unchecked_lovers_maps in sql READMEs

Tim Howe thowe at bendtel.net
Tue Nov 22 20:16:43 CET 2011


On Tue, 22 Nov 2011 20:08:32 +0100
Mark Martinec <Mark.Martinec+amavis at ijs.si> wrote:

> TimH,
> 
> > There is a new field in the policy table named "unchecked_lovers_maps"
> > in the schema provided in the sql readme files.  However, it seems like
> > it should be "unchecked_lover" in the schemas.  The release notes seem
> > to agree with me:
> > [...]
> > So, in order to work correctly, should the provided schemas be changed
> > so that the field name is "unchecked_lover"?
> > In a nutshell, shouldn't policy.unchecked_lovers_maps be
> > policy.unchecked_lover in the SQL schemas?
> 
> Indeed, a documentation (schema) bug.
> Fixed, thanks!

	Awesome, thanks for clearing that up.  Now if I could just find
that IPv6 patch for Net::Server...

> > Also, should users.local be a boolean instead of a char?
> 
> For historical reasons the boolean settings are represented
> in SQL fields as char(1), where a values like "N" or 0 are
> treated as false, and values like "Y" or 1 are true.
> Actually:
> 
>   # convert values 'N', 'F', '0', ' ' and "\000" to 0
>   # to allow value to be used directly as a Perl boolean
>   $match = 0  if $match =~ /^([NnFf ]|0+|\000+)\ *\z/;
> 
> I believe you may declare these fields as booleans and the
> above logic would still work correctly, but I haven't tried it.

	Ah.  Thanks for the response.

Cheers!
--TimH


More information about the amavis-users mailing list