Just a mail to eventually awaken consciousness about what is coming towards us and initiate a process where we could debate what to do, eventually come out with a list a ticket and potentially target a next release dedicated to it, (eventually a 0.10).
One first very important task may be to already configure Travis CI to use these new versions !
Mongo 2.8 is already in release candidate phase (rc2 is out, rc4 is expected for X-Mas) so expected release is very beginning of the year 2015
Here are the development release notes: http://docs.mongodb. org/manual/release-notes/2.8/
We should support following:
- add tests for new storage driver (also compressed)
- add possibility to define storage for collections and indexes
- Add SCRAM-SHA-1 authentication mechanism
- Investigate what we could get more or not from the new serverStatus
PyMongo 2.8 is in front of the door, rc1 is out, due date is 06/Jan/15
Here are the release notes: http://api.mongodb.org/ python/current/changelog.html
We should try to support following
- Index name to hint()
- Verify hint is supported in count()
- Evaluate performance improvement possibilities of decode_iter()
- Support $numberLong (new field ?)
Shortly after should be release PyMongo 3.0, which was earlier thought to be the next release, before a 2.8 was decided, actual due date is 30/Jan/15 (but probably only later)
Here we should:
- Adapt to breaking API changes: https://jira.mongodb.org/
browse/PYTHON-526 - BSON int64 https://jira.mongodb.
org/browse/PYTHON-707 - Performance test for the rewritten BSON https://jira.mongodb.org/
browse/PYTHON-346 - Make sure we use only MongoClient (all the other connection methods are deprecated)
I could eventually create issues for these points in order to know on which PR to work in the future, but I wanted to discuss here what is the desired approach to work on it, if it is maybe preferred to not create too many tickets and if a special label (eventually a release version) could be good.
Thanks for your feedback,
댓글 없음:
댓글 쓰기