view tests/test-fastannotate-revmap.py @ 44363:f7459da77f23

nodemap: introduce an option to use mmap to read the nodemap mapping

The performance and memory benefit is much greater if we don't have to copy all
the data in memory for each information. So we introduce an option (on by
default) to read the data using mmap.

This changeset is the last one definition the API for index support nodemap
data. (they have to be able to use the mmaping).

Below are some benchmark comparing the best we currently have in 5.3 with the
final step of this series (using the persistent nodemap implementation in
Rust). The benchmark run `hg perfindex` with various revset and the following
variants:

Before:

 * do not use the persistent nodemap
 * use the CPython implementation of the index for nodemap
 * use mmapping of the changelog index

After:

 * use the MixedIndex Rust code, with the NodeTree object for nodemap access
   (still in review)
 * use the persistent nodemap data from disk
 * access the persistent nodemap data through mmap
 * use mmapping of the changelog index

The persistent nodemap greatly speed up  most operation on very large
repositories. Some of the previously very fast lookup end up a bit slower because
the persistent nodemap has to be setup. However the absolute slowdown is very
small and won't matters in the big picture.

Here are some numbers (in seconds) for the reference copy of mozilla-try:

Revset            Before   After     abs-change speedup

-10000:           0.004622 0.005532   0.000910  ×   0.83
-10:              0.000050 0.000132   0.000082  ×   0.37
tip               0.000052 0.000085   0.000033  ×   0.61
0 + (-10000:)     0.028222 0.005337  -0.022885  ×   5.29
0                 0.023521 0.000084  -0.023437  × 280.01
(-10000:) + 0     0.235539 0.005308  -0.230231  ×  44.37
(-10:) + :9       0.232883 0.000180  -0.232703  ×1293.79
(-10000:) + (:99) 0.238735 0.005358  -0.233377  ×  44.55
:99 + (-10000:)   0.317942 0.005593  -0.312349  ×  56.84
:9 + (-10:)       0.313372 0.000179  -0.313193  ×1750.68
:9                0.316450 0.000143  -0.316307  ×2212.93


On smaller repositories, the cost of nodemap related operation is not as big, so
the win is much more modest. Yet it helps shaving a handful of millisecond here
and there.

Here are some numbers (in seconds) for the reference copy of mercurial:

Revset            Before   After     abs-change speedup
-10:              0.000065 0.000097   0.000032  × 0.67
tip               0.000063 0.000078   0.000015  × 0.80
0                 0.000561 0.000079  -0.000482  × 7.10
-10000:           0.004609 0.003648  -0.000961  × 1.26
0 + (-10000:)     0.005023 0.003715  -0.001307  × 1.35
(-10:) + :9       0.002187 0.000108  -0.002079  ×20.25
(-10000:) + 0     0.006252 0.003716  -0.002536  × 1.68
(-10000:) + (:99) 0.006367 0.003707  -0.002660  × 1.71
:9 + (-10:)       0.003846 0.000110  -0.003736  ×34.96
:9                0.003854 0.000099  -0.003755  ×38.92
:99 + (-10000:)   0.007644 0.003778  -0.003866  × 2.02

Differential Revision: https://phab.mercurial-scm.org/D7894

from __future__ import absolute_import, print_function

import os
import tempfile

from mercurial import (
    pycompat,
    util,
)

from hgext.fastannotate import error, revmap

if pycompat.ispy3:
    xrange = range


def genhsh(i):
    return pycompat.bytechr(i) + b'\0' * 19


def gettemppath():
    fd, path = tempfile.mkstemp()
    os.close(fd)
    os.unlink(path)
    return path


def ensure(condition):
    if not condition:
        raise RuntimeError('Unexpected')


def testbasicreadwrite():
    path = gettemppath()

    rm = revmap.revmap(path)
    ensure(rm.maxrev == 0)
    for i in xrange(5):
        ensure(rm.rev2hsh(i) is None)
    ensure(rm.hsh2rev(b'\0' * 20) is None)

    paths = [
        b'',
        b'a',
        None,
        b'b',
        b'b',
        b'c',
        b'c',
        None,
        b'a',
        b'b',
        b'a',
        b'a',
    ]
    for i in xrange(1, 5):
        ensure(rm.append(genhsh(i), sidebranch=(i & 1), path=paths[i]) == i)

    ensure(rm.maxrev == 4)
    for i in xrange(1, 5):
        ensure(rm.hsh2rev(genhsh(i)) == i)
        ensure(rm.rev2hsh(i) == genhsh(i))

    # re-load and verify
    rm.flush()
    rm = revmap.revmap(path)
    ensure(rm.maxrev == 4)
    for i in xrange(1, 5):
        ensure(rm.hsh2rev(genhsh(i)) == i)
        ensure(rm.rev2hsh(i) == genhsh(i))
        ensure(bool(rm.rev2flag(i) & revmap.sidebranchflag) == bool(i & 1))

    # append without calling save() explicitly
    for i in xrange(5, 12):
        ensure(
            rm.append(genhsh(i), sidebranch=(i & 1), path=paths[i], flush=True)
            == i
        )

    # re-load and verify
    rm = revmap.revmap(path)
    ensure(rm.maxrev == 11)
    for i in xrange(1, 12):
        ensure(rm.hsh2rev(genhsh(i)) == i)
        ensure(rm.rev2hsh(i) == genhsh(i))
        ensure(rm.rev2path(i) == paths[i] or paths[i - 1])
        ensure(bool(rm.rev2flag(i) & revmap.sidebranchflag) == bool(i & 1))

    os.unlink(path)

    # missing keys
    ensure(rm.rev2hsh(12) is None)
    ensure(rm.rev2hsh(0) is None)
    ensure(rm.rev2hsh(-1) is None)
    ensure(rm.rev2flag(12) is None)
    ensure(rm.rev2path(12) is None)
    ensure(rm.hsh2rev(b'\1' * 20) is None)

    # illformed hash (not 20 bytes)
    try:
        rm.append(b'\0')
        ensure(False)
    except Exception:
        pass


def testcorruptformat():
    path = gettemppath()

    # incorrect header
    with open(path, 'wb') as f:
        f.write(b'NOT A VALID HEADER')
    try:
        revmap.revmap(path)
        ensure(False)
    except error.CorruptedFileError:
        pass

    # rewrite the file
    os.unlink(path)
    rm = revmap.revmap(path)
    rm.append(genhsh(0), flush=True)

    rm = revmap.revmap(path)
    ensure(rm.maxrev == 1)

    # corrupt the file by appending a byte
    size = os.stat(path).st_size
    with open(path, 'ab') as f:
        f.write(b'\xff')
    try:
        revmap.revmap(path)
        ensure(False)
    except error.CorruptedFileError:
        pass

    # corrupt the file by removing the last byte
    ensure(size > 0)
    with open(path, 'wb') as f:
        f.truncate(size - 1)
    try:
        revmap.revmap(path)
        ensure(False)
    except error.CorruptedFileError:
        pass

    os.unlink(path)


def testcopyfrom():
    path = gettemppath()
    rm = revmap.revmap(path)
    for i in xrange(1, 10):
        ensure(
            rm.append(genhsh(i), sidebranch=(i & 1), path=(b'%d' % (i // 3)))
            == i
        )
    rm.flush()

    # copy rm to rm2
    rm2 = revmap.revmap()
    rm2.copyfrom(rm)
    path2 = gettemppath()
    rm2.path = path2
    rm2.flush()

    # two files should be the same
    ensure(len(set(util.readfile(p) for p in [path, path2])) == 1)

    os.unlink(path)
    os.unlink(path2)


class fakefctx(object):
    def __init__(self, node, path=None):
        self._node = node
        self._path = path

    def node(self):
        return self._node

    def path(self):
        return self._path


def testcontains():
    path = gettemppath()

    rm = revmap.revmap(path)
    for i in xrange(1, 5):
        ensure(rm.append(genhsh(i), sidebranch=(i & 1)) == i)

    for i in xrange(1, 5):
        ensure(((genhsh(i), None) in rm) == ((i & 1) == 0))
        ensure((fakefctx(genhsh(i)) in rm) == ((i & 1) == 0))
    for i in xrange(5, 10):
        ensure(fakefctx(genhsh(i)) not in rm)
        ensure((genhsh(i), None) not in rm)

    # "contains" checks paths
    rm = revmap.revmap()
    for i in xrange(1, 5):
        ensure(rm.append(genhsh(i), path=(b'%d' % (i // 2))) == i)
    for i in xrange(1, 5):
        ensure(fakefctx(genhsh(i), path=(b'%d' % (i // 2))) in rm)
        ensure(fakefctx(genhsh(i), path=b'a') not in rm)


def testlastnode():
    path = gettemppath()
    ensure(revmap.getlastnode(path) is None)
    rm = revmap.revmap(path)
    ensure(revmap.getlastnode(path) is None)
    for i in xrange(1, 10):
        hsh = genhsh(i)
        rm.append(hsh, path=(b'%d' % (i // 2)), flush=True)
        ensure(revmap.getlastnode(path) == hsh)
        rm2 = revmap.revmap(path)
        ensure(rm2.rev2hsh(rm2.maxrev) == hsh)


testbasicreadwrite()
testcorruptformat()
testcopyfrom()
testcontains()
testlastnode()