On Tue, Jul 11, 2006 at 03:38:38AM -0700, Nathaniel Smith wrote:
Now that we have diff -p support on mainline, is there any reason we
shouldn't make it the default?
[...]
AFAIK it's still compatible with patch(1) and the various other tools
out there.
FTR, I just checked patch(1), patchutils
On Thu, Jul 13, 2006 at 05:24:16PM +1000, Daniel Carosone wrote:
On Thu, Jul 13, 2006 at 12:16:15AM -0700, Nathaniel Smith wrote:
OTOH, while the manual is kept up to date at all times, I think the
last time the FAQ was _updated_ also predates the wiki by a
considerable amount. It's quite
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 00:32:13 -0700, Nathaniel
Smith [EMAIL PROTECTED] said:
njs Oh, good point, I understand the whole ssh into venge.net and run
njs emacs on the html files in the webroot thing is a little low tech by
njs the standards of kids today.
njs
njs Any
On Thu, Jul 13, 2006 at 09:47:57AM +0200, Richard Levitte - VMS Whacker wrote:
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 00:32:13 -0700, Nathaniel
Smith [EMAIL PROTECTED] said:
njs Oh, good point, I understand the whole ssh into venge.net and run
njs emacs on the html files in the
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Nathaniel Smith schreef:
On Thu, Jul 13, 2006 at 09:47:57AM +0200, Richard Levitte - VMS Whacker wrote:
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 00:32:13 -0700, Nathaniel
Smith [EMAIL PROTECTED] said:
njs Oh, good point, I understand the
On Wed, Jul 12, 2006 at 01:09:19AM -0500, Timothy Brownawell wrote:
Everything seems to work now, and all the buildbots except win32 (which
has a longstanding unit_tests failure, and what looks like a
canonicalization issue in one of the new tests) are green.
Indeed. Great work!
The new
Nathaniel Smith [EMAIL PROTECTED] writes:
On Wed, Jul 12, 2006 at 09:49:13PM +0100, Bruce Stephens wrote:
A restrictions version of pluck would do much what I imagine you'd
expect update to do. I suspect that would be easily accepted and
understood by users.
Committed to mainline last
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 01:03:04 -0700, Nathaniel
Smith [EMAIL PROTECTED] said:
njs On Thu, Jul 13, 2006 at 09:47:57AM +0200, Richard Levitte - VMS Whacker
wrote:
njs Do you want it done by crontab or through note_netsync_end()?
njs
njs Crontab, I'd think, or even a
On Thu, Jul 13, 2006 at 10:07:35AM +0200, Koen Kooi wrote:
Do you want it done by crontab or through note_netsync_end()?
Crontab, I'd think, or even a 'while sleep 600; ...' script run under
daemontools supervision :-). (This is how the CIA script works now,
to avoid crontab
Hi there,
I think about to tweak GUITONE in that way, that it can display the
branches on a server. It may be very handy to know what branches are out
there and then pull it in your local db the graphically way...
Is there something like 'mtn list branches venge.net' ?
Some news on the
Hi Ingo!
I think about to tweak GUITONE in that way, that it can display the
branches on a server. It may be very handy to know what branches are out
there and then pull it in your local db the graphically way...
Is there something like 'mtn list branches venge.net' ?
$ mtn ls branches
On Thursday 13 July 2006 13:32, Thomas Keller wrote:
$ mtn ls branches
which works on the current workspace' db or via --db on any database.
I'm unsure ATM if you can query any data from a database which is *not*
locally available...
And yeah, there is no automation interface yet for almost
Hi Thomas,
On Thursday 13 July 2006 13:32, Thomas Keller wrote:
$ mtn ls branches
which works on the current workspace' db or via --db on any database.
I'm unsure ATM if you can query any data from a database which is *not*
locally available...
And yeah, there is no automation interface yet
usecase:
2 branches, 'one' and 'two', 'one' propagates to 'two' on a regular
basis.
in a distant past file 'onlyhere.txt' was dropped from 'two' thus
leaving it only in 'one'
making a change to 'onlyhere.txt' in 'one' doesnt propagate that
change to 'two'
This is probably correct
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 14:51:01 +0200, Marcel van
der Boom [EMAIL PROTECTED] said:
marcel 2 branches, 'one' and 'two', 'one' propagates to 'two' on a regular
marcel basis.
marcel in a distant past file 'onlyhere.txt' was dropped from 'two' thus
marcel leaving it
There have been many requests for peeking of all sorts, like to know
what you would get from a server if you pulled, or what you would send
during a push, or what files would be changed/dropped/renamed by an
update, or...
To cover all those cases, would it be a very bad idea to have a
--dryrun
On 13 jul 2006, at 15:11, Richard Levitte - VMS Whacker wrote:
marcel In practice this all happens in silence. It would be
comfortable to
marcel be able to break that silence somehow, either by letting
the merge
marcel output something informational, or having a way to list the
files
Richard Levitte - VMS Whacker wrote:
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 01:03:04 -0700, Nathaniel Smith
[EMAIL PROTECTED] said:
njs On Thu, Jul 13, 2006 at 09:47:57AM +0200, Richard Levitte - VMS Whacker
wrote:
njs Do you want it done by crontab or through note_netsync_end()?
Richard Levitte - VMS Whacker wrote:
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 14:51:01 +0200, Marcel van der
Boom [EMAIL PROTECTED] said:
[...]
marcel In practice this all happens in silence. It would be comfortable to
marcel be able to break that silence somehow, either by letting
Whatever happened to the optional ANCESTOR argument to explicit_merge?
It no longer appears to be supported by monotone, although the manual
still claims its existence.
Craig
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
On Thu, 2006-07-13 at 11:36 -0400, Craig Lennox wrote:
Whatever happened to the optional ANCESTOR argument to explicit_merge?
Our merge algorithms improved, so it isn't really needed any more. We
don't need an ancestor at all unless there's a content conflict, and
then we now pick the best
I'd like to throw out another possibility for testsuite improvement:
Kill boost::unit_test; move everything into the Lua testsuite.
This may sound completely crazy, but think about it. Wouldn't it be
nice if we didn't have to recompile the entire codebase in order to
run the test suite? The
On Thu, Jul 13, 2006 at 08:55:35AM -0700, Zack Weinberg wrote:
I'd like to throw out another possibility for testsuite improvement:
Kill boost::unit_test; move everything into the Lua testsuite.
This may sound completely crazy, but think about it. Wouldn't it be
nice if we didn't have to
Hallo,
On 7/13/06, Zack Weinberg [EMAIL PROTECTED] wrote:
I'd like to throw out another possibility for testsuite improvement:
Kill boost::unit_test; move everything into the Lua testsuite.
What I've done here at work is to write all the unit test
functions as Lua CFunctions, export then
On Thu, Jul 13, 2006 at 12:15:07PM +0200, [EMAIL PROTECTED] wrote:
On Thu, Jul 13, 2006 at 07:16:38PM +1000, Daniel Carosone wrote:
The best system would be one that touched a flag file or similar (ie,
I usually use a fifo for that kind of things - that way, it's echo fifo
on the sender and
On Thu, Jul 13, 2006 at 04:38:50PM +0200, Marcel van der Boom wrote:
Yeah, this is _sometimes_ a solution. Sometimes, because it's not
always possible to construct such output apparently. In our case for
example the msg appears:
mtn: warning: restriction excludes directory 'blah/blah/1'
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Daniel Carosone schreef:
On Thu, Jul 13, 2006 at 10:07:35AM +0200, Koen Kooi wrote:
Do you want it done by crontab or through note_netsync_end()?
Crontab, I'd think, or even a 'while sleep 600; ...' script run under
daemontools supervision :-).
On Thu, Jul 13, 2006 at 07:26:52PM +0200, Koen Kooi wrote:
Daniel Carosone schreef:
The note_* hooks Richard was talking about are exactly these kinds of
triggers.
An important point about these triggers is that you don't want to do
much work directly inside of them, and that of course
On 13 jul 2006, at 19:20, Nathaniel Smith wrote:
On Thu, Jul 13, 2006 at 04:38:50PM +0200, Marcel van der Boom wrote:
mtn: warning: restriction excludes directory 'blah/blah/1'
mtn: warning: restriction excludes directory 'blah/blah/2'
mtn: misuse: invalid restriction excludes required
Hallo,
On 7/13/06, Koen Kooi [EMAIL PROTECTED] wrote:
but I might be biased by my general dislike of python[1].
And I thought I was the only one.
--
-alex
http://www.ventonegro.org/
___
Monotone-devel mailing list
Monotone-devel@nongnu.org
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 10:18:02 -0500, Matthew A.
Nicholson [EMAIL PROTECTED] said:
matt or mtn diff -r h:one -r h:two
Oh, yeah, right, had forgotten about that one...
-
Please consider sponsoring my work on free software.
See http://www.free.lp.se/sponsoring.html
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 10:14:42 -0500, Matthew A.
Nicholson [EMAIL PROTECTED] said:
matt The debian package from the n.v.m.debian branch also contains
matt something like this.
Hmm? Something that mirrors a collection and updates content
somewhere? Where? For the
Nathaniel Smith [EMAIL PROTECTED] writes:
Other VCSes have historically overloaded the 'update' command, so it
performed a bunch of logically unrelated functions -- moving your
work in progress, doing merges, reverting files, you name it.
Does mtn update merge changes into files modified
On Thu, 2006-07-13 at 11:03 -0700, Steven E. Harris wrote:
Nathaniel Smith [EMAIL PROTECTED] writes:
Other VCSes have historically overloaded the 'update' command, so it
performed a bunch of logically unrelated functions -- moving your
work in progress, doing merges, reverting files, you
Timothy Brownawell [EMAIL PROTECTED] writes:
Yes.
Good; that's what I thought. I was confused by Nathaniel Smith's
comparison of monotone's simple 'update' with the more complex command
in other tools.
--
Steven E. Harris
___
Monotone-devel
On Thu, Jul 13, 2006 at 07:36:01PM +0200, Marcel van der Boom wrote:
On 13 jul 2006, at 19:20, Nathaniel Smith wrote:
On Thu, Jul 13, 2006 at 04:38:50PM +0200, Marcel van der Boom wrote:
mtn: warning: restriction excludes directory 'blah/blah/1'
mtn: warning: restriction excludes directory
On Thu, Jul 13, 2006 at 12:19:16PM -0700, Steven E. Harris wrote:
Timothy Brownawell [EMAIL PROTECTED] writes:
Yes.
Good; that's what I thought. I was confused by Nathaniel Smith's
comparison of monotone's simple 'update' with the more complex command
in other tools.
Ah, I just meant
Alex Queiroz [EMAIL PROTECTED] writes:
[...]
What I've done here at work is to write all the unit test functions
as Lua CFunctions, export then to Lua and run the unit tests from a
Lua script that also generates a report. Very handy.
Sounds to me like the costs would outweigh the benefits
Richard Levitte - VMS Whacker [EMAIL PROTECTED] writes:
There have been many requests for peeking of all sorts, like to know
what you would get from a server if you pulled, or what you would send
during a push, or what files would be changed/dropped/renamed by an
update, or...
To cover all
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 20:42:07 +0100, Bruce
Stephens [EMAIL PROTECTED] said:
monotone I believe that was the intention, yes. Hmm, I thought the idea was
monotone described in the wiki, but I don't see it. Anyway, my impression was
monotone that Nathaniel et alles
Richard Levitte - VMS Whacker [EMAIL PROTECTED] writes:
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 20:42:07 +0100, Bruce
Stephens [EMAIL PROTECTED] said:
monotone I believe that was the intention, yes. Hmm, I thought the idea was
monotone described in the wiki, but I don't see it.
Hallo,
On 7/13/06, Bruce Stephens [EMAIL PROTECTED] wrote:
Easier, but still not trivial. It's surely still four or five lines
per function?
This is getting off-topic, but still... What I did was to create
some macros for the developers to allow them to write the testsuite
without any
Currently the knowledge of which characters are not allowed in a
pathname is split between paths.cc and constants.cc.
paths.cc:has_bad_chars is the sole user of
constants.cc:illegal_path_bytes, but adds more to the set (notably
backslash). I note also that this code is all marked as must be
On Thu, 2006-07-13 at 01:43 -0700, Nathaniel Smith wrote:
On Wed, Jul 12, 2006 at 01:09:19AM -0500, Timothy Brownawell wrote:
The new testsuite seems to be done now, but some things aren't optimal
yet. In particular I'd like to make tester fairly independent from the
rest of monotone by
Okay I just went though a huge nightmare of an update...first update, some path to create already existed, and monotone complained that could not create it.I moved that path away from beign in the way, and tried to update again.
directories and files were left in _MTN/tmp which I had to get rid
Richard Levitte - VMS Whacker wrote:
In message [EMAIL PROTECTED] on Thu, 13 Jul 2006 10:14:42 -0500, Matthew A.
Nicholson [EMAIL PROTECTED] said:
matt The debian package from the n.v.m.debian branch also contains
matt something like this.
Hmm? Something that mirrors a collection and updates
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I've modified the scripts and monotonerc at
http://www.venge.net/monotone/wiki/UnixAttribsAndSymlinks to use the
lua hooks for attr_init_functions on 'mtn add'. This is much, much
faster than calling 'mtn attr get' and 'mtn attr set' from an outside
47 matches
Mail list logo
