view mercurial/debugcommands.py @ 30432:3633403888ae

bdiff: give slight preference to appending lines

[This change could be folded into the previous changeset to minimize the repo
churn ...]

The general preference to matches in the middle of bdiff ranges helps getting
balanced recursion and efficient computation. But, as previous changes have
shown, it might also give diffs that seems "obviously wrong".

To mitigate that: If the best match on the A side starts at the beginning of
the bdiff range, don't aim for the middle-most B side match but for the
earliest.

This will make the matches balanced (by both sides being "early") even though
the bisection will be less balanced. Still, this case only apply if the *best*
and middle-most match was fully unbalanced on the A side. Each recursion will
thus even in this worst case reduce the problem significantly and we are not
re-introducing the problem that was fixed in f1ca249696ed.

The bundle size for 4.0 (hg bundle --base null -r 4.0 x.hg) happens to go from
22806817 to 22807275 bytes - a 0.002% increase.

This make the recent test-bdiff.py changes give a more pretty output ... but
they no longer show that the recursion is around middle matches (because it in
these cases isn't).

# debugcommands.py - command processing for debug* commands
#
# Copyright 2005-2016 Matt Mackall <[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 os

from .i18n import _
from .node import (
    hex,
)
from . import (
    cmdutil,
    commands,
    context,
    dagparser,
    error,
    lock as lockmod,
    revlog,
    scmutil,
    simplemerge,
)

release = lockmod.release

# We reuse the command table from commands because it is easier than
# teaching dispatch about multiple tables.
command = cmdutil.command(commands.table)

@command('debugancestor', [], _('[INDEX] REV1 REV2'), optionalrepo=True)
def debugancestor(ui, repo, *args):
    """find the ancestor revision of two revisions in a given index"""
    if len(args) == 3:
        index, rev1, rev2 = args
        r = revlog.revlog(scmutil.opener(os.getcwd(), audit=False), index)
        lookup = r.lookup
    elif len(args) == 2:
        if not repo:
            raise error.Abort(_('there is no Mercurial repository here '
                                '(.hg not found)'))
        rev1, rev2 = args
        r = repo.changelog
        lookup = repo.lookup
    else:
        raise error.Abort(_('either two or three arguments required'))
    a = r.ancestor(lookup(rev1), lookup(rev2))
    ui.write('%d:%s\n' % (r.rev(a), hex(a)))

@command('debugbuilddag',
    [('m', 'mergeable-file', None, _('add single file mergeable changes')),
    ('o', 'overwritten-file', None, _('add single file all revs overwrite')),
    ('n', 'new-file', None, _('add new file at each rev'))],
    _('[OPTION]... [TEXT]'))
def debugbuilddag(ui, repo, text=None,
                  mergeable_file=False,
                  overwritten_file=False,
                  new_file=False):
    """builds a repo with a given DAG from scratch in the current empty repo

    The description of the DAG is read from stdin if not given on the
    command line.

    Elements:

     - "+n" is a linear run of n nodes based on the current default parent
     - "." is a single node based on the current default parent
     - "$" resets the default parent to null (implied at the start);
           otherwise the default parent is always the last node created
     - "<p" sets the default parent to the backref p
     - "*p" is a fork at parent p, which is a backref
     - "*p1/p2" is a merge of parents p1 and p2, which are backrefs
     - "/p2" is a merge of the preceding node and p2
     - ":tag" defines a local tag for the preceding node
     - "@branch" sets the named branch for subsequent nodes
     - "#...\\n" is a comment up to the end of the line

    Whitespace between the above elements is ignored.

    A backref is either

     - a number n, which references the node curr-n, where curr is the current
       node, or
     - the name of a local tag you placed earlier using ":tag", or
     - empty to denote the default parent.

    All string valued-elements are either strictly alphanumeric, or must
    be enclosed in double quotes ("..."), with "\\" as escape character.
    """

    if text is None:
        ui.status(_("reading DAG from stdin\n"))
        text = ui.fin.read()

    cl = repo.changelog
    if len(cl) > 0:
        raise error.Abort(_('repository is not empty'))

    # determine number of revs in DAG
    total = 0
    for type, data in dagparser.parsedag(text):
        if type == 'n':
            total += 1

    if mergeable_file:
        linesperrev = 2
        # make a file with k lines per rev
        initialmergedlines = [str(i) for i in xrange(0, total * linesperrev)]
        initialmergedlines.append("")

    tags = []

    wlock = lock = tr = None
    try:
        wlock = repo.wlock()
        lock = repo.lock()
        tr = repo.transaction("builddag")

        at = -1
        atbranch = 'default'
        nodeids = []
        id = 0
        ui.progress(_('building'), id, unit=_('revisions'), total=total)
        for type, data in dagparser.parsedag(text):
            if type == 'n':
                ui.note(('node %s\n' % str(data)))
                id, ps = data

                files = []
                fctxs = {}

                p2 = None
                if mergeable_file:
                    fn = "mf"
                    p1 = repo[ps[0]]
                    if len(ps) > 1:
                        p2 = repo[ps[1]]
                        pa = p1.ancestor(p2)
                        base, local, other = [x[fn].data() for x in (pa, p1,
                                                                     p2)]
                        m3 = simplemerge.Merge3Text(base, local, other)
                        ml = [l.strip() for l in m3.merge_lines()]
                        ml.append("")
                    elif at > 0:
                        ml = p1[fn].data().split("\n")
                    else:
                        ml = initialmergedlines
                    ml[id * linesperrev] += " r%i" % id
                    mergedtext = "\n".join(ml)
                    files.append(fn)
                    fctxs[fn] = context.memfilectx(repo, fn, mergedtext)

                if overwritten_file:
                    fn = "of"
                    files.append(fn)
                    fctxs[fn] = context.memfilectx(repo, fn, "r%i\n" % id)

                if new_file:
                    fn = "nf%i" % id
                    files.append(fn)
                    fctxs[fn] = context.memfilectx(repo, fn, "r%i\n" % id)
                    if len(ps) > 1:
                        if not p2:
                            p2 = repo[ps[1]]
                        for fn in p2:
                            if fn.startswith("nf"):
                                files.append(fn)
                                fctxs[fn] = p2[fn]

                def fctxfn(repo, cx, path):
                    return fctxs.get(path)

                if len(ps) == 0 or ps[0] < 0:
                    pars = [None, None]
                elif len(ps) == 1:
                    pars = [nodeids[ps[0]], None]
                else:
                    pars = [nodeids[p] for p in ps]
                cx = context.memctx(repo, pars, "r%i" % id, files, fctxfn,
                                    date=(id, 0),
                                    user="debugbuilddag",
                                    extra={'branch': atbranch})
                nodeid = repo.commitctx(cx)
                nodeids.append(nodeid)
                at = id
            elif type == 'l':
                id, name = data
                ui.note(('tag %s\n' % name))
                tags.append("%s %s\n" % (hex(repo.changelog.node(id)), name))
            elif type == 'a':
                ui.note(('branch %s\n' % data))
                atbranch = data
            ui.progress(_('building'), id, unit=_('revisions'), total=total)
        tr.close()

        if tags:
            repo.vfs.write("localtags", "".join(tags))
    finally:
        ui.progress(_('building'), None)
        release(tr, lock, wlock)