tests/test-lrucachedict.py

match: making visitdir() deal with non-recursive entries

Primarily as an optimization to avoid recursing into directories that will
never have a match inside, this classifies each matcher pattern's root as
recursive or non-recursive (erring on the side of keeping it recursive,
which may lead to wasteful directory or manifest walks that yield no matches).

I measured the performance of "rootfilesin" in two repos:
- The Firefox repo with tree manifests, with
  "hg files -r . -I rootfilesin:browser".
  The browser directory contains about 3K files across 249 subdirectories.
- A specific Google-internal directory which contains 75K files across 19K
  subdirectories, with "hg files -r . -I rootfilesin:REDACTED".

I tested with both cold and warm disk caches. Cold cache was produced by
running "sync; echo 3 > /proc/sys/vm/drop_caches". Warm cache was produced
by re-running the same command a few times.

These were the results:

               Cold cache           Warm cache
             Before   After      Before   After
firefox      0m5.1s   0m2.18s   0m0.22s  0m0.14s
google3 dir  2m3.9s   0m1.57s   0m8.12s  0m0.16s

Certain extensions, notably narrowhg, can depend on this for correctness
(not trying to recurse into directories for which it has no information).

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()