win32utils devel - Jan 2010 - Strange nmake bug with Sapphire, Vista, VC 9

I see this when I try to compile Sapphire on Vista with VC 9:

compiling Win32API

Microsoft (R) Program Maintenance Utility Version 9.00.21022.08
Copyright (C) Microsoft Corporation.  All rights reserved.

makefile(1) : fatal error U1007: double quotation mark not allowed in name
Stop.
NMAKE : fatal error U1077: ''.\miniruby.exe'' : return code
''0x1''
Stop.

Sapphire is based on ruby-1.8.6-p383, though a fresh download of that source
builds fine. I don''t see this behavior on OS X, either.

Any ideas? Could it be this commit?

http://github.com/djberg96/Sapphire/commit/8bfbabe633b270af655709fd030c9556f
bf2c6a0

Otherwise, I''m not sure what the heck I did and I''ll to revert
one commit at
a time until I find the culprit. :(

Regards,

Dan

git://github.com/djberg96/Sapphire.git

Hi,

2010/1/24 Daniel Berger <djberg96 at gmail.com>:
> I see this when I try to compile Sapphire on Vista with VC 9:
>
> compiling Win32API
>
> Microsoft (R) Program Maintenance Utility Version 9.00.21022.08
> Copyright (C) Microsoft Corporation. ?All rights reserved.
>
> makefile(1) : fatal error U1007: double quotation mark not allowed in name
> Stop.
> NMAKE : fatal error U1077: ''.\miniruby.exe'' : return code
''0x1''
> Stop.
>
> Sapphire is based on ruby-1.8.6-p383, though a fresh download of that
source
> builds fine. I don''t see this behavior on OS X, either.
>
> Any ideas? Could it be this commit?
>
>
http://github.com/djberg96/Sapphire/commit/8bfbabe633b270af655709fd030c9556f
> bf2c6a0
>
> Otherwise, I''m not sure what the heck I did and I''ll to
revert one commit at
> a time until I find the culprit. :(
>
The culprit is Array#to_s, of course.
Here is a patch for lib/mkmf.rb:

--- mkmf.rb.old 2010-01-24 18:35:22.000000000 +0900
+++ mkmf.rb     2010-01-24 18:35:01.000000000 +0900
@@ -1236,7 +1236,7 @@
   dllib = target ? "$(TARGET).#{CONFIG[''DLEXT'']}" :
""
   staticlib = target ? "$(TARGET).#$LIBEXT" : ""
   mfile = open("Makefile", "wb")
-  mfile.print configuration(srcprefix)
+  mfile.print configuration(srcprefix).join("")
   mfile.print "
 libpath = #{($DEFLIBPATH|$LIBPATH).join(" ")}
 LIBPATH = #{libpath}
@@ -1427,7 +1427,7 @@
       mfile.print ".SUFFIXES: .", suffixes.uniq.join(" ."),
"\n\n"
     end
     mfile.print "$(OBJS): $(RUBY_EXTCONF_H)\n\n" if $extconf_h
-    mfile.print depout
+    mfile.print depout.join("")
   else
     headers = %w[ruby.h defines.h]
     if RULE_SUBST


Regards,

Park Heesob

On Sun, Jan 24, 2010 at 2:43 AM, Heesob Park <phasis at gmail.com>
wrote:
> Hi,
>
> 2010/1/24 Daniel Berger <djberg96 at gmail.com>:
>> I see this when I try to compile Sapphire on Vista with VC 9:
>>
>> compiling Win32API
>>
>> Microsoft (R) Program Maintenance Utility Version 9.00.21022.08
>> Copyright (C) Microsoft Corporation. ?All rights reserved.
>>
>> makefile(1) : fatal error U1007: double quotation mark not allowed in
name
>> Stop.
>> NMAKE : fatal error U1077: ''.\miniruby.exe'' : return
code ''0x1''
>> Stop.
>>
>> Sapphire is based on ruby-1.8.6-p383, though a fresh download of that
source
>> builds fine. I don''t see this behavior on OS X, either.
>>
>> Any ideas? Could it be this commit?
>>
>>
http://github.com/djberg96/Sapphire/commit/8bfbabe633b270af655709fd030c9556f
>> bf2c6a0
>>
>> Otherwise, I''m not sure what the heck I did and I''ll
to revert one commit at
>> a time until I find the culprit. :(
>>
> The culprit is Array#to_s, of course.
> Here is a patch for lib/mkmf.rb:
>
> --- mkmf.rb.old 2010-01-24 18:35:22.000000000 +0900
> +++ mkmf.rb ? ? 2010-01-24 18:35:01.000000000 +0900
> @@ -1236,7 +1236,7 @@
> ? dllib = target ?
"$(TARGET).#{CONFIG[''DLEXT'']}" : ""
> ? staticlib = target ? "$(TARGET).#$LIBEXT" : ""
> ? mfile = open("Makefile", "wb")
> - ?mfile.print configuration(srcprefix)
> + ?mfile.print configuration(srcprefix).join("")
> ? mfile.print "
> ?libpath = #{($DEFLIBPATH|$LIBPATH).join(" ")}
> ?LIBPATH = #{libpath}
> @@ -1427,7 +1427,7 @@
> ? ? ? mfile.print ".SUFFIXES: .", suffixes.uniq.join("
."), "\n\n"
> ? ? end
> ? ? mfile.print "$(OBJS): $(RUBY_EXTCONF_H)\n\n" if $extconf_h
> - ? ?mfile.print depout
> + ? ?mfile.print depout.join("")
> ? else
> ? ? headers = %w[ruby.h defines.h]
> ? ? if RULE_SUBST
Thanks!

I ultimately decided to undo the change that originally caused this
error instead.

Regards,

Dan