Author ncoghlan
Recipients eli.bendersky, ethan.furman, flox, jcea, jkloth, ncoghlan, python-dev, scoder
Date 2013-08-27.22:06:21
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <CADiSq7c_R5cdUneFDGW+SEt+Y7bJuum8E5C02nWSvtDgyyYPPg@mail.gmail.com>
In-reply-to <1377613343.57.0.834334476055.issue17741@psf.upfronthosting.co.za>
Content
Stefan, your proposed merged design isn't going to happen. Two alternative
ways of using the one class is far more complicated to explain to users
than exposing a helper that composes the existing API components to address
a particular use case (particularly when the helper is hiding the fact that
the theoretical underlying composition doesn't currently work the way it
should according to the public APIs).

As a higher level helper, users that are comfortable with and prefer the
lower level API are then clearly free to ignore. This recasting of the
nature of the API as a composition of the lower level components is a
*direct* response to your concern about it being a capability that the
lower level API should support directly.

You clearly don't like that design choice, but this isn't a democracy.
Arguing further on this point will just be ranting for the sake of ranting,
rather than having any chance of changing Eli's mind as module maintainer.

As far as naming goes, I still think it's better to drop "Parser" from the
name entirely. Call it "XMLEventReader" or something like that. The point
is to use it when you want to get the events out without blocking when no
events have been triggered and don't want to set up your own class to
receive event notifications through the target push API. The docs should
make it clear that this API is "pull only" - if people need callbacks or
other push triggers, then that's still what custom targets are for.
History
Date User Action Args
2013-08-27 22:06:21ncoghlansetrecipients: + ncoghlan, jcea, scoder, jkloth, eli.bendersky, flox, ethan.furman, python-dev
2013-08-27 22:06:21ncoghlanlinkissue17741 messages
2013-08-27 22:06:21ncoghlancreate