win32: clarify comment regarding use of fdopen

win32: allow catching of both pywintypes.error and WindowsError

atomictempfile: delegate to posixfile instead of inheriting from it

merge with mpm

inotify: inotify.server.walk() simplify control flow

inotify: inotify.server.walk*() remove unnecessary var Remove hginside var and the test it relates to: not( top or not hginside ) == (not top) and hginside, so the only case when nothing will be yielded is when hginside is True and top is False. Because of the returns placed upstream, this case will not happen anymore. We can then safely remove hginside and the (if)s

inotify: inotify.server.walk*() cleanup When not in root repo, if we meet a .hg, bail out. break was used, but return can be safely used instead: if we go through break then (top or not hginside) == False -> nothing is done after the for loop

inotify: inotify.server.walkrepodirs() simplify walking Do not 'yield tuple, boolean' to filter later on the boolean. Test the boolean first, and yield tuple only if needed.

inotify: inotify.server.walkrepodirs() simplify compute string concatenation only once

inotify: inotify.server.walk() simplify algorithm Do not yield (bool, tuple) in a subfunction and check later: 'if not bool: yield tuple'. Instead simplify so the tuple doesnt get yielded on the first time, to avoid systematic checks.

inotify: inotify.server.walk cleanups * Removing useless path computation * moving 'reporoot' computation down to the first call to avoid confusion with the parameter used in walkit()

revlog: add cache priming for reconstructing delta chains

revlog: use chunk cache to avoid rereading when splitting inline files

revlog: clean up the chunk caching code

revlog: use index to find index size

revlog: preread revlog .i file Smaller revlogs can be read with a single read, do it on open.