 used to display messages. Looks better now,
if you ask me. What do you think?

I failed to do this using css. If I understand
http://www.w3.org/TR/CSS21/text.html#propdef-white-space correctly, a 

table.messages td pre 
{ white-space: pre-wrap } 

..should do the trick, but it doesn't work for some unknown reason.

----------
status: chatting -> resolved

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 21:04:16 2007
From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Roch=C3=A9_Compaan?=)
Date: Mon, 15 Jan 2007 20:04:16 -0000
Subject: [Tracker-discuss] [issue75] Improve wrapping of comment text
In-Reply-To: <1168890907.7.0.0333780649992.issue75@psf.upfronthosting.co.za>
Message-ID: <1168891332.6709.27.camel@kwaaitjie>

Roch? Compaan added the comment:

On Mon, 2007-01-15 at 19:55 +0000, Erik Forsberg wrote:
> Erik Forsberg  added the comment:
> 
> I added a wrap="true" to the  used to display messages. Looks better now,
> if you ask me. What do you think?
> 
> I failed to do this using css. If I understand
> http://www.w3.org/TR/CSS21/text.html#propdef-white-space correctly, a 
> 
> table.messages td pre 
> { white-space: pre-wrap } 

I've been here before. A slight improvement to make it cross browser
compatible:

pre {
 white-space: pre-wrap;       /* css-3 */
 white-space: -moz-pre-wrap;  /* Mozilla, since 1999 */
 white-space: -pre-wrap;      /* Opera 4-6 */
 white-space: -o-pre-wrap;    /* Opera 7 */
 word-wrap: break-word;       /* Internet Explorer 5.5+ */
}

See:
http://myy.helia.fi/~karte/pre-wrap-css3-mozilla-opera-ie.html

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 21:16:43 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Mon, 15 Jan 2007 20:16:43 -0000
Subject: [Tracker-discuss] [issue75] Improve wrapping of comment text
In-Reply-To: <1168891332.6709.27.camel@kwaaitjie> (Roché Compaan's message of "Mon, 15 Jan 2007 20:04:16 -0000")
Message-ID: <877ivo3v11.fsf@uterus.efod.se>

Erik Forsberg added the comment:

Roch? Compaan  writes:

> I've been here before. A slight improvement to make it cross browser
> compatible:

Ah. Thankyou!

The wonderful world of browsers.. 

\EF
-- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 21:26:18 2007
From: metatracker at psf.upfronthosting.co.za (Richard Jones)
Date: Mon, 15 Jan 2007 20:26:18 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
In-Reply-To: <8764b85fqx.fsf@uterus.efod.se>
Message-ID: 

Richard Jones added the comment:

On 16/01/2007, at 5:03 AM, Erik Forsberg wrote:
> Right now, issue numbers are allocated by adding 1 to the number of
> the last issue already present in the database. As the sourceforge
> bugs imported had high numbers, newly created bugs will also get high
> numbers.
>
> I'm assuming we want to keep the old numbers for old bugs. I don't
> know how roundup will behave when a taken number appears if we start
> at a lower number, if it is at all possible.

At the end of import you should be telling the hyperdb what the next  
id sequence is and then it can allocate new ids correctly. You must  
definitely NOT start allocating at lower number - it'll then  
eventually reallocate the ids already used. The API call to do this is:

    self.db.setid(classname, str(maxid+1))

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 21:33:07 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Mon, 15 Jan 2007 20:33:07 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
In-Reply-To: 
	(Richard Jones's message of "Mon, 15 Jan 2007 20:26:18 -0000")
Message-ID: <87y7o42fpa.fsf@uterus.efod.se>

Erik Forsberg added the comment:

Richard Jones  writes:

> At the end of import you should be telling the hyperdb what the next  
> id sequence is and then it can allocate new ids correctly. 

This already is in place, as the last lines of [1] shows.

> You must definitely NOT start allocating at lower number - it'll
> then eventually reallocate the ids already used.

OK - this answers the second question - if it's possible at all to use
lower numbers while keeping the sourceforge bug IDs. 

[1] http://svn.python.org/view/tracker/importer/sfxml2roundup.py?rev=53081&view=auto

Cheers,
\EF
-- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 21:58:40 2007
From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Roch=C3=A9_Compaan?=)
Date: Mon, 15 Jan 2007 20:58:40 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
In-Reply-To: <87y7o42fpa.fsf@uterus.efod.se>
Message-ID: <1168894595.6709.47.camel@kwaaitjie>

Roch? Compaan added the comment:

On Mon, 2007-01-15 at 20:33 +0000, Erik Forsberg wrote:
> Erik Forsberg added the comment:
> 
> Richard Jones  writes:
> 
> > At the end of import you should be telling the hyperdb what the next  
> > id sequence is and then it can allocate new ids correctly. 
> 
> This already is in place, as the last lines of [1] shows.
> 
> > You must definitely NOT start allocating at lower number - it'll
> > then eventually reallocate the ids already used.
> 
> OK - this answers the second question - if it's possible at all to use
> lower numbers while keeping the sourceforge bug IDs. 

Is it not an option to use a prefix for sourceforge bug ids, say SF, and
then restart with a lower number for bugs generated in Roundup?

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 21:59:43 2007
From: metatracker at psf.upfronthosting.co.za (Brett C.)
Date: Mon, 15 Jan 2007 20:59:43 -0000
Subject: [Tracker-discuss] [issue68] email addresses showing
Message-ID: <1168894783.79.0.466073006291.issue68@psf.upfronthosting.co.za>

Brett C.  added the comment:

Emails do show up when you enter a person's email address and request a password
reminder/reset.

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 22:07:08 2007
From: metatracker at psf.upfronthosting.co.za (Brett C.)
Date: Mon, 15 Jan 2007 21:07:08 -0000
Subject: [Tracker-discuss] [issue76] Want just any user to be able to edit
	nosy list?
Message-ID: <1168895228.43.0.0383309268963.issue76@psf.upfronthosting.co.za>

New submission from Brett C. :

I noticed that any user can edit the nosy list on any bug report.  Is there any
worry of random users coming along and editing the list in a malicious way
(e.g., adding every person to single issue)?  I can see why the developers want
to be able to add people to the nosy list, but I don't know if regular users
need to.

Perhaps it would be enough to just allow the submitter of a bug to be able to
assign the bug initially to get the attention of someone (assuming that if an
issue is assigned to someone they are implicitly added to the nosy list or at
least notified of the assignment).  Occasionally developers will tell someone to
create a bug or patch item and assign it to them.  That is probably enough to
get someone's initial attention.  After that it is up to the developers, I
think, to pull in who needs to come in to help with the issue.

----------
messages: 363
nosy: brettcannon
priority: bug
status: unread
title: Want just any user to be able to edit nosy list?

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 22:11:24 2007
From: metatracker at psf.upfronthosting.co.za (Brett C.)
Date: Mon, 15 Jan 2007 21:11:24 -0000
Subject: [Tracker-discuss] [issue44] Set up python-dev tracker
Message-ID: <1168895484.49.0.973577635603.issue44@psf.upfronthosting.co.za>

Brett C.  added the comment:

I think requiring Type is fine.  Plus I don't know if "behavior" is the right
word.  I personally prefer "semantics".  Anyone else agree?

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 22:13:21 2007
From: metatracker at psf.upfronthosting.co.za (Brett C.)
Date: Mon, 15 Jan 2007 21:13:21 -0000
Subject: [Tracker-discuss] [issue76] Restricting what non-submitter,
	non-developer can edit
Message-ID: <1168895601.46.0.472922586154.issue76@psf.upfronthosting.co.za>

Brett C.  added the comment:

Making this a more broad issue.

Do we really want people who are not the submitter or a developer able to edit
anything on an issue in terms of metadata on the issue?  Leaving a comment and
uploading a file is fine, but perhaps they shouldn't be able to do anything else.

----------
status: unread -> chatting
title: Want just any user to be able to edit nosy list? -> Restricting what non-submitter, non-developer can edit

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 22:24:02 2007
From: metatracker at psf.upfronthosting.co.za (Brett C.)
Date: Mon, 15 Jan 2007 21:24:02 -0000
Subject: [Tracker-discuss] [issue77] Assigned To list not sorted
Message-ID: <1168896242.11.0.101010921388.issue77@psf.upfronthosting.co.za>

New submission from Brett C. :

The Assigned To fields drop-down list is not sorted on the tracker.  With the
list so long that definitely needs to be fixed.

----------
messages: 366
nosy: brettcannon
priority: bug
status: unread
title: Assigned To list not sorted

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 22:25:38 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Mon, 15 Jan 2007 21:25:38 -0000
Subject: [Tracker-discuss] [issue76] Want just any user to be able to edit
	nosy list?
In-Reply-To: <1168895228.43.0.0383309268963.issue76@psf.upfronthosting.co.za>
	(Brett C.'s message of "Mon, 15 Jan 2007 21:07:08 -0000")
Message-ID: <87tzys2d9q.fsf@uterus.efod.se>

Erik Forsberg added the comment:

"Brett C."  writes:

> worry of random users coming along and editing the list in a malicious way
> (e.g., adding every person to single issue)?  I can see why the developers want
> to be able to add people to the nosy list, but I don't know if regular users
> need to.

I for sure want to be able to add myself to the nosy list of issues
that I'm interested in, to get updates when something happens with a
bug I'm having trouble with in daily work.

This is of course could be implemented with an finer access control
than the one currently in place - users without developer privileges
could be limited to add/remove only their own username from the nosy
list.

\EF
-- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9

----------
title: Restricting what non-submitter, non-developer can edit -> Want just any user to be able to edit nosy list?

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 22:30:02 2007
From: metatracker at psf.upfronthosting.co.za (Brett C.)
Date: Mon, 15 Jan 2007 21:30:02 -0000
Subject: [Tracker-discuss] [issue78] Make use of keywords more obvious
Message-ID: <1168896602.03.0.530314090964.issue78@psf.upfronthosting.co.za>

New submission from Brett C. :

Not sure best way to go about this, but I did not get how to use the keywords
field until I added a new one and noticed it became a list.  The solution might
be enough to just have more than a single keyword to begin with so that it is
obvious it is a list (probably shouldn't start with Py3K anyway but make it a
version number or something).

I figured the keywords would be used for things like branches or something;
stuff that is really transitive and not consistent in Python like bugs on the
stdlib or the interpreter core.

----------
messages: 368
nosy: brettcannon
priority: wish
status: unread
title: Make use of keywords more obvious

______________________________________________________
Meta Tracker 

______________________________________________________

From brett at python.org  Mon Jan 15 22:36:13 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 15 Jan 2007 13:36:13 -0800
Subject: [Tracker-discuss] clarification on Components and Keywords fields
	usage
Message-ID: 

I just wanted to get straight in my head what we are planning to use
keywords for compared to the components field.  For instance, the
components field lists "Windows", "XML", "Regular Expressions",
"Tkinter", etc.  I personally feel that is too specific to be in the
components list and instead should be treated as a keyword if someone
wants that much detail as to what it affects.

But is that what we are thinking to use the keywords field for?  Or is
there some other planned use?

-Brett

From brett at python.org  Mon Jan 15 22:39:33 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 15 Jan 2007 13:39:33 -0800
Subject: [Tracker-discuss] Managing the tracker metadata options
Message-ID: 

I just wanted to find out who is going to be able to make users a
Developer?  Or how about removing keywords, adding new versions, or
editing the components list?

Is all of that going to need to go through you guys (Erik, Stefan, and
Paul)?  I assume that is all schema stuff and thus you guys are the
only ones who can play with that.  That's fine with me but I just
wanted to double-check.

-Brett

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 22:45:57 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Mon, 15 Jan 2007 21:45:57 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
In-Reply-To: <1168894595.6709.47.camel@kwaaitjie> (Roché Compaan's message of "Mon, 15 Jan 2007 20:58:40 -0000")
Message-ID: <87irf82cbw.fsf@uterus.efod.se>

Erik Forsberg added the comment:

> Roch? Compaan added the comment:

> Is it not an option to use a prefix for sourceforge bug ids, say SF, and
> then restart with a lower number for bugs generated in Roundup?

Here's one idea:

1) Add a new property to the issue class, sf_id=Number()

2) Rewrite importer to set sf_id to the original sourceforge id of
   each issue, but don't set the roundup ID to the sourceforge id -
   instead, allocate issue IDs on the go, beginning with 1, ending
   with about 9940 (that's the number of issues currently in the sf
   database).

3) Add ability to search for sourceforge ID in search interface.

4) Make sf bug redirector aware of the mechanism, so that
   python.org/sf/ redirects to the correct issue.

This will break the one to one mapping between the sourceforge
artifact IDs and the new roundup issue ids, but it might be a better
solution for day to day use.

Comments? 

\EF
-- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 22:50:06 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Mon, 15 Jan 2007 21:50:06 -0000
Subject: [Tracker-discuss] [issue68] email addresses showing
In-Reply-To: <1168894783.79.0.466073006291.issue68@psf.upfronthosting.co.za>
Message-ID: 

Paul Dubois added the comment:

I think that isn't a problem, since it is sent just to them.

On 1/15/07, Brett C.  wrote:
>
>
> Brett C.  added the comment:
>
> Emails do show up when you enter a person's email address and request a
> password
> reminder/reset.
>
> ______________________________________________________
> Meta Tracker 
> 
> ______________________________________________________
>

______________________________________________________
Meta Tracker 

______________________________________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20070115/00c9044a/attachment.html 

From brett at python.org  Mon Jan 15 22:55:38 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 15 Jan 2007 13:55:38 -0800
Subject: [Tracker-discuss] [issue73] Issue numbering
In-Reply-To: <87irf82cbw.fsf@uterus.efod.se>
References: <1168894595.6709.47.camel@kwaaitjie>
	<87irf82cbw.fsf@uterus.efod.se>
Message-ID: 

On 1/15/07, Erik Forsberg  wrote:
>
> Erik Forsberg added the comment:
>
> > Roch? Compaan added the comment:
>
> > Is it not an option to use a prefix for sourceforge bug ids, say SF, and
> > then restart with a lower number for bugs generated in Roundup?
>
> Here's one idea:
>
> 1) Add a new property to the issue class, sf_id=Number()
>
> 2) Rewrite importer to set sf_id to the original sourceforge id of
>    each issue, but don't set the roundup ID to the sourceforge id -
>    instead, allocate issue IDs on the go, beginning with 1, ending
>    with about 9940 (that's the number of issues currently in the sf
>    database).
>
> 3) Add ability to search for sourceforge ID in search interface.
>
> 4) Make sf bug redirector aware of the mechanism, so that
>    python.org/sf/ redirects to the correct issue.
>
> This will break the one to one mapping between the sourceforge
> artifact IDs and the new roundup issue ids, but it might be a better
> solution for day to day use.
>
> Comments?

So if I am following this correctly this would internally store the SF
number but externally number starting at 1 for Roundup numbers?  As
long as you still have www.python.org/sf/ still do its redirect
properly (or your suggestion of a query field) that seems good to me
as eventually we won't care too much about the SF numbers.

-Brett

From pfdubois at gmail.com  Mon Jan 15 22:58:40 2007
From: pfdubois at gmail.com (Paul Dubois)
Date: Mon, 15 Jan 2007 13:58:40 -0800
Subject: [Tracker-discuss] Fwd: Failed issue tracker submission
In-Reply-To: <20070115215417.D203678065@psf.localdomain>
References: <20070115215417.D203678065@psf.localdomain>
Message-ID: 

I got the error Martin got. I don't understand this, in all the years we've
used Roundup on my project this has never happened.

The Re: [Tracker-discuss] is supposed to be stripped. The hypen in the
tracker name maybe?

---------- Forwarded message ----------
From: Meta Tracker 
Date: Jan 15, 2007 1:54 PM
Subject: Failed issue tracker submission
To: pfdubois at gmail.com

The class name you identified in the subject line ("Tracker-discuss") does
not exist in the
database.

Valid class names are: file, issue, keyword, msg, priority, query, status,
user
Subject was: "Re: [Tracker-discuss] [issue73] Issue numbering"

Mail Gateway Help
=================
Incoming messages are examined for multiple parts:
. In a multipart/mixed message or part, each subpart is extracted and
   examined. The text/plain subparts are assembled to form the textual
   body of the message, to be stored in the file associated with a "msg"
   class node. Any parts of other types are each stored in separate files
   and given "file" class nodes that are linked to the "msg" node.
. In a multipart/alternative message or part, we look for a text/plain
   subpart and ignore the other parts.

Summary
-------
The "summary" property on message nodes is taken from the first non-quoting
section in the message body. The message body is divided into sections by
blank lines. Sections where the second and all subsequent lines begin with
a ">" or "|" character are considered "quoting sections". The first line of
the first non-quoting section becomes the summary of the message.

Addresses
---------
All of the addresses in the To: and Cc: headers of the incoming message are
looked up among the user nodes, and the corresponding users are placed in
the "recipients" property on the new "msg" node. The address in the From:
header similarly determines the "author" property of the new "msg"
node. The default handling for addresses that don't have corresponding
users is to create new users with no passwords and a username equal to the
address. (The web interface does not permit logins for users with no
passwords.) If we prefer to reject mail from outside sources, we can simply
register an auditor on the "user" class that prevents the creation of user
nodes with no passwords.

Actions
-------
The subject line of the incoming message is examined to determine whether
the message is an attempt to create a new item or to discuss an existing
item. A designator enclosed in square brackets is sought as the first thing
on the subject line (after skipping any "Fwd:" or "Re:" prefixes).

If an item designator (class name and id number) is found there, the newly
created "msg" node is added to the "messages" property for that item, and
any new "file" nodes are added to the "files" property for the item.

If just an item class name is found there, we attempt to create a new item
of that class with its "messages" property initialized to contain the new
"msg" node and its "files" property initialized to contain any new "file"
nodes.

Triggers
--------
Both cases may trigger detectors (in the first case we are calling the
set() method to add the message to the item's spool; in the second case we
are calling the create() method to create a new node). If an auditor raises
an exception, the original message is bounced back to the sender with the
explanatory message given in the exception.

$Id: mailgw.py,v 1.180 2006/12/19 01:13:31 richard Exp $
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20070115/3428dee8/attachment.html 
-------------- next part --------------
Return-Path: 
X-Original-To: metatracker at psf.upfronthosting.co.za
Delivered-To: metatracker at psf.upfronthosting.co.za
Received: from an-out-0708.google.com (an-out-0708.google.com [209.85.132.249])
	by psf.localdomain (Postfix) with ESMTP id 47A4A78054
	for ; Mon, 15 Jan 2007 21:54:17 +0000 (UTC)
Received: by an-out-0708.google.com with SMTP id c35so653776anc
        for ; Mon, 15 Jan 2007 13:54:15 -0800 (PST)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=beta;
        h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references;
        b=sryuAMk+OY2uaS6DKO+p2rO389PdsTZRLxNa49FpWIRwCxRUseV5UfLBWwn5Xq2BcH2kyxveVQRUPgnh/2bpTvUKkXKj7gMFNrTgoALq6w1ZsiM/7NIcJ5VldT7D3fnRGwNYVTYK/YTKACw/e1iyy0hgnyxvsq7CqrscfThTKI4=
Received: by 10.35.69.11 with SMTP id w11mr8505894pyk.1168898054773;
        Mon, 15 Jan 2007 13:54:14 -0800 (PST)
Received: by 10.35.40.8 with HTTP; Mon, 15 Jan 2007 13:54:14 -0800 (PST)
Message-ID: 
Date: Mon, 15 Jan 2007 13:54:14 -0800
From: "Paul Dubois" 
To: "Meta Tracker" 
Subject: Re: [Tracker-discuss] [issue73] Issue numbering
In-Reply-To: <87irf82cbw.fsf at uterus.efod.se>
MIME-Version: 1.0
Content-Type: multipart/alternative; 
	boundary="----=_Part_42543_16268805.1168898054310"
References: <1168894595.6709.47.camel at kwaaitjie>
	 <87irf82cbw.fsf at uterus.efod.se>

------=_Part_42543_16268805.1168898054310
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

This also has the virtue that at some point we could drop it. I like it.

On 1/15/07, Erik Forsberg  wrote:
>
>
> Erik Forsberg added the comment:
>
> > Roch=E9 Compaan added the comment:
>
> > Is it not an option to use a prefix for sourceforge bug ids, say SF, an=
d
> > then restart with a lower number for bugs generated in Roundup?
>
> Here's one idea:
>
> 1) Add a new property to the issue class, sf_id=3DNumber()
>
> 2) Rewrite importer to set sf_id to the original sourceforge id of
>    each issue, but don't set the roundup ID to the sourceforge id -
>    instead, allocate issue IDs on the go, beginning with 1, ending
>    with about 9940 (that's the number of issues currently in the sf
>    database).
>
> 3) Add ability to search for sourceforge ID in search interface.
>
> 4) Make sf bug redirector aware of the mechanism, so that
>    python.org/sf/ redirects to the correct issue.
>
> This will break the one to one mapping between the sourceforge
> artifact IDs and the new roundup issue ids, but it might be a better
> solution for day to day use.
>
> Comments?
>
> \EF
> --
> Erik Forsberg                 http://efod.se
> GPG/PGP Key: 1024D/0BAC89D9
>
> ______________________________________________________
> Meta Tracker 
> 
> ______________________________________________________
> _______________________________________________
> Tracker-discuss mailing list
> Tracker-discuss at python.org
> http://mail.python.org/mailman/listinfo/tracker-discuss
>

------=_Part_42543_16268805.1168898054310
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

This also has the virtue that at some point we could drop it. I like it.
On 1/15/07, Erik Forsberg <
metatracker at psf.upfronthosting.co.za> wrote:
Erik Forsberg added the commen=
t:

> Roch=E9 Compaan added the comment:

> Is it not an op=
tion to use a prefix for sourceforge bug ids, say SF, and
> then rest=
art with a lower number for bugs generated in Roundup?

Here's on=
e idea:

1) Add a new property to the issue class, sf_id=3DNumber()

2=
) Rewrite importer to set sf_id to the original sourceforge id of
 =
  each issue, but don't set the roundup ID to the sourceforge id -=

   instead, allocate issue IDs on the go, beginning with 1, e=
nding

   with about 9940 (that's the number of issues currently=
 in the sf
   database).

3) Add ability to search for s=
ourceforge ID in search interface.

4) Make sf bug redirector aware o=
f the mechanism, so that

   python.org/sf/<id=
> redirects to the correct issue.

This will break the one to one =
mapping between the sourceforge
artifact IDs and the new roundup issue i=
ds, but it might be a better

solution for day to day use.

Comments?

\EF
--
Erik =
Forsberg           &=
nbsp;     http://efod.se=

GPG/PGP Key: 1024D/0BAC89D9

____________________________________=
__________________

Meta Tracker <http://psf.upfronthosting.co.za=
/roundup/meta/issue73
>
______________________________________________________
_____=
__________________________________________
Tracker-discuss mailing list<=
br>Tracker-discuss at python.org

http://mail.python.org/mailman/listinfo/tracker-discuss

------=_Part_42543_16268805.1168898054310--

From pfdubois at gmail.com  Mon Jan 15 23:15:06 2007
From: pfdubois at gmail.com (Paul Dubois)
Date: Mon, 15 Jan 2007 14:15:06 -0800
Subject: [Tracker-discuss] [issue76] Want just any user to be able to
	edit nosy list?
In-Reply-To: <87tzys2d9q.fsf@uterus.efod.se>
References: <1168895228.43.0.0383309268963.issue76@psf.upfronthosting.co.za>
	<87tzys2d9q.fsf@uterus.efod.se>
Message-ID: 

Actually, an anonymous (unsigned-in user) cannot edit an issue. A signed-in
user who did something malicious would leave an audit trail. So it isn't
that big a worry.

On the plus side, I often add someone else to an issue's nosy list when I
add a comment that I want them to see or if I know they should be aware of
the issue. So it's a feature.

On 1/15/07, Erik Forsberg  wrote:
>
>
> Erik Forsberg added the comment:
>
> "Brett C."  writes:
>
> > worry of random users coming along and editing the list in a malicious
> way
> > (e.g., adding every person to single issue)?  I can see why the
> developers want
> > to be able to add people to the nosy list, but I don't know if regular
> users
> > need to.
>
> I for sure want to be able to add myself to the nosy list of issues
> that I'm interested in, to get updates when something happens with a
> bug I'm having trouble with in daily work.
>
> This is of course could be implemented with an finer access control
> than the one currently in place - users without developer privileges
> could be limited to add/remove only their own username from the nosy
> list.
>
> \EF
> --
> Erik Forsberg                 http://efod.se
> GPG/PGP Key: 1024D/0BAC89D9
>
> ----------
> title: Restricting what non-submitter, non-developer can edit -> Want just
> any user to be able to edit nosy list?
>
> ______________________________________________________
> Meta Tracker 
> 
> ______________________________________________________
> _______________________________________________
> Tracker-discuss mailing list
> Tracker-discuss at python.org
> http://mail.python.org/mailman/listinfo/tracker-discuss
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20070115/d88758fa/attachment.htm 

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 23:15:12 2007
From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Martin_v._L=C3=B6wis?=)
Date: Mon, 15 Jan 2007 22:15:12 -0000
Subject: [Tracker-discuss] [issue56] Make bugs.python.org point to the
	roundup tracker
In-Reply-To: <87ac0k5fzz.fsf@uterus.efod.se>
Message-ID: <45ABFCEB.1040609@v.loewis.de>

Martin v. L?wis added the comment:

Erik Forsberg schrieb:
> Thinking about the idea of hosting more than the python-dev tracker,
> we should decide if we want the url of the tracker we're currently
> working on to be 'http://issues.python.org/' (or
> 'http://bugs.python.org/'), or if it should perhaps be something like
> 'http://issues.python.org/python-dev/', which would leave room for
> other trackers to appear in a natural way (i.e.,
> 'http://issues.python.org/jython', etc.).
> 
> Just a thought. Personally, I don't care - I'm not a daily user of the
> python issue database.

The canonical URLs should be as short as possible. I also think nobody
will remember that it's "issues.python.org"; people will prefer
"bugs.python.org" even though it's not technically correct (I personally
dislike "issues.python.org" *because* that is technically correct;
I despise correctness for merely political reasons (i.e. PC)).

Regards,
Martin

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 15 23:31:00 2007
From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Martin_v._L=C3=B6wis?=)
Date: Mon, 15 Jan 2007 22:31:00 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
Message-ID: <1168900260.93.0.0538823480138.issue73@psf.upfronthosting.co.za>

Martin v. L?wis  added the comment:

I like issue IDs to be really the same, so I would not encourage introduction of
a new field. If that means that one has to start numbering at the highest value,
so be it. There still seems to be an issue, though: I very much doubt that the
highest number after import was in the 999449 range. In fact (contrary to what I
suggested), the highest number after import is 1619060, i.e. *above* the numbers
that roundup then starts allocating. Could it be that the maximum-forming uses
the textual (decimal) representation, rather than the numeric value?

Still, I would like to see roundup changed to support low numbers in some form.
For example, importing currently starts at ID 207608. If new issues would start
at 1000 (say), it would take *years* until we get to 200000 (given that we
imported 9941 issues). So just raising an error if a new issue ID is above
200000 might be enough: if we ever get to that number (likely not in my
lifetime), we can skip over the imported numbers (going straight from 200000 to
2000000).

Others may feel differently about this, of course, and prefer that the numbers
increase in strictly historical order.

______________________________________________________
Meta Tracker 

______________________________________________________

From martin at v.loewis.de  Mon Jan 15 23:44:28 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Mon, 15 Jan 2007 23:44:28 +0100
Subject: [Tracker-discuss] [issue76] Want just any user to be able to
 edit	nosy list?
In-Reply-To: <1168895228.43.0.0383309268963.issue76@psf.upfronthosting.co.za>
References: <1168895228.43.0.0383309268963.issue76@psf.upfronthosting.co.za>
Message-ID: <45AC03CC.4070105@v.loewis.de>

> I noticed that any user can edit the nosy list on any bug report.  Is there any
> worry of random users coming along and editing the list in a malicious way
> (e.g., adding every person to single issue)?  I can see why the developers want
> to be able to add people to the nosy list, but I don't know if regular users
> need to.

If that's "any authenticated user", then it's fine with me. We will have
logs what the authenticated users do, and we can ban anybody who
misbehaves (but nobody will, because ding-dong ditches are fun only
if you don't get caught).

Regards,
Martin

From martin at v.loewis.de  Mon Jan 15 23:46:45 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Mon, 15 Jan 2007 23:46:45 +0100
Subject: [Tracker-discuss] [issue78] Make use of keywords more obvious
In-Reply-To: <1168896602.03.0.530314090964.issue78@psf.upfronthosting.co.za>
References: <1168896602.03.0.530314090964.issue78@psf.upfronthosting.co.za>
Message-ID: <45AC0455.2040403@v.loewis.de>

Brett C. schrieb:
> (probably shouldn't start with Py3K anyway but make it a
> version number or something).

We decided against that: the version field indicates past versions
(where the error occurred in), not future versions. The py3k tag
indicates a future version.

We need the tag to be able to filter/filter-out py3k issues.

Regards,
Martin

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 00:07:46 2007
From: metatracker at psf.upfronthosting.co.za (Richard Jones)
Date: Mon, 15 Jan 2007 23:07:46 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
In-Reply-To: <1168894595.6709.47.camel@kwaaitjie>
Message-ID: <655FDBC7-5C68-43BB-9409-7928FDA21E2E@commonground.com.au>

Richard Jones added the comment:

On 16/01/2007, at 7:58 AM, Roch? Compaan wrote:
> Is it not an option to use a prefix for sourceforge bug ids, say  
> SF, and
> then restart with a lower number for bugs generated in Roundup?

No. IDs are strictly numbers.

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 00:10:45 2007
From: metatracker at psf.upfronthosting.co.za (Richard Jones)
Date: Mon, 15 Jan 2007 23:10:45 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
In-Reply-To: <1168900260.93.0.0538823480138.issue73@psf.upfronthosting.co.za>
Message-ID: 

Richard Jones added the comment:

On 16/01/2007, at 9:31 AM, Martin v. L?wis wrote:
> Still, I would like to see roundup changed to support low numbers  
> in some form. For example, importing currently starts at ID 207608.  
> If new issues would start at 1000 (say), it would take *years*  
> until we get to 200000 (given that we imported 9941 issues). So  
> just raising an error if a new issue ID is above 200000 might be  
> enough: if we ever get to that number (likely not in my lifetime),  
> we can skip over the imported numbers (going straight from 200000  
> to 2000000).

Given that you're using postgres as the backend it should raise a key  
uniqueness error if that condition was actually met, BTW.

______________________________________________________
Meta Tracker 

______________________________________________________

From brett at python.org  Tue Jan 16 00:41:06 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 15 Jan 2007 15:41:06 -0800
Subject: [Tracker-discuss] [issue76] Want just any user to be able to
	edit nosy list?
In-Reply-To: 
References: <1168895228.43.0.0383309268963.issue76@psf.upfronthosting.co.za>
	<87tzys2d9q.fsf@uterus.efod.se>

Message-ID: 

On 1/15/07, Paul Dubois  wrote:
> Actually, an anonymous (unsigned-in user) cannot edit an issue. A signed-in
> user who did something malicious would leave an audit trail. So it isn't
> that big a worry.
>
> On the plus side, I often add someone else to an issue's nosy list when I
> add a comment that I want them to see or if I know they should be aware of
> the issue. So it's a feature.
>

Fair enough.  As long as it is not too hard to undo things then I can
live with this.

-Brett

>
> On 1/15/07, Erik Forsberg
>  wrote:
> >
> >
> > Erik Forsberg added the comment:
> >
> > "Brett C."  writes:
> >
> > > worry of random users coming along and editing the list in a malicious
> way
> > > (e.g., adding every person to single issue)?  I can see why the
> developers want
> > > to be able to add people to the nosy list, but I don't know if regular
> users
> > > need to.
> >
> > I for sure want to be able to add myself to the nosy list of issues
> > that I'm interested in, to get updates when something happens with a
> > bug I'm having trouble with in daily work.
> >
> > This is of course could be implemented with an finer access control
> > than the one currently in place - users without developer privileges
> > could be limited to add/remove only their own username from the nosy
> > list.
> >
> > \EF
> > --
> > Erik Forsberg                 http://efod.se
> > GPG/PGP Key: 1024D/0BAC89D9
> >
> > ----------
> > title: Restricting what non-submitter, non-developer can edit -> Want just
> any user to be able to edit nosy list?
> >
> > ______________________________________________________
> > Meta Tracker < metatracker at psf.upfronthosting.co.za>
> > 
> > ______________________________________________________
> > _______________________________________________
> > Tracker-discuss mailing list
> > Tracker-discuss at python.org
> > http://mail.python.org/mailman/listinfo/tracker-discuss
> >
>
>
> _______________________________________________
> Tracker-discuss mailing list
> Tracker-discuss at python.org
> http://mail.python.org/mailman/listinfo/tracker-discuss
>
>
>

From brett at python.org  Tue Jan 16 00:41:40 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 15 Jan 2007 15:41:40 -0800
Subject: [Tracker-discuss] [issue78] Make use of keywords more obvious
In-Reply-To: <45AC0455.2040403@v.loewis.de>
References: <1168896602.03.0.530314090964.issue78@psf.upfronthosting.co.za>
	<45AC0455.2040403@v.loewis.de>
Message-ID: 

On 1/15/07, "Martin v. L?wis"  wrote:
> Brett C. schrieb:
> > (probably shouldn't start with Py3K anyway but make it a
> > version number or something).
>
> We decided against that: the version field indicates past versions
> (where the error occurred in), not future versions. The py3k tag
> indicates a future version.
>
> We need the tag to be able to filter/filter-out py3k issues.
>

Ah, OK.  I figured I was forgetting something.

-Brett

> Regards,
> Martin
> _______________________________________________
> Tracker-discuss mailing list
> Tracker-discuss at python.org
> http://mail.python.org/mailman/listinfo/tracker-discuss
>

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 05:16:03 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Tue, 16 Jan 2007 04:16:03 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
Message-ID: <1168920963.16.0.653713022206.issue62@psf.upfronthosting.co.za>

Paul Dubois  added the comment:

I may have this all wrong but on the meta tracker somehow an extra
[Tracker-discuss] is getting into the subject line where it doesn't belong,
unless we have modified mailgw.py in the roundup runtime. I think Roundup is
prepared natively to handle Re: [issue76]. At least that is what the documents
say. The code is re-hard for a scientist to read.

Here's the error msg again:

Valid class names are: file, issue, keyword, msg, priority, query, status, user
Subject was: "Re: [Tracker-discuss] [issue76] Want just any user to be able to
edit nosy list?"

----------
status: resolved -> chatting

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 05:16:48 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Tue, 16 Jan 2007 04:16:48 -0000
Subject: [Tracker-discuss] [issue68] email addresses showing
Message-ID: <1168921008.05.0.614062134578.issue68@psf.upfronthosting.co.za>

Paul Dubois  added the comment:

On this one I'm waiting to see if it gets fixed when roundup is restarted.

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 05:24:03 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Tue, 16 Jan 2007 04:24:03 -0000
Subject: [Tracker-discuss] [issue80] Need controls for stopping the tracker
	for maintenance.
Message-ID: <1168921443.88.0.965287333781.issue80@psf.upfronthosting.co.za>

New submission from Paul Dubois :

To maintain the tracker we will need to be able to start and stop it ourselves.

When we stop it, we need to have something appear saying the tracker is down for
maintenance.

We also need to be able to start roundup on a side port so that we can test it
before making it live again. The number of users being so large, we must assume
the tracker is always being modifed when it is available. Ideally, the procedure
would ensure that the main tracker was halted, then halt the alternative when it
was time to switch back. 

This is possibly overkill but I don't think so. I've had times where I exported
a tracker to a test site, everything worked, but when I put it in the 'real'
tracker something failed. 

At the least, we need the ability to stop/start (a la server-ctl) *ourselves*
and to have something nice appear when things are stopped.

----------
messages: 378
nosy: brettcannon, dubois, forsberg, izak, roche, stefan
priority: urgent
status: unread
title: Need controls for stopping the tracker for maintenance.

______________________________________________________
Meta Tracker 

______________________________________________________

From roche at upfrontsystems.co.za  Tue Jan 16 06:40:21 2007
From: roche at upfrontsystems.co.za (=?ISO-8859-1?Q?Roch=E9?= Compaan)
Date: Tue, 16 Jan 2007 07:40:21 +0200
Subject: [Tracker-discuss] Fwd: Failed issue tracker submission
In-Reply-To: 
References: <20070115215417.D203678065@psf.localdomain>

Message-ID: <1168926022.6709.54.camel@kwaaitjie>

On Mon, 2007-01-15 at 13:58 -0800, Paul Dubois wrote:
> I got the error Martin got. I don't understand this, in all the years
> we've used Roundup on my project this has never happened.
> 
> The Re: [Tracker-discuss] is supposed to be stripped. The hypen in the
> tracker name maybe? 

This has always failed for me. Roundup matches the first piece of text
between brackets on the subject line as the class name. The regular
expression should confirm this.

-- 
Roch? Compaan
Upfront Systems                   http://www.upfrontsystems.co.za

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 07:29:16 2007
From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Martin_v._L=C3=B6wis?=)
Date: Tue, 16 Jan 2007 06:29:16 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
In-Reply-To: <1168920963.16.0.653713022206.issue62@psf.upfronthosting.co.za>
Message-ID: <45AC70B8.5040508@v.loewis.de>

Martin v. L?wis added the comment:

Paul Dubois schrieb:
> I may have this all wrong but on the meta tracker somehow an extra
> [Tracker-discuss] is getting into the subject line where it doesn't belong,
> unless we have modified mailgw.py in the roundup runtime. 

This extra header is getting into the subject because all messages are
also posted to tracker-discuss at python.org now, and mailman inserts that
as the header.

It would be possible to fix this in roundup: it would need to look at
*all* [foo] blocks, not just the first one, and be happy if exactly one
of them mentions a class name.

For the status quo, there seem to be three alternatives:
1 don't add a mailman mailing list to the normal recipients of new
  issues. This would be unfortunate for the Python tracker, as we
  used to send all new issues to python-bugs-list at python.org, and
  would like to continue to do so.
2 educate users not to follow up by email
3 educate users to edit the subject when following up by email

While 3 may work for some users, I guess most would move to 2 over
time (i.e. after they forgot to edit the headers a few times).

Regards,
Martin

______________________________________________________
Meta Tracker 

______________________________________________________

From forsberg at efod.se  Tue Jan 16 07:54:57 2007
From: forsberg at efod.se (Erik Forsberg)
Date: Tue, 16 Jan 2007 07:54:57 +0100
Subject: [Tracker-discuss] Managing the tracker metadata options
In-Reply-To: 
	(Brett Cannon's message of "Mon, 15 Jan 2007 13:39:33 -0800")
References: 
Message-ID: <87ejpv31ha.fsf@uterus.efod.se>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"Brett Cannon"  writes:

> I just wanted to find out who is going to be able to make users a
> Developer?  Or how about removing keywords, adding new versions, or
> editing the components list?

In addition to the 'User' and 'Developer' roles, there is also the
'Coordinator' role, which, if I read the permissions part of the
schema correctly, can edit for example keywords, versions, etc. 

Stefan did most of the schema design, so he'll have to correct me if
I'm wrong.

When it comes to being able to add roles to user, I'm not entirely
sure how it works right now. We need to investigate that, but I think
it's a good idea to make it possible for Coordinators to add the
Developer role to people that are only User. 

> Is all of that going to need to go through you guys (Erik, Stefan, and
> Paul)? 

I hope not :-).

> I assume that is all schema stuff and thus you guys are the
> only ones who can play with that.

Nope, most of it is not schema, but instances of classes, editable
through the web given the right permissions. Permissions coding (i.e.,
"Coordinator should be able to do that and that") is schema stuff,
though, but that will probably not need change very often. 

\EF
- -- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iD8DBQFFrHa/rJurFAusidkRAkqmAJwOtDtxbVGf5JKJMsJ62+Q77Agy1ACcD1jJ
KCyUfPM2gReNcSAv9pHnD0M=
=nQOo
-----END PGP SIGNATURE-----

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 07:55:42 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Tue, 16 Jan 2007 06:55:42 -0000
Subject: [Tracker-discuss] [issue68] email addresses showing
In-Reply-To: <1168921008.05.0.614062134578.issue68@psf.upfronthosting.co.za>
	(Paul Dubois's message of "Tue, 16 Jan 2007 04:16:48 -0000")
Message-ID: <87ac0j31g3.fsf@uterus.efod.se>

Erik Forsberg added the comment:

Paul Dubois  writes:

> On this one I'm waiting to see if it gets fixed when roundup is restarted.

Well, go ahead and restart. 'sudo /etc/init.d/roundup restart'

\EF
-- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9

______________________________________________________
Meta Tracker 

______________________________________________________

From forsberg at efod.se  Tue Jan 16 07:57:26 2007
From: forsberg at efod.se (Erik Forsberg)
Date: Tue, 16 Jan 2007 07:57:26 +0100
Subject: [Tracker-discuss] Fwd: Failed issue tracker submission
In-Reply-To: <1168926022.6709.54.camel@kwaaitjie> (=?iso-8859-1?Q?Roch=E9?=
	Compaan's message of "Tue, 16 Jan 2007 07:40:21 +0200")
References: <20070115215417.D203678065@psf.localdomain>

	<1168926022.6709.54.camel@kwaaitjie>
Message-ID: <8764b731d5.fsf@uterus.efod.se>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Roch? Compaan  writes:

> On Mon, 2007-01-15 at 13:58 -0800, Paul Dubois wrote:
>> I got the error Martin got. I don't understand this, in all the years
>> we've used Roundup on my project this has never happened.
>> 
>> The Re: [Tracker-discuss] is supposed to be stripped. The hypen in the
>> tracker name maybe? 
>
> This has always failed for me. Roundup matches the first piece of text
> between brackets on the subject line as the class name. The regular
> expression should confirm this.

Richard confirmed that this is the case on the roundup user list a few
hours ago.

I'll see if I can figure out some solution. It is going to be a
problem otherwise. 

\EF
- -- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iD8DBQFFrHdWrJurFAusidkRAhOMAKCnXim1BL+N8Czm/oyVndSy7Mh/ygCguxjQ
OhfFh0TKnfvPabd/PWzh0xI=
=0i76
-----END PGP SIGNATURE-----

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 08:01:31 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Tue, 16 Jan 2007 07:01:31 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
In-Reply-To: <45AC70B8.5040508@v.loewis.de> (Martin v. Löwis's message of "Tue, 16 Jan 2007 06:29:16 -0000")
Message-ID: <871wlv316d.fsf@uterus.efod.se>

Erik Forsberg added the comment:

"Martin v. L?wis"  writes:

> This extra header is getting into the subject because all messages are
> also posted to tracker-discuss at python.org now, and mailman inserts that
> as the header.
>
> It would be possible to fix this in roundup: it would need to look at
> *all* [foo] blocks, not just the first one, and be happy if exactly one
> of them mentions a class name.

Indeed.

I agree that this is a problem, and I think it needs to be fixed
before we go live, or the usefulness of the e-mail interface will be
much lower. 

> For the status quo, there seem to be three alternatives:
> 1 don't add a mailman mailing list to the normal recipients of new
>   issues. This would be unfortunate for the Python tracker, as we
>   used to send all new issues to python-bugs-list at python.org, and
>   would like to continue to do so.
> 2 educate users not to follow up by email
> 3 educate users to edit the subject when following up by email

There is a fourth alternative - modify the lists to not add [listname]
on the subject line. I guess that is not an alternative? 

\EF
-- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9

______________________________________________________
Meta Tracker 

______________________________________________________

From seefeld at sympatico.ca  Tue Jan 16 07:56:36 2007
From: seefeld at sympatico.ca (Stefan Seefeld)
Date: Tue, 16 Jan 2007 01:56:36 -0500
Subject: [Tracker-discuss] Managing the tracker metadata options
In-Reply-To: <87ejpv31ha.fsf@uterus.efod.se>
References: 
	<87ejpv31ha.fsf@uterus.efod.se>
Message-ID: <45AC7724.3040104@sympatico.ca>

Erik Forsberg wrote:
> "Brett Cannon"  writes:
> 
>>> I just wanted to find out who is going to be able to make users a
>>> Developer?  Or how about removing keywords, adding new versions, or
>>> editing the components list?
> 
> In addition to the 'User' and 'Developer' roles, there is also the
> 'Coordinator' role, which, if I read the permissions part of the
> schema correctly, can edit for example keywords, versions, etc. 
> 
> Stefan did most of the schema design, so he'll have to correct me if
> I'm wrong.

You are correct, the intend was to let coordinators make additions
to the various enumerations (such as keywords, types, versions, etc.).

> When it comes to being able to add roles to user, I'm not entirely
> sure how it works right now. We need to investigate that, but I think
> it's a good idea to make it possible for Coordinators to add the
> Developer role to people that are only User. 

I agree. And probably also to assign the 'Coordinator' role to others.
(Who else would be able to do that ? :-) )

Regards,
		Stefan

-- 

      ...ich hab' noch einen Koffer in Berlin...

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 08:05:25 2007
From: metatracker at psf.upfronthosting.co.za (Stefan Seefeld)
Date: Tue, 16 Jan 2007 07:05:25 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
In-Reply-To: <871wlv316d.fsf@uterus.efod.se>
Message-ID: <45AC77CB.9020500@sympatico.ca>

Stefan Seefeld added the comment:

Erik Forsberg wrote:
> Erik Forsberg added the comment:
> 
> "Martin v. L?wis"  writes:
> 
>> This extra header is getting into the subject because all messages are
>> also posted to tracker-discuss at python.org now, and mailman inserts that
>> as the header.
>>
>> It would be possible to fix this in roundup: it would need to look at
>> *all* [foo] blocks, not just the first one, and be happy if exactly one
>> of them mentions a class name.
> 
> Indeed.
> 
> I agree that this is a problem, and I think it needs to be fixed
> before we go live, or the usefulness of the e-mail interface will be
> much lower. 

We encountered this problem in our company-internal trackers. We have
a fix, though I thought it also went into the roundup mainline. May be
I'm mistaken. I'll look for it and send it, when I get a chance. (Will
talk to our admin tomorrow morning.)

-- 

      ...ich hab' noch einen Koffer in Berlin...

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 08:06:42 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Tue, 16 Jan 2007 07:06:42 -0000
Subject: [Tracker-discuss] [issue80] Need controls for stopping the tracker
	for maintenance.
Message-ID: <1168931202.7.0.450271859579.issue80@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

We are able to start/stop the tracker - 'sudo /etc/init.d/roundup
restart' does the trick. 

>From what I can see in the exports from sourceforge, the amount of
changes per minute is not that high, so perhaps some time of downtime
for maintenance/upgrade now and then is acceptable, making the need
for "side trackers" etc. less urgent?

----------
status: unread -> chatting

______________________________________________________
Meta Tracker 

______________________________________________________

From pfdubois at gmail.com  Tue Jan 16 09:03:38 2007
From: pfdubois at gmail.com (Paul Dubois)
Date: Tue, 16 Jan 2007 00:03:38 -0800
Subject: [Tracker-discuss] Managing the tracker metadata options
In-Reply-To: <45AC7724.3040104@sympatico.ca>
References: 
	<87ejpv31ha.fsf@uterus.efod.se> <45AC7724.3040104@sympatico.ca>
Message-ID: 

The normal way one gets a new role is that admin (or to be precise, a user
with edit permissions on class User) goes to their user record and edits the
field 'Roles' to add to the comma-delimited list of roles.

There is no way for the user to do this themselves nor should there be
because gaining a role gains permissions.

This is, for a normal tracker, no big deal and doing things like this was
one of the things I expected to do as a member of the Gang of Four.

However, the inability to get to a user record quickly at the moment would
make this a tough thing to do very much of.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20070116/13ee341b/attachment.htm 

From seefeld at sympatico.ca  Tue Jan 16 09:21:41 2007
From: seefeld at sympatico.ca (Stefan Seefeld)
Date: Tue, 16 Jan 2007 03:21:41 -0500
Subject: [Tracker-discuss] Managing the tracker metadata options
In-Reply-To: 
References: 	
	<87ejpv31ha.fsf@uterus.efod.se> <45AC7724.3040104@sympatico.ca>

Message-ID: <45AC8B15.2070803@sympatico.ca>

Paul Dubois wrote:
> The normal way one gets a new role is that admin (or to be precise, a
> user with edit permissions on class User) goes to their user record and
> edits the field 'Roles' to add to the comma-delimited list of roles.
> 
> There is no way for the user to do this themselves nor should there be
> because gaining a role gains permissions.
> 
> This is, for a normal tracker, no big deal and doing things like this
> was one of the things I expected to do as a member of the Gang of Four.
> 
> However, the inability to get to a user record quickly at the moment
> would make this a tough thing to do very much of.
> 

I'm not sure in what sense you use the term 'normal' here. I certainly
don't want admins to be required to play any role in the process of
adding users, or giving new previleges to them.

Regards,
		Stefan

-- 

      ...ich hab' noch einen Koffer in Berlin...

From pfdubois at gmail.com  Tue Jan 16 16:41:29 2007
From: pfdubois at gmail.com (Paul Dubois)
Date: Tue, 16 Jan 2007 07:41:29 -0800
Subject: [Tracker-discuss] Managing the tracker metadata options
In-Reply-To: <45AC8B15.2070803@sympatico.ca>
References: 
	<87ejpv31ha.fsf@uterus.efod.se> <45AC7724.3040104@sympatico.ca>

	<45AC8B15.2070803@sympatico.ca>
Message-ID: 

To give someone a role, you have to log in as admin in Roundup. A user can
register themselves, but (obviously) can't give themselves privileges
(roles). I didn't mean the system administrators.

On 1/16/07, Stefan Seefeld  wrote:
>
> Paul Dubois wrote:
> > The normal way one gets a new role is that admin (or to be precise, a
> > user with edit permissions on class User) goes to their user record and
> > edits the field 'Roles' to add to the comma-delimited list of roles.
> >
> > There is no way for the user to do this themselves nor should there be
> > because gaining a role gains permissions.
> >
> > This is, for a normal tracker, no big deal and doing things like this
> > was one of the things I expected to do as a member of the Gang of Four.
> >
> > However, the inability to get to a user record quickly at the moment
> > would make this a tough thing to do very much of.
> >
>
> I'm not sure in what sense you use the term 'normal' here. I certainly
> don't want admins to be required to play any role in the process of
> adding users, or giving new previleges to them.
>
> Regards,
>                 Stefan
>
> --
>
>       ...ich hab' noch einen Koffer in Berlin...
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20070116/314ba38d/attachment.html 

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 17:33:56 2007
From: metatracker at psf.upfronthosting.co.za (Brett C.)
Date: Tue, 16 Jan 2007 16:33:56 -0000
Subject: [Tracker-discuss] [issue80] Need controls for stopping the tracker
	for maintenance.
In-Reply-To: <1168931202.7.0.450271859579.issue80@psf.upfronthosting.co.za>
Message-ID: 

Brett C. added the comment:

With the amount of downtime that SF already has it is not a huge issue
if we just have a "down for maintenance" page up if it is going to be
mroe than just a couple of minutes.

On 1/15/07, Erik Forsberg  wrote:
>
> Erik Forsberg added the comment:
>
> We are able to start/stop the tracker - 'sudo /etc/init.d/roundup
> restart' does the trick.
>
> From what I can see in the exports from sourceforge, the amount of
> changes per minute is not that high, so perhaps some time of downtime
> for maintenance/upgrade now and then is acceptable, making the need
> for "side trackers" etc. less urgent?
>
> ----------
> status: unread -> chatting
>
> ______________________________________________________
> Meta Tracker 
> 
> ______________________________________________________
>

______________________________________________________
Meta Tracker 

______________________________________________________

From brett at python.org  Tue Jan 16 17:35:56 2007
From: brett at python.org (Brett Cannon)
Date: Tue, 16 Jan 2007 08:35:56 -0800
Subject: [Tracker-discuss] Managing the tracker metadata options
In-Reply-To: <45AC7724.3040104@sympatico.ca>
References: 
	<87ejpv31ha.fsf@uterus.efod.se> <45AC7724.3040104@sympatico.ca>
Message-ID: 

On 1/15/07, Stefan Seefeld  wrote:
> Erik Forsberg wrote:
> > "Brett Cannon"  writes:
> >
> >>> I just wanted to find out who is going to be able to make users a
> >>> Developer?  Or how about removing keywords, adding new versions, or
> >>> editing the components list?
> >
> > In addition to the 'User' and 'Developer' roles, there is also the
> > 'Coordinator' role, which, if I read the permissions part of the
> > schema correctly, can edit for example keywords, versions, etc.
> >
> > Stefan did most of the schema design, so he'll have to correct me if
> > I'm wrong.
>
> You are correct, the intend was to let coordinators make additions
> to the various enumerations (such as keywords, types, versions, etc.).
>
> > When it comes to being able to add roles to user, I'm not entirely
> > sure how it works right now. We need to investigate that, but I think
> > it's a good idea to make it possible for Coordinators to add the
> > Developer role to people that are only User.
>
> I agree. And probably also to assign the 'Coordinator' role to others.
> (Who else would be able to do that ? :-) )
>

Whomever is currently an admin on the SF tracker right now plus maybe
some other people (e.g., me  =).

-Brett

From pfdubois at gmail.com  Tue Jan 16 20:29:37 2007
From: pfdubois at gmail.com (Paul Dubois)
Date: Tue, 16 Jan 2007 11:29:37 -0800
Subject: [Tracker-discuss] [issue80] Need controls for stopping the
	tracker for maintenance.
In-Reply-To: <1168931202.7.0.450271859579.issue80@psf.upfronthosting.co.za>
References: <1168931202.7.0.450271859579.issue80@psf.upfronthosting.co.za>
Message-ID: 

I think I will start a small document to capture what is in your brains
about how we do our business as roundup maintainers. Perhaps I am not the
only one who does not know Erik's secrets like the one below -- but even if
I am, those who come after us need a place to learn the drill.

On 1/15/07, Erik Forsberg  wrote:
>
>
> Erik Forsberg added the comment:
>
> We are able to start/stop the tracker - 'sudo /etc/init.d/roundup
> restart' does the trick.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20070116/adcafe7a/attachment.html 

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 21:24:08 2007
From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Martin_v._L=C3=B6wis?=)
Date: Tue, 16 Jan 2007 20:24:08 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
In-Reply-To: <871wlv316d.fsf@uterus.efod.se>
Message-ID: <45AD3465.4020200@v.loewis.de>

Martin v. L?wis added the comment:

Erik Forsberg schrieb:
> There is a fourth alternative - modify the lists to not add [listname]
> on the subject line. I guess that is not an alternative? 

I'm not sure: if it would be possible technically, I wouldn't mind doing
it. However, I don't have access to the mailman configuration of that
particular list; Barry Warsaw should be able to help if necessary.

Regards,
Martin

______________________________________________________
Meta Tracker 

______________________________________________________

From barry at python.org  Tue Jan 16 21:33:05 2007
From: barry at python.org (Barry Warsaw)
Date: Tue, 16 Jan 2007 15:33:05 -0500
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
In-Reply-To: <45AD3465.4020200@v.loewis.de>
References: <45AD3465.4020200@v.loewis.de>
Message-ID: 

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Jan 16, 2007, at 3:24 PM, Martin v. L?wis wrote:

> Erik Forsberg schrieb:
>> There is a fourth alternative - modify the lists to not add  
>> [listname]
>> on the subject line. I guess that is not an alternative?
>
> I'm not sure: if it would be possible technically, I wouldn't mind  
> doing
> it. However, I don't have access to the mailman configuration of that
> particular list; Barry Warsaw should be able to help if necessary.

Tell me which list on python.org you want to do this and I (or other  
postmaster at python.org-ers) can easily make this happen.

- -Barry

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iQCVAwUBRa02gXEjvBPtnXfVAQIX+AP/ZAbD941+MbbVIMhqEBig4MshmfcS51G0
Ro6oo22mjBhJja8bZL1FJgZKVk/JBmoP3MdYN5P4ifh3aAo958XnQQf3XncEMmTn
6eiNnJoBGAkp5GZ1ZpDMXBZGcYDl7Bu891xZPL0Cgthi59zlLBhH60MFdkJrQHhI
FAIVbx6wLcE=
=CdA/
-----END PGP SIGNATURE-----

From metatracker at psf.upfronthosting.co.za  Tue Jan 16 21:38:00 2007
From: metatracker at psf.upfronthosting.co.za (Georg Brandl)
Date: Tue, 16 Jan 2007 20:38:00 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
In-Reply-To: <1168920963.16.0.653713022206.issue62@psf.upfronthosting.co.za>
Message-ID: <45AD37A0.2070807@gmx.net>

Georg Brandl added the comment:

Paul Dubois schrieb:
> Paul Dubois  added the comment:
> 
> I may have this all wrong but on the meta tracker somehow an extra
> [Tracker-discuss] is getting into the subject line where it doesn't belong,
> unless we have modified mailgw.py in the roundup runtime. I think Roundup is
> prepared natively to handle Re: [issue76]. At least that is what the documents
> say. The code is re-hard for a scientist to read.

While we're at it, perhaps you'll want to accept other prefixes as well. From
the top of my head I can think of "AW:" which is used by a German MS Outlook.
(Okay, probably there's no Outlook user among Python developers, but just a
hint.)

Georg

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Wed Jan 17 00:57:06 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Tue, 16 Jan 2007 23:57:06 -0000
Subject: [Tracker-discuss] [issue71] py3k bug set on all newly created issues
Message-ID: <1168991826.78.0.447984557364.issue71@psf.upfronthosting.co.za>

Paul Dubois added the comment:

I see two alternatives to fixing this bug (and I do think it is a bug):

a. Have a 'not selected' on the drop down list.

b. Change it to be like the nosy list, that is, a comma-delimited list. One can
check for valid entries in an auditor. 

I do not like relying on highlights in a list to tell me which keywords are on
an issue. A nice text list is easy to understand, and it doesn't have the
'default' problem, since the default is blank.

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Wed Jan 17 00:59:29 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Tue, 16 Jan 2007 23:59:29 -0000
Subject: [Tracker-discuss] [issue71] py3k bug set on all newly created issues
Message-ID: <1168991969.19.0.357647862637.issue71@psf.upfronthosting.co.za>

Paul Dubois added the comment:

To explain better, I'm thinking the original classic scheme was just fine. Call
it keywords if you want, but there was nothing wrong with the original. You can
control the ability to create keywords in page.html, restricting it to
Coordinators or Developers as desired.

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Wed Jan 17 08:25:27 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Wed, 17 Jan 2007 07:25:27 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
In-Reply-To: <45AD37A0.2070807@gmx.net>
Message-ID: <200701170825.23351.forsberg@efod.se>

Erik Forsberg added the comment:

> While we're at it, perhaps you'll want to accept other prefixes as well.
> From the top of my head I can think of "AW:" which is used by a German MS
> Outlook. (Okay, probably there's no Outlook user among Python developers,
> but just a hint.)

"AW:" is already on the list of accepted prefixes, among with a whole bunch of 
other similar strings. 

\EF
-- 
http://efod.se/

______________________________________________________
Meta Tracker 

______________________________________________________

From izak at upfrontsystems.co.za  Wed Jan 17 09:09:15 2007
From: izak at upfrontsystems.co.za (Izak Burger)
Date: Wed, 17 Jan 2007 10:09:15 +0200
Subject: [Tracker-discuss] [issue80] Need controls for stopping the
 tracker for maintenance.
In-Reply-To: 
References: 
Message-ID: <45ADD9AB.3060602@upfrontsystems.co.za>

Hi all,

While catching up on some mail, thought I'd reply on this issue:

Brett C. wrote:
> With the amount of downtime that SF already has it is not a huge issue
> if we just have a "down for maintenance" page up if it is going to be
> mroe than just a couple of minutes.

We have apache running in front of roundup-server, with apache serving 
the requests for static content while passing everything else to 
roundup-server.  If you want to take the server down for a while, you 
could do this by changing the apache config.

This being a debian host, the virtual hosts are set up in 
/etc/apache2/sites-available and then symlinked into 
/etc/apache2/sites-enabled.  I think a quick solution to this might be 
to keep two configurations in /etc/apache2/sites-available, one that 
forwards to roundup-server (as it does at the moment) and another that 
simply serves a page saying that it is down for maintenance.  When you 
go down for maintenance, you would then:

1. Symlink the "other" config in /etc/apache2/sites-available to 
/etc/apache2/sites-enabled

2. Reload apache config: /etc/init.d/apache2 reload

3. You can still reach roundup-server by port-forwarding localhost:8080 
over ssh.

4. When you're done, repeat steps 1 and 2 to symlink the "live" config 
into place again.

You need root privileges to do the above of course, but the guys with 
sudo rights should have everything they need to do the above.

regards,
Izak

From metatracker at psf.upfronthosting.co.za  Wed Jan 17 19:44:40 2007
From: metatracker at psf.upfronthosting.co.za (Brett C.)
Date: Wed, 17 Jan 2007 18:44:40 -0000
Subject: [Tracker-discuss] [issue71] py3k bug set on all newly created issues
Message-ID: <1169059480.27.0.819954164546.issue71@psf.upfronthosting.co.za>

Brett C. added the comment:

I personally like the multi-selection list since you then know exactly what
keywords are available.  The problem is that Py3K is the only option available
so the use of the keyword list is non-obvious.

We either add a "(none)" option as Paul suggested, we just add more than one
keyword at launch, or something.

The issue I have with doing a list is that I don't want to have to remember what
all the keywords are.  I assume the auditor will kick me back to the issue with
everything filled in sans the misspelled keyword, but that is still a pain.

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Thu Jan 18 09:07:31 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Thu, 18 Jan 2007 08:07:31 -0000
Subject: [Tracker-discuss] [issue68] email addresses showing
Message-ID: <1169107651.79.0.420819418659.issue68@psf.upfronthosting.co.za>

Paul Dubois added the comment:

This seems to be fixed.

----------
status: chatting -> resolved

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Thu Jan 18 09:10:39 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Thu, 18 Jan 2007 08:10:39 -0000
Subject: [Tracker-discuss] [issue76] Want just any user to be able to edit
	nosy list?
Message-ID: <1169107839.3.0.162171886468.issue76@psf.upfronthosting.co.za>

Paul Dubois added the comment:

To sum up this discussion, I think one can say there is a good reason why it is
the way it is, and that if experience proves this isn't right we can change it
later. Lots of existing experience says it is right. Again note that only logged
in users can change issues and they leave an audit trail. 

It should not hold up the release so I'm moving it to feature.

----------
priority: bug -> feature

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Thu Jan 18 09:14:58 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Thu, 18 Jan 2007 08:14:58 -0000
Subject: [Tracker-discuss] [issue19] Add search interface for users
Message-ID: <1169108098.52.0.534101126875.issue19@psf.upfronthosting.co.za>

Paul Dubois added the comment:

It is imperative that we add a search option or the Coordinators will have a
terrible time. I'll try to do it.

----------
assignedto:  -> dubois
nosy: +dubois
priority: feature -> bug
status: unread -> chatting

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Thu Jan 18 20:02:54 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Thu, 18 Jan 2007 19:02:54 -0000
Subject: [Tracker-discuss] [issue77] Assigned To list not sorted
Message-ID: <1169146974.54.0.213502161303.issue77@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

Fixed.

----------
assignedto:  -> forsberg
nosy: +forsberg
status: unread -> resolved

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Sat Jan 20 18:45:43 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Sat, 20 Jan 2007 17:45:43 -0000
Subject: [Tracker-discuss] [issue19] Add search interface for users
Message-ID: <1169315143.18.0.372018267492.issue19@psf.upfronthosting.co.za>

Paul Dubois added the comment:

We need some facility for the Coordinators to get to a specific user's record in
order to edit it. I have ideas on how to do this but my TAL isn't strong enough.
I've reread the document but there is something I'm not getting because I can't
make any of them work.

Here are some ways to solve the problem for the moment. If anyone knows how to
do this please tell me and I'll do the work.

1. As in 'Show issue' make a similar section in page.html so that Coordinators
can enter a username in a "Show user" box and get to that user's record. That
will alleviate most of the trouble.

2. Change the User List link so that it brings up something like the search
stuff we get when clicking on the (list) link for the Nosy list. This would be
ideal. Or, change the User List link so it goes to a page with more generic
search form.

----------
assignedto: dubois -> 
nosy: +brettcannon, richard, stefan

______________________________________________________
Meta Tracker 

______________________________________________________

From martin at v.loewis.de  Sat Jan 20 19:58:21 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Sat, 20 Jan 2007 19:58:21 +0100
Subject: [Tracker-discuss] [issue19] Add search interface for users
In-Reply-To: <1169315143.18.0.372018267492.issue19@psf.upfronthosting.co.za>
References: <1169315143.18.0.372018267492.issue19@psf.upfronthosting.co.za>
Message-ID: <45B2664D.3050208@v.loewis.de>

Paul Dubois schrieb:
> 1. As in 'Show issue' make a similar section in page.html so that Coordinators
> can enter a username in a "Show user" box and get to that user's record. That
> will alleviate most of the trouble.

Not sure what you've been trying; to find a user by username, you should
use an URL like

http://psf.upfronthosting.co.za/roundup/tracker/user?username=loewis&@action=search

I don't think you need any TAL for that if you use

http://psf.upfronthosting.co.za/roundup/tracker/user

as the action for the form; setting action="user" (ie. with a relative
URL) should work as well.

Regards,
Martin

From metatracker at psf.upfronthosting.co.za  Mon Jan 22 19:28:09 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Mon, 22 Jan 2007 18:28:09 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
Message-ID: <1169490489.31.0.322782691144.issue73@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

There were indeed something fishy going on with the issue numbering - due to a
bug in the importer, the number for the next issue to be created was not
correctly set.

This has now been corrected, with the result that the next issue created has a
number somewhere around 1619061, which is of course ridicoulusly high. The
lowest number imported is 207608, and just as Martin points out, there's a
really large bunch of issue numbers available between 1000 and 207608. If 10
issues are added each day, it'll take about 56 years to get there. 

I'm beginning to think that starting at 1000 is a very good idea, so unless
there's any strong opinions against it, that's what the importer will instruct
roundup to do.

----------
assignedto:  -> forsberg
nosy: +forsberg

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Mon Jan 22 20:23:13 2007
From: metatracker at psf.upfronthosting.co.za (Paul Dubois)
Date: Mon, 22 Jan 2007 19:23:13 -0000
Subject: [Tracker-discuss] [issue19] Add search interface for users
Message-ID: <1169493793.92.0.0423095106795.issue19@psf.upfronthosting.co.za>

Paul Dubois added the comment:

Despite an email from Martin that meant to help me but which I really didn't
understand, I cannot do this job.

If someone else will do this job I'll be the 'fix user issues' person (such as
role changing, etc.) once we get going. I think the section of the customizing
document on web interfaces just assumes too much prior understanding for the
likes of me.

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Wed Jan 24 07:35:04 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Wed, 24 Jan 2007 06:35:04 -0000
Subject: [Tracker-discuss] [issue19] Add search interface for users
Message-ID: <1169620504.68.0.575099666115.issue19@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

Fixed, although the layout might need some tweaks for better looks.

----------
assignedto:  -> forsberg
status: chatting -> resolved

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Wed Jan 24 20:07:58 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Wed, 24 Jan 2007 19:07:58 -0000
Subject: [Tracker-discuss] [issue71] py3k bug set on all newly created issues
Message-ID: <1169665678.82.0.0372451646709.issue71@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

I made the selectbox 3 items high instead of one, which I think makes the
problem less severe.

----------
assignedto:  -> forsberg
status: chatting -> resolved

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Sat Jan 27 18:26:03 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Sat, 27 Jan 2007 17:26:03 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
Message-ID: <1169918763.1.0.573153442135.issue62@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

FYI: I've been working on modifications on the roundup mail parser to make it
less picky. Currently waiting for implementation feedback from Richard.

----------
assignedto: stefan -> forsberg
nosy: +richard
status: chatting -> in-progress

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Sat Jan 27 18:33:09 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Sat, 27 Jan 2007 17:33:09 -0000
Subject: [Tracker-discuss] [issue73] Issue numbering
Message-ID: <1169919189.65.0.608313651622.issue73@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

No strong opinions have been heard, the importer will now configure roundup to
create its first issue with id 1000.

I've tested the behaviour when there's an id collision - there's a traceback,
which I think is OK. We'll have to deal with it when (if?) it happens.

----------
status: chatting -> resolved

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Sat Jan 27 19:42:45 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Sat, 27 Jan 2007 18:42:45 -0000
Subject: [Tracker-discuss] [issue56] Make bugs.python.org point to the
	roundup tracker
Message-ID: <1169923365.82.0.976853859058.issue56@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

I've added configuration for a test hostname, http://pybugs.efod.se and changed
roundup configuration for the test tracker. Seems to work nice. Very easy to
switch to http://bugs.python.org when the time comes.

----------
status: chatting -> in-progress

______________________________________________________
Meta Tracker 

______________________________________________________

From forsberg at efod.se  Sat Jan 27 20:05:46 2007
From: forsberg at efod.se (Erik Forsberg)
Date: Sat, 27 Jan 2007 20:05:46 +0100
Subject: [Tracker-discuss] SSL?
Message-ID: <87bqkk5lyt.fsf@uterus.efod.se>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

Should the tracker run SSL (perhaps only required for logins etc.),
and if so, is there a way to get a certificate that's not self-signed? 

Cheers,
\EF
- -- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iD8DBQFFu6KKrJurFAusidkRApZvAJ9obGfHX5Xh6tWpyZXLUWIE5wGeNwCgrEca
hBGVE6X/VsvyTU/KDCfy028=
=yGuF
-----END PGP SIGNATURE-----

From brett at python.org  Sat Jan 27 20:08:25 2007
From: brett at python.org (Brett Cannon)
Date: Sat, 27 Jan 2007 11:08:25 -0800
Subject: [Tracker-discuss] SSL?
In-Reply-To: <87bqkk5lyt.fsf@uterus.efod.se>
References: <87bqkk5lyt.fsf@uterus.efod.se>
Message-ID: 

On 1/27/07, Erik Forsberg  wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi!
>
> Should the tracker run SSL (perhaps only required for logins etc.),

Probably.

> and if so, is there a way to get a certificate that's not self-signed?
>

=)  Ah, the trick.  We can have the PSF pay for a cert from someone
like Thawte.  Trick is getting the purchase made.  I think Barry has
handled all of them for python.org, but he has been rather swamped and
has not updated pydotorg's cert so who knows how long it would be
until he can get a whole new one.

But yes, there should be a way.

-Brett

> Cheers,
> \EF
> - --
> Erik Forsberg                 http://efod.se
> GPG/PGP Key: 1024D/0BAC89D9
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.2.2 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.8+ 
>
> iD8DBQFFu6KKrJurFAusidkRApZvAJ9obGfHX5Xh6tWpyZXLUWIE5wGeNwCgrEca
> hBGVE6X/VsvyTU/KDCfy028=
> =yGuF
> -----END PGP SIGNATURE-----
> _______________________________________________
> Tracker-discuss mailing list
> Tracker-discuss at python.org
> http://mail.python.org/mailman/listinfo/tracker-discuss
>

From metatracker at psf.upfronthosting.co.za  Sat Jan 27 22:46:39 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Sat, 27 Jan 2007 21:46:39 -0000
Subject: [Tracker-discuss] [issue44] Set up python-dev tracker
In-Reply-To: <1168849811.35.0.843033803697.issue44@psf.upfronthosting.co.za> (Martin v. Löwis's message of "Mon, 15 Jan 2007 08:30:11 -0000")
Message-ID: <871wlg5eir.fsf@uterus.efod.se>

Erik Forsberg added the comment:

"Martin v. L?wis"  writes:

> Martin v. L?wis  added the comment:
>
>> * Can an old user (with sourceforge mail address) recover password? 
>
> I've confirmed that this is possible. The procedure I used requires
> two email roundtrips:
>
> 1. Follow "Lost your login?"
> 2. Fill in "Username:" field (second form) with SF id, hit
>    "Request password reset"
> 3. After reception of email 1 (Confirm reset of password
>    for Tracker), follow the link in this email. Do not
>    put any data into the form on that page.

I've modified the information shown to the user in this situation - it
no longer shows the input forms in this step, but instead just the
message in the green box. That will hopefully make the procedure less
confusing.

Cheers,
\EF
-- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Sun Jan 28 16:55:37 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Sun, 28 Jan 2007 15:55:37 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
In-Reply-To: <1169918763.1.0.573153442135.issue62@psf.upfronthosting.co.za>
	(Erik Forsberg's message of "Sat, 27 Jan 2007 17:26:03 -0000")
Message-ID: <87k5z7403s.fsf@uterus.efod.se>

Erik Forsberg added the comment:

Erik Forsberg  writes:

> Erik Forsberg added the comment:
>
> FYI: I've been working on modifications on the roundup mail parser
> to make it less picky. Currently waiting for implementation feedback
> from Richard.

This has now been commited both to roundup-HEAD and to the source code
tree used by our tracker(s). Let's see if this message gets through.. 

Cheers,
\EF
-- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Sun Jan 28 16:56:03 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Sun, 28 Jan 2007 15:56:03 -0000
Subject: [Tracker-discuss] [issue62] Can't follow-up per email
Message-ID: <1169999763.94.0.643941750958.issue62@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

Good. Considering this problem fixed.

----------
status: in-progress -> resolved

______________________________________________________
Meta Tracker 

______________________________________________________

From forsberg at efod.se  Mon Jan 29 20:05:43 2007
From: forsberg at efod.se (Erik Forsberg)
Date: Mon, 29 Jan 2007 20:05:43 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
Message-ID: <87bqkh3b7c.fsf@uterus.efod.se>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

With the list of showstoppers in the meta database reduced to two
issues, both of which are of the type which can't be closed until
we've gone into production, I think it's time to start thinking about
going live. 

Some thoughts;

* Do we need a public beta period to allow people not on this list to
  test the tracker?

  In my opinion, we should avoid this as I'm afraid it will to lots of
  issues that are seen as showstoppers by people. Better handle these
  problems as they come - after all, this tracker is supposed to be
  much better than the current tracker, so we should start using it
  ASAP.

* What kind of documentation needs to be written? And who volunteers
  to do it? :-)

* In what steps, and when, do we run the final conversion? The
  conversion process looks something like this:

  0. (Only in final conversion) Lock sf tracker down, if possible. 

  1. Get a fresh copy of the xml backup from sf.net. Takes 10-15 minutes.

  2. Transfer to psf. 5 minutes.

  3. Run conversion. Takes about an hour.

  4. Done. 

  Some hostname fiddling somewhere inbetween, or before, to get
  bugs.python.org to point to the right IP.

  I'm willing to do the conversion work. You python-dev people will
  have to come with some feedback on when and how to do it.

Regards,
\EF
- -- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iD8DBQFFvkWHrJurFAusidkRApw3AKDewp7H8Z+vvr3A4aaXJ+HgPZ7cuwCgvghs
WHrutANjnV4l+0oHAZTtLRc=
=7QM2
-----END PGP SIGNATURE-----

From amk at amk.ca  Mon Jan 29 20:42:21 2007
From: amk at amk.ca (A.M. Kuchling)
Date: Mon, 29 Jan 2007 14:42:21 -0500
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <87bqkh3b7c.fsf@uterus.efod.se>
References: <87bqkh3b7c.fsf@uterus.efod.se>
Message-ID: <20070129194221.GA16440@localhost.localdomain>

On Mon, Jan 29, 2007 at 08:05:43PM +0100, Erik Forsberg wrote:
> * Do we need a public beta period to allow people not on this list to
>   test the tracker?

Um, have I missed some announcements of the tracker status on
python-dev?  Searching for 'tracker' in the December and January
python-dev archive didn't turn anything up.  In the absence of such an
announcement, I doubt many python-dev readers are aware of the current
status of the tracker work, or even of its URL.

--amk

From brett at python.org  Mon Jan 29 20:49:24 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 29 Jan 2007 11:49:24 -0800
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <20070129194221.GA16440@localhost.localdomain>
References: <87bqkh3b7c.fsf@uterus.efod.se>
	<20070129194221.GA16440@localhost.localdomain>
Message-ID: 

On 1/29/07, A.M. Kuchling  wrote:
> On Mon, Jan 29, 2007 at 08:05:43PM +0100, Erik Forsberg wrote:
> > * Do we need a public beta period to allow people not on this list to
> >   test the tracker?
>
> Um, have I missed some announcements of the tracker status on
> python-dev?  Searching for 'tracker' in the December and January
> python-dev archive didn't turn anything up.  In the absence of such an
> announcement, I doubt many python-dev readers are aware of the current
> status of the tracker work, or even of its URL.
>

Go one month back to November and you will find the announcement of
tracker-discuss on python-dev along with some initial cross-posts.  I
honestly think most people on python-dev just don't care that much and
are just expecting to adapt.  Martin and Georg use the tracker the
most (I suspect) and they have both contributed input on all of this.

-Brett

From brett at python.org  Mon Jan 29 20:59:06 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 29 Jan 2007 11:59:06 -0800
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <87bqkh3b7c.fsf@uterus.efod.se>
References: <87bqkh3b7c.fsf@uterus.efod.se>
Message-ID: 

On 1/29/07, Erik Forsberg  wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi!
>
> With the list of showstoppers in the meta database reduced to two
> issues, both of which are of the type which can't be closed until
> we've gone into production, I think it's time to start thinking about
> going live.
>
> Some thoughts;
>
> * Do we need a public beta period to allow people not on this list to
>   test the tracker?
>
>   In my opinion, we should avoid this as I'm afraid it will to lots of
>   issues that are seen as showstoppers by people. Better handle these
>   problems as they come - after all, this tracker is supposed to be
>   much better than the current tracker, so we should start using it
>   ASAP.
>

I agree.  We will just get bogged down in people screaming for some
little feature they want.  If people want something they can post here
and we can work it out.  Plus we told people who were interested to
join this list when it was created.

At worst we can tell people they have until the launch to log any
complaints but that it must be hashed through this list and it cannot
hold up the launch.

> * What kind of documentation needs to be written? And who volunteers
>   to do it? :-)
>

Basic guidelines of how we expect new bugs and patches to be handled
in terms of how to set the various fields (both for people reporting
and developers).  We can initially write them in the python.org wiki
so we can all contribute.

> * In what steps, and when, do we run the final conversion? The
>   conversion process looks something like this:
>
>   0. (Only in final conversion) Lock sf tracker down, if possible.
>

I think Martin said this is doable.

>   1. Get a fresh copy of the xml backup from sf.net. Takes 10-15 minutes.
>
>   2. Transfer to psf. 5 minutes.
>
>   3. Run conversion. Takes about an hour.
>
>   4. Done.
>
>   Some hostname fiddling somewhere inbetween, or before, to get
>   bugs.python.org to point to the right IP.
>

Do we have a way to lock down the Roundup instance during the
conversion so people don't jump the gun?  Or do we just take the whole
thing down and not bring it back up until it's ready to accept new
issues?

And how about setting the developer bit on people's accounts?  Are we
doing that on a request basis or just scraping the SF page and letting
all of those accounts get set up?  Either way we need to doc
something; how to request the bit or recover your SF account as a
Roundup account, respectively.

>   I'm willing to do the conversion work. You python-dev people will
>   have to come with some feedback on when and how to do it.
>

When we feel we have the process down and the docs written I will ask
python-dev when they are up for doing the transfer (have to make sure
it doesn't happen around 2.5.1's release).

I'm all excited that this is actually going to happen!  Hopefully we
can have this done by PyCon so it can be said we all got this finished
in less than a year since the infrastructure committee was given the
task to do this.  But it is no big deal if we don't; I do not want to
rush this.

-Brett

From forsberg at efod.se  Mon Jan 29 21:29:03 2007
From: forsberg at efod.se (Erik Forsberg)
Date: Mon, 29 Jan 2007 21:29:03 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: 
	(Brett Cannon's message of "Mon, 29 Jan 2007 11:59:06 -0800")
References: <87bqkh3b7c.fsf@uterus.efod.se>

Message-ID: <877iv537cg.fsf@uterus.efod.se>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"Brett Cannon"  writes:

> Do we have a way to lock down the Roundup instance during the
> conversion so people don't jump the gun?  Or do we just take the whole
> thing down and not bring it back up until it's ready to accept new
> issues?

I'd say we should take the whole thing down, possibly replacing any
requests with a placeholer page telling people to take it easy. 

> And how about setting the developer bit on people's accounts?  Are we
> doing that on a request basis or just scraping the SF page and letting
> all of those accounts get set up?  Either way we need to doc
> something; how to request the bit or recover your SF account as a
> Roundup account, respectively.

The importer currently:

1) Sets up accounts for all accounts that has contributed to the sf
   tracker in any way, with the same username as the sf username. This
   includes accounts ranging from people who've made one bug report to
   people doing all kinds of work on a daily basis. All accounts get
   the 'User' role.

2) Assigns the 'Developer' role to people who have at least one
   bug/rfe/patch assigned to them in the sf tracker.

There is also a 'Coordinator' role who has permission to assign roles
(except the Administrator role) to other users. That should be a
handful of python developers, probably the same people that handles
the svn permissions today. 

All accounts are currently assigned a random password and need to use
the "Lost login" feature (Currently at
http://pybugs.efod.se/user?@template=forgotten) using their
sourceforge username. This will mail their sourceforge mail address
(@users.sourceforge.net), so it is important that people
have a correct address setup so that the password reset mail comes
through.

Coordinators can edit this mail address if the mail fails to deliver
for some people. Of course, users can also, once logged in, edit their
own mail adress to set it to something that doesn't involve
sourceforge.

New accounts can also be created, and requires e-mail address confirmation. 

> When we feel we have the process down and the docs written I will ask
> python-dev when they are up for doing the transfer (have to make sure
> it doesn't happen around 2.5.1's release).

Sounds like a good idea.

> I'm all excited that this is actually going to happen!  Hopefully we
> can have this done by PyCon so it can be said we all got this finished
> in less than a year since the infrastructure committee was given the
> task to do this.  But it is no big deal if we don't; I do not want to
> rush this.

23-25/2 - well, it doesn't sound completely undoable :-).

Cheers,
\EF
- -- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iD8DBQFFvlkOrJurFAusidkRAo3uAKCtkFm/3W1hR5PkFiCAr+7ReQZ59ACgzjB2
AP6owMgnBThtbMMij/CyKrE=
=51Za
-----END PGP SIGNATURE-----

From martin at v.loewis.de  Mon Jan 29 21:31:39 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Mon, 29 Jan 2007 21:31:39 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: 
References: <87bqkh3b7c.fsf@uterus.efod.se>

Message-ID: <45BE59AB.1030405@v.loewis.de>

Brett Cannon schrieb:
>> * In what steps, and when, do we run the final conversion? The
>>   conversion process looks something like this:
>>
>>   0. (Only in final conversion) Lock sf tracker down, if possible.
>>
> 
> I think Martin said this is doable.

Actually, I think I said the contrary. I can take the tracker from
the front page, but I can't lock it. So all links that point to
it will continue working.

I can also put a message on top of the submission page that the
further submissions will get ignored.

Regards,
Martin

From martin at v.loewis.de  Mon Jan 29 21:25:24 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Mon, 29 Jan 2007 21:25:24 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <87bqkh3b7c.fsf@uterus.efod.se>
References: <87bqkh3b7c.fsf@uterus.efod.se>
Message-ID: <45BE5834.3070508@v.loewis.de>

Erik Forsberg schrieb:
> * Do we need a public beta period to allow people not on this list to
>   test the tracker?
> 
>   In my opinion, we should avoid this as I'm afraid it will to lots of
>   issues that are seen as showstoppers by people. Better handle these
>   problems as they come - after all, this tracker is supposed to be
>   much better than the current tracker, so we should start using it
>   ASAP.

I think there should be a beta period. I understand you want to see
this done, but I still think people should be given a chance to object.
They don't to be given too much time (e.g. two weeks), and issues
can be dismissed as non-showstoppers easily. However, if real
showstoppers are detected, I'd like to know before.

As for showstoppers: it seems that the machine is down at
the moment (Jan 29, 20:07 GMT), no?

> * What kind of documentation needs to be written? And who volunteers
>   to do it? :-)

The following pages refer to SF currently:

http://docs.python.org/lib/reporting-bugs.html
http://docs.python.org/api/reporting-bugs.html
[this is the 2.5 documentation; SF is referenced in every version
 of the manual, both online and installed]
http://www.python.org/search/
[link to SF search page]
http://www.python.org/doc/faq/general/#how-do-i-submit-bug-reports-and-patches-for-python
http://www.python.org/doc/faq/it/general/#come-posso-segnalare-l-esistenza-di-bug-e-patch-per-python
http://www.python.org/doc/faq/es/general/#c-mo-env-o-reportes-de-error-y-parches-para-python
http://www.python.org/download/
[link to bug manager, patch manager]
http://www.python.org/dev/patches/
http://www.python.org/dev/peps/pep-0001/
[--- many more PEPs ---]
http://www.python.org/dev/
http://www.python.org/about/help/

Regards,
Martin

From forsberg at efod.se  Mon Jan 29 21:41:01 2007
From: forsberg at efod.se (Erik Forsberg)
Date: Mon, 29 Jan 2007 21:41:01 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <45BE5834.3070508@v.loewis.de> (Martin v. =?iso-8859-1?Q?L=F6?=
	=?iso-8859-1?Q?wis's?= message of
	"Mon, 29 Jan 2007 21:25:24 +0100")
References: <87bqkh3b7c.fsf@uterus.efod.se> <45BE5834.3070508@v.loewis.de>
Message-ID: <87wt351s82.fsf@uterus.efod.se>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"Martin v. L?wis"  writes:

> Erik Forsberg schrieb:
>> * Do we need a public beta period to allow people not on this list to
>>   test the tracker?
>> 
>>   In my opinion, we should avoid this as I'm afraid it will to lots of
>>   issues that are seen as showstoppers by people. Better handle these
>>   problems as they come - after all, this tracker is supposed to be
>>   much better than the current tracker, so we should start using it
>>   ASAP.
>
> I think there should be a beta period. I understand you want to see
> this done, but I still think people should be given a chance to object.
> They don't to be given too much time (e.g. two weeks), and issues
> can be dismissed as non-showstoppers easily. However, if real
> showstoppers are detected, I'd like to know before.

> As for showstoppers: it seems that the machine is down at
> the moment (Jan 29, 20:07 GMT), no?

Not for me. DNS trouble? 

\EF
- -- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iD8DBQFFvlvdrJurFAusidkRArIFAKCK4nlfrnMe6/DY9WVQnx2kVR3RQgCeOSge
LvlAS0e80st9iMTqipIWUL4=
=ftRA
-----END PGP SIGNATURE-----

From martin at v.loewis.de  Mon Jan 29 21:58:37 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Mon, 29 Jan 2007 21:58:37 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <877iv537cg.fsf@uterus.efod.se>
References: <87bqkh3b7c.fsf@uterus.efod.se>	
	<877iv537cg.fsf@uterus.efod.se>
Message-ID: <45BE5FFD.10809@v.loewis.de>

Erik Forsberg schrieb:
> There is also a 'Coordinator' role who has permission to assign roles
> (except the Administrator role) to other users. That should be a
> handful of python developers, probably the same people that handles
> the svn permissions today. 

I think that should be Neal Norwitz, Andrew Kuchling, Tim Peters,
and myself.

Regards,
Martin

From martin at v.loewis.de  Mon Jan 29 22:01:14 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Mon, 29 Jan 2007 22:01:14 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <87wt351s82.fsf@uterus.efod.se>
References: <87bqkh3b7c.fsf@uterus.efod.se> <45BE5834.3070508@v.loewis.de>
	<87wt351s82.fsf@uterus.efod.se>
Message-ID: <45BE609A.1030603@v.loewis.de>

Erik Forsberg schrieb:
>> As for showstoppers: it seems that the machine is down at
>> the moment (Jan 29, 20:07 GMT), no?
> 
> Not for me. DNS trouble? 

No, it resolves fine, just some packets fail to route or
come back properly (assuming 88.198.142.26 is the
right address):

martin at mira:~/work/py2.6$ traceroute psf.upfronthosting.co.za
traceroute to psf.upfronthosting.co.za (88.198.142.26), 30 hops max, 40
byte packets
 1  * * *
 2  PC1-201.bln2-g.mcbone.net (62.104.193.97)  52.807 ms  52.878 ms
52.932 ms
 3  lo0-0.lpz2-j2.mcbone.net (62.104.191.208)  59.668 ms  59.402 ms
58.391 ms
 4  lo0-0.lpz2-j.mcbone.net (62.104.191.207)  60.187 ms  57.041 ms
56.825 ms
 5  L0.nbg2-g2.mcbone.net (62.104.191.136)  61.636 ms  62.918 ms  61.975 ms
 6  L0.nbg2-g.mcbone.net (62.104.191.135)  62.615 ms  62.353 ms  63.981 ms
 7  GigE-0.Hetzner.DHK.N-IX.net (195.85.217.16)  62.484 ms  62.391 ms
62.652 ms
 8  hos-bb2.rs8000.rz5.hetzner.de (213.239.240.132)  62.961 ms  63.877
ms  62.995 ms
 9  et.1.16.rs3k6.rz5.hetzner.de (213.239.247.168)  62.758 ms  64.386 ms
 61.625 ms
10  * * *
11  * * *
12  * * *
13  * * *
14  * * *

Regards,
Martin

From g.brandl at gmx.net  Mon Jan 29 22:06:34 2007
From: g.brandl at gmx.net (Georg Brandl)
Date: Mon, 29 Jan 2007 21:06:34 +0000
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: 
References: <87bqkh3b7c.fsf@uterus.efod.se>	<20070129194221.GA16440@localhost.localdomain>

Message-ID: <45BE61DA.5080707@gmx.net>

Brett Cannon schrieb:
> On 1/29/07, A.M. Kuchling  wrote:
>> On Mon, Jan 29, 2007 at 08:05:43PM +0100, Erik Forsberg wrote:
>> > * Do we need a public beta period to allow people not on this list to
>> >   test the tracker?
>>
>> Um, have I missed some announcements of the tracker status on
>> python-dev?  Searching for 'tracker' in the December and January
>> python-dev archive didn't turn anything up.  In the absence of such an
>> announcement, I doubt many python-dev readers are aware of the current
>> status of the tracker work, or even of its URL.
>>
> 
> Go one month back to November and you will find the announcement of
> tracker-discuss on python-dev along with some initial cross-posts.  I
> honestly think most people on python-dev just don't care that much and
> are just expecting to adapt.  Martin and Georg use the tracker the
> most (I suspect) and they have both contributed input on all of this.

I'd also favor going live immediately, I can't recall any problem that's really
urgent.

Georg

From amk at amk.ca  Mon Jan 29 22:30:13 2007
From: amk at amk.ca (A.M. Kuchling)
Date: Mon, 29 Jan 2007 16:30:13 -0500
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <45BE5FFD.10809@v.loewis.de>
References: <87bqkh3b7c.fsf@uterus.efod.se>

	<877iv537cg.fsf@uterus.efod.se> <45BE5FFD.10809@v.loewis.de>
Message-ID: <20070129213013.GA19482@localhost.localdomain>

On Mon, Jan 29, 2007 at 09:58:37PM +0100, "Martin v. Loewis" wrote:
> I think that should be Neal Norwitz, Andrew Kuchling, Tim Peters,
> and myself.

Raymond might be a better choice than me -- I certainly don't have
admin privileges on the current bug tracker.

--amk

From pfdubois at gmail.com  Mon Jan 29 22:55:33 2007
From: pfdubois at gmail.com (Paul Dubois)
Date: Mon, 29 Jan 2007 13:55:33 -0800
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <45BE5FFD.10809@v.loewis.de>
References: <87bqkh3b7c.fsf@uterus.efod.se>

	<877iv537cg.fsf@uterus.efod.se> <45BE5FFD.10809@v.loewis.de>
Message-ID: 

I volunteer to be the person who fixes account problems, emails, etc. I am
off 5 days a week so I think I can get it done faster than most of you. I'm
for letting it rip.

I never got an answer to the question of how the patches would be marked so
that I can make that report on the weekly.

On 1/29/07, "Martin v. L?wis"  wrote:
>
> Erik Forsberg schrieb:
> > There is also a 'Coordinator' role who has permission to assign roles
> > (except the Administrator role) to other users. That should be a
> > handful of python developers, probably the same people that handles
> > the svn permissions today.
>
> I think that should be Neal Norwitz, Andrew Kuchling, Tim Peters,
> and myself.
>
> Regards,
> Martin
>
> _______________________________________________
> Tracker-discuss mailing list
> Tracker-discuss at python.org
> http://mail.python.org/mailman/listinfo/tracker-discuss
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20070129/d65d3e94/attachment.html 

From martin at v.loewis.de  Mon Jan 29 23:12:35 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Mon, 29 Jan 2007 23:12:35 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: 
References: <87bqkh3b7c.fsf@uterus.efod.se>	

	<877iv537cg.fsf@uterus.efod.se> <45BE5FFD.10809@v.loewis.de>

Message-ID: <45BE7153.2060307@v.loewis.de>

Paul Dubois schrieb:
> I volunteer to be the person who fixes account problems, emails, etc. I
> am off 5 days a week so I think I can get it done faster than most of
> you. I'm for letting it rip.
> 
> I never got an answer to the question of how the patches would be marked
> so that I can make that report on the weekly.

I think this still is a show-stopper.

Regards,
Martin

From martin at v.loewis.de  Mon Jan 29 23:16:37 2007
From: martin at v.loewis.de (=?ISO-8859-15?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Mon, 29 Jan 2007 23:16:37 +0100
Subject: [Tracker-discuss] Cannot create new issues
Message-ID: <45BE7245.5090606@v.loewis.de>

It seems that I can't create any new issues whatsoever on
neither the meta tracker nor the Python tracker.

Regards,
Martin

From brett at python.org  Tue Jan 30 00:02:24 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 29 Jan 2007 15:02:24 -0800
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: <45BE7153.2060307@v.loewis.de>
References: <87bqkh3b7c.fsf@uterus.efod.se>

	<877iv537cg.fsf@uterus.efod.se> <45BE5FFD.10809@v.loewis.de>

	<45BE7153.2060307@v.loewis.de>
Message-ID: 

On 1/29/07, "Martin v. L?wis"  wrote:
> Paul Dubois schrieb:
> > I volunteer to be the person who fixes account problems, emails, etc. I
> > am off 5 days a week so I think I can get it done faster than most of
> > you. I'm for letting it rip.
> >
> > I never got an answer to the question of how the patches would be marked
> > so that I can make that report on the weekly.
>
> I think this still is a show-stopper.
>

Agreed.

Should we have a 'patch' keyword?  Don't remember if anyone logged in
can change the keywords, though, let alone attach a file to an issue.

-Brett

From pfdubois at gmail.com  Tue Jan 30 01:06:48 2007
From: pfdubois at gmail.com (Paul Dubois)
Date: Mon, 29 Jan 2007 16:06:48 -0800
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: 
References: <87bqkh3b7c.fsf@uterus.efod.se>

	<877iv537cg.fsf@uterus.efod.se> <45BE5FFD.10809@v.loewis.de>

	<45BE7153.2060307@v.loewis.de>

Message-ID: 

It is completely our choice who can change the keywords associated with an
issue. Similar to nosy lists, it is probably best not to be restrictive,
allowing anyone who can post an update (ie, in our case a logged-in user) to
do so.

There was a discussion on this topic that I think came to a conclusion
although I don't know what it was; I thought it was that we would create
appropriate keyword(s).

This should NOT be a showstopper in that whatever the scheme it should not
take me much time to modify the weekly reporter appropriately, once I know
the scheme. If in a worst case I missed a week or two the sky would not
fall. The basic report works now.

On 1/29/07, Brett Cannon  wrote:
>
> On 1/29/07, "Martin v. L?wis"  wrote:
> > Paul Dubois schrieb:
> > > I volunteer to be the person who fixes account problems, emails, etc.
> I
> > > am off 5 days a week so I think I can get it done faster than most of
> > > you. I'm for letting it rip.
> > >
> > > I never got an answer to the question of how the patches would be
> marked
> > > so that I can make that report on the weekly.
> >
> > I think this still is a show-stopper.
> >
>
> Agreed.
>
> Should we have a 'patch' keyword?  Don't remember if anyone logged in
> can change the keywords, though, let alone attach a file to an issue.
>
> -Brett
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20070129/823ad62c/attachment.htm 

From brett at python.org  Tue Jan 30 01:49:43 2007
From: brett at python.org (Brett Cannon)
Date: Mon, 29 Jan 2007 16:49:43 -0800
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: 
References: <87bqkh3b7c.fsf@uterus.efod.se>

	<877iv537cg.fsf@uterus.efod.se> <45BE5FFD.10809@v.loewis.de>

	<45BE7153.2060307@v.loewis.de>

Message-ID: 

On 1/29/07, Paul Dubois  wrote:
> It is completely our choice who can change the keywords associated with an
> issue. Similar to nosy lists, it is probably best not to be restrictive,
> allowing anyone who can post an update (ie, in our case a logged-in user) to
> do so.
>

Yeah, this would be good.  Sometimes people find a bug who are not the
OP and write up a patch.  If they could just attach it to the bug that
would be great.  Just as long as there is a way to flag that an issue
has a patch and not just some log output or canonical bug example.

> There was a discussion on this topic that I think came to a conclusion
> although I don't know what it was; I thought it was that we would create
> appropriate keyword(s).
>

I don't remember either.  =)  Probably a keyword since that makes the
most sense.

> This should NOT be a showstopper in that whatever the scheme it should not
> take me much time to modify the weekly reporter appropriately, once I know
> the scheme. If in a worst case I missed a week or two the sky would not
> fall. The basic report works now.
>

I think Martin means it is a showstopper if we don't have a way to
mark items as containing a patch, not for your email summary, Paul.
It is a rather critical thing to be able to get a list of all
available patches.

-Brett

>
> On 1/29/07, Brett Cannon  wrote:
> > On 1/29/07, "Martin v. L?wis"  wrote:
> > > Paul Dubois schrieb:
> > > > I volunteer to be the person who fixes account problems, emails, etc.
> I
> > > > am off 5 days a week so I think I can get it done faster than most of
> > > > you. I'm for letting it rip.
> > > >
> > > > I never got an answer to the question of how the patches would be
> marked
> > > > so that I can make that report on the weekly.
> > >
> > > I think this still is a show-stopper.
> > >
> >
> > Agreed.
> >
> > Should we have a 'patch' keyword?  Don't remember if anyone logged in
> > can change the keywords, though, let alone attach a file to an issue.
> >
> > -Brett
> >
>
>

From martin at v.loewis.de  Tue Jan 30 07:20:20 2007
From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=)
Date: Tue, 30 Jan 2007 07:20:20 +0100
Subject: [Tracker-discuss] Ready for Prime Time?
In-Reply-To: 
References: <87bqkh3b7c.fsf@uterus.efod.se>	

	<877iv537cg.fsf@uterus.efod.se> <45BE5FFD.10809@v.loewis.de>	

	<45BE7153.2060307@v.loewis.de>	

Message-ID: <45BEE3A4.2030509@v.loewis.de>

Paul Dubois schrieb:
> This should NOT be a showstopper in that whatever the scheme it should
> not take me much time to modify the weekly reporter appropriately, once
> I know the scheme. If in a worst case I missed a week or two the sky
> would not fall. The basic report works now.

Sure. However, it *is* a show-stopper wrt. the import: if we lose the
information what a patch is on import, we won't ever get it back
afterwards (i.e. I don't want to go manually through all issue and
add a patch tag).

Regards,
Martin

From forsberg at efod.se  Tue Jan 30 07:32:21 2007
From: forsberg at efod.se (Erik Forsberg)
Date: Tue, 30 Jan 2007 07:32:21 +0100
Subject: [Tracker-discuss] Cannot create new issues
In-Reply-To: <45BE7245.5090606@v.loewis.de> (Martin v. =?iso-8859-1?Q?L=F6?=
	=?iso-8859-1?Q?wis's?= message of
	"Mon, 29 Jan 2007 23:16:37 +0100")
References: <45BE7245.5090606@v.loewis.de>
Message-ID: <87ps8x10ui.fsf@uterus.efod.se>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"Martin v. L?wis"  writes:

> It seems that I can't create any new issues whatsoever on
> neither the meta tracker nor the Python tracker.

That is just to convince you that there's no showstoppers left.. ;-)

I have the trackeback. Will take a look.

\EF
- -- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iD8DBQFFvuZ1rJurFAusidkRAqznAKDLeZibQ87tdCRZa9+x0asGchDzcACgoGdQ
g7k7cbV2MQbxe00qmDlGupQ=
=KKuM
-----END PGP SIGNATURE-----

From metatracker at psf.upfronthosting.co.za  Tue Jan 30 07:56:32 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Tue, 30 Jan 2007 06:56:32 -0000
Subject: [Tracker-discuss] [issue83] Unable to create new issues
Message-ID: <1170140192.9.0.0606052321013.issue83@psf.upfronthosting.co.za>

New submission from Erik Forsberg:

Hmm.. does creating new messages fail?

----------
status: unread -> chatting

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Tue Jan 30 07:58:34 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Tue, 30 Jan 2007 06:58:34 -0000
Subject: [Tracker-discuss] [issue84] Testing creation of new issue,
	with message
Message-ID: <1170140314.34.0.285243632505.issue84@psf.upfronthosting.co.za>

New submission from Erik Forsberg:

Seems to have been a bug in roundupdb.py caused by the fix for issue68.

----------
messages: 410
nosy: forsberg
priority: bug
status: unread
title: Testing creation of new issue, with message

______________________________________________________
Meta Tracker 

______________________________________________________

From metatracker at psf.upfronthosting.co.za  Tue Jan 30 08:01:13 2007
From: metatracker at psf.upfronthosting.co.za (Erik Forsberg)
Date: Tue, 30 Jan 2007 07:01:13 -0000
Subject: [Tracker-discuss] [issue83] Unable to create new issues
Message-ID: <1170140473.61.0.282094370621.issue83@psf.upfronthosting.co.za>

Erik Forsberg added the comment:

Problem was due to an incorrect format string introduced with the fix for
issue68.  The reason this was not detected earlier was that the bug was only
triggered when creating new issues. I guess all testing were made on existing
issues.

----------
assignedto:  -> forsberg
nosy: +forsberg
status: chatting -> resolved

______________________________________________________
Meta Tracker 

______________________________________________________

From forsberg at efod.se  Tue Jan 30 08:02:01 2007
From: forsberg at efod.se (Erik Forsberg)
Date: Tue, 30 Jan 2007 08:02:01 +0100
Subject: [Tracker-discuss] Cannot create new issues
In-Reply-To: <87ps8x10ui.fsf@uterus.efod.se> (Erik Forsberg's message of "Tue, 
	30 Jan 2007 07:32:21 +0100")
References: <45BE7245.5090606@v.loewis.de> <87ps8x10ui.fsf@uterus.efod.se>
Message-ID: <87lkjl0zh2.fsf@uterus.efod.se>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Erik Forsberg  writes:

> I have the trackeback. Will take a look.

Fixed. 

\EF
- -- 
Erik Forsberg                 http://efod.se
GPG/PGP Key: 1024D/0BAC89D9
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iD8DBQFFvu1prJurFAusidkRAtNxAKCOSDTGI7BlBWzq5fA7zpMqtGfpoQCgzCbC
YXT8tkjk3yG1uJTRIVI1XvQ=
=m3lB
-----END PGP SIGNATURE-----