The latest changes to the redirect handler couldn't possibly have been
tested, because they did not compute a newurl and failed with a
NameError.  The __name__ == "__main__": block has a test for
redirects.

Also, fix SF bug 723831.  A urlopen() that failed because the host was
not found raised a socket.gaierror unlike earlier versions of
urllib2.  The problem is that httplib actually establishes the
connection at a different point starting with Python 2.2.  Move the
try/except to endheaders(), which is where the connection gets
established.

  and adds.

* Note that dictionary iteration is negatively impacted by additional
  sparseness.

Added some docstrings.

only.  Repaired, and added new tests to test_file.py.

including the "Callback Options" section).

Remove first-person usage from it.  Various wording tweaks.

Particularly important since they now store True and False rather than 1
and 0!

This should be turned into a footnote by someone who actually
understands OSS and its relationship to Linux, FreeBSD, OpenBSD, NetBSD,
etc.

Remove whitespace around em-dashes.

correctly.  (Closes SF #731689.)  Use \emph{} in a few places.

be taken, and define NI_MAX{HOST|SERV} if necessary.

there or where to find it.

  the itertoolsmodule.
* Taught itertools.repeat(obj, n) to treat negative repeat counts as
  zero.  This behavior matches that for sequences and prevents
  infinite loops.

  this causes some browsers to offer descriptive tooltips when the
  mouse hovers over the linked icon
- clean up the code in this area

(Jack: maybe this could/should be done in Framework.py instead, that way
apps don't have to think about it. PackMan springs to mind...)

based threads, release the GIL when the event loop goes to sleep,
and acquire it again when the event loop wakes up again. OSX-only.

effect of an import. (This is one step towards threading support in
the IDE.)

Simplify test_main().

a bug in urllib2 this caused pimp to stop working. Fixed the URL to
make it work again.

- implement custom_REL_hook() so LaTeX2HTML uses currently recommended
  link types; those work fine with tools like the Mozilla Site
  Navigation Bar

It is better to be explicit and just allow stop to be None.