[TYPO3-core] RFC: Feature Request #5530: Hook in TCEmain after all database operations

Ingmar Schlecht ingmar at typo3.org
Sat Nov 24 20:20:57 CET 2007


Hi Masi,

Martin Kutschker schrieb:
> Ingmar Schlecht schrieb:
>>
>> I read Nils' and Oliver Klee's replies before your post, which is why I
>> just pointed out to them that hooks are not allowed for 4.1.
> 
> 4.1 is Michael Stucki's "baby". He always objected to new hooks. Please
> ask again when he's back.

No, 4.0 was Stucki's release. I was the 4.1 release manager and am still
responsible for that branch. We settled these responsibilities on the
last core dev meeting.

> The main point was that an end user must not know the TYPO3 patch level
> for any new features. And if an extension requires a hook, it's a
> feature. We could add numerous mini features otherwise.

The main point I see in not allowing new features for patch level
versions is that the stability might suffer when you add new features
there - yet I really don't see how hooks affect the stability if they're
as well tested as this one and only get active if used.

Anyway, the point that the "end user must not know the TYPO3 patch
level" might be a valid one, even though when if it's about hooks I
would clearly favour adding hooks in stable branches to having lots of
XCLASSes in the TER.

> To me it seems it would be better to realease TYPO3 more often (eg 10
> months instead of 14), than start the add-new-feature-policy-discussion
> again and again. If I look at PHP I shudder. Features creep in any time,
> so you have not to as for 5.1.x but for 5.1.2 or 5.1.7 etc which really
> sucks.

Hmm, what's wrong with checking for > 4.1.4 for using a hook and fall
back to XCLASSes for lower versions in an extension?

I think we can live with a -1 for this hook in 4.1, but I would like to
come back to this topic at the next core dev meeting (developer days in
may 2008).

cheers
Ingmar

-- 
Ingmar Schlecht
TYPO3 Association Active Member


More information about the TYPO3-team-core mailing list