Tres Seaver wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Martin Aspeli wrote:
Hi guys,
After a lot of "is-this-a-bug" type discussions with Rob and Wichert,
I've come to feel pretty strongly about the following:
When you load an extension profile for the first time in GS, it looks to
see if it has any import steps (in import_steps.xml) that are not
already "known". If so, it adds them to the import step registry and
then runs all steps in the registry.
Any time a profile (the same, or another one) is run subsequently, that
import step will be run too. In Rob's words, the import steps and the
profile that they came from (import_steps.xml-wise) get decoupled.
We normally don't mind too much, because steps are run in the context of
a particular profile, and if that context has no matching XML file,
nothing is loaded. At least so we assume - there's of course nothing to
stop the setuphandler from having side effects even if the XML file is
not found. We *do* see a problem with importVarious-type hacks, though.
Here, people resort to checking for a "flag" (normally an empty text
file) to determine if the setuphandler should be run.
'importVarious' is a brutal hack: better to focus efforts on making it
disappear. The entire point of the tool is to externalize configuration
as declarative data in the profile; accomodating imperative
configuration is not something I care to support.
The 'upgrade' stuff Rob ported in from CPS is the only sane case I know
of for imperative action.
while i agree with this in principal, i think the truth of the matter is that
plenty of relatively inexperienced developers will be coding GenericSetup
profiles. these developers are the ones most likely to write imperative
import steps (b/c they won't understand how to write a new XML parser, and it
will seem like a lot of extra work compared to just twiddling a setting
somewhere). they are also the _least_ likely to understand the subtleties of
GS's behaviour.
the problem is exacerbated by the symptoms; it's not that there will
necesarily be problems running the profile in question. it's when you try to
import OTHER profiles that there will be problems, because these naively coded
steps will be performing actions when they shouldn't be.
this has already shown itself to be the most misunderstood aspect of
GenericSetup, i've had to explain the behaviour over and over again. the
documentation can be improved, sure, but we all know that people will be
bitten by this no matter how clearly this issue is spelled out.
i'm pretty much in support of the proposal as it's evolved so far. here's my
understanding of it at this point:
- the import step registry will record the profile from which a given import
step is included. if multiple profiles include the same step in their
import_steps.xml, they will both be recorded as related to that import step.
- when a profile is imported, only the import steps from the base profile and
all transitively dependent profiles will be run. (clearly this depends on
actually finishing the dependency support.)
- when a profile is exported (or a snapshot is generated), all of the
registered steps will be used.
this way, a naively coded import handler can only cause a problem when the
profile it is associated with (or any other profile that explicitly depends on
the problematic profile) is run. writing imperative handlers should still be
discouraged, but the damage that they can cause is contained.
-r
_______________________________________________
Zope-CMF maillist - Zope-CMF@lists.zope.org
http://mail.zope.org/mailman/listinfo/zope-cmf
See http://collector.zope.org/CMF for bug reports and feature requests
