mercurial/pure/bdiff.py

pycompat: change argv conversion semantics

Use of os.fsencode() to convert Python's sys.argv back to bytes
was not correct because it isn't the logically inverse operation
from what CPython was doing under the hood.

This commit changes the logic for doing the str -> bytes
conversion. This required a separate implementation for
POSIX and Windows.

The Windows behavior is arguably not ideal. The previous
behavior on Windows was leading to failing tests, such as
test-http-branchmap.t, which defines a utf-8 branch name
via a command argument. Previously, Mercurial's argument
parser looked to be receiving wchar_t bytes in some cases.
After this commit, behavior on Windows is compatible with
Python 2, where CPython did not implement `int wmain()` and
Windows was performing a Unicode to ANSI conversion on the
wchar_t native command line.

Arguably better behavior on Windows would be for Mercurial to
preserve the original Unicode sequence coming from Python and
to wrap this in a bytes-like type so we can round trip safely.
But, this would be new, backwards incompatible behavior. My
goal for this commit was to converge Mercurial behavior on
Python 3 on Windows to fix busted tests. And I believe I was
successful, as this commit fixes 9 tests on my Windows
machine and 14 tests in the AWS CI environment!

Differential Revision: https://phab.mercurial-scm.org/D8337

# bdiff.py - Python implementation of bdiff.c
#
# Copyright 2009 Matt 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
    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