Updating blobs in save points could cause spurious "invalidations
out of order" errors.  https://bugs.launchpad.net/zodb/+bug/509801

(Thanks to Christian Zagrodnick for chasing this down.)

2.7. :(

created temporarily (not passed to the constructor).

in. When no changes are passed to the constructor, then they are
closed when the storage is closed.

Changed the behavior for closing changes when changes weren't passed
in. When no changes are passed to the constructor, then they are
closed when the storage is closed.

Fixed a test that wasn't testing what it though it was (or anything really).

This required using manuel to get the footnote
feature. Unfortunately, manuel imports zope.testing.doctest. :)
Hpefully, this will be fixed soon.

Updated the testhistoricalconnections.py tearDown to clear conflict
resolution class cache. The cache was causing tests to fail when tests
were run multiple times.

(Also use standard doctest.)

  Applications for this include compression and encryption.  An
  example wrapper storage implementation, ZODB.tests.hexstorage, was
  included for testing.

It is important that storage implementations not assume that
  storages contain pickles.  Renamed IStorageDB to IStorageWrapper and
  expanded it to provide methods for transforming and untransforming
  data records.  Storages implementations should use these methods to
  get pickle data from stored records.

- Deprecated ZODB.interfaces.StorageStopIteration.  Storage
  iterator implementations should just raise StopIteration, which
  means they can now be implemented as generators.

thought this support already existed, but found out I was wrong when I
tried to make it work. :/) Applications for this include compression
and encryption.  An example wrapper storage implementation,
ZODB.tests.hexstorage, was included for testing.

It is important that storage implementations not assume that
storages contain pickles.  Renamed IStorageDB to IStorageWrapper and
expanded it to provide methods for transforming and untransforming
data records.  Storages implementations should use these methods to
get pickle data from stored records.

can fail if the pickle has been transformed and a calling storage
doesn't account for that.  Printing is certainly not the right way to
deal with the problem.

Also cleaned up some whitespace.

iterator implementations should just raise StopIteration, which
means they can now be implemented as generators.

  The filestorage packer configuration option noe accepts calues of
  the form ``modname:expression``, allowing the use of packer
  factories with options.

  The file-storage backup script, repoze, will now create a backup
  index file if an output file name is given via the --output/-o
  option.

(Merged the tseaver-repozo_index branch.)

sucking up a lot of time, as shown in some profile output.

This is to tr to warn people of likely design mistakes.  There is a
new option (large_record_size/large-record-size) to control the
Index: src/CHANGES.txt

methods for transforming and untransforming data records. This was
needed to make storage wrappers that transform data work with
conflict resolution.

- Can use it as a decorator

- Can omit the label, and get the label from the assertion function.

- You can now pass storage keyword arguments to ZODB.DB and
  ZODB.connection.

- You can now pass None (rather than a storage or file name) to get
  a database with a mapping storage.

(transaction-manager attempts method) introduced in the
``transaction`` 1.1.0 release.

made moot by the introduction of multi-version concurrency control
several years ago.

  BTree sets and tree sets didn't correctly check values passed to
  update or to constructors, causing Python to exit under certain
  circumstances.