win32utils devel - Oct 2008 - Problems with win32-dir and create_junction

Hi,

I don''t know when this started happening, but I can''t
win32-dir tests to pass
any more for the create_junction method.

It looks like the junction is created, but it''s not valid.
Here''s the results of
the tests on Vista using the latest from CVS:

Started
....................E.E.................................

Finished in 0.477 seconds.

   1) Error:
test_create_junction_ascii(TC_Win32_Dir):
Errno::EINVAL: Invalid argument - test_to_directory
     ./test/test_dir.rb:38:in `open''
     ./test/test_dir.rb:38:in `entries''
     ./test/test_dir.rb:38:in `test_create_junction_ascii''

   2) Error:
test_create_junction_unicode(TC_Win32_Dir):
Errno::EINVAL: Invalid argument - ??????????
     ./test/test_dir.rb:44:in `open''
     ./test/test_dir.rb:44:in `entries''
     ./test/test_dir.rb:44:in `test_create_junction_unicode''

56 tests, 111 assertions, 0 failures, 2 errors, 0 pendings, 0 omissions, 0 
notifications
rake aborted!
Command failed with status (1): [C:/Ruby/bin/ruby -w -Ilib
"C:/Ruby/lib/rub...]

(See full trace by running task with --trace)

Any ideas?

Thanks,

Dan

Hi,

2008/10/2 Daniel Berger <djberg96 at gmail.com>:
> Hi,
>
> I don''t know when this started happening, but I can''t
win32-dir tests to
> pass any more for the create_junction method.
>
> It looks like the junction is created, but it''s not valid.
Here''s the
> results of the tests on Vista using the latest from CVS:
>
> Started
> ....................E.E.................................
>
> Finished in 0.477 seconds.
>
>  1) Error:
> test_create_junction_ascii(TC_Win32_Dir):
> Errno::EINVAL: Invalid argument - test_to_directory
>    ./test/test_dir.rb:38:in `open''
>    ./test/test_dir.rb:38:in `entries''
>    ./test/test_dir.rb:38:in `test_create_junction_ascii''
>
>  2) Error:
> test_create_junction_unicode(TC_Win32_Dir):
> Errno::EINVAL: Invalid argument - ??????????
>    ./test/test_dir.rb:44:in `open''
>    ./test/test_dir.rb:44:in `entries''
>    ./test/test_dir.rb:44:in `test_create_junction_unicode''
>
> 56 tests, 111 assertions, 0 failures, 2 errors, 0 pendings, 0 omissions, 0
> notifications
> rake aborted!
> Command failed with status (1): [C:/Ruby/bin/ruby -w -Ilib
> "C:/Ruby/lib/rub...]
>
> (See full trace by running task with --trace)
>
> Any ideas?
>
It is due to the change of multi_to_wide method.
Trailing \0''s and wide_string.size mismatch problem.

Modify line#111 of dir.rb
   wide_string = multi_to_wide(buf_target)
To
   wide_string = multi_to_wide(buf_target)[0..-3]
works for me in case of ascii junction.

In case of the unicode, more work is needed.

Regards,

Park Heesob

> -----Original Message-----
> From: win32utils-devel-bounces at rubyforge.org 
> [mailto:win32utils-devel-bounces at rubyforge.org] On Behalf Of 
> Heesob Park
> Sent: Thursday, October 02, 2008 7:59 AM
> To: Development and ideas for win32utils projects
> Subject: Re: [Win32utils-devel] Problems with win32-dir and 
> create_junction
> 
> Hi,
> 
> 2008/10/2 Daniel Berger <djberg96 at gmail.com>:
> > Hi,
> >
> > I don''t know when this started happening, but I
can''t
> win32-dir tests 
> > to pass any more for the create_junction method.
> >
> > It looks like the junction is created, but it''s not valid. 
> Here''s the 
> > results of the tests on Vista using the latest from CVS:
> >
> > Started
> > ....................E.E.................................
> >
> > Finished in 0.477 seconds.
> >
> >  1) Error:
> > test_create_junction_ascii(TC_Win32_Dir):
> > Errno::EINVAL: Invalid argument - test_to_directory
> >    ./test/test_dir.rb:38:in `open''
> >    ./test/test_dir.rb:38:in `entries''
> >    ./test/test_dir.rb:38:in `test_create_junction_ascii''
> >
> >  2) Error:
> > test_create_junction_unicode(TC_Win32_Dir):
> > Errno::EINVAL: Invalid argument - ??????????
> >    ./test/test_dir.rb:44:in `open''
> >    ./test/test_dir.rb:44:in `entries''
> >    ./test/test_dir.rb:44:in `test_create_junction_unicode''
> >
> > 56 tests, 111 assertions, 0 failures, 2 errors, 0 pendings, 0 
> > omissions, 0 notifications rake aborted!
> > Command failed with status (1): [C:/Ruby/bin/ruby -w -Ilib 
> > "C:/Ruby/lib/rub...]
> >
> > (See full trace by running task with --trace)
> >
> > Any ideas?
> >
> It is due to the change of multi_to_wide method.
> Trailing \0''s and wide_string.size mismatch problem.
> 
> Modify line#111 of dir.rb
>    wide_string = multi_to_wide(buf_target) To
>    wide_string = multi_to_wide(buf_target)[0..-3] works for 
> me in case of ascii junction.
Ah, ok. Thanks.

Do you think we should automatically strip the trailing "\000\000" in
the multi_to_wide method then? I think this would work because I modified
wide_to_multi to automatically append "\000\000" to the incoming
string if they weren''t already there.
 
> In case of the unicode, more work is needed.
Any suggestions?

Regards,

Dan


This communication is the property of Qwest and may contain confidential or
privileged information. Unauthorized use of this communication is strictly 
prohibited and may be unlawful.  If you have received this communication 
in error, please immediately notify the sender by reply e-mail and destroy 
all copies of the communication and any attachments.

> -----Original Message-----
> From: win32utils-devel-bounces at rubyforge.org 
> [mailto:win32utils-devel-bounces at rubyforge.org] On Behalf Of 
> Berger, Daniel
> Sent: Thursday, October 02, 2008 8:12 AM
> To: Development and ideas for win32utils projects
> Subject: Re: [Win32utils-devel] Problems with win32-dir and 
> create_junction
<snip>
> Do you think we should automatically strip the trailing 
> "\000\000" in the multi_to_wide method then? I think this 
> would work because I modified wide_to_multi to automatically 
> append "\000\000" to the incoming string if they weren''t
> already there.
On second thought, I think that might cause wide character functions to
fail.

Regards,

Dan


This communication is the property of Qwest and may contain confidential or
privileged information. Unauthorized use of this communication is strictly 
prohibited and may be unlawful.  If you have received this communication 
in error, please immediately notify the sender by reply e-mail and destroy 
all copies of the communication and any attachments.

2008/10/2 Berger, Daniel <Daniel.Berger at
qwest.com>:
>> -----Original Message-----
>> From: win32utils-devel-bounces at rubyforge.org
>> [mailto:win32utils-devel-bounces at rubyforge.org] On Behalf Of
>> Berger, Daniel
>> Sent: Thursday, October 02, 2008 8:12 AM
>> To: Development and ideas for win32utils projects
>> Subject: Re: [Win32utils-devel] Problems with win32-dir and
>> create_junction
>
> <snip>
>
>> Do you think we should automatically strip the trailing
>> "\000\000" in the multi_to_wide method then? I think this
>> would work because I modified wide_to_multi to automatically
>> append "\000\000" to the incoming string if they
weren''t
>> already there.
>
> On second thought, I think that might cause wide character functions to
> fail.
>
I think it would be better to define new method for getting the
wide_string length
than to use String#size method.

Regards,

Park Heesob

Heesob Park wrote:
> 2008/10/2 Berger, Daniel <Daniel.Berger at qwest.com>:
>>> -----Original Message-----
>>> From: win32utils-devel-bounces at rubyforge.org
>>> [mailto:win32utils-devel-bounces at rubyforge.org] On Behalf Of
>>> Berger, Daniel
>>> Sent: Thursday, October 02, 2008 8:12 AM
>>> To: Development and ideas for win32utils projects
>>> Subject: Re: [Win32utils-devel] Problems with win32-dir and
>>> create_junction
>> <snip>
>>
>>> Do you think we should automatically strip the trailing
>>> "\000\000" in the multi_to_wide method then? I think this
>>> would work because I modified wide_to_multi to automatically
>>> append "\000\000" to the incoming string if they
weren''t
>>> already there.
>> On second thought, I think that might cause wide character functions to
>> fail.
>>
> I think it would be better to define new method for getting the
> wide_string length
> than to use String#size method.
I have a very alpha "UString" class I''ve got in CVS (in the
"shards" project,
not win32utils) where I''ve been tinkering with a Windows-only subclass
of
String. It redefines methods like length, size, etc. I''ve pasted what
I''ve done
so far below.

Or do you think that''s overkill?

Regards,

Dan

# ustring.rb - ALPHA

class UString < String
    include Windows::Unicode
    include Windows::MSVCRT::String

    VERSION = ''0.1.0''

    ACP  = CP_ACP
    UTF7 = CP_UTF7
    UTF8 = CP_UTF8

    attr_accessor :encoding

    alias :at :[]

    def initialize(string, encoding = UTF8)
       ustring = multi_to_wide(string, encoding)
       @encoding = encoding
       super(ustring)
    end

    def inspect
       wide_to_multi(self, @encoding)
    end

    def to_s
       wide_to_multi(self, @encoding)
    end

    def to_str
       wide_to_multi(self, @encoding)
    end

    # Length in chars
    def length
       wcslen(self)
    end

    # Size in bytes
    def size
       strlen(wide_to_multi(self, @encoding))
    end

    # Returns the character (not the numeric value)
    def [](n)
       wide_to_multi(self, @encoding).at(n, 1)
    end
end

Hi, 
> -----Original Message-----
> From: win32utils-devel-bounces at rubyforge.org 
> [mailto:win32utils-devel-bounces at rubyforge.org] On Behalf Of 
> Heesob Park
> Sent: Thursday, October 02, 2008 8:25 AM
> To: Development and ideas for win32utils projects
> Subject: Re: [Win32utils-devel] Problems with win32-dir and 
> create_junction
> 
> 2008/10/2 Berger, Daniel <Daniel.Berger at qwest.com>:
> >> -----Original Message-----
> >> From: win32utils-devel-bounces at rubyforge.org
> >> [mailto:win32utils-devel-bounces at rubyforge.org] On Behalf 
> Of Berger, 
> >> Daniel
> >> Sent: Thursday, October 02, 2008 8:12 AM
> >> To: Development and ideas for win32utils projects
> >> Subject: Re: [Win32utils-devel] Problems with win32-dir and 
> >> create_junction
> >
> > <snip>
> >
> >> Do you think we should automatically strip the trailing 
> "\000\000" in 
> >> the multi_to_wide method then? I think this would work because I 
> >> modified wide_to_multi to automatically append
"\000\000" to the
> >> incoming string if they weren''t already there.
> >
> > On second thought, I think that might cause wide character 
> functions 
> > to fail.
> >
> I think it would be better to define new method for getting 
> the wide_string length than to use String#size method.
Here''s a minimal WideString class:

   require ''windows/unicode''
   require ''windows/msvcrt/string''

   class WideString < String
      include Windows::Unicode
      include Windows::MSVCRT::String
   
      ACP  = CP_ACP
      UTF7 = CP_UTF7
      UTF8 = CP_UTF8
   
      attr_accessor :encoding
   
      def initialize(string, encoding = UTF8)
         super(multi_to_wide(string, encoding))
         @encoding = encoding
      end
   
      def length
         wcslen(self) * 2      
      end

      alias size length   
   end

Another approach is to monkey-patch String:

   require ''windows/msvcrt/string''

   class String
      include Windows::MSVCRT::String
      def wlength
         wcslen(self) * 2
      end
      alias wsize wlength
   end

What do you think?

Regards,

Dan


This communication is the property of Qwest and may contain confidential or
privileged information. Unauthorized use of this communication is strictly 
prohibited and may be unlawful.  If you have received this communication 
in error, please immediately notify the sender by reply e-mail and destroy 
all copies of the communication and any attachments.