[#5143] Win32API segfault in 1.8.3p1 — Nathaniel Talbott <ntalbott@...>
I'm on Windows XP, using VC7 to compile. I've previously gotten a good
Hi,
[#5151] COPY and INSTALL on Windows — Nathaniel Talbott <ntalbott@...>
1.8.3p1 has changed the defaults for the COPY and INSTALL Makefile
[#5152] 1.8.3 p1 segfault in array.c- bccwin32 - bcc5.5 (free) compiler bug — "daz" <dooby@...10.karoo.co.uk>
[#5160] Alternative for win32\ifchange.bat — "daz" <dooby@...10.karoo.co.uk>
[#5179] Cannot build HEAD on OS X 10.4.1 — Eric Hodel <drbrain@...7.net>
Somehow the rb_fd_init macro is conflicting with the definition of
Hi,
Hi,
[#5188] Re: IO#read — Yukihiro Matsumoto <matz@...>
Hi,
[#5190] Resolv and TTL — Eric Hodel <drbrain@...7.net>
I would like to retrieve the TTL values from Resolv, but they seem to
[#5206] Object#inspect() doesn't return; uses 100% cpu — Andrew Walrond <andrew@...>
Is this something I could have caused by overriding some method on the
[#5211] ruby 1.8 CVS do not work with --enable-pthread configure option — noreply@...
Bugs item #2038, was opened at 2005-06-16 13:57
[#5215] Hackers Guide Translation Request! — "Charles E. Thornton" <ruby-core@...>
I have recently discovered RUBY and want to understand it a deep level -
[#5219] Segmentation fault in timeout.rb — Michel Pastor <K@...>
Hi,
On Fri, 17 Jun 2005 05:03:18 +0900
Hi,
On Fri, 17 Jun 2005 11:51:07 +0900
Hi,
On Sat, 18 Jun 2005 10:28:53 +0900
Hi,
On Sun, 19 Jun 2005 23:05:56 +0900
[#5233] event_hook shows weirdness when invoked on mixed in methods — Ryan Davis <ryand-ruby@...>
The following attachment, when run, shows the following behavior:
[#5264] XMLRPC vulnerabilities? — Hugh Sasse <hgs@...>
I've just seen this (by RSS)
[#5267] RubyUnit Test Ordering — Jordan Gilliland <jordan@...>
I'm using ruby 1.8.2 (2004-12-25) [i686-linux] and I've noticed that the
[#5277] Macros in win32.h — james@...
win32.h defines a load of macros. This means any C or C++ program which embeds
[#5288] committing rdoc additions corrections to head? — Ryan Davis <ryand-ruby@...>
There is some discussion on ruby-doc about people documenting core
[#5296] Subversion — Shugo Maeda <shugo@...>
Hi,
Shugo Maeda wrote:
Curt Hibbs wrote:
On 6/30/05, Nikolai Weibull
Austin Ziegler wrote:
On 6/30/05, Nikolai Weibull
Austin Ziegler wrote:
On 6/30/05, mathew <meta@pobox.com> wrote:
Austin Ziegler wrote:
On 7/1/05, mathew <meta@pobox.com> wrote:
Austin Ziegler wrote:
On 7/1/05, Nikolai Weibull
Austin Ziegler wrote:
On 7/1/05, mathew <meta@pobox.com> wrote:
Austin Ziegler wrote:
On Thursday 30 June 2005 19:19, mathew wrote:
"Sean E. Russell" <ser@germane-software.com> writes:
On 30 Jun 2005, at 08:19, Shugo Maeda wrote:
Hi,
Where to put non-GNU build system components
VMS doesn't have a complete GNU environment yet. (See gnv.sf.net for the status of this work.) Thus, I have some files that would ordinarily be build products from running ./configure and make that need to be part of the VMS port. Now, it seems the cleanest thing to do would be to locate these all in ruby/vms, but I'm not sure exactly how. I have one .opt file (linker options) and one .mms file (VMS's own non-GNU makefile) per directory. One idea I am considering would be to put an entire skeletal build tree in CVS under ruby/vms/build and just deposit the .opt and .mms files here. The top-level descrip.mms file in ruby/vms/build would then copy all source files into the build tree for the build, and clean them up again afterwards, along with all other build products. Using this plan, we would have in CVS: ruby/vms/build ruby/vms/build/descrip.mms (top-level Makefile) ... ruby/vms/build/ext/socket/descrip.mms ruby/vms/build/ext/socket/link.opt ... I am wary, though, of creating such a large directory structure in CVS, given that deletion of CVS directories is impossible, and ultimately we would like this "scaffolding" for a proper GNU tools based build system to disappear. Alternatively, I could include a single patchfile in CVS: ruby/vms/build.diff This patchfile could be used to create the build structure. However, maintaining this patchfile would be unpleasant. Another alternative is that the .mms and .opt files could be generated somehow. But at this stage in the port, I don't know if I could write such a generator, except for the very simple .opt files, without extra data files per directory to support all of the specific needs of each module. Finally, if I felt that GNV would very soon provide a complete build system so I could do away with these extra files, I could simply maintain all of these files outside of CVS, with a view to replacing them with a proper GNU-based build system as soon as GNV is ready. However, I don't have faith that GNV will finish this work soon, and besides it would make it difficult for any others to obtain the VMS port and try to build it themselves. This is my least favourite alternative, but is in fact what I am doing today. Thoughts, anyone? Ben