Gregory Szorc <gregory.szorc@gmail.com> [Tue, 06 Feb 2018 11:08:36 -0800] rev 36015
sshpeer: initial definition and implementation of new SSH protocol
The existing SSH protocol has several design flaws. Future commits
will elaborate on these flaws as new features are introduced
to combat these flaws. For now, hopefully you can take me for my
word that a ground up rewrite of the SSH protocol is needed.
This commit lays the foundation for a new SSH protocol by defining
a mechanism to upgrade the SSH transport channel away from the
default (version 1) protocol to something modern (which we'll call
"version 2" for now).
This upgrade process is detailed in the internals documentation
for the wire protocol. The gist of it is the client sends a
request line preceding the "hello" command/line which basically
says "I'm requesting an upgrade: here's what I support." If the
server recognizes that line, it processes the upgrade request and
the transport channel is switched to use the new version of the
protocol. If not, it sends an empty response, which is how all
Mercurial SSH servers from the beginning of time reacted to unknown
commands. The upgrade request is effectively ignored and the client
continues to use the existing version of the protocol as if nothing
happened.
The new version of the SSH protocol is completely identical to
version 1 aside from the upgrade dance and the bytes that follow.
The immediate bytes that follow the protocol switch are defined to
be a length framed "capabilities: " line containing the remote's
advertised capabilities. In reality, this looks very similar to
what the "hello" response would look like. But it will evolve
quickly.
The methodology by which the protocol will evolve is important.
I'm not going to introduce the new protocol all at once. That would
likely lead to endless bike shedding and forward progress would
stall. Instead, I intend to tricle out new features and diversions
from the existing protocol in small, incremental changes.
To support the gradual evolution of the protocol, the on-the-wire
advertised protocol name contains an "exp" to denote "experimental"
and a 4 digit field to capture the sub-version of the protocol.
Whenever we make a BC change to the wire protocol, we can increment
this version and lock out all older clients because it will appear
as a completely different protocol version. This means we can incur
as many breaking changes as we want. We don't have to commit to
supporting any one feature or idea for a long period of time. We
can even evolve the handshake mechanism, because that is defined
as being an implementation detail of the negotiated protocol version!
Hopefully this lowers the barrier to accepting changes to the
protocol and for experimenting with "radical" ideas during its
development.
In core, sshpeer received most of the attention. We haven't even
implemented the server bits for the new protocol in core yet.
Instead, we add very primitive support to our test server, mainly
just to exercise the added code paths in sshpeer.
Differential Revision: https://phab.mercurial-scm.org/D2061
# no-check-commit because of required foo_bar naming
Gregory Szorc <gregory.szorc@gmail.com> [Tue, 06 Feb 2018 10:51:15 -0800] rev 36014
internals: refactor wire protocol documentation
Upcoming work will introduce a new version of the HTTP and SSH
transports. The differences will be significant enough to consider
them new transports. So, we now attach a version number to each
transport.
In addition, having the handshake documented after the transport
and in a single shared section made it harder to follow the flow
of the connection. The handshake documentation is now moved to the
protocol section it describes. We now have a generic section about
the purpose of the handshake, which was rewritten significantly.
Differential Revision: https://phab.mercurial-scm.org/D2060
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 18:04:40 +0100] rev 36013
revlog: rename 'self.checkinlinesize' into '_enforceinlinesize'
The name change has two motivations:
1) The function has no external caller, so we move it to protected space.
2) the function does more than checking it also split the data if we have more
data than 'inline' supports.
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 17:57:29 +0100] rev 36012
revlog: add a _datareadfp context manager for data access needs
The helper handles:
1) is there a file handle already open that we shall just reuse,
2) is the revlog inlined or not.
Using a context manager for all read access will help setting up file pointer
caching in later changesets.
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 17:35:14 +0100] rev 36011
revlog: use context manager for data file lifetime in checksize
This is clearer, safer and more modern.
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 17:34:57 +0100] rev 36010
revlog: use context manager for index file lifetime in checkinlinesize
This is clearer, safer and more modern.
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 17:34:47 +0100] rev 36009
revlog: use context manager for data file lifetime in checkinlinesize
This is clearer, safer and more modern.
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 17:34:19 +0100] rev 36008
revlog: use context manager for index file life time in __init__
This is clearer, safer and more modern.
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 17:22:13 +0100] rev 36007
revlog: move index file opening in a method
Having file operation centralized into a single spot help to factor common
logic out (eg: special flag handling according to the mode).
It is also the first step to simplify file handling during batch operation
(eg: revlog cloning). However, that part does not seems to be a hotspot yet.
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 17:03:56 +0100] rev 36006
revlog: move datafile opening in a method
Having file operation centralized into a single spot help to factor common
logic out.
It is also the first step to simplify file handling during batch operation
(eg: revlog cloning). However, that part does not seems to be a hotspot yet.
Boris Feld <boris.feld@octobus.net> [Tue, 06 Feb 2018 12:25:11 +0100] rev 36005
parseindex: implement context manager method on the wrapper
This is needed for incoming cleanups.
Boris Feld <boris.feld@octobus.net> [Tue, 06 Feb 2018 11:52:37 +0100] rev 36004
parseindex: also forward keyword argument in a debug wrapper
Otherwise, it gets in the way of a coming refactoring.
Boris Feld <boris.feld@octobus.net> [Tue, 06 Feb 2018 11:51:39 +0100] rev 36003
dumprevlog: handle being passed a mode parameter
This makes it closer to an actual file opener. Otherwise, it gets in the way
of a coming refactoring.
Boris Feld <boris.feld@octobus.net> [Tue, 06 Feb 2018 11:08:41 +0100] rev 36002
changelog: add the missing 'closed' property on 'appender' object
This mimic file object further.
Boris Feld <boris.feld@octobus.net> [Mon, 05 Feb 2018 20:36:51 +0100] rev 36001
changelog: implement context manager method for 'appender' object
This object mimics a file and was missing the context manager API.
Joerg Sonnenberger <joerg@bec.de> [Sun, 04 Feb 2018 00:29:22 +0100] rev 36000
diff: improve ui.write performance when not coloring on Windows
Differential Revision: https://phab.mercurial-scm.org/D2022
Augie Fackler <augie@google.com> [Tue, 06 Feb 2018 05:25:36 -0500] rev 35999
hgsh: enable clang-format
Nothing looks awful, so we can just turn it on.
Differential Revision: https://phab.mercurial-scm.org/D2059
Augie Fackler <augie@google.com> [Tue, 06 Feb 2018 05:24:28 -0500] rev 35998
chg: enable clang-format on all .c and .h files
Nothing in here looks awful, so I think we may as well just do it.
# skip-blame because it's just reformatting with no functionality change
Differential Revision: https://phab.mercurial-scm.org/D2058
Augie Fackler <augie@google.com> [Wed, 07 Feb 2018 06:28:43 -0500] rev 35997
python3: whitelist an additional 11 tests
I think these are mostly the result of Pulkit's recent work. Thanks!
Differential Revision: https://phab.mercurial-scm.org/D2078
Yuya Nishihara <yuya@tcha.org> [Mon, 05 Feb 2018 20:48:51 +0900] rev 35996
ui: add explicit path to write prompt text bypassing buffers
The prompt= flag was added at e35d7f131483, when colorui had its own write()
function. Since we've merged colorui to ui, we can simply call the unbuffered
write() function.
Yuya Nishihara <yuya@tcha.org> [Mon, 05 Feb 2018 20:40:49 +0900] rev 35995
ui: write prompt text in ui.prompt(), not in ui._readline()
self.label() is replaced by label= option, which should make it clearer why
we can't pass the text to raw_input(prompt).
Yuya Nishihara <yuya@tcha.org> [Sun, 21 Jan 2018 14:00:56 +0900] rev 35994
logcmdutil: mark changesetprinter.showpatch() as private
Yuya Nishihara <yuya@tcha.org> [Sun, 21 Jan 2018 14:28:03 +0900] rev 35993
logcmdutil: drop default arguments from changesetdisplayer/templater() calls
Yuya Nishihara <yuya@tcha.org> [Sun, 21 Jan 2018 13:47:06 +0900] rev 35992
logcmdutil: make default parameters of changesetprinters consistent
Yuya Nishihara <yuya@tcha.org> [Sun, 04 Feb 2018 10:33:14 +0900] rev 35991
mdiff: use str.startswith/endswith() instead of slicing
Yuya Nishihara <yuya@tcha.org> [Sun, 04 Feb 2018 10:28:03 +0900] rev 35990
patch: unify check_binary and binary flags
Follows up 079b27b5a869. If opts.text=True, check_binary is ignored, so we
can just pass the binary flag to unidiff().
perfunidiff now takes any inputs as text files, which I think is a desired
behavior.
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 07 Feb 2018 13:49:02 +0530] rev 35989
py3: make sure we open the file in bytes mode
# skip-blame because we are just adding b
Differential Revision: https://phab.mercurial-scm.org/D2077
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 07 Feb 2018 13:47:03 +0530] rev 35988
py3: use bytes instead of str
Differential Revision: https://phab.mercurial-scm.org/D2076
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 07 Feb 2018 13:34:51 +0530] rev 35987
py3: add b'' prefixes to string literals in test files
# skip-blame because we are just adding b''
Differential Revision: https://phab.mercurial-scm.org/D2075
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 07 Feb 2018 13:21:06 +0530] rev 35986
py3: use range instead of xrange
xrange is not available on Python 3.
Differential Revision: https://phab.mercurial-scm.org/D2074
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 07 Feb 2018 13:19:19 +0530] rev 35985
py3: use open() instead of file()
file() is not present in Python 3.
Differential Revision: https://phab.mercurial-scm.org/D2073
Pulkit Goyal <7895pulkit@gmail.com> [Wed, 07 Feb 2018 13:17:19 +0530] rev 35984
py3: use pycompat.bytestr to convert _b85chars to bytes
The tranformer does append b'' to the value and make that a bytes but bytes in
Python 3 returns the ascii value on getting characters using indexing.
Characters of this string are queried using indexing multiple times in the file
and to support that we use pycompat.bytestr which returns the bytechrs using
indexing.
Differential Revision: https://phab.mercurial-scm.org/D2072