tests/test-cappedreader.py

copies: do full filtering at end of _changesetforwardcopies()

As mentioned earlier, pathcopies() is very slow when copies are stored
in the changeset. Most of the cost comes from calling _chain() for
every changeset, which is slow because it needs to read manifests. It
needs to read manifests to be able to filter out copies that are were
created in one commit and then deleted. (It also filters out copies
that were created from a file that didn't exist in the starting
revision, but that's a fixed revision across calls to _chain(), so
it's much cheaper.)

This patch changes from _chainandfilter() to just _chain() in the main
loop in _changesetforwardcopies(). It instead removes copies that have
subsequently been removed by using ctx.filesremoved(). We thus rely on
that to be fast.

It timed this command in mozilla-unified:

  hg debugpathcopies FIREFOX_59_0b3_BUILD2 FIREFOX_BETA_59_END

It took 18s before and 1.1s after. It's still faster when copy
information is stored in filelogs: 0.70s. It also still gets slow when
there are merge commits involved, because we read manifests there
too. We'll deal with that later.

Differential Revision: https://phab.mercurial-scm.org/D6419

from __future__ import absolute_import, print_function

import io
import unittest

from mercurial import (
    util,
)

class CappedReaderTests(unittest.TestCase):
    def testreadfull(self):
        source = io.BytesIO(b'x' * 100)

        reader = util.cappedreader(source, 10)
        res = reader.read(10)
        self.assertEqual(res, b'x' * 10)
        self.assertEqual(source.tell(), 10)
        source.seek(0)

        reader = util.cappedreader(source, 15)
        res = reader.read(16)
        self.assertEqual(res, b'x' * 15)
        self.assertEqual(source.tell(), 15)
        source.seek(0)

        reader = util.cappedreader(source, 100)
        res = reader.read(100)
        self.assertEqual(res, b'x' * 100)
        self.assertEqual(source.tell(), 100)
        source.seek(0)

        reader = util.cappedreader(source, 50)
        res = reader.read()
        self.assertEqual(res, b'x' * 50)
        self.assertEqual(source.tell(), 50)
        source.seek(0)

    def testreadnegative(self):
        source = io.BytesIO(b'x' * 100)

        reader = util.cappedreader(source, 20)
        res = reader.read(-1)
        self.assertEqual(res, b'x' * 20)
        self.assertEqual(source.tell(), 20)
        source.seek(0)

        reader = util.cappedreader(source, 100)
        res = reader.read(-1)
        self.assertEqual(res, b'x' * 100)
        self.assertEqual(source.tell(), 100)
        source.seek(0)

    def testreadmultiple(self):
        source = io.BytesIO(b'x' * 100)

        reader = util.cappedreader(source, 10)
        for i in range(10):
            res = reader.read(1)
            self.assertEqual(res, b'x')
            self.assertEqual(source.tell(), i + 1)

        self.assertEqual(source.tell(), 10)
        res = reader.read(1)
        self.assertEqual(res, b'')
        self.assertEqual(source.tell(), 10)
        source.seek(0)

        reader = util.cappedreader(source, 45)
        for i in range(4):
            res = reader.read(10)
            self.assertEqual(res, b'x' * 10)
            self.assertEqual(source.tell(), (i + 1) * 10)

        res = reader.read(10)
        self.assertEqual(res, b'x' * 5)
        self.assertEqual(source.tell(), 45)

    def readlimitpasteof(self):
        source = io.BytesIO(b'x' * 100)

        reader = util.cappedreader(source, 1024)
        res = reader.read(1000)
        self.assertEqual(res, b'x' * 100)
        self.assertEqual(source.tell(), 100)
        res = reader.read(1000)
        self.assertEqual(res, b'')
        self.assertEqual(source.tell(), 100)

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