Re: 1.10alpha 'release'

[Johannes S. <ya...@gm...>]

Michael Ströder wrote:
> 
> 1. Testing:
> I tested the version of Johannes and it seems to work. I tested all
> operations (search, modify, add, modrdn, delete). Hmm, I'm thinking
> about writing a Python script which implements the BLITS test suite
> on a defined data set. This is a pretty boring job but would ease
> testing in the future.

Please also test error conditions/exceptions. It would be nasty
if a missing INCREF/surplus DECREF would make it dump core on
error conditions.

> Can anybody else say something about how stable the current version
> is?

I just sent some minor corrections to David.

> I'm still not confident regarding the naming. IMHO the C-module
> should still be called ldapmodule.so and fog's stuff should be
> called ldaplib or whatever to be able to distribute both modules
> separately without having to mess with different ldap.py. Different
> ldap.py will cause nothing but troubles and boring FAQs by users.

IMHO _ldapmodule.so + ldap.py is more flexible because it makes
it easier to add stuff to "the ldap module" in python.
ldapmodule.so requires enhancements to be implemented in C.

But Michael's right: You *MUST* include a warning about deleting
the old ldapmodule.so (or ldap.dll) in the INSTALL/README
document. It may even be a good idea to add a micro-FAQ to the
distribution.

> And I know that there's "from _ldap import *" in ldap.py but this
> does not import _ldap.__version__ and using "from module import *"
> is bad behaviour anyway.

"from module import *" is bad in user code because it a) makes it
diffcult to see which symbol was imported from where, and b)
makes namespace collisions more likely.
But it is IMHO OK (and a common idiom) for C-wrapper/interface
modules. The added flexibility for future enhancements is
invaluable.

Johannes