[PATCH] Turn enumeration parameter to boolean

[Simo Sorce]

Since we enumerate in background and use the member/memberof scheme you
can't do group enumerations without also doing a user enumeration first.

So turn the enumeration option into a boolean with warnings if the old
syntax is used.

Simo.

[Dmitri Pal]

Simo Sorce wrote:
> Since we enumerate in background and use the member/memberof scheme you
> can't do group enumerations without also doing a user enumeration first.
>
> So turn the enumeration option into a boolean with warnings if the old
> syntax is used.
>
> Simo.

Simo, I think this patch is missing the part where it reads the INI file
and sticks initial data into the database.
It should have same logic:

* try to read as int, if success then probably stick as int in the sysdb
* otherwise try to read as bool and stick bool in.
> sssd-devel mailing list
> [e-mail]
> https://fedorahosted.org/mailman/listinfo/sssd-devel

[Stephen Gallagher]

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

On 08/31/2009 02:56 PM, Dmitri Pal wrote:
> Simo Sorce wrote:
> > Since we enumerate in background and use the member/memberof scheme you
> > can't do group enumerations without also doing a user enumeration first.
> >
> > So turn the enumeration option into a boolean with warnings if the old
> > syntax is used.
> >
> > Simo.
>
> Simo, I think this patch is missing the part where it reads the INI file
> and sticks initial data into the database.
> It should have same logic:
> try to read as int, if success then probably stick as int in the sysdb
> otherwise try to read as bool and stick bool in.

We discussed this, but currently we read all entries as strings into the
confdb (since that's how they're stored internally) and then we do the
validation when we query the confdb instead.

Once the INI parser supports schema validation, we'll move these checks
there.
> > sssd-devel mailing list
> > [e-mail]
> > https://fedorahosted.org/mailman/listinfo/sssd-devel

- -- 
Stephen Gallagher
RHCE 804006346421761

Looking to carve out IT costs?
www.redhat.com/carveoutcosts/
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.9 (GNU/Linux)
> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
>
> iEYEARECAAYFAkqcHccACgkQeiVVYja6o6PRdgCeJ3h5vDKUx2CrQ5HbY33iIUtt
> hI0AoK9cukGUiPimIbNUyghkBrjPcJ4l
> =uPxu
> -----END PGP SIGNATURE-----

[Dmitri Pal]

Stephen Gallagher wrote:
> On 08/31/2009 02:56 PM, Dmitri Pal wrote:
> > Simo Sorce wrote:
> > > Since we enumerate in background and use the member/memberof scheme you
> > > can't do group enumerations without also doing a user enumeration
>
> first.
> > > So turn the enumeration option into a boolean with warnings if the old
> > > syntax is used.
> > >
> > > Simo.
> >
> > Simo, I think this patch is missing the part where it reads the INI file
> > and sticks initial data into the database.
> > It should have same logic:
> > try to read as int, if success then probably stick as int in the sysdb
> > otherwise try to read as bool and stick bool in.
>
> We discussed this, but currently we read all entries as strings into the
> confdb (since that's how they're stored internally) and then we do the
> validation when we query the confdb instead.
>
> Once the INI parser supports schema validation, we'll move these checks
> there.

Ah, Ok, makes sense then.
I do not know the code in details so I do not feel myself qualified to
ack it but looks Ok at the first glance.
> > >
> > > sssd-devel mailing list
> > > [e-mail]
> > > https://fedorahosted.org/mailman/listinfo/sssd-devel

_______________________________________________
sssd-devel mailing list
[e-mail]
https://fedorahosted.org/mailman/listinfo/sssd-devel

[Stephen Gallagher]

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

On 08/31/2009 03:04 PM, Dmitri Pal wrote:
> Stephen Gallagher wrote:
> > On 08/31/2009 02:56 PM, Dmitri Pal wrote:
> > > Simo Sorce wrote:
> > > > Since we enumerate in background and use the member/memberof scheme you
> > > > can't do group enumerations without also doing a user enumeration
> >
> > first.
> > > > So turn the enumeration option into a boolean with warnings if the old
> > > > syntax is used.
> > > >
> > > > Simo.
> > >
> > > Simo, I think this patch is missing the part where it reads the INI file
> > > and sticks initial data into the database.
> > > It should have same logic:
> > > try to read as int, if success then probably stick as int in the sysdb
> > > otherwise try to read as bool and stick bool in.
> >
> > We discussed this, but currently we read all entries as strings into the
> > confdb (since that's how they're stored internally) and then we do the
> > validation when we query the confdb instead.
> >
> > Once the INI parser supports schema validation, we'll move these checks
> > there.
>
> Ah, Ok, makes sense then.
> I do not know the code in details so I do not feel myself qualified to
> ack it but looks Ok at the first glance.
> > > >
> > > > sssd-devel mailing list
> > > > [e-mail]
> > > > https://fedorahosted.org/mailman/listinfo/sssd-devel
>
> sssd-devel mailing list
> [e-mail]
> https://fedorahosted.org/mailman/listinfo/sssd-devel

Ack and pushed to master with one minor typo correction. (Enumeration
instead of enunmeration)

- -- 
Stephen Gallagher
RHCE 804006346421761

Looking to carve out IT costs?
www.redhat.com/carveoutcosts/
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.9 (GNU/Linux)
> Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
>
> iEYEARECAAYFAkqcIBcACgkQeiVVYja6o6PynwCfc0mysvVpR5dfScoDDzq2WKk+
> V88AoJ2aIh2FQAtAOjLTNr+222+M1cT6
> =ps6T
> -----END PGP SIGNATURE-----