What the meaning of " (!0) " in ILNumerics Library?

Question

What the meaning of " (!0) " in ILNumerics just like this

return (!0)ILMathInternal.exp(this.m_mu);

and like this,

return 0.5 + (!0)ILMathInternal.log(this.m_sigma) + this.m_mu + (!0)ILMathInternal.log(ILMathInternal.sqrt(2.0 * ILMathInternal.pi));

Why we use this kind of expression---"(!0)" ?

I don't think this is related with C#. `!0` does not generate a type for explicit casting. — Soner Gönül, Mar 08 '16 at 11:51
Unless you post your source for this code, this question is useless. The code you've posted here exists nowhere else on the web (according to everyone's favourite search engine) — spender, Mar 08 '16 at 12:04
@Abbondanza from a look at the library's website, the "Computing Engine" is indeed for .NET: C# and VB — ardila, Mar 08 '16 at 12:07
@ardila, but the underlying source code that OP has posted here can still be C++. — Good Night Nerd Pride, Mar 08 '16 at 12:10

Jodrell · Accepted Answer · 2016-03-08T12:12:56.757

5

(!0) has no meaning, it is not valid C#

If I had to guess, which I do, I suspect the (!0) is how a decompiler has chosen to represent some IL code that has no C# representation.

edited Mar 08 '16 at 12:12

answered Mar 08 '16 at 12:02

Jodrell

But if the ILNumerics source is C# it somehow has to be valid C#, right? Maybe the operator `!` has been overloaded? – Good Night Nerd Pride Mar 08 '16 at 12:06
1

@Abbondanza, I suspect that this is more likely, an issue with decompilation. The `(!0)` being a representation of some IL operation which has no C# representation. – Jodrell Mar 08 '16 at 12:10
2

@Abbondanza It seems to be [Generic attribute](http://stackoverflow.com/questions/31559861/why-is-0-a-type-in-microsoft-intermediate-language-msil) – PTwr Mar 08 '16 at 12:14

1 Answers1