On 04/25/2015 03:48 AM, Branko Čibej wrote:
On 24.04.2015 10:46, Alexander Thomas wrote:
On 04/24/2015 12:47 PM, Branko Čibej wrote:
Can you try compiling with the following patch, please, and report the
results?
ndings/javahl
Here is result after patching ...
[[[
subversion/bindings
On 04/24/2015 12:47 PM, Branko Čibej wrote:
Can you try compiling with the following patch, please, and report the
results?
Here is result after patching ...
[[[
subversion/bindings/javahl/native/jniwrapper/jni_io_stream.cpp: In function
`svn_error_t* Javaglobal_stream_close_input(voi
On 04/24/2015 12:47 PM, Branko Čibej wrote:
if (contents != NULL)
{
Java::Env env;
SVN_JAVAHL_CATCH(Java::Env(), ...);
}
if (contents != NULL)
{
Java::Env env;
SVN_JAVAHL_CATCH(env,...);
}
This I tried earlier even before I tried upgrading GCC
Compiling 1.9.0-rc1 JavaHL in RHEL3 32bit with GCC version 3.2.3-53 and
libstdc++ version 5.0.3 fails with multiple compilation error in
javahl/native/EditorProxy.cpp. In addition to line number 213, line 360
also fails with same error.
However I was able to successfully compile against GCC 4.
; was my initial workaround, but this tends to fails
later while compiling bindings (IIRC).
--
Alexander Thomas (AT)
CollabNet <> www.collab.net <> Enterprise Cloud Development
signature.asc
Description: OpenPGP digital signature
On Friday 08 March 2013 06:38 PM, Philip Martin wrote:
Alexander Thomas writes:
svn@trunk compiled in Linux x86_64 with libc-iconv, segfaults when executed
with an argument. I don't understand libsvn_subr/utf.c code well enough.
Here I'm attaching gdb traceback for "svn -
Wed Jul 12 23:32:02 EDT 2006
x86_64 x86_64 x86_64 GNU/Linux
* subversion trunk@1452694
* glibc-2.3.4
* apr-1.4.6
* apr-util-1.5.1
Thanks
-Alexander Thomas (AT)
GNU gdb Red Hat Linux (6.3.0.0-1.132.EL4rh)
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU Ge
Don't break 'make install-tools' if TOOLSDIR is same as BINDIR.
* Makefile.in
(INSTALL_EXTRA_TOOLS): Check that target and link_name are not the
same before creating symlink to svnmucc in BINDIR.
Patch by: Alexander Thomas
]]]
Thanks
-Alexander Thomas (AT
nk: cannot determine absolute directory name of `no/lib'
]]]
[[[
Fixes build failure when configured with '--without-zlib'.
* build/ac-macros/zlib.m4
(SVN_LIB_Z): Configure error-outs if executed with '--without-zlib'
or '--with-zlib=no' options.
Patch by: Alexan
[[[
On the 'gpg-agent-password-store' branch:
* configure.ac:
(SVN_HAVE_GPG_AGENT): Defines when --with-gpg-agent option passed to
'configure'
]]]
Index: configure.ac
===
--- configure.ac (revision 1005331)
+++ configure.ac
This happens in Solaris10 SPARC, works well in Linux. After executing
the SVN testsuite I tried removing the svn-test-work folder using 'rm
-rf' which failed with
"rm: Unable to remove directory : File exists" errors.
and following directories not deleted from
'svn-test-work/working_copies'
On Sat, 2010-06-12 at 13:04 +0200, Stefan Sperling wrote:
> On Fri, Jun 11, 2010 at 02:30:09PM +0530, Alexander Thomas wrote:
> > On Fri, 2010-06-11 at 09:53 +0100, Philip Martin wrote:
> > > Alexander Thomas writes:
> > >
> > > > Yes SUNWuiu8 is the one th
On Fri, 2010-06-11 at 09:53 +0100, Philip Martin wrote:
> Alexander Thomas writes:
>
> > Yes SUNWuiu8 is the one that comes default with Solaris10 and 646/ASCII
> > conversion to UTF-8 works fine. But for other reasons I should continue
> > using gnu-libiconv.
>
&g
On Fri, 2010-06-11 at 08:17 +0100, Philip Martin wrote:
> Alexander Thomas writes:
>
> > Code snippet produced "646".
> >
> > "iconv -f 646 -t UTF-8 /dev/null" returned
> > "iconv: conversion from `646' is not supported"
On Thu, 2010-06-10 at 18:23 +0200, Stefan Sperling wrote:
> On Wed, Jun 09, 2010 at 01:31:54PM +0200, Stefan Sperling wrote:
> > > PASS: prop_tests.py 22: test prop* handle invalid property names
> >
> > Hmmm... I will try to debug this during the hackathon.
>
> This is not a problem in Subver
On Wed, 2010-06-09 at 12:36 +0200, Stefan Sperling wrote:
> On Wed, Jun 09, 2010 at 02:03:01PM +0530, Alexander Thomas wrote:
> > > The test used to check for an unprintable ascii character (backspace).
> > > With this patch, it checks for an empty string.
> > >
On Wed, 2010-06-09 at 09:51 +0200, Stefan Sperling wrote:
> On Wed, Jun 09, 2010 at 12:55:26PM +0530, Alexander Thomas wrote:
> > [[[
> > Replace existing invalid propname to make test pass in Solaris.
> >
> > * subversion/tests/cmdline/prop_tests.py
> > (invali
[[[
Replace existing invalid propname to make test pass in Solaris.
* subversion/tests/cmdline/prop_tests.py
(invalid_propnames): Changed propname to empty char.
]]]
-Alexander Thomas (AT)
Index: subversion/tests/cmdline/prop_tests.py
On Mon, 2009-12-14 at 13:47 +, Philip Martin wrote:
> Alexander Thomas writes:
>
> > [[[
> > Allow build against the full path to neon-config.
> >
> > * build/ac-macros/neon.m4
> > (SVN_LIB_NEON): Modified to accept full path to neon-config
[[[
Allow build against the full path to neon-config.
* build/ac-macros/neon.m4
(SVN_LIB_NEON): Modified to accept full path to neon-config. Fixed doc
string accordingly.
]]]
-Alexander Thomas(AT)
Index: build/ac-macros/neon.m4
20 matches
Mail list logo
