view tests/test-lrucachedict.py @ 38661:8ac0c9cd4c48

fncache: avoid loading the filename cache when not actually modifying it

With time, fncache can become very large. The mozilla-central repo for example,
has a 31M and growing fncache file. Loading this file takes time (280ms for the
mozilla-central repository).

In many scenarios, we don't need to load fncache at all. For example, when
committing changes to existing files, or pushing such commits to another clone.
This patch detects when a name is added via store.vfs(), and only loads the
cache if a) the data metadata file doesn't already exist, or b) when opening
for appending, the data or metadata file exists but has size  (a transaction
rollback leaves behind such files).

Benchmarks (run on Macos 10.13 on a 2017-model Macbook Pro with Core i7 2.9GHz
and flash drive), each test without and with patch run 5 times:

* committing to an existing file, against the mozilla-central repository.
  Baseline real time average 2.3736, with patch 1.9884.

* unbundling a large changeset consisting *only* of existing-file modifications
  (159 revisions, 1050 modifications, mozilla-central
  4a250a0e4f29:beea9ac7d823), into a clone limited to the ancestor revision of
  that revset). Baseline real time average 1.5048, with patch 1.3108.

from __future__ import absolute_import, print_function

from mercurial import (
    util,
)

def printifpresent(d, xs, name='d'):
    for x in xs:
        present = x in d
        print("'%s' in %s: %s" % (x, name, present))
        if present:
            print("%s['%s']: %s" % (name, x, d[x]))

def test_lrucachedict():
    d = util.lrucachedict(4)
    d['a'] = 'va'
    d['b'] = 'vb'
    d['c'] = 'vc'
    d['d'] = 'vd'

    # all of these should be present
    printifpresent(d, ['a', 'b', 'c', 'd'])

    # 'a' should be dropped because it was least recently used
    d['e'] = 've'
    printifpresent(d, ['a', 'b', 'c', 'd', 'e'])

    assert d.get('a') is None
    assert d.get('e') == 've'

    # touch entries in some order (get or set).
    d['e']
    d['c'] = 'vc2'
    d['d']
    d['b'] = 'vb2'

    # 'e' should be dropped now
    d['f'] = 'vf'
    printifpresent(d, ['b', 'c', 'd', 'e', 'f'])

    d.clear()
    printifpresent(d, ['b', 'c', 'd', 'e', 'f'])

    # Now test dicts that aren't full.
    d = util.lrucachedict(4)
    d['a'] = 1
    d['b'] = 2
    d['a']
    d['b']
    printifpresent(d, ['a', 'b'])

    # test copy method
    d = util.lrucachedict(4)
    d['a'] = 'va3'
    d['b'] = 'vb3'
    d['c'] = 'vc3'
    d['d'] = 'vd3'

    dc = d.copy()

    # all of these should be present
    print("\nAll of these should be present:")
    printifpresent(dc, ['a', 'b', 'c', 'd'], 'dc')

    # 'a' should be dropped because it was least recently used
    print("\nAll of these except 'a' should be present:")
    dc['e'] = 've3'
    printifpresent(dc, ['a', 'b', 'c', 'd', 'e'], 'dc')

    # contents and order of original dict should remain unchanged
    print("\nThese should be in reverse alphabetical order and read 'v?3':")
    dc['b'] = 'vb3_new'
    for k in list(iter(d)):
        print("d['%s']: %s" % (k, d[k]))

if __name__ == '__main__':
    test_lrucachedict()