On 10 Mar 2003, Robert Collins wrote:
On Wed, 2003-03-05 at 11:41, Igor Pechtchanski wrote:
On 5 Mar 2003, Robert Collins wrote:
On Wed, 2003-03-05 at 08:19, Igor Pechtchanski wrote:
On 5 Mar 2003, Robert Collins wrote:
Using the packages as dependencies we can build the same
On Mon, 2003-03-10 at 15:33, Robert Collins wrote:
* Going back, changing the local cache dir, and going forward again
does not
correctly reparse inis in the new dir.
Reproduction: Choose an incorrect cache dir, go back, correct it.
Chooser shows all packages in Misc.
Fixed,
Oh poo, I used the wrong subject didn't I? ;-)
On a serious note, I am willing to take over maintainership of this cygwin
package.
Regards,
Elfyn McBratney
[EMAIL PROTECTED]
www.exposure.org.uk
* Toggling from All Default to All Install takes *way* too long.
This takes under 1/10th of a second on my PC.
I need a testcase, or more info.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
Description: This is a digitally signed message part
On Mon, 2003-03-10 at 16:18, Igor Pechtchanski wrote:
Sure they can. All it takes is two maintainers marking the other script
as a requirement.
Oh, of course you can *have* circular dependences, but you can't *honor*
them. Script dependences mean that some script has to be run before
Ok, a few pragmatic thoughts:
* When do we parse it?
It's not trivial:
- if we parse on startup, and the user changes the cygwin / path during
runtime, should we revert to defaults?
- if we wait for the / path to be chosen, we can't store any pre- /
dialog settings.
- if we parse on startup, and
On Mon, 2003-03-10 at 15:48, Igor Pechtchanski wrote:
Rob,
There was a follow-up message: http://cygwin.com/ml/cygwin-apps/2002-10/msg00152.html
That eliminates #1. I don't know how to fix #2 -- could have something to
do with cvs checkout timestamps. I don't believe it was addressed,
I've observed that dynamic registration always seems to give folks
pause, so I've added a couple of classes to setup to begin the user
persistence codefest...
UserSettings: Registry of UserSetting.
UserSetting: A user setting. should be the entry point for the logic of
the parser and
Ok, I've fixed
* compress_bz::error called appears in the log every time a package is
installed, even if everything works.
So, setup-3.324 (now 325 due to the UserSetting addition) (snapshot
uploading :}) should be considered a release candidate. Pending
confirmation of the correctness of the
On 10 Mar 2003, Robert Collins wrote:
On Wed, 2003-03-05 at 11:52, Igor Pechtchanski wrote:
On 5 Mar 2003, Robert Collins wrote:
Other than that, please expand to address all scripts.
Rob
Rob,
There is a design issue here that I'd like to address before I work more
on this. I
On Mon, 2003-03-10 at 16:03, Igor Pechtchanski wrote:
Sounds fine. I got the first part mostly done (it'll be a run script,
output to file idiom). The second might take some time... I'll probably
have to use mktemp() or something similar.
Just use the postinstall.log you where planning to
On 2401, there was a commit to setup/zlib, log message Eliminate \r
in line endings.
I hypothesize that the repository files might have been d2u-ed at some
point, because there are *no* \r's in any revision.
This change also stripped whitespace at the end of lines across many
files. I'd like
1. grace
date : 25 Nov 2002
version: 5.1.12-1
status : updated package available for review
notes : http://www.cygwin.com/ml/cygwin-apps/2002-11/msg00322.html
reviews: http://www.cygwin.com/ml/cygwin-apps/2003-03/msg00254.html
votes : 2 (Lapo and Robert)
url:
On Mon, 2003-03-10 at 21:30, Max Bowsher wrote:
On 2401, there was a commit to setup/zlib, log message Eliminate \r
in line endings.
I hypothesize that the repository files might have been d2u-ed at some
point, because there are *no* \r's in any revision.
This change also stripped
On Mon, 10 Mar 2003, Robert Collins wrote:
On Mon, 2003-03-10 at 21:30, Max Bowsher wrote:
On 2401, there was a commit to setup/zlib, log message Eliminate \r
in line endings.
I hypothesize that the repository files might have been d2u-ed at some
point, because there are *no* \r's
On Mon, 2003-03-10 at 22:04, Max Bowsher wrote:
NB: I have verified this will NOT introduce \r characters into the files.
MM, it would be easier for you to use patch -l :].
1) patch won't be involved in the process at all
Huh? Then I've misunderstood.
2) I think gratuitous
On 10 Mar 2003, Robert Collins wrote:
Ok, a few pragmatic thoughts:
* When do we parse it?
It's not trivial:
- if we parse on startup, and the user changes the cygwin / path during runtime,
should we revert to defaults?
- if we wait for the / path to be chosen, we can't store any pre-/
Robert Collins wrote:
* Toggling from All Default to All Install takes *way* too long.
This takes under 1/10th of a second on my PC.
I need a testcase, or more info.
I can no longer reproduce. Suggest deleting from TODO, on the assumption it
has been fixed by coincidental changes to setup.
Igor Pechtchanski wrote:
On 10 Mar 2003, Robert Collins wrote:
Ok, a few pragmatic thoughts:
* When do we parse it?
It's not trivial:
- if we parse on startup, and the user changes the cygwin / path
during runtime, should we revert to defaults?
- if we wait for the / path to be chosen, we
Just botched the log message on a checkin.
Please could a member of cvsadmin do the following in setup:
cvs admin -m 2.89:Regenerate after adding foreign to AUTOMAKE_OPTIONS
Makefile.in
cvs admin -m 2.29:Add wishlist item: hold version README
Thanks,
Max.
Joshua Daniel Franklin wrote:
With the talk of a /etc/setuprc file and the email below on cygwin@,
I thought I'd suggest a 'hold' option for the setup.exe wishlist.
There are certain packages (like postgresql in this case) that some
users *never* want to upgrade as long as the old version
On Mon, 10 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
On Sun, 9 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
There have been bits of persistent state that people wanted to store
for setup. I'm proposing an Xdefaults-style setup.cfg in
/etc/setup. If we compile a
Robert Collins wrote:
On Mon, 2003-03-10 at 22:04, Max Bowsher wrote:
NB: I have verified this will NOT introduce \r characters into the
files.
MM, it would be easier for you to use patch -l :].
1) patch won't be involved in the process at all
Huh? Then I've misunderstood.
cvs import
cvs
Robert Collins wrote:
On Mon, 2003-03-10 at 15:48, Igor Pechtchanski wrote:
Rob,
There was a follow-up message:
http://cygwin.com/ml/cygwin-apps/2002-10/msg00152.html That
eliminates #1. I don't know how to fix #2 -- could have something
to
do with cvs checkout timestamps. I don't
Igor Pechtchanski wrote:
On Mon, 10 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
On Sun, 9 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
There have been bits of persistent state that people wanted to
store for setup. I'm proposing an Xdefaults-style setup.cfg in
On Mon, 10 Mar 2003, Brian Keener wrote:
Igor Pechtchanski wrote:
Still:
* Not tested on Win95. It's a fairly low level change, and I have 'this
much' (holds thumb and forefinger very close together) confidence that
we won't get surprised.
Uh, can someone test this on Win95? I
On Mon, 10 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
On Mon, 10 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
On Sun, 9 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
There have been bits of persistent state that people wanted to
store for setup. I'm
Igor Pechtchanski wrote:
On Mon, 10 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
On Mon, 10 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
On Sun, 9 Mar 2003, Max Bowsher wrote:
Igor Pechtchanski wrote:
There have been bits of persistent state that people wanted to
On Mon, Mar 10, 2003 at 07:24:17AM -, Elfyn McBratney wrote:
Oh poo, I used the wrong subject didn't I? ;-)
Mail to me, to some other mailing list, or use of a different subject
will immediately disqualify you from this special offer.
Maybe you thought I was kidding. I wasn't, really.
This
On Tue, 2003-03-11 at 03:36, Max Bowsher wrote:
Robert Collins wrote:
On Mon, 2003-03-10 at 22:04, Max Bowsher wrote:
NB: I have verified this will NOT introduce \r characters into the
files.
MM, it would be easier for you to use patch -l :].
1) patch won't be involved in the
On Tue, 2003-03-11 at 02:28, Max Bowsher wrote:
Robert Collins wrote:
* Toggling from All Default to All Install takes *way* too long.
This takes under 1/10th of a second on my PC.
I need a testcase, or more info.
I can no longer reproduce. Suggest deleting from TODO, on the
On Tue, 2003-03-11 at 03:37, Max Bowsher wrote:
Ok, well I'm going to practice delegation here: Max - when you look at
zlib, can you ensure that the timestamps on zlib's autotool files are
correctly ordered?
Delegation accepted, but I think it's potentially more complicated than
that.
On Tue, 2003-03-11 at 03:54, Max Bowsher wrote:
Either we disallow comments, or we keep the order, IMO.
^^^
My vote.
Ok, we are starting down a rabbit hole.
Whomever puts the time in can decide whether it's user editable or not.
I've already listed the
On Tue, 2003-03-11 at 07:39, Robert Collins wrote:
I'll add another incentive: I'll accept 'progress' patches, that
refactor the code to use the emerging framework, as long as the current
*behaviour* doesn't change. (i.e. we still use last-mirror etc).
The first patch that starts to use
On Tue, Mar 11, 2003 at 07:24:17AM +1100, Robert Collins wrote:
On Tue, 2003-03-11 at 02:28, Max Bowsher wrote:
Robert Collins wrote:
* Toggling from All Default to All Install takes *way* too long.
This takes under 1/10th of a second on my PC.
I need a testcase, or more info.
I can
On Tue, 2003-03-11 at 07:50, Christopher Faylor wrote:
Yep, please move to done. (I'm assuming that until the cvsadmin commands
you requests, I shouldn't commit stuff?)
Ok. I've made this crucial change.
For the record, I don't want to get involved whenever people screw up
their commit
Christopher Faylor wrote:
On Tue, Mar 11, 2003 at 07:24:17AM +1100, Robert Collins wrote:
On Tue, 2003-03-11 at 02:28, Max Bowsher wrote:
Robert Collins wrote:
* Toggling from All Default to All Install takes *way* too long.
This takes under 1/10th of a second on my PC.
I need a testcase,
Robert Collins wrote:
On Tue, 2003-03-11 at 02:28, Max Bowsher wrote:
Robert Collins wrote:
* Toggling from All Default to All Install takes *way* too long.
This takes under 1/10th of a second on my PC.
I need a testcase, or more info.
I can no longer reproduce. Suggest deleting from
On Mon, Mar 10, 2003 at 09:31:49PM -, Max Bowsher wrote:
Thanks Chris, and yes, I'm sufficiently embarrased to remember very
clearly not to do that again!
There's no need to be embarrassed, IMO. I do this all the time. I think the
ChangeLog is more important that the checkin messages,
Igor Pechtchanski wrote:
Thanks a lot for testing! Most postinstall scripts run silently, so what
you're seeing is the output of printf's. If you could try it with
scripts that actually output something (e.g., iu-config.sh, gcc-mingw.sh,
post-texmf.sh, profile.sh), you'd see if the output is
On Mon, 10 Mar 2003, Brian Keener wrote:
Igor Pechtchanski wrote:
Thanks a lot for testing! Most postinstall scripts run silently, so what
you're seeing is the output of printf's. If you could try it with
scripts that actually output something (e.g., iu-config.sh, gcc-mingw.sh,
Replies inline below. Incidentally, why did your message arrive twice
from two different From: addresses? And which should the
replies go to?
---
One address is subscribed to the list, the other is not.
Normally, if one responds to 'all', it goes to the OP and the list.
If one just
42 matches
Mail list logo
