This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author r.david.murray
Recipients BreamoreBoy, fdrake, jjlee, kxroberto, orsenthil, r.david.murray, terry.reedy
Date 2010-08-24.13:13:39
SpamBayes Score 6.3524574e-05
Marked as misclassified No
Message-id <1282655621.13.0.0372652370816.issue1486713@psf.upfronthosting.co.za>
In-reply-to
Content
2.6 is now in security-fix-only mode.  Since this is a new feature, it can only go into 3.2.

Can you provide a patch against py3k trunk?

I've only glanced at the patch briefly, but one thing that concerns me is 'warning file'.  I suppose that either the logging module or perhaps the warnings module should be used instead.  We should look at how other stdlib modules handle this kind of thing.  Or perhaps warnings shouldn't be generated at all, since the default will be strict and therefore the programmer has consciously selected tolerant mode.

One stdlib model we could follow is the model of the email module: have a 'defects' attribute that collects the errors.  email6, by the way, is going to have both 'tolerant' and 'strict' modes, and in that case the default is tolerant (and always has been) in respect for Postel's law, which is enshrined in the email RFCs.  If the HTTP standards have a similar recommendation to accept "dirty" input when possible, we could make an argument for changing HTMLParser's default to tolerant.
History
Date User Action Args
2010-08-24 13:13:41r.david.murraysetrecipients: + r.david.murray, fdrake, terry.reedy, jjlee, orsenthil, kxroberto, BreamoreBoy
2010-08-24 13:13:41r.david.murraysetmessageid: <1282655621.13.0.0372652370816.issue1486713@psf.upfronthosting.co.za>
2010-08-24 13:13:39r.david.murraylinkissue1486713 messages
2010-08-24 13:13:39r.david.murraycreate