:52 PM, Th3x0d3r wrote:

Thanks ! but math it's not my field, and frankly I'm quite lost, can you
show me some how to do that ?

--
You received this message because you are subscribed to the Google Groups
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to golang-nuts+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.