[Mailman-Users] admin notifications for held messages

[Richard Barrett]

On Thursday, October 9, 2003, at 05:55  pm, Saurav Pathak wrote:

> i am still struggling to find out what caused the error.  after all
> i have just added one line to Default.py.  why should it complain
> about the version number?  i would appreciate it if someone could
> give me some pointers.
>

I have been rather busy today at my paid job hence not earlier response.

I tried applying the patch I suggested to you to a freshly extracted MM 
2.1.1 distribution and  got the following terminal output. When you 
tried using the patch did you get a result like this?

bash-2.05a$ zcat distributions/mailman-2.1.1.tgz | tar xf -
bash-2.05a$ mv mailman-2.1.1 mailman-2.1.1-mod
bash-2.05a$ cd mailman-2.1.1-mod/
bash-2.05a$ patch -p1 < ../patches/current/modinc-2.1.3-0.1.patch
patching file Mailman/Defaults.py.in
Hunk #1 succeeded at 921 with fuzz 2 (offset -18 lines).
patching file Mailman/Gui/General.py
patching file Mailman/Handlers/Hold.py
Hunk #1 succeeded at 268 (offset -1 lines).
patching file Mailman/MailList.py
Hunk #1 succeeded at 314 (offset -14 lines).
patching file Mailman/Version.py
bash-2.05a$

This indicates the patch applied OK. The offsets are due to other 
differences in the files but do not invalidate the patch.

Looking at the files changed the patch has applied OK.

As regards the problem you are having after applying the patch, I think 
is because the lines defining values for the variables True, False, 
Yes, No, On and Off do not appear in the Defaults.py.in of MM 2.1.1 
source code; they are in the MM 2.1.2 and MM 2.1.3 source.

Try changing the line in the patch from:

DEFAULT_MODERATE_INC_BODY = Yes

to:

DEFAULT_MODERATE_INC_BODY = 1

> anyway, this would be a nice feature in mailman.  i hope it will no
> longer be a patch in the next version.
>
> thanks,
> saurav
>
>
>
> Thus spake Saurav Pathak:
>
> +
> +
> +
> +  Thus spake Richard Barrett:
> +
> +  +  As it stands, the patch has to be applied to the Mailman source
> +  +  distribution and there are only versions for MM 2.1.2 and MM 
> 2.1.3
> +  +
> +  +  That said, it will probably work with earlier versions of MM 2.1 
> or can
> +  +  be trivially adapted to work with them. The code changes 
> themselves are
> +  +  really quite small and could be added by hand to an installed 
> system if
> +  +  you really wanted to.
> +
> +  thanks for the patch.  i added the patch by hand, but got the
> +  following error while trying to access the admin webpage:
> +
> +  --------------
> +  Bug in Mailman version <undetermined>
> +
> +  We're sorry, we hit a bug!
> +
> +  If you would like to help us identify the problem, please email a
> +  copy of this page to the webmaster for this site with a description
> +  of what happened. Thanks!
> +  Traceback:
> +
> +  Traceback (most recent call last):
> +    File "/var/mailman/scripts/driver", line 114, in print_traceback
> +      from Mailman.mm_cfg import VERSION
> +  ImportError: cannot import name VERSION
> +  --------------
> +
> +  i donot understand why i should be getting this.
> +
> +  instead of the patches in all the files, I decided to add it in the
> +  Default.py alone.  With all the other files as they were, and only
> +  the Default.py patched, i still get the error.
> +
> +  saurav
> +  ps: the patch i am applying is:
> +
> +  --------------
> +  diff -r -u -P mailman-2.1.2/Mailman/Defaults.py.in
> +  mailman-2.1.2-mod/Mailman/Defaults.py.in
> +  --- mailman-2.1.2/Mailman/Defaults.py.in        Sun Apr 20 05:02:03 
> 2003
> +  +++ mailman-2.1.2-mod/Mailman/Defaults.py.in    Wed Jun 25 15:49:37 
> 2003
> +  @@ -934,7 +934,9 @@
> +   # Check for administrivia in messages sent to the main list?
> +   DEFAULT_ADMINISTRIVIA = Yes
> +
> +  -
> +  +# Should the body of a message requiring moderation be included in 
> the
> +  +# request to the moderator(s)
> +  +DEFAULT_MODERATE_INC_BODY = Yes
> +   ^L
> +   #####
> +   # Digestification defaults.  Same caveat applies here as with list 
> defaults.
> +  --------------
> -- 
> saurav
>
>
-----------------------------------------------------------------------
Richard Barrett                               http://www.openinfo.co.uk