.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone-devel
will become less of an issue.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone
before 0.19.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone-devel
was _only_ used to sync against each
developer's local database, but even then you would need a way to
coordinate access to that shared database, and it's questionable whether
SQLite's locking will work with any given NFS server[0].
[0] http://www.sqlite.org/faq.html#q7
Cheers,
-mjg
--
Matthew
upstream.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone-devel
if that would be a
workable solution to our problem.
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org
is officially unsupported by Microsoft (or is soon to be), and has
roughly the same level of C++ standards compliance as a Perl
interpreter. ;-)
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
is the domain of autotest scriptlets.)
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo
--revision t:release-0.9' to diff the current working
copy against whichever revision was tagged 'release-0.9'.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel
At 2005-08-27T18:49:43-0700, Howard Spindel wrote:
I've been corresponding with Matthew Gregan about some continued
problems I've been having with monotone, but enough new stuff has
come to light that I thought a post to the group as a whole might
stir some useful thoughts.
I just sent you
to mainline (rev. fc87df) to work around this
problem.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http
with the MinGW C++ support runtime, which is used by the
official Win32 binaries.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel
when running with mismatched C++ libraries before.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http
in
1fead0fcebb8d252b733110cd7a1b1e4c8adbd77.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org
, instead:
$ monotone serve 0.0.0.0 '*'
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman
of the Cygwin environment.
So the ls command works fine. Oh and it is not the fact
that ls is part of msys. If you replace ls with notepad.exe
the expansion behaviour is correct as well.
notepad.exe, like most native Windows applications, performs its own
glob expansion.
Cheers,
-mjg
--
Matthew
At 2005-10-12T13:19:11+0200, Wim Oudshoorn wrote:
Matthew Gregan [EMAIL PROTECTED] writes:
[Interesting explanation how expansion on MS windows work deleted]
So the ls command works fine. Oh and it is not the fact
that ls is part of msys. If you replace ls with notepad.exe
, this could help:
'automate keys' is new in mainline--it provides a more detailed, basic_io
formatted, version of the 'list keys' output.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
version 0.23.
Not that this really bothers me, but for new users it can look intimidating.
Thanks for the bug report. This was fixed by Nathaniel in mainline
(revision e893fff9bd48f5bec46a1fb338d4b36ab8c898cb) a few days ago.
Cheers,
-mjg
--
Matthew Gregan
It seems it can detect that the DB is locked but the error
message the user sees is not very clear.
Thanks,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing
in 45510da1bc74f0164cf0fd09feb5de75689de721 on
mainline that will allow monotone to detect a locked database much earlier
and provide a better error message to the user.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
initialization problem (we've seen these before on
different platforms when doing ports of monotone), but I haven't had time to
dig into it yet.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone-devel
to be portable across many platforms, and it requires a casual user
who is compiling from source to know _more_ about their particular system.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
very soon, will contain this fix.
As a workaround, delete the lines from the dump output that perform the
create and inserts on sqlite_stat1, then load it as you would normally.
Thanks,
-mjg
--
Matthew Gregan |/
/|[EMAIL
Makefile.am) -- this should
work fine, as Botan also collects entropy via CryptoAPI on Windows.
- improve Win32_EntropySource::do_slow_poll() to detect at runtime and use
only those APIs available on the platform. OpenSSL's RAND_poll might be
useful for ideas.
Cheers,
-mjg
--
Matthew
: the line '' in your .mt-ignore file caused exception 'Empty
expression' while matching filename 'cryptopp/.deps
/lib3rdparty_a-algebra.Po', ignoring this regex for all remaining files.
Thanks,
-mjg
--
Matthew Gregan |/
/|[EMAIL
into this over the next few days. Out
of curiousity, have you tried reproducing this on other versions of Windows?
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone
key ID (you shouldn't have mulitple differing
keys with the same ID anyway). You can set a per-project default key by
adding it to MT/options.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone-devel
option.
On the other hand, for security reasons, if you have 25 separate monotone
repositories that should be as independent as possible, you probably do want
separate logins for each project you want to serve--but it's not a
requirement monotone imposes.
Cheers,
-mjg
--
Matthew Gregan
database and containing directory are
writeable
Windows doesn't strip the single quote characters before they're passed into
main() via argv[], so monotone sees the literal string: 'i:'.
You can reproduce this with bash like so:
$ monotone automate select \'i:\'
Cheers,
-mjg
--
Matthew Gregan
,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone-devel
At 2005-11-30T16:21:09-0200, Alex Queiroz wrote:
Hallo all,
Attached is a patch against the current head. It has the
beginnings of a translation to Brazilian Portuguese.
Thanks, applied in 04d89ccc7f3045f6c4782d66762d686cc5ccedbf.
Cheers,
-mjg
--
Matthew Gregan
--either by building Lua as a C++ library, or by builting it
with the equivalent of -fexceptions.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone-devel
forgetting right now).
For MinGW, see here: http://venge.net/monotone/wiki/BuildingOnWindows
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone
is a
fresh pull with 0.25. The third is a fresh pull with 0.26pre1.
131150848 2005-12-05 22:35 monotone.db.old
89192448 2006-01-10 14:59 monotone.db
87205888 2006-01-14 00:16 monotone.db.rosters
Cheers,
-mjg
--
Matthew Gregan
At 2006-01-19T10:51:36+1300, wrote:
Oops.
Was the semistatic binary built on a modern machine, and you're not trying
^^^now
--
Matthew Gregan |/
/|[EMAIL
of filenames that
monotone considers valid.
Any chance you can file a bug for this in our bug tracker?
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
libraries would get
upset; it would work fine in a subdirectory, eg X:\work
That is bug #16442, and was fixed in e980c5630f40fbeda04ddd11a799837936658fc1.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
'--debug' and '--dump=...' specified?
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman
/~kinetik/files/mtn.exe.bz2 if you'd like to give it a
shot.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http
somewhere, I'm running out of ideas for now...
I assume this is only occuring on this particular machine, and you don't
have a recipe to reproduce this elsewhere? It'd obviously be a lot easier
to track down if I could reproduce it locally. :-)
Cheers,
-mjg
--
Matthew Gregan
in an option to add monotone to the path as part of this--it should be
included in 0.27.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
works, and won't harm other platforms (we already have a
specialization for unsigned int anyway), so I've committed a similar fix.
Thanks for chasing this up!
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
determine which file is the problem from the hash key? mtn list
certs key reports no match.
% mtn automate get_file HASH
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
seem to
have left some hardcoded paths in the ISS script.
I've asked Nathaniel to upload an updated Win32 installer. The new
installer will be named monotone-0.27.1-setup.exe, so you can be sure which
copy you've got.
Cheers,
-mjg
--
Matthew Gregan
on the repository
directly, but I won't describe how unless you really want to know. :-)
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel
]: *** [all-recursive] Error 1
make: *** [all] Error 2
The problem is in hash_map.hh: apparently _T is defined in
/usr/include/ctype.h.
Thanks for the bug report. I've committed a fix to hash_map.hh in
917af11f03689b3dcc4338b27b5c0b3b77170205 on mainline.
Cheers,
-mjg
--
Matthew Gregan
OPT_FOO in our existing
code), I tried to use the program_options::option objects themselves. In
retrospect I'm not sure it was a good idea, because it made handling
ownership a bit messy.
Cheers,
-mjg
--
Matthew Gregan
something from upstream, and the layout rules that were in use were not
consistent between blocks.
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
to stay within eighty columns, but looking at
it now, it obviously was committed that way... my Emacs window must have
been wider than I realised.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
to the wiki.
[0]
http://colabti.de/irclogger/irclogger_log/monotone?date=2006-10-19,Thusel=97#l206
[1] http://venge.net/monotone/wiki/MtnSummit
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
.
key [EMAIL PROTECTED]
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo
or Graydon to get access to
post the release files on venge.net.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
, it's a new test that is expected to fail but
hadn't been marked as xfailed initially.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
an existing XFAILed test
(db_kill_rev_and_recommit).
(and see also:
http://colabti.de/irclogger/irclogger_log/monotone?date=2007-02-26,Monsel=110#l215)
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
have some
time.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/monotone
wouldn't worry about it for nvm.ssh-agent for now, as long as MinGW is okay.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel
need
to be plucked onto the release candidate line of development.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel
with conflicts and you must resolve them as they
are presented). Work on in-workspace conflict presentation and
resolution is planned for the future.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
be failing with an invariant here, we
should handle this case and report it to the user as a misuse for now. Once
we have the non-content conflict resolution, we should be able to deal with
this type of merge nicely.
Cheers,
-mjg
--
Matthew Gregan
to be a mentor? This basically means helping read
student apps and potentially being a main point of contact for a
student. (It also means you can't apply as a student.)
Count me in, please.
Cheers,
-mjg
--
Matthew Gregan
prefer to have clone require a --db argument be supplied.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http
convenient
behaviour myself.
It seems that mtn add --unknown is now broken under windows.
Can you be more specific? It seems to work for me and the testsuite.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
difference with monotone.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo
intermittant
failure in the netxx_pipe unit tests.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org
stable release and go from there.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http://lists.nongnu.org/mailman
evidence that it was even used, so it was completely removed in 0.33.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
up correctly. The
instructions were updated a couple of months ago when I got localization
working properly: http://www.venge.net/mtn-wiki/BuildingOnWindows
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
installed)? If
not, then this patch can be committed.
Both MinGW and MSVC need libintl-8 for i18n.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing
for this, but I'm not
immediately sure what form they would take.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
http
us know. In the meantime, feel free to
download it from:
http://flim.org/~kinetik/monotone/monotone-0.36-setup.exe
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
be able to run that buildbot for quite a
while. That goes for the Debian and OS X ones I ran as well, but I think we
have decent coverage there already.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
slightly more hassle for the Visual Studio builds, but not a big deal.
Worst case (and probably not necessary), we can keep generated copies of the
files in the tree under visualc/ like we do with a few other files.
Cheers,
-mjg
--
Matthew Gregan
At 2007-12-02T02:44:04+1300, Matthew Gregan wrote:
I'm seeing this too, except the exit code is 5 in my case. What's
happening is that we're crashing while in the dtor code of
std::vectorBotan::RSA_PublicKey when destroying ssh_agent as the program
returns from cpp_main.
The problem turned
socketpair.c into our tree, and replace the existing pipe code with a cross
platform implementation that uses socketpair().
I started working on doing exactly this a while back but got busy with other
stuff. I'll try to find some time to work on it again soon.
Cheers,
-mjg
--
Matthew Gregan
to it, but it will just cause socketpair() to return an
error because its own attempt to connect to the listening socket will fail.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED
. The unix/
implementation can just call the real socketpair() directly like the !WIN32
case in socketpair.c does.
I assume this should be done on a branch;
n.v.m.experimental.win32_pipes ?
It can be--branches are free, after all.
Cheers,
-mjg
--
Matthew Gregan
At 2007-12-04T15:18:18-0800, Zack Weinberg wrote:
On Dec 4, 2007 12:11 PM, Matthew Gregan [EMAIL PROTECTED] wrote:
Right, that code is an implementation of socketpair() for Windows because it
is not provided by the Windows socket API. Cross platform refers to the
rest of the code--all
in overlapped mode even
though the caller isn't expecting them to. Chaos ensues.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone-devel mailing list
Monotone-devel
at it as well.
So I think we are back to the fundamental problem; how do we do
non-blocking IO on Win32 stdio?
I have a working standalone proof of concept for this approach already, so I
know we can make it work with monotone.
Cheers,
-mjg
--
Matthew Gregan
debug code. Maybe I could start with yours.
Attached. Try not to be too shocked, it's a fearsome trainwreck of
write-only test code.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
/* test ncm's socketpair for win32
the Win32 installer, and I was in the process of moving job/city
around the time of the release. 0.38 is up now.
Cheers,
-mjg
--
Matthew Gregan |/
/|[EMAIL PROTECTED]
___
Monotone
-fexceptions
CXXFLAGS = -g -O2 -Wall -W -Wno-unused -fno-strict-aliasing
...which are pretty much what you're using, so the performance regression
might be real and needs further investigation.
Cheers,
-mjg
--
Matthew Gregan
::ServerBase::bind_to().
It never made it to the main branch. It's still on
net.venge.monotone.experimental.win32_pipes, but I don't think the approach
ever ended up working properly on Win32. Stephen Leake did all of the work,
so he'd know more.
Cheers,
-mjg
--
Matthew Gregan
87 matches
Mail list logo