added hexString() and hexBytes() functions

[ghost]

as suggested by W.B in http://forum.dlang.org/thread/me4njq$1cul$1@digitalmars.com

[nomad-software]

I'd perhaps add a template constraint to isHexLiteral and maybe rename:

static bool isHex(T)(in T hexData) if (isSomeString!T)

and shouldn't this go in std.string along with isNumeric?

[ghost]

std.ascii already has isHexDigit, in isHexLiteral the point is that whites chars are valid.

[JakobOvrum]

What is the purpose of this? It already returns early if it's not a hex literal.

edit:

Nevermind, I see now it's not that simple. I still think the code is seriously confusing. bool is not supposed to be treated like an 8-bit integer.

[ghost]

the last return statement is not about isWhite or isHexDigit it's about the parity of the hex digits count (i & 1), confere with https://github.com/D-Programming-Language/dmd/blob/master/src/lexer.c#L1186

[JakobOvrum]

I like the compile-time templates, but I don't think we should commit to the runtime functions yet. They're pretty inefficient and are hard-coded to using the GC. If we make them public, it's only a matter of time before someone complains that they don't accept ranges, don't accept an output range, aren't @nogc, aren't lazy; heck, here I am, already complaining :)

[ghost]

The idea is to use the same private template for the compile time and run time versions. I wait for more comments about this. The initial idea of Walter is to replace a single feature, but i recognize that i've been a bit overzealous in the commit.

[JakobOvrum]

The idea is to use the same private template for the compile time and run time versions.

And I'm fine with that, but I think making the runtime functions public is a mistake, as they are too specific.

[JakobOvrum]

These should be value templates, not function templates:

template hexString(string hexData, C = char)
    if (hexData.isHexLiteral && isSomeChar!C)
{
    immutable hexString = hexStrImpl!C(hexData);
}

[dnadlinger]

regardless

[WalterBright]

Why was this closed?

[dnadlinger]

It seems like @bbasile has deleted his fork of Phobos.

[WalterBright]

Seems a shame. Looks like I'll adopt and resurrect it myself.

[ghost]

I've deleted the source fork because i forgot to create a brach for the patch. Later i didnt re-fork it because i thought the initial PR would fall into a black hole.