From djberg96 at gmail.com  Sat Apr  7 08:21:50 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Sat, 7 Apr 2012 02:21:50 -0600
Subject: [Win32utils-devel] FFI and msvcrt
Message-ID: <CAGSpib=RvMbwetDB2GJDtGyykSXG7cPObrk8wdiyEK27sWppvg@mail.gmail.com>

Hi,

I've been using FFI with a Ruby 1.9.3 built with MSVC++ and it's been
working well. One thing I've run into though is this:

ffi_lib :msvcrt

But that's not the runtime I want. But I don't want to hard code the
runtime name either. I realize I could parse it out of RbConfig, but I
was hoping for something nicer.

Is there a way we could create an analogue to FFI::Library::LIBC for
the MSVC runtime? Something like FFI::Library::MSVCRT ?

Regards,

Dan

From phasis at gmail.com  Sat Apr  7 14:47:50 2012
From: phasis at gmail.com (Heesob Park)
Date: Sat, 7 Apr 2012 23:47:50 +0900
Subject: [Win32utils-devel] FFI and msvcrt
In-Reply-To: <CAGSpib=RvMbwetDB2GJDtGyykSXG7cPObrk8wdiyEK27sWppvg@mail.gmail.com>
References: <CAGSpib=RvMbwetDB2GJDtGyykSXG7cPObrk8wdiyEK27sWppvg@mail.gmail.com>
Message-ID: <CAHw=FdvhUvop=CXK1yc7-RKO6NPNUNb4LJx50Dpp+zMJ6=tqTw@mail.gmail.com>

Hi,


2012/4/7 Daniel Berger <djberg96 at gmail.com>

> Hi,
>
> I've been using FFI with a Ruby 1.9.3 built with MSVC++ and it's been
> working well. One thing I've run into though is this:
>
> ffi_lib :msvcrt
>
> But that's not the runtime I want. But I don't want to hard code the
> runtime name either. I realize I could parse it out of RbConfig, but I
> was hoping for something nicer.
>
> Is there a way we could create an analogue to FFI::Library::LIBC for
> the MSVC runtime? Something like FFI::Library::MSVCRT ?
>
>
I can find msvcrt runtime like this:

def find_msvcrt
require 'rbconfig'
 RbConfig::CONFIG['RUBY_SO_NAME'].split('-')[-2]+'.dll'
end

Did you mean this seems ugly?

Well, I cannot find a nicer method than above one.

Regards,

Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120407/3a51251f/attachment-0001.html>

From luislavena at gmail.com  Sat Apr  7 15:10:45 2012
From: luislavena at gmail.com (Luis Lavena)
Date: Sat, 7 Apr 2012 12:10:45 -0300
Subject: [Win32utils-devel] FFI and msvcrt
In-Reply-To: <CAHw=FdvhUvop=CXK1yc7-RKO6NPNUNb4LJx50Dpp+zMJ6=tqTw@mail.gmail.com>
References: <CAGSpib=RvMbwetDB2GJDtGyykSXG7cPObrk8wdiyEK27sWppvg@mail.gmail.com>
	<CAHw=FdvhUvop=CXK1yc7-RKO6NPNUNb4LJx50Dpp+zMJ6=tqTw@mail.gmail.com>
Message-ID: <CAN_J9aupAFJL3zJwqO12DQoPfVP8tzBCgs4XJyzVsWeyP91NBA@mail.gmail.com>

On Sat, Apr 7, 2012 at 11:47 AM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
>
> 2012/4/7 Daniel Berger <djberg96 at gmail.com>
>>
>> Hi,
>>
>> I've been using FFI with a Ruby 1.9.3 built with MSVC++ and it's been
>> working well. One thing I've run into though is this:
>>
>> ffi_lib :msvcrt
>>
>> But that's not the runtime I want. But I don't want to hard code the
>> runtime name either. I realize I could parse it out of RbConfig, but I
>> was hoping for something nicer.
>>
>> Is there a way we could create an analogue to FFI::Library::LIBC for
>> the MSVC runtime? Something like FFI::Library::MSVCRT ?
>>
>
> I can find msvcrt runtime like this:
>
> def find_msvcrt
> require 'rbconfig'
> RbConfig::CONFIG['RUBY_SO_NAME'].split('-')[-2]+'.dll'
> end
>
> Did you mean this seems ugly?
>
> Well, I cannot find a nicer method than above one.
>

Maybe something that correct the MSVCRT issue instead?

Here is the source of FFI::Library::MSVCRT:
https://github.com/ffi/ffi/blob/master/lib/ffi/platform.rb#L92-98

Perhaps something that uses RbConfig::CONFIG["RUBY_SO_NAME"] and
extract from it?

I bet Wayne (from FFI project) would love the contribution.

-- 
Luis Lavena
AREA 17
-
Perfection in design is achieved not when there is nothing more to add,
but rather when there is nothing more to take away.
Antoine de Saint-Exup?ry

From djberg96 at gmail.com  Sat Apr  7 17:21:12 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Sat, 7 Apr 2012 11:21:12 -0600
Subject: [Win32utils-devel] FFI and msvcrt
In-Reply-To: <CAHw=FdvhUvop=CXK1yc7-RKO6NPNUNb4LJx50Dpp+zMJ6=tqTw@mail.gmail.com>
References: <CAGSpib=RvMbwetDB2GJDtGyykSXG7cPObrk8wdiyEK27sWppvg@mail.gmail.com>
	<CAHw=FdvhUvop=CXK1yc7-RKO6NPNUNb4LJx50Dpp+zMJ6=tqTw@mail.gmail.com>
Message-ID: <CAGSpibmEQcaL3K0o6reWQ_tePrSpfPkYN6p6DFJxJRhbXOoOiQ@mail.gmail.com>

On Sat, Apr 7, 2012 at 8:47 AM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
>
> 2012/4/7 Daniel Berger <djberg96 at gmail.com>
>>
>> Hi,
>>
>> I've been using FFI with a Ruby 1.9.3 built with MSVC++ and it's been
>> working well. One thing I've run into though is this:
>>
>> ffi_lib :msvcrt
>>
>> But that's not the runtime I want. But I don't want to hard code the
>> runtime name either. I realize I could parse it out of RbConfig, but I
>> was hoping for something nicer.
>>
>> Is there a way we could create an analogue to FFI::Library::LIBC for
>> the MSVC runtime? Something like FFI::Library::MSVCRT ?
>>
>
> I can find msvcrt runtime like this:
>
> def find_msvcrt
> require 'rbconfig'
> RbConfig::CONFIG['RUBY_SO_NAME'].split('-')[-2]+'.dll'
> end
>
> Did you mean this seems ugly?
>
> Well, I cannot find a nicer method than above one.

Ok, thanks. I thought maybe there was a better way.

I've submitted a pull request: https://github.com/ffi/ffi/pull/199

It does feel a little funny overloading LIBC like that, but I guess
I'm ok with it. :)

Thanks again.

Dan

From djberg96 at gmail.com  Tue Apr 10 15:29:43 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Tue, 10 Apr 2012 09:29:43 -0600
Subject: [Win32utils-devel] FFI and DWORD
Message-ID: <CAGSpibkBtKxr1a3pQCFCcPxG_YcwNMZ_Kr-TB5dzka84HD20iQ@mail.gmail.com>

Hi,

A DWORD is declared as: typedef unsigned long DWORD;

However, the docs also say that it's a 32-bit unsigned integer. The
range is 0 through 4294967295 decimal.

When using FFI should I declare a DWORD as :ulong or :uint?

Regards,

Dan

From phasis at gmail.com  Tue Apr 10 15:48:41 2012
From: phasis at gmail.com (Heesob Park)
Date: Wed, 11 Apr 2012 00:48:41 +0900
Subject: [Win32utils-devel] FFI and DWORD
In-Reply-To: <CAGSpibkBtKxr1a3pQCFCcPxG_YcwNMZ_Kr-TB5dzka84HD20iQ@mail.gmail.com>
References: <CAGSpibkBtKxr1a3pQCFCcPxG_YcwNMZ_Kr-TB5dzka84HD20iQ@mail.gmail.com>
Message-ID: <CAHw=Fdu65tYsx7her1wDJqv4MzoA0PBTtqmtn6vDPrT27HL0kA@mail.gmail.com>

Hi,

2012/4/11 Daniel Berger <djberg96 at gmail.com>

> Hi,
>
> A DWORD is declared as: typedef unsigned long DWORD;
>
> However, the docs also say that it's a 32-bit unsigned integer. The
> range is 0 through 4294967295 decimal.
>
> When using FFI should I declare a DWORD as :ulong or :uint?
>
>
>
I think DWORD should be :ulong.

In the header file WinDef.h of Platform SDK, I can see

typedef unsigned long       DWORD;

Also refer to python's wintypes.py
http://coverage.livinglogic.de/Lib/ctypes/wintypes.py.html


Regards,

Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120411/8e023b0b/attachment.html>

From djberg96 at gmail.com  Wed Apr 11 10:02:25 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Wed, 11 Apr 2012 04:02:25 -0600
Subject: [Win32utils-devel] FFI and DWORD
In-Reply-To: <CAHw=Fdu65tYsx7her1wDJqv4MzoA0PBTtqmtn6vDPrT27HL0kA@mail.gmail.com>
References: <CAGSpibkBtKxr1a3pQCFCcPxG_YcwNMZ_Kr-TB5dzka84HD20iQ@mail.gmail.com>
	<CAHw=Fdu65tYsx7her1wDJqv4MzoA0PBTtqmtn6vDPrT27HL0kA@mail.gmail.com>
Message-ID: <CAGSpibn5YwXhAkZXg5O+9cHtie04vbrXYoZj_Av0WOSmwqVPHg@mail.gmail.com>

On Tue, Apr 10, 2012 at 9:48 AM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/4/11 Daniel Berger <djberg96 at gmail.com>
>>
>> Hi,
>>
>> A DWORD is declared as: typedef unsigned long DWORD;
>>
>> However, the docs also say that it's a 32-bit unsigned integer. The
>> range is 0 through 4294967295 decimal.
>>
>> When using FFI should I declare a DWORD as :ulong or :uint?
>>
>>
>
> I think DWORD should be :ulong.
>
> In the header file WinDef.h of Platform SDK, I can see
>
> typedef unsigned long ? ? ? DWORD;
>
> Also refer to python's
> wintypes.py?http://coverage.livinglogic.de/Lib/ctypes/wintypes.py.html

Ok, thanks, that's what I've been using.

Regards,

Dan

From djberg96 at gmail.com  Wed Apr 11 16:16:48 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Wed, 11 Apr 2012 10:16:48 -0600
Subject: [Win32utils-devel] FFI and errno on Windows
Message-ID: <CAGSpibn6EYNYE3j1HDEoS_Vx-+viG_PohvO2Tfog10TzJLbdRg@mail.gmail.com>

I'm a bit confused by error handling for posixy functions on Windows.
Consider the following code where I'm intentionally passing a bad
template to _mktemp. The docs say it should return EINVAL. My first
problem is not knowing exactly what the docs mean when they talk about
EINVAL on Windows. My second problem is that _get_errno returns a
different value that FFI.errno, and FFI.errno seems to return the same
value as GetLastError().

require 'ffi'
class Windows
  extend FFI::Library
  ffi_lib FFI::Library::LIBC

  attach_function :_mktemp, [:pointer], :string
  attach_function :_get_errno, [:pointer], :int

  ffi_lib :kernel32

  attach_function :GetLastError, [], :int

  def self.temp(template)
    result = _mktemp(template)

    err = get_err_num # 22
    #err = FFI.errno # 158
    #err = GetLastError() # 158

    if result.nil?
      raise SystemCallError, err, '_mktemp'
    end

    result
  end

  def self.get_err_num
    ptr = FFI::MemoryPointer.new(:int)

    if _get_errno(ptr) != 0
      raise SystemCallError, FFI.errno, '_get_errno'
    end

    ptr.read_int
  end
end

Windows.temp('xx')

Any insight on the subject of GetLastError vs _get_errno would be
greatly appreciated. My google skills seem to be failing.

Regards,

Dan

From phasis at gmail.com  Sun Apr 15 14:29:42 2012
From: phasis at gmail.com (Heesob Park)
Date: Sun, 15 Apr 2012 23:29:42 +0900
Subject: [Win32utils-devel] FFI and errno on Windows
In-Reply-To: <CAGSpibn6EYNYE3j1HDEoS_Vx-+viG_PohvO2Tfog10TzJLbdRg@mail.gmail.com>
References: <CAGSpibn6EYNYE3j1HDEoS_Vx-+viG_PohvO2Tfog10TzJLbdRg@mail.gmail.com>
Message-ID: <CAHw=FdtG24Xqg8TZPmqr+VgXdfEAgUbZkY5JsR4=QBZVbt=oYg@mail.gmail.com>

Hi,

2012/4/12 Daniel Berger <djberg96 at gmail.com>

> I'm a bit confused by error handling for posixy functions on Windows.
> Consider the following code where I'm intentionally passing a bad
> template to _mktemp. The docs say it should return EINVAL. My first
> problem is not knowing exactly what the docs mean when they talk about
> EINVAL on Windows. My second problem is that _get_errno returns a
> different value that FFI.errno, and FFI.errno seems to return the same
> value as GetLastError().
>
> require 'ffi'
> class Windows
>  extend FFI::Library
>  ffi_lib FFI::Library::LIBC
>
>  attach_function :_mktemp, [:pointer], :string
>  attach_function :_get_errno, [:pointer], :int
>
>  ffi_lib :kernel32
>
>  attach_function :GetLastError, [], :int
>
>  def self.temp(template)
>    result = _mktemp(template)
>
>    err = get_err_num # 22
>    #err = FFI.errno # 158
>    #err = GetLastError() # 158
>
>    if result.nil?
>      raise SystemCallError, err, '_mktemp'
>    end
>
>    result
>  end
>
>  def self.get_err_num
>    ptr = FFI::MemoryPointer.new(:int)
>
>    if _get_errno(ptr) != 0
>      raise SystemCallError, FFI.errno, '_get_errno'
>    end
>
>    ptr.read_int
>  end
> end
>
> Windows.temp('xx')
>
> Any insight on the subject of GetLastError vs _get_errno would be
> greatly appreciated. My google skills seem to be failing.
>
>
As you know, _mktemp is a function of C Runtime Library and provides
compatibility with mktemp of Linux.
According to the Linux manual page (
http://www.kernel.org/doc/man-pages/online/pages/man3/mktemp.3.html):

 EINVAL means that the last six characters of template were not XXXXXX.


FFI.errno is same to GetLastError() on Windows by the following lines of
LastError.c

void
rbffi_save_errno(void)
{
    int error = 0;

#ifdef _WIN32
    error = GetLastError();
#else
    error = errno;
#endif

    thread_data_get()->td_errno = error;
}

I think errno and GetLastError must be separated Windows.

The doc of mktemp mentioned only errno value. Thus GetLastError() value is
undetermined in this case.

Regards,

Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120415/6b0e51ac/attachment.html>

From djberg96 at gmail.com  Mon Apr 16 15:06:59 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Mon, 16 Apr 2012 11:06:59 -0400
Subject: [Win32utils-devel] FFI and errno on Windows
In-Reply-To: <CAHw=FdtG24Xqg8TZPmqr+VgXdfEAgUbZkY5JsR4=QBZVbt=oYg@mail.gmail.com>
References: <CAGSpibn6EYNYE3j1HDEoS_Vx-+viG_PohvO2Tfog10TzJLbdRg@mail.gmail.com>
	<CAHw=FdtG24Xqg8TZPmqr+VgXdfEAgUbZkY5JsR4=QBZVbt=oYg@mail.gmail.com>
Message-ID: <CAGSpib=iHdzjsvpAxHR5Bt3Tx=HiJiFgfoJzDjXqeVNa+dUG-Q@mail.gmail.com>

On Sun, Apr 15, 2012 at 10:29 AM, Heesob Park <phasis at gmail.com> wrote:
> Hi,
>
> 2012/4/12 Daniel Berger <djberg96 at gmail.com>
>>
>> I'm a bit confused by error handling for posixy functions on Windows.
>> Consider the following code where I'm intentionally passing a bad
>> template to _mktemp. The docs say it should return EINVAL. My first
>> problem is not knowing exactly what the docs mean when they talk about
>> EINVAL on Windows. My second problem is that _get_errno returns a
>> different value that FFI.errno, and FFI.errno seems to return the same
>> value as GetLastError().
>>
>> require 'ffi'
>> class Windows
>> ?extend FFI::Library
>> ?ffi_lib FFI::Library::LIBC
>>
>> ?attach_function :_mktemp, [:pointer], :string
>> ?attach_function :_get_errno, [:pointer], :int
>>
>> ?ffi_lib :kernel32
>>
>> ?attach_function :GetLastError, [], :int
>>
>> ?def self.temp(template)
>> ? ?result = _mktemp(template)
>>
>> ? ?err = get_err_num # 22
>> ? ?#err = FFI.errno # 158
>> ? ?#err = GetLastError() # 158
>>
>> ? ?if result.nil?
>> ? ? ?raise SystemCallError, err, '_mktemp'
>> ? ?end
>>
>> ? ?result
>> ?end
>>
>> ?def self.get_err_num
>> ? ?ptr = FFI::MemoryPointer.new(:int)
>>
>> ? ?if _get_errno(ptr) != 0
>> ? ? ?raise SystemCallError, FFI.errno, '_get_errno'
>> ? ?end
>>
>> ? ?ptr.read_int
>> ?end
>> end
>>
>> Windows.temp('xx')
>>
>> Any insight on the subject of GetLastError vs _get_errno would be
>> greatly appreciated. My google skills seem to be failing.
>>
>
> As you know, _mktemp is a function of C Runtime Library and provides
> compatibility with mktemp of Linux.
> According to the Linux manual page
> (http://www.kernel.org/doc/man-pages/online/pages/man3/mktemp.3.html):
>
> ?EINVAL means that the last six characters of template were not XXXXXX.
>
>
> FFI.errno is same to GetLastError() on Windows by the following lines of
> LastError.c
>
> void
> rbffi_save_errno(void)
> {
> ? ? int error = 0;
>
> #ifdef _WIN32
> ? ? error = GetLastError();
> #else
> ? ? error = errno;
> #endif
>
> ? ? thread_data_get()->td_errno = error;
> }
>
> I think errno and GetLastError must be separated Windows.
>
> The doc of mktemp mentioned only errno value.?Thus GetLastError() value is
> undetermined in this case.

I didn't quite understand you. Do you mean the current code for
FFI.errno is wrong? Should it be replaced with _get_errno?

Or is it simply up to library authors to know which one to use?

Regards,

Dan

From phasis at gmail.com  Tue Apr 17 00:38:54 2012
From: phasis at gmail.com (Heesob Park)
Date: Tue, 17 Apr 2012 09:38:54 +0900
Subject: [Win32utils-devel] FFI and errno on Windows
In-Reply-To: <CAGSpib=iHdzjsvpAxHR5Bt3Tx=HiJiFgfoJzDjXqeVNa+dUG-Q@mail.gmail.com>
References: <CAGSpibn6EYNYE3j1HDEoS_Vx-+viG_PohvO2Tfog10TzJLbdRg@mail.gmail.com>
	<CAHw=FdtG24Xqg8TZPmqr+VgXdfEAgUbZkY5JsR4=QBZVbt=oYg@mail.gmail.com>
	<CAGSpib=iHdzjsvpAxHR5Bt3Tx=HiJiFgfoJzDjXqeVNa+dUG-Q@mail.gmail.com>
Message-ID: <CAHw=FduYZfmRo7S3X66177+6htFLKznANMF+4UHnEL540SKnAA@mail.gmail.com>

Hi,

2012/4/17 Daniel Berger <djberg96 at gmail.com>

> On Sun, Apr 15, 2012 at 10:29 AM, Heesob Park <phasis at gmail.com> wrote:
> > Hi,
> >
> > 2012/4/12 Daniel Berger <djberg96 at gmail.com>
> >>
> >> I'm a bit confused by error handling for posixy functions on Windows.
> >> Consider the following code where I'm intentionally passing a bad
> >> template to _mktemp. The docs say it should return EINVAL. My first
> >> problem is not knowing exactly what the docs mean when they talk about
> >> EINVAL on Windows. My second problem is that _get_errno returns a
> >> different value that FFI.errno, and FFI.errno seems to return the same
> >> value as GetLastError().
> >>
> >> require 'ffi'
> >> class Windows
> >>  extend FFI::Library
> >>  ffi_lib FFI::Library::LIBC
> >>
> >>  attach_function :_mktemp, [:pointer], :string
> >>  attach_function :_get_errno, [:pointer], :int
> >>
> >>  ffi_lib :kernel32
> >>
> >>  attach_function :GetLastError, [], :int
> >>
> >>  def self.temp(template)
> >>    result = _mktemp(template)
> >>
> >>    err = get_err_num # 22
> >>    #err = FFI.errno # 158
> >>    #err = GetLastError() # 158
> >>
> >>    if result.nil?
> >>      raise SystemCallError, err, '_mktemp'
> >>    end
> >>
> >>    result
> >>  end
> >>
> >>  def self.get_err_num
> >>    ptr = FFI::MemoryPointer.new(:int)
> >>
> >>    if _get_errno(ptr) != 0
> >>      raise SystemCallError, FFI.errno, '_get_errno'
> >>    end
> >>
> >>    ptr.read_int
> >>  end
> >> end
> >>
> >> Windows.temp('xx')
> >>
> >> Any insight on the subject of GetLastError vs _get_errno would be
> >> greatly appreciated. My google skills seem to be failing.
> >>
> >
> > As you know, _mktemp is a function of C Runtime Library and provides
> > compatibility with mktemp of Linux.
> > According to the Linux manual page
> > (http://www.kernel.org/doc/man-pages/online/pages/man3/mktemp.3.html):
> >
> >  EINVAL means that the last six characters of template were not XXXXXX.
> >
> >
> > FFI.errno is same to GetLastError() on Windows by the following lines of
> > LastError.c
> >
> > void
> > rbffi_save_errno(void)
> > {
> >     int error = 0;
> >
> > #ifdef _WIN32
> >     error = GetLastError();
> > #else
> >     error = errno;
> > #endif
> >
> >     thread_data_get()->td_errno = error;
> > }
> >
> > I think errno and GetLastError must be separated Windows.
> >
> > The doc of mktemp mentioned only errno value. Thus GetLastError() value
> is
> > undetermined in this case.
>
> I didn't quite understand you. Do you mean the current code for
> FFI.errno is wrong? Should it be replaced with _get_errno?
>
> Or is it simply up to library authors to know which one to use?
>
>
>
I think FFI.errno should be errno regardless Linux or Windows and provide
another method for GetLastError() something like FFI.get_last_error.

Regards,

Park Heesob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/win32utils-devel/attachments/20120417/aafeb96d/attachment.html>

From djberg96 at gmail.com  Thu Apr 19 16:54:11 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Thu, 19 Apr 2012 12:54:11 -0400
Subject: [Win32utils-devel] FFI and wide character output buffers
Message-ID: <CAGSpib=Zy7GNeXL-RdsPK+v1qnWSQdrUVE_LDgKTGMFz4bGjXg@mail.gmail.com>

Hi all,

I've seen how to deal with wide input strings with FFI, but I'm not
sure how to deal with output buffers for wide character functions. I
tried this:

class Windows
  extend FFI::Library

  ffi_lib :kernel32

  attach_function :GetTempPathA, [:ulong, :pointer], :long
  attach_function :GetTempPathW, [:ulong, :pointer], :long

  def self.temp_path
    buf = 0.chr * 256
    buf.encode!("UTF-16LE")
    ptr = FFI::MemoryPointer.from_string(buf)

    # segfault or interpreter failure here
    if GetTempPathW(ptr.size, ptr) == 0
      raise SystemCallError, FFI.errno, "GetTempPathW"
    end

    p ptr.read_string
  end
end

Windows.temp_path

Any suggestions?

Regards,

Dan

From djberg96 at gmail.com  Thu Apr 19 21:23:34 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Thu, 19 Apr 2012 17:23:34 -0400
Subject: [Win32utils-devel] FFI and wide character output buffers
In-Reply-To: <CAGSpib=Zy7GNeXL-RdsPK+v1qnWSQdrUVE_LDgKTGMFz4bGjXg@mail.gmail.com>
References: <CAGSpib=Zy7GNeXL-RdsPK+v1qnWSQdrUVE_LDgKTGMFz4bGjXg@mail.gmail.com>
Message-ID: <CAGSpibmCfKb_M5E8Bh+bJxhHfFEE4BOdu745p+VzvpdNYWX2qA@mail.gmail.com>

On Thu, Apr 19, 2012 at 12:54 PM, Daniel Berger <djberg96 at gmail.com> wrote:
> Hi all,
>
> I've seen how to deal with wide input strings with FFI, but I'm not
> sure how to deal with output buffers for wide character functions. I
> tried this:
>
> class Windows
> ?extend FFI::Library
>
> ?ffi_lib :kernel32
>
> ?attach_function :GetTempPathA, [:ulong, :pointer], :long
> ?attach_function :GetTempPathW, [:ulong, :pointer], :long
>
> ?def self.temp_path
> ? ?buf = 0.chr * 256
> ? ?buf.encode!("UTF-16LE")
> ? ?ptr = FFI::MemoryPointer.from_string(buf)
>
> ? ?# segfault or interpreter failure here
> ? ?if GetTempPathW(ptr.size, ptr) == 0
> ? ? ?raise SystemCallError, FFI.errno, "GetTempPathW"
> ? ?end
>
> ? ?p ptr.read_string
> ?end
> end
>
> Windows.temp_path
>
> Any suggestions?

Disregard. Changing the second arg from :pointer to :buffer_out and
just using an encoded string worked.

Regards,

Dan

From djberg96 at gmail.com  Sat Apr 21 13:15:29 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Sat, 21 Apr 2012 09:15:29 -0400
Subject: [Win32utils-devel] RbConfig and CPPOUTFILE
Message-ID: <CAGSpibmyX9Sa7FvbhfFMPQbhajt2FLF_V5+W-K8xA7jgGkVx=Q@mail.gmail.com>

Hi,

For some reason with my version of Ruby built with MSVC++ 10 the
RbConfig::CONFIG['CPPOUTFILE'] value is set to "-P". Shouldn't it be
set to an actual file? With mingw it's set to "-o conftest.i"

I guess I was expecting something like "/Fo conftest.i".

I ran into this with mkmf-lite if you're curious, since I was using
the result of CPPOUTFILE to help create a build command. Or am I
misusing this?

Regards,

Dan

From djberg96 at gmail.com  Sat Apr 21 14:14:15 2012
From: djberg96 at gmail.com (Daniel Berger)
Date: Sat, 21 Apr 2012 10:14:15 -0400
Subject: [Win32utils-devel] RbConfig and CPPOUTFILE
In-Reply-To: <CAGSpibmyX9Sa7FvbhfFMPQbhajt2FLF_V5+W-K8xA7jgGkVx=Q@mail.gmail.com>
References: <CAGSpibmyX9Sa7FvbhfFMPQbhajt2FLF_V5+W-K8xA7jgGkVx=Q@mail.gmail.com>
Message-ID: <CAGSpib=VPecs-udiJV2ZYv_fKeKh2Rv8nZtpjSejM+CCcMT75g@mail.gmail.com>

On Sat, Apr 21, 2012 at 9:15 AM, Daniel Berger <djberg96 at gmail.com> wrote:
> Hi,
>
> For some reason with my version of Ruby built with MSVC++ 10 the
> RbConfig::CONFIG['CPPOUTFILE'] value is set to "-P". Shouldn't it be
> set to an actual file? With mingw it's set to "-o conftest.i"
>
> I guess I was expecting something like "/Fo conftest.i".
>
> I ran into this with mkmf-lite if you're curious, since I was using
> the result of CPPOUTFILE to help create a build command. Or am I
> misusing this?

Ok, I think I got a bit confused there. MS just defaults to using the
input file name + ".i" with the -P option, so I guess it's why it's
set like that.

Although, now I'm wondering why CPPOUTFILE is creating a .i file for
gcc, since that's typically used for preprocessed files and not
executables, right?

Regards,

Dan