From: "Faizel Dakri" <webobjects-dev@wocommunity.org>
Received-SPF: pass
 receiver=post.selbstdenker.com; client-ip=64.147.123.20; envelope-from=listfez@dakri.com
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (1.0)
Subject: Re: [WO-DEV] ERXGenericRecord takeValueForKey
Date: Mon, 30 Aug 2021 09:28:55 -0500
Message-Id: <548DCB9C-77E0-4D13-A278-B92DE085F706@dakri.com>
References: <list-26355634@post.selbstdenker.com>
In-Reply-To: <list-26355634@post.selbstdenker.com>
To: WebObjects & WOnder Development <webobjects-dev@wocommunity.org>

If your EOs are inheriting from EOGenericRecord, they most likely are implem=
enting EOKeyValueCoding, which is an extension to NSKeyValueCoding that uses=
 a couple of special methods to set/retrieve model properties:

	- takeStoredValueForKey()
	- storedValueForKey()

If you want your EO to behave using standard NSKeyValueCoding rules, try ove=
rriding canAccessFieldsDirectly() in your EO class (static method) to return=
 false.=20

F

--=20
Faizel Dakri


> On Aug 25, 2021, at 8:50 PM, webobjects-dev@wocommunity.org wrote:
>=20
> =EF=BB=BFHi there,
>=20
> I must be missing something darn obvious; looks like I've missed EOF 101 s=
omehow :/
>=20
> My EO class, which extends ERXGenericRecord, does not implement modelled p=
roperty setters. Thus, I assumed takeValueForKey would always fall through h=
andleTakeValueForUnboundKey. Seems not.
>=20
> I've overridden the methods like this:
>=20
> =3D=3D=3D
> class OCSEnterpriseObject extends ERXGenericRecord {
>  static boolean canAccessFieldsDirectly { NO }
>  void takeValueForKey(value,String key) {
>    println "???takeValueForKey $key"
>    //super.takeValueForKey(value, key) // precisely same behaviour as with=
 the thing below, no difference
>    NSKeyValueCoding.DefaultImplementation.takeValueForKey(this,value,key)
>  }
>  void handleTakeValueForUnboundKey(value,String key) {
>    println "???handleTakeValueForUnboundKey $key"
>    // ... some stuff to simulate setter (eventually using takeStoredValueFo=
rKey), not relevant, for it is never called
>  }
>  ...
> }
> =3D=3D=3D
>=20
> It does NOT work as (I've na=C3=AFvely) presumed.
>=20
> When eo.takeValueForKey(o,key) is called, I can always see =E2=80=9C???tak=
eValueForKey=E2=80=9D (naturally), but I *never ever* see =E2=80=9C???handle=
TakeValueForUnboundKey=E2=80=9C for any key amongst modelled attributes. Not=
hing other (known to me) gets called =E2=80=94 not even takeStoredValueForKe=
y, which I've also overridden to log =E2=80=94, but the value is set all rig=
ht in the EO.
>=20
> Only if foo is a random (non-existing, non-modelled) string, I can see =E2=
=80=9C???handleTakeValueForUnboundKey=E2=80=9C.
>=20
> Looks like there's some trick in NSKeyValueCoding.DefaultImplementation.ta=
keValueForKey which recognises modelled attributes in EOGenericRecord and fi=
ll their values directly, without falling to handleTakeValueForUnboundKey?!?=
 Weird.
>=20
> Or I must be doing something very far wrong.
>=20
> Is this documented somewhere? Far as I can say, neither ERXGenericRecord n=
or EOGenericRecord nor NSKeyValueCoding.DefaultImplementation mentions anyth=
ing special; NSKeyValueCoding.DefaultImplementation.takeValueForKey should s=
imply check for the accessor (either with or without the _ prefix) and if th=
ere is none (which there indeed is not), fall to handleTakeValueForUnboundKe=
y. What am I missing here?
>=20
> Thanks,
> OC
>=20
>=20
>=20
> #############################################################
> This message is sent to you because you are subscribed to
>  the mailing list <webobjects-dev@wocommunity.org>.
> To unsubscribe, E-mail to: <webobjects-dev-off@wocommunity.org>
> To switch to the DIGEST mode, E-mail to <webobjects-dev-digest@wocommunity=
.org>
> To switch to the INDEX mode, E-mail to <webobjects-dev-index@wocommunity.o=
rg>
> Send administrative queries to  <webobjects-dev-request@wocommunity.org>
>=20
>=20