view tests/test-linelog.py @ 41277:61f9ef23a12f

dagop: minor python optimization to `headrevs`

Less lookup and less function call never hurt. This provides a small speedup
on various run of the 'heads()' revset. This also buys back some of the slow
down we observed in the previous changesets for single value lookup.

Performance number:
0) before dagop.headrevs usage
1) after dagop.headrevs usage
2) after this change

revset: heads(all())
   plain         min           max           first         last          reverse       rev..rst      rev..ast      sort          sor..rst      sor..ast
0) 0.036503      0.032564      0.030024      0.032378      0.030887      0.036367      0.031713      0.032205      0.036467      0.032286      0.030300
1) 0.036668      0.035347 108% 0.035611 118% 0.035358 109% 0.035726 115% 0.036411      0.035261 111% 0.036096 112% 0.036052      0.035095 108% 0.035792 118%
2) 0.034254  93% 0.034482      0.035003      0.034353      0.033754  94% 0.034689      0.034361      0.035059      0.034636      0.034662      0.035465

revset: heads(-10000:-1)
   plain         min           max           first         last          reverse       rev..rst      rev..ast      sort          sor..rst      sor..ast
0) 0.003936      0.003218      0.003227      0.003302      0.003328      0.003848      0.003305      0.003252      0.003839      0.003306      0.003279
1) 0.003870      0.003785 117% 0.003821 118% 0.003780 114% 0.003769 113% 0.003776      0.003792 114% 0.003805 117% 0.003810      0.003798 114% 0.003840 117%
2) 0.003666  94% 0.003577  94% 0.003632      0.003644      0.003614      0.003638      0.003652      0.003632      0.003661      0.003660      0.003658

revset: (-5000:-1000) and heads(-10000:-1)
   plain         min           max           first         last          reverse       rev..rst      rev..ast      sort          sor..rst      sor..ast
0) 0.004244      0.003368      0.003313      0.003367      0.003327      0.004325      0.003401      0.003379      0.004310      0.003359      0.003396
1) 0.003969  93% 0.003862 114% 0.003834 115% 0.003810 113% 0.003822 114% 0.003940  91% 0.003908 114% 0.003814 112% 0.003986  92% 0.003954 117% 0.003816 112%
2) 0.003728  93% 0.003638  94% 0.003659      0.003685      0.003628  94% 0.003716  94% 0.003653  93% 0.003655      0.003748  94% 0.003740  94% 0.003686

revset: heads(matching(tip, "author"))
   plain         min           max           first         last          reverse       rev..rst      rev..ast      sort          sor..rst      sor..ast
0) 7.574666      7.545950      7.570743      7.578697      7.525725      7.509929      7.443854      7.488442      7.452880      7.445411      7.689107
1) 7.549390      7.389162      7.529790      7.536297      7.450467      7.555347      7.404586      7.514948      7.542794      7.524787      7.536918
2) 7.568294      7.479326      7.578624      7.380375      7.440102      7.454218      7.515189      7.556511      7.524585      7.537566      7.507418

revset: heads(matching(tip, "author")) and -10000:-1
   plain         min           max           first         last          reverse       rev..rst      rev..ast      sort          sor..rst      sor..ast
0) 7.512533      7.605877      7.382894      7.462109      7.420086      7.575034      7.448452      7.549374      7.457880      7.450308      7.515019
1) 7.548677      7.551832      7.629598      7.494857      7.550554      7.521838      7.451794      error         7.321781      7.546885      7.557523
2) 7.451985      7.541044      7.506563      7.470928      7.512618      7.474988      7.498887      7.547930      7.560276      7.618599      7.465442

revset: (-10000:-1) and heads(matching(tip, "author"))
   plain         min           max           first         last          reverse       rev..rst      rev..ast      sort          sor..rst      sor..ast
0) 7.465419      7.570089      7.439594      7.521221      7.498716      7.492922      7.479108      7.552397      7.407888      error         7.468264
1) 7.539866      7.548045      7.491761      7.517170      7.469824      7.501990      7.579102      7.502568      7.578102      7.555754      7.567622
2) 7.370463      7.514712      7.497024      7.679428      7.638138      7.490775      7.472273      7.652587      7.584139      7.511893      7.466384

from __future__ import absolute_import, print_function

import difflib
import random
import unittest

from mercurial import linelog

vecratio = 3 # number of replacelines / number of replacelines_vec
maxlinenum = 0xffffff
maxb1 = 0xffffff
maxdeltaa = 10
maxdeltab = 10

def _genedits(seed, endrev):
    lines = []
    random.seed(seed)
    rev = 0
    for rev in range(0, endrev):
        n = len(lines)
        a1 = random.randint(0, n)
        a2 = random.randint(a1, min(n, a1 + maxdeltaa))
        b1 = random.randint(0, maxb1)
        b2 = random.randint(b1, b1 + maxdeltab)
        usevec = not bool(random.randint(0, vecratio))
        if usevec:
            blines = [(random.randint(0, rev), random.randint(0, maxlinenum))
                      for _ in range(b1, b2)]
        else:
            blines = [(rev, bidx) for bidx in range(b1, b2)]
        lines[a1:a2] = blines
        yield lines, rev, a1, a2, b1, b2, blines, usevec

class linelogtests(unittest.TestCase):
    def testlinelogencodedecode(self):
        program = [linelog._eof(0, 0),
                   linelog._jge(41, 42),
                   linelog._jump(0, 43),
                   linelog._eof(0, 0),
                   linelog._jl(44, 45),
                   linelog._line(46, 47),
                   ]
        ll = linelog.linelog(program, maxrev=100)
        enc = ll.encode()
        # round-trips okay
        self.assertEqual(linelog.linelog.fromdata(enc)._program, ll._program)
        self.assertEqual(linelog.linelog.fromdata(enc), ll)
        # This encoding matches the encoding used by hg-experimental's
        # linelog file, or is supposed to if it doesn't.
        self.assertEqual(enc, (b'\x00\x00\x01\x90\x00\x00\x00\x06'
                               b'\x00\x00\x00\xa4\x00\x00\x00*'
                               b'\x00\x00\x00\x00\x00\x00\x00+'
                               b'\x00\x00\x00\x00\x00\x00\x00\x00'
                               b'\x00\x00\x00\xb1\x00\x00\x00-'
                               b'\x00\x00\x00\xba\x00\x00\x00/'))

    def testsimpleedits(self):
        ll = linelog.linelog()
        # Initial revision: add lines 0, 1, and 2
        ll.replacelines(1, 0, 0, 0, 3)
        self.assertEqual([(l.rev, l.linenum) for l in ll.annotate(1)],
                         [(1, 0),
                          (1, 1),
                          (1, 2),
                         ])
        # Replace line 1 with a new line
        ll.replacelines(2, 1, 2, 1, 2)
        self.assertEqual([(l.rev, l.linenum) for l in ll.annotate(2)],
                         [(1, 0),
                          (2, 1),
                          (1, 2),
                         ])
        # delete a line out of 2
        ll.replacelines(3, 1, 2, 0, 0)
        self.assertEqual([(l.rev, l.linenum) for l in ll.annotate(3)],
                         [(1, 0),
                          (1, 2),
                         ])
        # annotation of 1 is unchanged
        self.assertEqual([(l.rev, l.linenum) for l in ll.annotate(1)],
                         [(1, 0),
                          (1, 1),
                          (1, 2),
                         ])
        ll.annotate(3) # set internal state to revision 3
        start = ll.getoffset(0)
        end = ll.getoffset(1)
        self.assertEqual(ll.getalllines(start, end), [
            (1, 0),
            (2, 1),
            (1, 1),
        ])
        self.assertEqual(ll.getalllines(), [
            (1, 0),
            (2, 1),
            (1, 1),
            (1, 2),
        ])

    def testparseclinelogfile(self):
        # This data is what the replacements in testsimpleedits
        # produce when fed to the original linelog.c implementation.
        data = (b'\x00\x00\x00\x0c\x00\x00\x00\x0f'
                b'\x00\x00\x00\x00\x00\x00\x00\x02'
                b'\x00\x00\x00\x05\x00\x00\x00\x06'
                b'\x00\x00\x00\x06\x00\x00\x00\x00'
                b'\x00\x00\x00\x00\x00\x00\x00\x07'
                b'\x00\x00\x00\x06\x00\x00\x00\x02'
                b'\x00\x00\x00\x00\x00\x00\x00\x00'
                b'\x00\x00\x00\t\x00\x00\x00\t'
                b'\x00\x00\x00\x00\x00\x00\x00\x0c'
                b'\x00\x00\x00\x08\x00\x00\x00\x05'
                b'\x00\x00\x00\x06\x00\x00\x00\x01'
                b'\x00\x00\x00\x00\x00\x00\x00\x05'
                b'\x00\x00\x00\x0c\x00\x00\x00\x05'
                b'\x00\x00\x00\n\x00\x00\x00\x01'
                b'\x00\x00\x00\x00\x00\x00\x00\t')
        llc = linelog.linelog.fromdata(data)
        self.assertEqual([(l.rev, l.linenum) for l in llc.annotate(1)],
                         [(1, 0),
                          (1, 1),
                          (1, 2),
                         ])
        self.assertEqual([(l.rev, l.linenum) for l in llc.annotate(2)],
                         [(1, 0),
                          (2, 1),
                          (1, 2),
                         ])
        self.assertEqual([(l.rev, l.linenum) for l in llc.annotate(3)],
                         [(1, 0),
                          (1, 2),
                         ])
        # Check we emit the same bytecode.
        ll = linelog.linelog()
        # Initial revision: add lines 0, 1, and 2
        ll.replacelines(1, 0, 0, 0, 3)
        # Replace line 1 with a new line
        ll.replacelines(2, 1, 2, 1, 2)
        # delete a line out of 2
        ll.replacelines(3, 1, 2, 0, 0)
        diff = '\n   ' + '\n   '.join(difflib.unified_diff(
            ll.debugstr().splitlines(), llc.debugstr().splitlines(),
            'python', 'c', lineterm=''))
        self.assertEqual(ll._program, llc._program, 'Program mismatch: ' + diff)
        # Done as a secondary step so we get a better result if the
        # program is where the mismatch is.
        self.assertEqual(ll, llc)
        self.assertEqual(ll.encode(), data)

    def testanothersimplecase(self):
        ll = linelog.linelog()
        ll.replacelines(3, 0, 0, 0, 2)
        ll.replacelines(4, 0, 2, 0, 0)
        self.assertEqual([(l.rev, l.linenum) for l in ll.annotate(4)],
                         [])
        self.assertEqual([(l.rev, l.linenum) for l in ll.annotate(3)],
                         [(3, 0), (3, 1)])
        # rev 2 is empty because contents were only ever introduced in rev 3
        self.assertEqual([(l.rev, l.linenum) for l in ll.annotate(2)],
                         [])

    def testrandomedits(self):
        # Inspired by original linelog tests.
        seed = random.random()
        numrevs = 2000
        ll = linelog.linelog()
        # Populate linelog
        for lines, rev, a1, a2, b1, b2, blines, usevec in _genedits(
                seed, numrevs):
            if usevec:
                ll.replacelines_vec(rev, a1, a2, blines)
            else:
                ll.replacelines(rev, a1, a2, b1, b2)
            ar = ll.annotate(rev)
            self.assertEqual(ll.annotateresult, lines)
        # Verify we can get back these states by annotating each rev
        for lines, rev, a1, a2, b1, b2, blines, usevec in _genedits(
                seed, numrevs):
            ar = ll.annotate(rev)
            self.assertEqual([(l.rev, l.linenum) for l in ar], lines)

    def testinfinitebadprogram(self):
        ll = linelog.linelog.fromdata(
            b'\x00\x00\x00\x00\x00\x00\x00\x02'  # header
            b'\x00\x00\x00\x00\x00\x00\x00\x01'  # JUMP to self
        )
        with self.assertRaises(linelog.LineLogError):
            # should not be an infinite loop and raise
            ll.annotate(1)

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