view mercurial/pure/bdiff.py @ 48178:f12a19d03d2c

fix: reduce number of tool executions

By grouping together (path, ctx) pairs according to the inputs they would
provide to fixer tools, we can deduplicate executions of fixer tools to
significantly reduce the amount of time spent running slow tools.

This change does not handle clean files in the working copy, which could still
be deduplicated against the files in the checked out commit. It's a little
harder to do that because the filerev is not available in the workingfilectx
(and it doesn't exist for added files).

Anecdotally, this change makes some real uses cases at Google 10x faster. I
think we were originally hesitant to do this because the benefits weren't
obvious, and implementing it efficiently is kind of tricky. If we simply
memoized the formatter execution function, we would be keeping tons of file
content in memory.

Also included is a regression test for a corner case that I broke with my first
attempt at optimizing this code.

Differential Revision: https://phab.mercurial-scm.org/D11280

# bdiff.py - Python implementation of bdiff.c
#
# Copyright 2009 Olivia Mackall <[e-mail]> and others
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

from __future__ import absolute_import

import difflib
import re
import struct


def splitnewlines(text):
    '''like str.splitlines, but only split on newlines.'''
    lines = [l + b'\n' for l in text.split(b'\n')]
    if lines:
        if lines[-1] == b'\n':
            lines.pop()
        else:
            lines[-1] = lines[-1][:-1]
    return lines


def _normalizeblocks(a, b, blocks):
    prev = None
    r = []
    for curr in blocks:
        if prev is None:
            prev = curr
            continue
        shift = 0

        a1, b1, l1 = prev
        a1end = a1 + l1
        b1end = b1 + l1

        a2, b2, l2 = curr
        a2end = a2 + l2
        b2end = b2 + l2
        if a1end == a2:
            while (
                a1end + shift < a2end and a[a1end + shift] == b[b1end + shift]
            ):
                shift += 1
        elif b1end == b2:
            while (
                b1end + shift < b2end and a[a1end + shift] == b[b1end + shift]
            ):
                shift += 1
        r.append((a1, b1, l1 + shift))
        prev = a2 + shift, b2 + shift, l2 - shift

    if prev is not None:
        r.append(prev)

    return r


def bdiff(a, b):
    a = bytes(a).splitlines(True)
    b = bytes(b).splitlines(True)

    if not a:
        s = b"".join(b)
        return s and (struct.pack(b">lll", 0, 0, len(s)) + s)

    bin = []
    p = [0]
    for i in a:
        p.append(p[-1] + len(i))

    d = difflib.SequenceMatcher(None, a, b).get_matching_blocks()
    d = _normalizeblocks(a, b, d)
    la = 0
    lb = 0
    for am, bm, size in d:
        s = b"".join(b[lb:bm])
        if am > la or s:
            bin.append(struct.pack(b">lll", p[la], p[am], len(s)) + s)
        la = am + size
        lb = bm + size

    return b"".join(bin)


def blocks(a, b):
    an = splitnewlines(a)
    bn = splitnewlines(b)
    d = difflib.SequenceMatcher(None, an, bn).get_matching_blocks()
    d = _normalizeblocks(an, bn, d)
    return [(i, i + n, j, j + n) for (i, j, n) in d]


def fixws(text, allws):
    if allws:
        text = re.sub(b'[ \t\r]+', b'', text)
    else:
        text = re.sub(b'[ \t\r]+', b' ', text)
        text = text.replace(b' \n', b'\n')
    return text