Re: [Zope-dev] Re: Debian and Zope

2004-08-23 Thread Chris Withers
Thaddeus H. Black wrote:
Can you tell us who that maintainer is?
Luca De Vitis [EMAIL PROTECTED].  Last known
Zope activity: 4 Nov 2003.
Hmmm, don't know him. Anyone else know him?
Oh well, I guess we'll just have to wait for someone with the necessary 
Zope and Debian Zen to emerge :-/

cheers,
Chris
--
Simplistix - Content Management, Zope  Python Consulting
   - http://www.simplistix.co.uk
___
Zope-Dev maillist  -  [EMAIL PROTECTED]
http://mail.zope.org/mailman/listinfo/zope-dev
**  No cross posts or HTML encoding!  **
(Related lists - 
http://mail.zope.org/mailman/listinfo/zope-announce
http://mail.zope.org/mailman/listinfo/zope )


[Zope-dev] Re: Debian and Zope

2004-08-23 Thread Tres Seaver
Thaddeus H. Black wrote:
This question regards the code in Zope's
zpasswd.py.  If you know zpasswd.py or are
involved in Debian, please reply.
Debian GNU/Linux 3.1 releases soon.  The Debian
Zope package (actually the Debian `zopectl'
package) has what Debian calls a Release
Critical Bug open against it
[http://bugs.debian.org/251038].  Normally,
Debian's Zope maintainer would handle such a
bug, but the maintainer seems to be inactive
this year, so it falls to me to fix it.  This is
okay.  The maintainer's work in earlier years is
appreciated.  The trouble is, I do not use Zope
myself, so I am fixing the bug with limited
understanding of the potential consequences of
the fix.
The part of the Debian fix which affects your
Zope source is in zpasswd.py.  At present, a
user can invoke zpasswd in either of two ways:
  (1) without command-line options, in which
  case zpasswd prompts the user on tty for
  username and password; or
  (2) with username and password supplied on the
  command line.
How the user cannot presently invoke zpasswd is
  (3) with username supplied on the command
  line, so that zpasswd prompts the user on tty
  for a password only.
The Zope installation procedure on Debian
requires option (3).  With (3), I can cleanly
eliminate the Release Critical Bug now and
prevent Zope from being dropped from Debian 3.1.
Thus I am adding (3) to Debian Zope.
Question, please.  Does my addition create some
subtle problem I should know about?  If Zope
were coded perfectly cleanly, then the addition
of option (3) should create no problem; but you
and I have both done enough coding to realize
that old code can sometimes depend in strange
ways on the odd behavior of even older code.
With my small addition, zpasswd now correctly
handles (3).  Is this okay, or can you think of
something specific it might break, something I
might not have noticed?  I have checked
everything I know how to check, but I do not
know Zope; you guys do.
Regrettably, Zope's public SVN seems to be down
as I write these words, so in the unlikely event
that someone had already added option (3) to
zpasswd.py, I would not know about it.  I do
understand and respect that, like any good
free-software developers, you will probably all
want me to update Debian Zope now to Zope's
latest version!  To this, I must plead that such
an update far exceeds the limits of my Zope
ignorance (also, it would violate Debian Project
policy, which permits me minimally to fix a
Release Critical Bug for an inactive maintainer
but forbids me from unilaterally hijacking his
Debian package).  The matter at hand now is not
how to include the latest Zope (it is not
Debian's usual practice to include the very
latest software in any event) but rather how to
keep Zope in Debian at all.  This is a good
cause and I am pleased to help.
I do not regularly subscribe to zope-dev, so
please copy your reply to me and to
[EMAIL PROTECTED] (the latter automatically
attaches your reply to the relevant official
Debian bug report, #251038).  If there were no
reply, this would be okay; I would just assume
that my fix were as good as it seems to be, and
would proceed accordingly.  Please reply by 23
August.
If any active Zope developer on this list also
happens to be a registered Debian Developer, he
is asked to step in and take charge of Debian
bug #251038 from here: one can find my full
zopectl/zope Debian patch attached to the Debian
bug report.  Otherwise, thank you for your
attention in this matter.  If you want my patch
to zpasswd.py, here it is.  You probably want to
apply it to your own SVN Zope source.
diff -u zope-2.6.4/zpasswd.py zope-2.6.4.new/zpasswd.py
--- zope-2.6.4/zpasswd.py	2004-08-03 20:34:53.0 +
+++ zope-2.6.4.new/zpasswd.py	2004-08-03 18:49:11.0 +
@@ -87,6 +87,16 @@
 
 import do; do.ch(ac_path, user, group)
 
+def get_password():
+while 1:
+password = getpass.getpass(Password: )
+verify = getpass.getpass(Verify password: )
+if verify == password:
+return password
+else:
+password = verify = ''
+print Password mismatch, please try again...
+
 
 def main(argv):
 short_options = ':u:p:e:d:'
@@ -150,7 +160,10 @@
 
 # Verify that we got what we need
 if not username or not password:
-raise CommandLineError
+if username:
+password = get_password()
+else:
+raise CommandLineError
 
 access_file.write(username + ':' +
   generate_passwd(password, encoding) +
@@ -163,14 +176,7 @@
 if username != '':
 break
 
-while 1:
-password = getpass.getpass(Password: )
-verify = getpass.getpass(Verify password: )
-if verify == password:
-break
-else:
-password = verify = ''
- 

Re: [Zope-dev] Re: Debian and Zope

2004-08-20 Thread Chris Withers
Hi Thaddeus,
Thaddeus H. Black wrote:
Please let me know if it doesn't show up
there...
I see your message there
[http://bugs.debian.org/251038] presently.
Cool, the bug's going to look a bit bloated due to the way the Zope 
Collector sends mails, but at least everyone will know where things are ;-)

might mention to them that Debian Zope has an
inactive maintainer and seems ripe for takeover.
Can you tell us who that maintainer is?
cheers,
Chris
--
Simplistix - Content Management, Zope  Python Consulting
   - http://www.simplistix.co.uk
___
Zope-Dev maillist  -  [EMAIL PROTECTED]
http://mail.zope.org/mailman/listinfo/zope-dev
**  No cross posts or HTML encoding!  **
(Related lists - 
http://mail.zope.org/mailman/listinfo/zope-announce
http://mail.zope.org/mailman/listinfo/zope )


Re: [Zope-dev] Re: Debian and Zope

2004-08-19 Thread Chris Withers
Thaddeus H. Black wrote:
This question regards the code in Zope's
zpasswd.py.  If you know zpasswd.py or are
involved in Debian, please reply.
I've attempted to add this to the Zope issue tracker and used the 
Debiuan bug's email address so it can be tracked there too.

Please let me know if it doesn't show up there...
Chris
--
Simplistix - Content Management, Zope  Python Consulting
   - http://www.simplistix.co.uk
___
Zope-Dev maillist  -  [EMAIL PROTECTED]
http://mail.zope.org/mailman/listinfo/zope-dev
**  No cross posts or HTML encoding!  **
(Related lists - 
http://mail.zope.org/mailman/listinfo/zope-announce
http://mail.zope.org/mailman/listinfo/zope )


Re: [Zope-dev] Re: Debian and Zope

2004-08-19 Thread Thaddeus H. Black
Chris wrote,

 I've attempted to add this to the Zope issue
 tracker and used the Debiuan bug's email
 address so it can be tracked there too.

Thank you for your expert response, advice and
intervention.

 Please let me know if it doesn't show up
 there...

I see your message there
[http://bugs.debian.org/251038] presently.
It looks good.  I and anyone else who reads
the #251038 bug log now understand that some
form of my zpasswd.py patch is to be considered
for the next official feature-adding Zope
release.

If any Debian guys happen to come around
zope-dev in the next six months, if they show a
serious interest, and if you think of it, you
might mention to them that Debian Zope has an
inactive maintainer and seems ripe for takeover.
Zope is somewhat a popular package among Debian
users.  While the Debian Project does not
normally take packages away from unreasonable
maintainers, it routinely takes them away from
missing maintainers who fail even to respond to
a takeover-proposing e-mail (but no such e-mail
has yet been sent, nor do I myself mean to send
one, nor do I call Debian's Zope maintainer
unreasonable: he has written me neither rudely
nor in any other manner; he has merely been
silent).  Sometimes sleepy Debian maintainers
wake up when a takeover is proposed; more often
in experience, they just keep sleeping.

I am no Zope user, or I would politely offer to
take the Debian package over myself.

(If the maintainer reads these words, I hope
that he will take no offense.  Presumably he has
just been busy this year in real life, which is
something we all respect.  If he would without
delay relieve me of handling #251038, I would
appreciate it.  His active stewardship of the
Debian package is welcome.)

Am unsubscribing from zope-dev now.  Thanks
guys.  If you should happen to want me
for any further purpose, related or unrelated
to #251038, please copy your message to me not
just to the list.

-- 
Thaddeus H. Black
508 Nellie's Cave Road, Blacksburg,
Virginia 24060, USA; +1 540 961 0920, [EMAIL PROTECTED]


pgp1TXJjLeqDL.pgp
Description: PGP signature
___
Zope-Dev maillist  -  [EMAIL PROTECTED]
http://mail.zope.org/mailman/listinfo/zope-dev
**  No cross posts or HTML encoding!  **
(Related lists - 
 http://mail.zope.org/mailman/listinfo/zope-announce
 http://mail.zope.org/mailman/listinfo/zope )


[Zope-dev] Re: Debian and Zope

2004-08-18 Thread Thaddeus H. Black
This question regards the code in Zope's
zpasswd.py.  If you know zpasswd.py or are
involved in Debian, please reply.

Debian GNU/Linux 3.1 releases soon.  The Debian
Zope package (actually the Debian `zopectl'
package) has what Debian calls a Release
Critical Bug open against it
[http://bugs.debian.org/251038].  Normally,
Debian's Zope maintainer would handle such a
bug, but the maintainer seems to be inactive
this year, so it falls to me to fix it.  This is
okay.  The maintainer's work in earlier years is
appreciated.  The trouble is, I do not use Zope
myself, so I am fixing the bug with limited
understanding of the potential consequences of
the fix.

The part of the Debian fix which affects your
Zope source is in zpasswd.py.  At present, a
user can invoke zpasswd in either of two ways:

  (1) without command-line options, in which
  case zpasswd prompts the user on tty for
  username and password; or

  (2) with username and password supplied on the
  command line.

How the user cannot presently invoke zpasswd is

  (3) with username supplied on the command
  line, so that zpasswd prompts the user on tty
  for a password only.

The Zope installation procedure on Debian
requires option (3).  With (3), I can cleanly
eliminate the Release Critical Bug now and
prevent Zope from being dropped from Debian 3.1.
Thus I am adding (3) to Debian Zope.

Question, please.  Does my addition create some
subtle problem I should know about?  If Zope
were coded perfectly cleanly, then the addition
of option (3) should create no problem; but you
and I have both done enough coding to realize
that old code can sometimes depend in strange
ways on the odd behavior of even older code.
With my small addition, zpasswd now correctly
handles (3).  Is this okay, or can you think of
something specific it might break, something I
might not have noticed?  I have checked
everything I know how to check, but I do not
know Zope; you guys do.

Regrettably, Zope's public SVN seems to be down
as I write these words, so in the unlikely event
that someone had already added option (3) to
zpasswd.py, I would not know about it.  I do
understand and respect that, like any good
free-software developers, you will probably all
want me to update Debian Zope now to Zope's
latest version!  To this, I must plead that such
an update far exceeds the limits of my Zope
ignorance (also, it would violate Debian Project
policy, which permits me minimally to fix a
Release Critical Bug for an inactive maintainer
but forbids me from unilaterally hijacking his
Debian package).  The matter at hand now is not
how to include the latest Zope (it is not
Debian's usual practice to include the very
latest software in any event) but rather how to
keep Zope in Debian at all.  This is a good
cause and I am pleased to help.

I do not regularly subscribe to zope-dev, so
please copy your reply to me and to
[EMAIL PROTECTED] (the latter automatically
attaches your reply to the relevant official
Debian bug report, #251038).  If there were no
reply, this would be okay; I would just assume
that my fix were as good as it seems to be, and
would proceed accordingly.  Please reply by 23
August.

If any active Zope developer on this list also
happens to be a registered Debian Developer, he
is asked to step in and take charge of Debian
bug #251038 from here: one can find my full
zopectl/zope Debian patch attached to the Debian
bug report.  Otherwise, thank you for your
attention in this matter.  If you want my patch
to zpasswd.py, here it is.  You probably want to
apply it to your own SVN Zope source.

diff -u zope-2.6.4/zpasswd.py zope-2.6.4.new/zpasswd.py
--- zope-2.6.4/zpasswd.py   2004-08-03 20:34:53.0 +
+++ zope-2.6.4.new/zpasswd.py   2004-08-03 18:49:11.0 +
@@ -87,6 +87,16 @@
 
 import do; do.ch(ac_path, user, group)
 
+def get_password():
+while 1:
+password = getpass.getpass(Password: )
+verify = getpass.getpass(Verify password: )
+if verify == password:
+return password
+else:
+password = verify = ''
+print Password mismatch, please try again...
+
 
 def main(argv):
 short_options = ':u:p:e:d:'
@@ -150,7 +160,10 @@
 
 # Verify that we got what we need
 if not username or not password:
-raise CommandLineError
+if username:
+password = get_password()
+else:
+raise CommandLineError
 
 access_file.write(username + ':' +
   generate_passwd(password, encoding) +
@@ -163,14 +176,7 @@
 if username != '':
 break
 
-while 1:
-password = getpass.getpass(Password: )
-verify = getpass.getpass(Verify password: )
-if verify == password:
-break
-else:
-password = verify = ''
-