mercurial/cffi/bdiff.py

revset: use phasecache.getrevset to calculate public()

Other revsets like secret(), draft(), _nonpublic() are using
phasescache.getrevset already. The latter is more efficient after D1606.
So let's migrate the public() revset function too.

Tested using:

  $ hg debugshell --hidden --cwd hg-committed`
  In [1]: %timeit len(repo.revs('public()'))

* Before D1606: 10 loops, best of 3: 22.5 ms per loop
* Before this change, after D1606: 10 loops, best of 3: 28.6 ms per loop
* After this change: 10 loops, best of 3: 20.2 ms per loop

Therefore `public()` revset becomes even slightly faster after the data
structure change by D1606. A similar performance win could also be observed
on a large repo.

A side effect is `phasecache.getrevset` needs to take a `subset` parameter.
That was added with a default value so it won't cause BC issues.

Differential Revision: https://phab.mercurial-scm.org/D1620

# bdiff.py - CFFI implementation of bdiff.c
#
# Copyright 2016 Maciej Fijalkowski <[e-mail]>
#
# 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 struct

from ..pure.bdiff import *
from . import _bdiff

ffi = _bdiff.ffi
lib = _bdiff.lib

def blocks(sa, sb):
    a = ffi.new("struct bdiff_line**")
    b = ffi.new("struct bdiff_line**")
    ac = ffi.new("char[]", str(sa))
    bc = ffi.new("char[]", str(sb))
    l = ffi.new("struct bdiff_hunk*")
    try:
        an = lib.bdiff_splitlines(ac, len(sa), a)
        bn = lib.bdiff_splitlines(bc, len(sb), b)
        if not a[0] or not b[0]:
            raise MemoryError
        count = lib.bdiff_diff(a[0], an, b[0], bn, l)
        if count < 0:
            raise MemoryError
        rl = [None] * count
        h = l.next
        i = 0
        while h:
            rl[i] = (h.a1, h.a2, h.b1, h.b2)
            h = h.next
            i += 1
    finally:
        lib.free(a[0])
        lib.free(b[0])
        lib.bdiff_freehunks(l.next)
    return rl

def bdiff(sa, sb):
    a = ffi.new("struct bdiff_line**")
    b = ffi.new("struct bdiff_line**")
    ac = ffi.new("char[]", str(sa))
    bc = ffi.new("char[]", str(sb))
    l = ffi.new("struct bdiff_hunk*")
    try:
        an = lib.bdiff_splitlines(ac, len(sa), a)
        bn = lib.bdiff_splitlines(bc, len(sb), b)
        if not a[0] or not b[0]:
            raise MemoryError
        count = lib.bdiff_diff(a[0], an, b[0], bn, l)
        if count < 0:
            raise MemoryError
        rl = []
        h = l.next
        la = lb = 0
        while h:
            if h.a1 != la or h.b1 != lb:
                lgt = (b[0] + h.b1).l - (b[0] + lb).l
                rl.append(struct.pack(">lll", (a[0] + la).l - a[0].l,
                                      (a[0] + h.a1).l - a[0].l, lgt))
                rl.append(str(ffi.buffer((b[0] + lb).l, lgt)))
            la = h.a2
            lb = h.b2
            h = h.next

    finally:
        lib.free(a[0])
        lib.free(b[0])
        lib.bdiff_freehunks(l.next)
    return "".join(rl)