|
Author: Chukharev
Date: 15-09-08 09:00
I remember that earlier FreshPorts had a reminder when a ports freeze was in effect. Now, I believe, ports are frozen, but no reminder is seen...
|
|
Reply To This Message
|
|
Author: Dan
Date: 15-09-08 13:56
Agreed.
If you can find me the commit that set the freeze, I'll review it tonight when I get home. :)
A URL into the FreeBSD archives would help me.
Thank you.
--
The Man Behind The Curtain
|
|
Reply To This Message
|
|
Author: Chukharev
Date: 16-09-08 08:42
http://www.freebsd.org/releng/index.html says about freeze status the following.
Subtree Status Contact Notes
ports/ Frozen portmgr@FreeBSD.org FreeBSD Ports Collection.
Let's look at the first of the upcoming releases.
http://www.freebsd.org/releases/7.1R/schedule.html indicates:
Action Expected Actual Description
Reminder announcement -- 22 August 2008 Release Engineers send announcement email to developers...
Announce the Ports Freeze -- 22 August 2008 Someone from portmgr@ should email freebsd-ports@...
Code freeze begins 29 August 2008 2 September 2008 After this date, all commits to RELENG_7 must be ...
Announce doc/ tree slush TBD -- Notification of the impending doc/ tree slush...
Ports tree frozen 8 September 2008 8 September 2008 Only approved commits will be permitted to the ports...
Ok, what was commited around Sep 8? The only commit I can connect to the freeze is
http://docs.freebsd.org/cgi/getmsg.cgi?fetch=766329+0+archive/2008/cvs-ports/20080914.cvs-ports
Hmm... Not much.
Aha! The annoncement in -ports@ had a typo in Subject:
http://docs.freebsd.org/cgi/getmsg.cgi?fetch=126396+0+archive/2008/freebsd-ports/20080914.freebsd-ports
Well, I don't know how helpful can be these findings, but here you are.
Gud luck!
Vladimir
|
|
Reply To This Message
|
|
Author: Dan
Date: 16-09-08 14:27
I think it's CVSROOT/approvers that I need to check.
--
The Man Behind The Curtain
|
|
Reply To This Message
|
|
Author: Dan
Date: 17-09-08 03:25
I have fixed the problem. The solution is in my development tree. I'll leave it there for further testing before checking it in and moving it to production.
--
The Man Behind The Curtain
|
|
Reply To This Message
|
|
Author: Chukharev
Date: 16-12-08 07:20
Dan wrote:
> The website now says it's a freeze. I think that's because it missed the commit/update/something.
I guessed that might be the case, and it was one of the reasons I started this again... Can you fight against missed commits by checking last commit diff from a cron job? Say, weekly job would limit the delay of status change reasonably.
> If you can see a way to tell....
Unfortunately, nothing better than looking into the commit message... And that is definitely too bad unless a special test can be added to the commit checking system.
Best regards,
Vladimir
|
|
Reply To This Message
|
|
Author: Dan
Date: 16-12-08 13:21
Chukharev wrote:
> Dan wrote:
> > The website now says it's a freeze. I think that's because it
> > missed the commit/update/something.
>
> I guessed that might be the case, and it was one of the reasons
> I started this again... Can you fight against missed commits by
> checking last commit diff from a cron job? Say, weekly job
> would limit the delay of status change reasonably.
It was not a missed commit. It was a file permission error.
As for looking for missed commits, the only strategy I can think of is comparing the message-ids from the emails against those stored in the database.
Also, the SVN revision IDs may be useful here.
If someone wishes to write a script (shell, python, or perl) that obtains a text file of IDs (messages-id or emails) from somewhere, then compares them against the database, that will come up with a list of missing IDs. Then we would obtain those missing emails.
Now that I think a bit more: it is better to use message-ids from the mailing list archives. That way, we can get the emails easily, which can them be parsed into XML and loaded into the system.
If we use SVN revision ids, we will have to create an 'email-like' message and parse it.
> > If you can see a way to tell....
>
> Unfortunately, nothing better than looking into the commit
> message... And that is definitely too bad unless a special test
> can be added to the commit checking system.
I don't want to get into parsing email message content. It will have false positives.
--
The Man Behind The Curtain
|
|
Reply To This Message
|
|
Author: Chukharev
Date: 17-04-09 12:14
(Well, you guess what I gonna say!)
Ports freeze is in effect but not indicated by freshports.org :(
|
|
Reply To This Message
|
|
If you buy from Amazon USA, please support us by using this link.