- removed all dependencies on that support

This should repair PySECURITY_ATTRIBUTES problems that
(mostly) Plone users report using recent Zopes on
Windows.

This change is in the 2.8.4 Windows installer I just
published on zope.org.  Alas, the 2.8.4 tarball there
(and the 2.8.4 tag) differs from the source used to build
the 2.8.4 Windows installer (by this one character,
bumping W32ALLVERSION in inst/WinBuilders/mk/python.mk).

- replaced url_quote by html_quote for add list values (pushing the add button escapes the value again)

there were pyc files around. Sigh.

Added tests for restrictions on augmented assignment and for handling
generator expressions.

object through an augmented assignment (aka inplace) operator.

object through an augmented assignment (aka inplace) operator.

Also added handling for generator expressions.

never need these again, but that's what I said the last time I had to
rediscover how this worked. :)

Collector #1926:  add tests for both not-yet-encrypted and pre-encrypted passwords handed to '_doAddUser'.

enabled.

No commit message

  non-existing or closed database connection next to the Connection ID
  dropdown and present it using red to increase its visibility.

      - Collector #1895: testrunner: omitting the 'var' from recursive
        directory walking

which was in the 2.8.2 release but not noted in CHANGES.txt.

      - OFS.Image.manage_FTPget() would str() it's .data attribute,
        potentially loading the whole file in memory as a
        string. Changed to use RESPONSE.write() iterating through the
        Pdata chain, just like index_html().

No commit message

o Forward-port from 2.7 branch.

Merge of fix for #1490 to 2.8 branch: new zope.conf option for controlling the encoding of data sent to ZPublisher without any specified encoding.

      - disabled ".. include" directive for all the ZReST product and the 
        reStructuredText package

File._read_data():  This uses savepoints just to get oids assigned
and to free memory, with no possibility of rollback.  Therefore
"optimistic" savepoints should be used:  there's no reason here to
insist that any data managers involved support rollback.