-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
NotDashEscaped: You need GnuPG to verify this message


1) fixes http://svn.kuehne.cn/dstress/nocompile/command_line_version_16.d

2) adds the identifiers "Unix", "darwin" and "IA64" (used by gdc-0.8)

3) adds "D_InlineAsm" (it's a compiler feature)

Thomas

--- ./109/dmd/src/dmd/debcond.c	2004-11-03 15:44:36.000000000 +0100
+++ ./109X/dmd/src/dmd/debcond.c	2004-12-07 18:17:18.375980136 +0100
@@ -112,11 +112,13 @@
 {
     static char* reserved[] =
     {
-	"DigitalMars", "X86", "AMD64",
+	"DigitalMars",
+	"X86", "AMD64", "IA64",
 	"Windows", "Win32", "Win64",
-	"linux",
+	"Unix", "linux", "darwin"
 	"LittleEndian", "BigEndian",
-	"none",
+	"D_InlineAsm",
+	"none", "all"
     };

     for (unsigned i = 0; i < sizeof(reserved) / sizeof(reserved[0]); i++)


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.9.13 (GNU/Linux)

iD8DBQFBteg33w+/yD4P9tIRAp5sAKCXy+xy1pYyPV35PYB2twdf8vyxxACfcXwv
5o4QW0nZgJHpqZUJCy2GckM=
=qSD6
-----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Thomas Kuehne schrieb am Tue, 7 Dec 2004 18:24:24 +0100:
> 3) adds "D_InlineAsm" (it's a compiler feature)

arg.. please remove this line, it's already taken care of

Thomas

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.9.13 (GNU/Linux)

iD8DBQFBtel63w+/yD4P9tIRAt/iAJ9Y9xk/F7WsuLkmsrM3PhnFYLewVQCfRML1
gFX6us7Zd5BfRc0URJKGksM=
=IhtX
-----END PGP SIGNATURE-----

Thomas Kuehne wrote:
<snip>
> 2) adds the identifiers "Unix", "darwin" and "IA64" (used by gdc-0.8)
<snip>

Did anyone particular decide "darwin" should be in lowercase?  FTM, does anyone have any idea why "linux" is lowercase, unlike the others?

Stewart.

-- 
My e-mail is valid but not my primary mailbox.  Please keep replies on the 'group where everyone may benefit.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
NotDashEscaped: You need GnuPG to verify this message

Stewart Gordon schrieb am Tue, 07 Dec 2004 17:58:32 +0000:
> Thomas Kuehne wrote:
><snip>
>> 2) adds the identifiers "Unix", "darwin" and "IA64" (used by gdc-0.8)
><snip>
>
> Did anyone particular decide "darwin" should be in lowercase?  FTM, does anyone have any idea why "linux" is lowercase, unlike the others?

Please, the lower case versus capital case issue for Linux has been beaten to death for ages now.

I think David Friedman was the first to use "darwin" for MacOS. http://home.earthlink.net/~dvdfrdmn/d/gdc-1g.tar.bz2

Thomas

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.9.13 (GNU/Linux)

iD8DBQFBtfZY3w+/yD4P9tIRAj/lAJ0UICV5ftHjiwb7AsfaO8C+10USQgCbBMjs
DVha0vqQ2+1uCmwNWsbef8k=
=LNtO
-----END PGP SIGNATURE-----

On Tue, 07 Dec 2004 18:24:24 +0100, Thomas Kuehne wrote:

> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> NotDashEscaped: You need GnuPG to verify this message
> 
> 
> 1) fixes http://svn.kuehne.cn/dstress/nocompile/command_line_version_16.d
> 
> 2) adds the identifiers "Unix", "darwin" and "IA64" (used by gdc-0.8)
> 

So, Unix is being used instead of Posix still?

Thomas Kuehne wrote:

>>Did anyone particular decide "darwin" should be in lowercase?

Walter.

>> FTM, does anyone have any idea why "linux" is lowercase, unlike the others?
> 
> Please, the lower case versus capital case issue for Linux has been beaten to
> death for ages now.

That doesn't make it stink less, though ?

Most newcomers notice the smell, even if
you get used to it after a while in here.

Think it had some obscure compiler reason...

> I think David Friedman was the first to use "darwin" for MacOS.

Since the `uname` is Darwin, I think it's just a "side effect" ?
(although David Friedman did a lot of Mac OS X porting efforts)

The `uname` for Linux is : Linux, which makes the D version
"linux". And since the `arch` is ppc, let's call it "PPC" ? :-P

It's like logic. In reverse. But "Windows" and "linux" it is.

--anders

John Reimer wrote:

>>2) adds the identifiers "Unix", "darwin" and "IA64" (used by gdc-0.8)
> 
> So, Unix is being used instead of Posix still?

AFAIK, the DMD compiler only defines "linux" ?
"Unix" is being used by the GDC compiler (now)

Should there be some kind of new directive,
then "Posix" would be a good standard name...


I think it's a little like : Windows / Win32 ?
("operating system" vs. "application interface)

Or something... Just HAVING a collective name for
e.g. Linux, Mac OS X and FreeBSD would be a start.

--anders

On Tue, 07 Dec 2004 20:16:55 +0100, Anders F Björklund wrote:

> John Reimer wrote:
> 
>>>2) adds the identifiers "Unix", "darwin" and "IA64" (used by gdc-0.8)
>> 
>> So, Unix is being used instead of Posix still?
> 
> AFAIK, the DMD compiler only defines "linux" ?
> "Unix" is being used by the GDC compiler (now)
> 

Okay, then I wonder why it uses "Unix"?  Why carry gdc's use of "Unix" back to dmd?  I assume it's to maintain compatibility between the two compiler systems.  Shouldn't this be discussed first before it's presumed to be the right thing to do?

> Should there be some kind of new directive,
> then "Posix" would be a good standard name...
> 

I agree.

> I think it's a little like : Windows / Win32 ?
> ("operating system" vs. "application interface)

I see what you are saying.  But I don't think "Unix" fits into this
picture correctly.  "FreeBSD", "Linux", "Darwin", "AIX" -> "Posix" do fit.

> Or something... Just HAVING a collective name for e.g. Linux, Mac OS X and FreeBSD would be a start.

I agree.  Unix is too broad /and/ ambiguous a term to be used in a any specific naming scheme; and API-based names should be used if a more broad scheme is necesary.

It almost looks like gdc's naming methods should be changed then.

I apologize for getting repetitive about this; I just see no
strict organization or methodology to where the version schemes going. :-(

Later,

John

John Reimer wrote:

> Okay, then I wonder why it uses "Unix"?

David Friedman would know for sure, but the README says:

> Supported Systems
> 
>     * GCC 3.3.x, 3.4.x
>     * Linux (tested on Red Hat 7.2, 8)
>     * Mac OS X 10.3.x
>     * FreeBSD 5.2.1
>     * Cygwin
> 
> Similar versions should work and other Unix platforms may work.  Although
> the compiler will probably work on most 32-bit architectures, the D runtime
> library will still need to be updated to support them.

Which does mention the "Unix platforms" that later show
up in the patched code for Phobos as: "version(Unix)"

Such as it is being used in std.system:

>     // Operating system family
>     enum Family
>     {
> 	Win32 = 1,	// Microsoft 32 bit Windows systems
> 	linux,		// all linux systems
> 	Unix,           // all other
> 	}
> 
>     version (Win32)
>     {
> 	Family family = Family.Win32;
>     }
>     else version (linux)
>     {
> 	Family family = Family.linux;
>     }
>     else version (Unix)
>     {
>       Family family = Family.Unix;
>     }
>     else
>     {
> 	static assert(0);
>     }

You can download the gdc distribution and check yourself.
http://home.earthlink.net/~dvdfrdmn/d/gdc-0.8.tar.bz2

> Why carry gdc's use of "Unix" back to dmd? 

Probably since it's in use in the patched code already,
but it's probably not that hard to search/replace... ?


>>I think it's a little like : Windows / Win32 ?
>>("operating system" vs. "application interface)
> 
> I see what you are saying.  But I don't think "Unix" fits into this
> picture correctly.  "FreeBSD", "Linux", "Darwin", "AIX" -> "Posix" do fit.

No argument there. "version (Posix)" would work nice I think ?


And I wouldn't cry if "Linux" and "Darwin" were added too,
in addition to the current "linux" and "darwin" versions...

As for architectures, the offical list is missing PPC and
PPC64, but I have already reported that shortcoming here.

It is also missing "GNU", as the "compiler vendor" for GDC...
(although I'm not sure if GDC is an official project just yet)


With official, I mean: http://www.digitalmars.com/d/version.html

--anders

On Tue, 07 Dec 2004 19:59:22 +0100, Anders F Björklund <afb@algonet.se> wrote:
> Thomas Kuehne wrote:
>
>>> Did anyone particular decide "darwin" should be in lowercase?
>
> Walter.
>
>>> FTM, does anyone have any idea why "linux" is lowercase, unlike the others?
>>  Please, the lower case versus capital case issue for Linux has been beaten to
>> death for ages now.
>
> That doesn't make it stink less, though ?
>
> Most newcomers notice the smell, even if
> you get used to it after a while in here.
>
> Think it had some obscure compiler reason...

IIRC it's a copying of the C compilers "habits". The idea being people are used to it.

IMHO they should be case insensitive, after all when are you ever going to want "Abc" or "ABC" or "abc" to mean different things!?

Regan