On Fri, 2003-03-28 at 17:44, Gary R. Van Sickle wrote:
Bigger and better than ever baby YEAH!
Changes from the last patch:
- Finish problem fixed.
- Back button on splash page disabled again.
- Bonus: two new icon resolutions!
Only known problem is the bottom separator line still goes
It would be nice, when something is unexpectedly installed, for users to
be able to do something like:
what-needed xfree86-base
and get back
libProplist.
If setup that could be a gui, of course.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
On Sat, 2003-03-29 at 17:24, Gary R. Van Sickle wrote:
it's too big for a single commit. I'd like you to start sending in the
bits that you can, in a
single-patch-does-a-single-conceptual-change style.
Those I can review and approve for Max or I to checkin.
Are you saying you
On Sat, 2003-03-29 at 14:16, amores nolikeyjunk wrote:
I'll bump this once then let it drop assuming no one is
interested.
(This is in regards to the two test installs I did, and
my considering trying other tests.)
Positive feedback is always good it gives a feeling for the failure rate
-
On Thu, 2003-03-27 at 22:10, Vince Hoffman wrote:
Ok, well this is 'not good' :}.
At what point does it crash?
At start before any window appears.
Shuttting down cygwin services made no difference.
let me know if there is any more i can do to help the problem.
Hmm, I don't think
Regarding XFree auto-installation...
__
Cygwin Package Information
...
libPropList 0.10.1-3
This requires XFree86-base. So setup is doing the right thing.
Rob
--
GPG key available at:
On Sat, 2003-03-29 at 15:13, amores nolikeyjunk wrote:
but when I reboot run setup again, it doesn't remember
the pkg directory. I guess I don't know if it ever did,
as this is the first time I'm not running the setup.exe
out of the pkg directory.
The first time through, there is a known
On Sat, 2003-03-29 at 15:39, amores nolikeyjunk wrote:
Now something is wrong.
The symptoms you describe aren't (AFAIK) something that a setup bug
could cause.
I'd suggest a VMWare or windows install problem is your most likely
culprit.
Cheers,
Rob
--
GPG key available at:
On Sat, 2003-03-29 at 15:41, amores nolikeyjunk wrote:
Robert Collins,
If you happen to be quite interested in this apparent problem
I've just found (altho I certainly don't know if it is
setup related or not) at this moment, and wish to discuss this via
Instant Messenger (eg, to give me
On Sat, 2003-03-29 at 15:50, amores nolikeyjunk wrote:
VM as in running Windows on top of say Mac OS/Linux?
Actually on top of Win2000. (Using vmware. It is the
only way I have time to test stuff on virgin machines.)
I really appreciate the testing effort you are putting in - it's great.
VMWare comes with tools to allow the client OS (the one inside the VM)
to interface with the host OS - i.e. mouse release so that the mouse can
go from inside the VM to outside easily. These tools are called.
vmtools.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
On Thu, 2003-03-27 at 23:44, Max Bowsher wrote:
Robert Collins wrote:
Max, I suggest making clean, then makeing with 'CXXFLAGS=-g -O0.
Then, the call stack may have more detail.
Anyway, as I can't reprodiuce this, I'm limited to throwing peanuts :}.
Unfortunately, I can't persuade
On Thu, 2003-03-27 at 18:45, Thomas Pfaff wrote:
Christopher Faylor wrote:
I assume that these patches are to clean up some warnings. Do they
make sense?
Yes.
I have recognized these warnings yesterday too and will apply the patch.
Thank you.
Rob
--
GPG key available at:
On Thu, 2003-03-27 at 22:11, Thomas Pfaff wrote:
Do not try to force me to comment every bit of this patch ;-)
2003-03-27 Thomas Pfaff [EMAIL PROTECTED]
* thread.h: Change class names, methods, members and local vars
according to the GNU coding style.
* thread.cc:
On Thu, 2003-03-27 at 23:52, Thomas Pfaff wrote:
2003-03-27 Thomas Pfaff [EMAIL PROTECTED]
* thread.cc: Change 1==foo equations to foo==1 throughout.
Thanks again - please apply.
I don't know if you meant to do this:
- return (pthread_equal ((*mutex)-owner, self)) 1 ==
On Thu, 2003-03-27 at 20:33, [EMAIL PROTECTED] wrote:
On the rockbox mailinglist i used a VALID email address
for_spam seemed invalid at first glance - it *appears* to be a black
hole address.
- and on my win32
sdk page is a VALID
email adress, too.
I didn't see that. I stumbled
On Thu, 2003-03-27 at 16:09, Christopher Faylor wrote:
What do you think? Do we still need the words at
http://cygwin.com/download.html to exist as a separate entity? If so,
I'll probably add another documentation link to the main web page for
people who are first downloading cygwin. I
On Fri, 2003-03-28 at 05:44, Christopher Faylor wrote:
On Thu, Mar 27, 2003 at 10:54:36AM +0100, Ronald Landheer-Cieslak wrote:
I can see it now:
I downloaded the abcdef RPM from my local LUG mirror and it didn't work -
why?
.. umm.. Linux executable?
I'm pretty sure that RPM won't let
On Fri, 2003-03-28 at 02:37, [EMAIL PROTECTED] wrote:
nobody is perfect ;)
now lets be friends again and make the world a better place with OpenSource
Software - ok ?
:)
Yep.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
Description: This is a
Max Bowsher wrote:
Here is the Dr. MinGW output for the at-startup crash being mentioned on
[EMAIL PROTECTED]
setup-2.340.2.3.debug.exe caused an Access Violation at location 78403330
Reading from location 78403330.
Registers:
eax=0077 ebx=0022f930 ecx=77f51b2b edx=004961bc esi=0022fe70
John Morrison wrote:
snip
FYI, once /etc/passwd and /etc/group exist, they will not be automatically
recreated by setup.
/snip
Igor et al wrt post install scripts... :)
Would it be unreasonable to have some scripts which
are _never_ renamed to .done? Doing that to the
[EMAIL PROTECTED]
Brian Keener [EMAIL PROTECTED]
Christopher Faylor [EMAIL PROTECTED]
Gary R. Van Sickle [EMAIL PROTECTED]
John Marshall [EMAIL PROTECTED]
Marcel Telka [EMAIL PROTECTED]
Max Bowsher [EMAIL PROTECTED]
Igor Pechtchanski [EMAIL PROTECTED]
Pavel Tsekov [EMAIL PROTECTED]
Robert
[EMAIL PROTECTED]
Brian Keener [EMAIL PROTECTED]
Christopher Faylor [EMAIL PROTECTED]
Gary R. Van Sickle [EMAIL PROTECTED]
John Marshall [EMAIL PROTECTED]
Marcel Telka [EMAIL PROTECTED]
Max Bowsher [EMAIL PROTECTED]
Igor Pechtchanski [EMAIL PROTECTED]
Pavel Tsekov [EMAIL PROTECTED]
Robert
On Thu, 2003-03-27 at 00:17, Joe Buehler wrote:
Robert Collins wrote:
USER-VISIBLE CHANGES:
Did you forget resizable dialogs
No.
or is that not done yet?
Bingo!
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
Description
On Thu, 2003-03-27 at 07:51, Andrew DeFaria wrote:
Marcel Telka wrote:
(Of course, going hand in hand with this would be a command line driven
setup.exe but perhaps we shouldn't go there just yet...)
Uhh, grab the source. It's ~ 95% there.
Rob
--
GPG key available at:
On Thu, 2003-03-27 at 02:20, Igor Pechtchanski wrote:
On Wed, 26 Mar 2003, Robert Collins wrote:
Just a thought,
It's a good one. Such scripts belong in
/bin
or
/usr/bin
:]
Rob
You mean along with ssh-user-config Co? Any particular reason why this
is any better than
On Thu, 2003-03-27 at 02:24, Vince Hoffman wrote:
Havent a clean system to try a full install. however on running it to
upgrade/add packages the first run crashed .
on the second run all was fine and componants were added/updated perfectly.
I tried running a few more times to reproduce the
On Thu, 2003-03-27 at 03:15, Rolf Campbell wrote:
Robert Collins wrote:
A new release of setup.exe is imminent. There are many changes, and to
reduce problems, feedback from YOU, is needed.
The new version is available from
http://www.cygwin.com/setup-snapshots/setup-2.340.2.3.exe
+ for (i = packages.begin (); i != packages.end (); ++i)
+{
+ packagemeta pkg = **i;
+ for_each (pkg.installed.scripts().begin(),
pkg.installed.scripts().end(),
+ RunScript(pkg.name, pkg.installed.scripts().size()));
+ ++k;
+ Progress.SetBar2 (k, numpkg);
+
On Wed, 2003-03-26 at 02:11, Igor Pechtchanski wrote:
Umm... Ping?.. Do I need to regenerate this against HEAD once Max checks
in the postinstall thread patch?
Yes, and I'll review it then.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
On Tue, 2003-03-25 at 09:48, Igor Pechtchanski wrote:
On 21 Mar 2003, Robert Collins wrote:
Rob,
I wanted (and still want) to keep the two bar progress on this. The
second bar would show progress through packages, and the first - progress
through the scripts *in the current package*.
Ok
On Mon, 2003-03-24 at 22:40, Klaus Friis Østergaard wrote:
What about adding PHP as major mode in emacs in the autoinstall and download
of files?
You could always package up the php mode and have that depend on emacs.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
On Mon, 2003-03-24 at 23:15, Patrick Eisenacher wrote:
Hi Jeff,
I'm waiting already for more than a month for this to happen. Needless
to say, I'm pretty frustrated. But despite Robert claiming to take care
of this in his initial response you cited below, no update was ever
published.
On Tue, 2003-03-25 at 09:35, Hannu E K Nevalainen (garbage mail) wrote:
(1)
I had the impression that setup.exe saved paths within it's *ini-file or
some such; thus forcing one to have *exactly* the same path at cache-disk
creation AND use time.
setup doesn't. :].
Rob
--
GPG key
On Tue, 2003-03-25 at 09:32, Elfyn McBratney wrote:
Thank you. I stepped down because I realised I wasn't fulfilling my
obligations and simply didn't want to get whipped to devote time I don't
have, on someone elses schedule, not because I wanted to see the package
removed (I don't).
: Robert Collins [EMAIL PROTECTED]
To: roland [EMAIL PROTECTED]
Sent: Tuesday, March 25, 2003 1:06 AM
Subject: Re: Another GPL violation: Re: MinimalisticBuild-Environmentforwin32 (~7.5MB)
---End Message---
signature.asc
Description: This is a digitally signed message part
On Fri, 2003-03-21 at 07:25, Max Bowsher wrote:
Christopher Faylor wrote:
On Thu, Mar 20, 2003 at 08:12:21PM -, Max Bowsher wrote:
I was just working on extracting the correct changelog entries for the
release notes for the upcoming setup release, when I discovered an
accidental
On Sun, 2003-03-23 at 22:07, Max Bowsher wrote:
That's good because I commited it three days ago!
Indeed. LOL.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
Description: This is a digitally signed message part
On Sat, 2003-03-22 at 22:19, Max Bowsher wrote:
The current librsync does *not* implement the same protocol as the rsync
program. Therefore I don't believe it can be useful to setup.
Protocol, no. sliding checksums, yes, AFAIK.
I therefore propose to remove the rsync directory and
On Sun, 2003-03-23 at 08:24, Pierre A. Humblet wrote:
Also doesn't setup
preset the choice depending if you have local Administrators privilege?
No. Patches Solicited to detect appropriate access and do so.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
On Sun, 2003-03-23 at 11:48, Pierre A. Humblet wrote:
At 11:21 AM 3/23/2003 +1100, Robert Collins wrote:
On Sun, 2003-03-23 at 08:24, Pierre A. Humblet wrote:
Also doesn't setup
preset the choice depending if you have local Administrators privilege?
No. Patches Solicited to detect
On Sun, 2003-03-23 at 07:26, Joshua Daniel Franklin wrote:
I have been messing with setup.exe's chooser documenting it this
morning and it hit me that a lot space is being used up with the
Bin and Src checkboxes. As far as I can tell the only reason for
them right now is the ability to install
On Sun, 2003-03-23 at 07:30, Christopher Faylor wrote:
On Sat, Mar 22, 2003 at 02:26:24PM -0600, Joshua Daniel Franklin wrote:
I have been messing with setup.exe's chooser documenting it this
morning and it hit me that a lot space is being used up with the
Bin and Src checkboxes. As far as I
On Sun, 2003-03-23 at 10:29, Joshua Daniel Franklin wrote:
I tend to download sources by hand and don't use setup for that
Me too. Also not that I don't like setup.exe, but usually I donwload
src to take a quick look at something and then delete the tree,
and so don't need it in
On Sun, 2003-03-23 at 11:55, Igor Pechtchanski wrote:
On 23 Mar 2003, Robert Collins wrote:
[snip]
* For most users, the Direct Connection method of downloading is the
best choice.
IMO the IE5 method is best. I've been considering making it the default.
The IE5 method will leverage
On Fri, 2003-03-21 at 13:14, Igor Pechtchanski wrote:
Hi,
There seem to be a couple of problems with the URLs in messages from the
cygwin-apps-cvs list:
- The initial revision provides r1=NONE parameter, which is not recognized
by cvsweb
Hmm, no comment here.
- cvsweb goes to the Attic
On Fri, 2003-03-21 at 23:12, Alex Tibbles wrote:
snip
I don't care whether you use std::string, or
String++ for any new code,
as long as:
1) You don't leak memory (std::string will leak if
you use c_str())
IIRC.
I've been unable to confirm this. I tried the attached
program
On Wed, 2003-03-19 at 14:27, CoolCold wrote:
in event log such record exists:
--
The description for Event ID ( 0 ) in Source ( Cygwin ) cannot be
found. The local computer may not have the necessary registry
information or message DLL files to display messages from a remote
computer.
On Thu, 2003-03-20 at 12:43, Christopher Faylor wrote:
On Thu, Mar 20, 2003 at 11:03:23AM +1100, Robert Collins wrote:
Thats cool. The offer remains open. arch would be great for this (as it
addresses all the points above :}), but I simply don't have time to port
it properly to cygwin.
arch
On Thu, 2003-03-20 at 19:07, Robert Collins wrote:
cgf
TCM
TCM?
Doh. found it in the other thread.
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
Description: This is a digitally signed message part
On Fri, 2003-03-21 at 01:46, Max Bowsher wrote:
Is the shudder over Tom, or arch ? If Tom, then I can understand ;]. If
arch itself - have you tried a recent version? I've found arch to be
incredibly useful, making some of the things I'd fought with CVS to make
happen trivial. (Notably
On Fri, 2003-03-21 at 04:49, Igor Pechtchanski wrote:
Same as above, but regenerated against HEAD. ChangeLog is the same.
The only thing about this patch that really makes me uncomfortable is
having to run through iterators/FindVisitors twice. Any suggestions for
improvement are welcome.
On Fri, 2003-03-21 at 07:12, Max Bowsher wrote:
I was just working on extracting the correct changelog entries for the
release notes for the upcoming setup release, when I discovered an
accidental deletion of an entry.
OK to apply this to the 2.249 branch?
Please do.
Thanks,
Rob
--
GPG
On Fri, 2003-03-21 at 08:29, Max Bowsher wrote:
cvsup requires a remote cvsup server, doesn't it? So short of convincing
overseers to install cvsup on sources.redhat.com, it isn't useful here. Or
have I misunderstood it?
cvsup was there on sources when I experimented with this.
Rob
--
GPG
On Fri, 2003-03-21 at 09:16, Max Bowsher wrote:
Oh! I didn't think to actually check. I just assumed it wasn't, since it
isn't publicized.
Max.
http://gcc.gnu.org/cvsup.html
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
Description: This is a
On Fri, 2003-03-21 at 09:36, Max Bowsher wrote:
We now face the job of summarizing 42kB of ChangeLog entries. It seems the
best way to do this would be to collaborate, so we need a collaboration
tool. Subject to approval from Chris and/or overseers, I could set up a mini
cvs repository in my
On Fri, 2003-03-21 at 09:46, Max Bowsher wrote:
What level of detail is required here?
For non-user-obvious bugfixes, can we just say Many miscellaneous
bugfixes?
Do we mention command line options at all, given that there is no way to
dump a list?
The ones that are useful and new, yes.
On Thu, 2003-03-20 at 18:50, Thomas Pfaff wrote:
On Wed, 19 Mar 2003, Robert Collins wrote:
I just followed the already existing method names in thread.h. Methods
like forEach, initMutex or isGoodObject were made by you.
Yes, I realise that.
I would suggest to commit my patch now and do
On Thu, 2003-03-20 at 18:28, Thomas Pfaff wrote:
On Wed, 19 Mar 2003, Robert Collins wrote:
On Thu, 2003-03-20 at 00:54, Thomas Pfaff wrote:
2003-03-19 Thomas Pfaff [EMAIL PROTECTED]
* pthread.cc (pthread_equal): Replacement for pthread_equal in
thread.cc.
* thread.cc
On Fri, 2003-03-21 at 08:01, Tron Thomas wrote:
I'm sorry. I misread what you wrote. I thought you were suggesting
that I try the #elif directive in simple coding example to see if the
compiler would accept it. This is was I used:
This program compiles just fine, and if I define the
On Wed, 2003-03-19 at 16:44, Christopher Faylor wrote:
Chris, would you like me to update the requires line as per
http://cygwin.com/ml/cygwin-apps/2003-03/msg00527.html ?
I've added base-passwd to cygwin's setup.hint.
Thanks Chris.
Which leaves:
To release we need to:
done: * update the
On Wed, 2003-03-19 at 20:47, Max Bowsher wrote:
Robert Collins wrote:
Max,
you seem quite familiar with CVS... do you have any insight why rtag
would assert? I suspected it might be the modules config for setup,
and tried a few variations thereof, but removing the libgetopt++
didn't
On Thu, 2003-03-20 at 02:35, Brian Gallew wrote:
Robert Collins wrote:
* Get some feedback from the users.
What does the Keep radio button for?
The Not Installed view totally rocks. I love it.
I don't suppose there will ever be a
never-install-so-don't-ever-show-it-to-me view
On Thu, 2003-03-20 at 09:05, Igor Pechtchanski wrote:
Just built a CVS HEAD of setup. When choosing Using IE5 settings, the
new setup.ini is apparently not downloaded at all. Switching to Direct
connection fixes the problem. IE5 itself works fine.
What version of IE do you have installed?
On Thu, 2003-03-20 at 07:58, Pierre A. Humblet wrote:
This is with a recent setup from CVS.
What version number? C'mon, thats got to be obvious.
When I initially look at a category it is
in default mode. If there are no new packages,
all packages are either keep or skip.
After cycling
On Thu, 2003-03-20 at 10:19, Max Bowsher wrote:
Is it OK to commit stuff now?
Did you find anything else out about this?
The branch looks successfully created to me.
Can you just confirm it's OK to proceed with commits?
Oh yes, please do.
The rtag command still fails
[EMAIL
On Thu, 2003-03-20 at 10:41, Max Bowsher wrote:
Thanks for the offer - but I won't be doing this:
I'm currently restricted to dialup internet until I return to university -
CVS operations are just too slow.
Plus, I already get the functionality of cvs merging, by keeping one working
On Thu, 2003-03-20 at 07:58, Pierre A. Humblet wrote:
This is with a recent setup from CVS.
When I initially look at a category it is
in default mode. If there are no new packages,
all packages are either keep or skip.
After cycling through Install/Reinstall/Uninstall/Default
on the
On Thu, 2003-03-20 at 10:49, Igor Pechtchanski wrote:
I tend to agree with Max here (even though I have a broadband connection).
I'm *really* wary of touching remote repositories. And we could use
mailing list archives for patch history. ;-)
mailing list archives are nowhere near as easy
On Fri, 2003-02-28 at 22:10, Thomas Pfaff wrote:
Reorganize the list handling of the pthreads objects by using the List
template class and remove a lot of duplicate code.
This looks good, except for for_each. can you do a proper for_each
template implementation?
Rob
--
GPG key available at:
On Mon, 2003-03-17 at 19:17, Thomas Pfaff wrote:
On Thu, 13 Mar 2003, Cygwin (Robert Collins) wrote:
This:
if (1 == InterlockedIncrement ((long *)lock_counter))
is not safe. You can only check for equal to 0, less than 0, and greater
than 0 with InterlockedIncrement | Decrement
On Wed, 2003-03-19 at 21:15, Thomas Pfaff wrote:
Enabled by default. Sure, as long as they aren't turned off again, or
someone builds without them to get 386 support... Please, use the
compatible test, it won't alter the code much. You can test for 0 and
0 safely.
The mutex stuff does
On Thu, 2003-03-20 at 00:54, Thomas Pfaff wrote:
2003-03-19 Thomas Pfaff [EMAIL PROTECTED]
* pthread.cc (pthread_equal): Replacement for pthread_equal in
thread.cc.
* thread.cc: Rename pthread_equal to pthread::equal throughout.
(pthread_equal): Remove.
*
Looks good to me. Chris, you happy with the winbase stuff?
Rob
On Thu, 2003-03-20 at 00:49, Thomas Pfaff wrote:
2003-03-19 Thomas Pfaff [EMAIL PROTECTED]
* thread.cc (pthread_cond::Wait): Remove typecasts for unsigned
long values when calling Interlocked functions. Use new UL
On Tue, 2003-03-18 at 19:22, Max Bowsher wrote:
OK to apply?
ChangeLog?
Rob
--
GPG key available at: http://users.bigpond.net.au/robertc/keys.txt.
signature.asc
Description: This is a digitally signed message part
Ok,
Once Max has the exit() changes ChangeLogged and commited, I'll be
branching setup for release.
I don't consider the md5 check to be release critical at this point.
Once we have a beta 'out there' we can get broader feedback.
Rob
--
GPG key available at:
FYI: I've checked in an update to expose the options in an OptionSet as
a collection.
This should form the basis for a setup.exe --help option that *works*
:}.
If you end up formatting the output, please consider refactoring the
other OptionSet code to reuse.paramterise that formatting logic.
On Wed, 2003-03-19 at 01:25, Igor Pechtchanski wrote:
Looks good.
Ok, can you:
Merge in the HEAD changes, so that the patch will apply cleanly.
Regenerate the patch.
Max, I'm going to be very busy for a few days - can you please merge
this patch in once Igor gets it to apply cleanly to HEAD.
On Wed, 2003-03-19 at 02:54, Max Bowsher wrote:
Several points:
1) In 5 of the 6, *.input directories, there is a spurious extra copy of the
directory inside itself. I've checked with librsync cvs - these directories
have never existed upstream.
I propose to cvs remove these.
Please do.
As we have multiple reviews of patches going on, overlapping patches
will naturally conflict...
So, I'm repeating an offer made here a long while back: cvs branches can
be used for developing patches.
i.e. maxb_UserSettings would be a branch of setup for Max to work on
user settings.
Scripts
Max,
you seem quite familiar with CVS... do you have any insight why rtag
would assert? I suspected it might be the modules config for setup, and
tried a few variations thereof, but removing the libgetopt++ didn't
correct it...
Rob
--
GPG key available at:
On Wed, 2003-03-19 at 10:17, Igor Pechtchanski wrote:
Good enough
for now, I guess.
Yep.
Done. Also upgraded to call the current interface instead of the
deprecated one.
Cool.
Same as above, but regenerated against cvs HEAD.
Thanks,
Thank you. I'm slating this for post-next-release
Ok,
there is a release candidate (2.340.2.1) up on
http://www.cygwin.com/setup-snapshots.
To release we need to:
* update the requires line for the cygwin package (otherwise new
installs will not setup info dirs correctly when /etc/passwd matters).
* Get some feedback from the users.
* Check we
On Wed, 2003-03-19 at 13:01, Igor Pechtchanski wrote:
Rob,
I don't know much about CVS, but I was getting some errors on the cfgaux
directory (which has been removed from CVS, but then recreated locally by
bootstrap.sh). CVS thinks it's still a valid directory, and tries to
recurse into
On Wed, 2003-03-19 at 13:22, Igor Pechtchanski wrote:
Well, cvs update -d -P took care of that. I was getting the following
error on cvs update, though:
cvs update: in directory setup/cfgaux:
cvs update: cannot open CVS/Entries for reading: No such file or directory
This happened with
The attached file is the result of visiting every node in my installed
cygwin packages via Tarjans strongly connected set identifier algortithm.
IOW it's a cycle-safe dependency ordered list of what I've got installed.
If we want to ensure that base-passwd runs before any other install
Igor Pechtchanski wrote:
Rob,
I don't think the attached list is correct (in terms of script ordering,
not in terms of the dependences). For example, every package requires
_update_info_dir (to pull it in), but its postinstall script should be run
*last*. In fact, that must have been the
On Tue, 2003-03-18 at 09:41, Igor Pechtchanski wrote:
On Tue, 18 Mar 2003, Robert Collins wrote:
File extraction completes before any postinstall script executes. So
any info files just dropped in will be fine.
Rob
I see. So this should work ok for now, as no info files are created
I've uploaded a new snapshot with the new postinstall dependency order
logic.
Enjoy.
A big thanks to Igor, for providing both the concept and the impetus :}.
Rob
--
Robert Collins [EMAIL PROTECTED]
signature.asc
Description: This is a digitally signed message part
and FilterVisitor.h
to inilint_SOURCES and setup_SOURCES.
Looks nice.
The strmatching logic should be a method of String++ IMO.
Also FilterVisitor isn't needed for inilint.
Cheers,
Rob
--
Robert Collins [EMAIL PROTECTED]
signature.asc
Description: This is a digitally signed message part
Free HTMLHelp compilers
around, or if we particularly care, but MS's is free for the downloading, and it
works OK in a build script.
There is a docbook-man tool, and IIRC a docbook-htmlHelp converter as
well. That would be wirth investigating IMO.
Rob
--
Robert Collins [EMAIL PROTECTED
On Sun, 2003-03-16 at 01:01, Max Bowsher wrote:
#ifndef _CYGWIN_
ExitProcess (1);
#else
exit (1);
#endif
I'm puzzled by the above. Shouldn't we be giving MSVCRT the opportunity to
run any shutdown it wishes to?
Well ExitProcess is the windows API exit call. Is there a MSVCRT
On Sun, 2003-03-16 at 20:34, Max Bowsher wrote:
Robert Collins wrote:
Max Bowsher wrote:
Robert,
Currently setup uses both stdexcept exceptions, and its own
std::exception-derived Exceptions.
I'm about to write a patch to convert the kludgy
if (error)
exit
On Sun, 2003-03-16 at 21:24, Pavel Tsekov wrote:
Just chaging exit() - throw Exception won't help much. There should be
an exception handler installed who takes care of processing these
exceptions i.e. shutdown setup with a simple message describing the
situation.
If you going to do this
On Sun, 2003-03-16 at 21:53, Max Bowsher wrote:
Makes sense. I'll wait for Robert to explain why setup has 2 different
exception paradigms, and which one can be removed, and if we keep Exception,
whether we can junk appErrNo, and replace it with derived classes as you
suggest.
appErrNo is
On Mon, 2003-03-17 at 01:44, Max Bowsher wrote:
I've been toying with some ideas for UserSetting persistence in setup.conf,
and have run into the following issue:
String, from String++.* lacks some functionality I need. std::string has the
necessary functionality.
Is one of String and
On Mon, 2003-03-17 at 02:24, Max Bowsher wrote:
Unfortunately, we will require multiple toplevel catch blocks: One
in each WndProc and thread initiation routine.
Actually, I've just read about set_terminate(), and I'm not so sure about
this now.
I wonder how exceptions interact with
On Mon, 2003-03-17 at 06:16, Igor Pechtchanski wrote:
Rob,
Will I be causing grief for you by applying this patch?
Not really, but let's bring your patch and mine closer, so that you could
reuse the good parts of mine (and I could avoid duplicating code). See
below.
Sure.
The
On Mon, 2003-03-17 at 06:51, Igor Pechtchanski wrote:
STL containers won't choke, as they only need a partial order, AFAIU.
The fail simply means it's not a strict weak ordering.
They can and will do the wrong thing.
because it's not a strict weak ordering, if we have:
foo: bar, zonk
gam:
On Mon, 2003-03-17 at 07:13, Igor Pechtchanski wrote:
And for clarity: my suggested tweak is also not sufficient to provide a
weak ordering.
Rob
Rob,
Your suggested tweak provides a total ordering. The unordered(x,y)
[!(x y) !(y x)] relation is false for any x != y [since either
401 - 500 of 2024 matches
Mail list logo
