view tests/test-bdiff.py @ 30764:e75463e3179f

protocol: send application/mercurial-0.2 responses to capable clients

With this commit, the HTTP transport now parses the X-HgProto-<N>
header to determine what media type and compression engine to use for
responses. So far, we only compress responses that are already being
compressed with zlib today (stream response types to specific
commands). We can expand things to cover additional response types
later.

The practical side-effect of this commit is that non-zlib compression
engines will be used if both ends support them. This means if both
ends have zstd support, zstd - not zlib - will be used to compress
data!

When cloning the mozilla-unified repository between a local HTTP
server and client, the benefits of non-zlib compression are quite
noticeable:

  engine     server CPU (s)   client CPU (s)    bundle size
zlib (l=6)      174.1            283.2         1,148,547,026
zstd (l=1)       99.2            267.3         1,127,513,841
zstd (l=3)      103.1            266.9         1,018,861,363
zstd (l=7)      128.3            269.7           919,190,278
zstd (l=10)     162.0               -            894,547,179
none             95.3            277.2         4,097,566,064

The default zstd compression level is 3. So if you deploy zstd
capable Mercurial to your clients and servers and CPU time on
your server is dominated by "getbundle" requests (clients cloning
and pulling) - and my experience at Mozilla tells me this is often
the case - this commit could drastically reduce your server-side
CPU usage *and* save on bandwidth costs!

Another benefit of this change is that server operators can install
*any* compression engine. While it isn't enabled by default, the
"none" compression engine can now be used to disable wire protocol
compression completely. Previously, commands like "getbundle" always
zlib compressed output, adding considerable overhead to generating
responses. If you are on a high speed network and your server is under
high load, it might be advantageous to trade bandwidth for CPU.
Although, zstd at level 1 doesn't use that much CPU, so I'm not
convinced that disabling compression wholesale is worthwhile. And, my
data seems to indicate a slow down on the client without compression.
I suspect this is due to a lack of buffering resulting in an increase
in socket read() calls and/or the fact we're transferring an extra 3 GB
of data (parsing HTTP chunked transfer and processing extra TCP packets
can add up). This is definitely worth investigating and optimizing. But
since the "none" compressor isn't enabled by default, I'm inclined to
punt on this issue.

This commit introduces tons of tests. Some of these should arguably
have been implemented on previous commits. But it was difficult to
test without the server functionality in place.

from __future__ import absolute_import, print_function
import collections
import struct
import unittest

import silenttestrunner

from mercurial import (
    bdiff,
    mpatch,
)

class diffreplace(
    collections.namedtuple('diffreplace', 'start end from_ to')):
    def __repr__(self):
        return 'diffreplace(%r, %r, %r, %r)' % self

class BdiffTests(unittest.TestCase):

    def assert_bdiff_applies(self, a, b):
        d = bdiff.bdiff(a, b)
        c = a
        if d:
            c = mpatch.patches(a, [d])
        self.assertEqual(
            c, b, ("bad diff+patch result from\n  %r to\n  "
                   "%r: \nbdiff: %r\npatched: %r" % (a, b, d, c[:200])))

    def assert_bdiff(self, a, b):
        self.assert_bdiff_applies(a, b)
        self.assert_bdiff_applies(b, a)

    def test_bdiff_basic(self):
        cases = [
            ("a\nc\n\n\n\n", "a\nb\n\n\n"),
            ("a\nb\nc\n", "a\nc\n"),
            ("", ""),
            ("a\nb\nc", "a\nb\nc"),
            ("a\nb\nc\nd\n", "a\nd\n"),
            ("a\nb\nc\nd\n", "a\nc\ne\n"),
            ("a\nb\nc\n", "a\nc\n"),
            ("a\n", "c\na\nb\n"),
            ("a\n", ""),
            ("a\n", "b\nc\n"),
            ("a\n", "c\na\n"),
            ("", "adjfkjdjksdhfksj"),
            ("", "ab"),
            ("", "abc"),
            ("a", "a"),
            ("ab", "ab"),
            ("abc", "abc"),
            ("a\n", "a\n"),
            ("a\nb", "a\nb"),
        ]
        for a, b in cases:
            self.assert_bdiff(a, b)

    def showdiff(self, a, b):
        bin = bdiff.bdiff(a, b)
        pos = 0
        q = 0
        actions = []
        while pos < len(bin):
            p1, p2, l = struct.unpack(">lll", bin[pos:pos + 12])
            pos += 12
            if p1:
                actions.append(a[q:p1])
            actions.append(diffreplace(p1, p2, a[p1:p2], bin[pos:pos + l]))
            pos += l
            q = p2
        if q < len(a):
            actions.append(a[q:])
        return actions

    def test_issue1295(self):
        cases = [
            ("x\n\nx\n\nx\n\nx\n\nz\n", "x\n\nx\n\ny\n\nx\n\nx\n\nz\n",
             ['x\n\nx\n\n', diffreplace(6, 6, '', 'y\n\n'), 'x\n\nx\n\nz\n']),
            ("x\n\nx\n\nx\n\nx\n\nz\n", "x\n\nx\n\ny\n\nx\n\ny\n\nx\n\nz\n",
             ['x\n\nx\n\n',
              diffreplace(6, 6, '', 'y\n\n'),
              'x\n\n',
              diffreplace(9, 9, '', 'y\n\n'),
              'x\n\nz\n']),
        ]
        for old, new, want in cases:
            self.assertEqual(self.showdiff(old, new), want)

    def test_issue1295_varies_on_pure(self):
            # we should pick up abbbc. rather than bc.de as the longest match
        got = self.showdiff("a\nb\nb\nb\nc\n.\nd\ne\n.\nf\n",
                            "a\nb\nb\na\nb\nb\nb\nc\n.\nb\nc\n.\nd\ne\nf\n")
        want_c = ['a\nb\nb\n',
                  diffreplace(6, 6, '', 'a\nb\nb\nb\nc\n.\n'),
                  'b\nc\n.\nd\ne\n',
                  diffreplace(16, 18, '.\n', ''),
                  'f\n']
        want_pure = [diffreplace(0, 0, '', 'a\nb\nb\n'),
                     'a\nb\nb\nb\nc\n.\n',
                     diffreplace(12, 12, '', 'b\nc\n.\n'),
                     'd\ne\n',
                     diffreplace(16, 18, '.\n', ''), 'f\n']
        self.assert_(got in (want_c, want_pure),
                     'got: %r, wanted either %r or %r' % (
                         got, want_c, want_pure))

    def test_fixws(self):
        cases = [
            (" \ta\r b\t\n", "ab\n", 1),
            (" \ta\r b\t\n", " a b\n", 0),
            ("", "", 1),
            ("", "", 0),
        ]
        for a, b, allws in cases:
            c = bdiff.fixws(a, allws)
            self.assertEqual(
                c, b, 'fixws(%r) want %r got %r (allws=%r)' % (a, b, c, allws))

    def test_nice_diff_for_trivial_change(self):
        self.assertEqual(self.showdiff(
            ''.join('<%s\n-\n' % i for i in range(5)),
            ''.join('>%s\n-\n' % i for i in range(5))),
                         [diffreplace(0, 3, '<0\n', '>0\n'),
                          '-\n',
                          diffreplace(5, 8, '<1\n', '>1\n'),
                          '-\n',
                          diffreplace(10, 13, '<2\n', '>2\n'),
                          '-\n',
                          diffreplace(15, 18, '<3\n', '>3\n'),
                          '-\n',
                          diffreplace(20, 23, '<4\n', '>4\n'),
                          '-\n'])

    def test_prefer_appending(self):
        # 1 line to 3 lines
        self.assertEqual(self.showdiff('a\n', 'a\n' * 3),
                         ['a\n', diffreplace(2, 2, '', 'a\na\n')])
        # 1 line to 5 lines
        self.assertEqual(self.showdiff('a\n', 'a\n' * 5),
                         ['a\n', diffreplace(2, 2, '', 'a\na\na\na\n')])

    def test_prefer_removing_trailing(self):
        # 3 lines to 1 line
        self.assertEqual(self.showdiff('a\n' * 3, 'a\n'),
                         ['a\n', diffreplace(2, 6, 'a\na\n', '')])
        # 5 lines to 1 line
        self.assertEqual(self.showdiff('a\n' * 5, 'a\n'),
                         ['a\n', diffreplace(2, 10, 'a\na\na\na\n', '')])

if __name__ == '__main__':
    silenttestrunner.main(__name__)