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 loewis
Recipients
Date 2006-04-25.06:21:57
SpamBayes Score
Marked as misclassified
Message-id
In-reply-to
Content
Logged In: YES 
user_id=21627

This implementation gives me many questions:
- what is the rationale for keeping a separate keys
dictionary? If applications wanted that, couldn't they
maintain a key dictionary themselves? It just seems strange
to rely on hashing inside a rbtree implementation.
- Currently, you need two RichCompare invocations for
insertion/contains. Couldn't that be reduced to just one
comparison if __cmp__ is used?
- for key functions, wouldn't it usually be better to cache
the key?
- shouldn't tp_clear also be implemented?
- this entire code needs many more comments. For example,
why does the forward iterator first follow the right-most
links? I would have expected that it needs to start with the
leftmost node first (and somehow, it does - but why?)
History
Date User Action Args
2007-08-23 15:44:13adminlinkissue1324770 messages
2007-08-23 15:44:13admincreate