view mercurial/rewriteutil.py @ 48687:f8f2ecdde4b5

branchmap: skip obsolete revisions while computing heads

It's time to make this part of core Mercurial obsolescence-aware.

Not considering obsolete revisions when computing heads is clearly what
Mercurial should do. But there are a couple of small issues:

- Let's say tip of the repo is obsolete. There are two ways of finding tiprev
  for branchcache (both are in use): looking at input data for update() and
  looking at computed heads after update(). Previously, repo tip would be
  tiprev of the branchcache. With this patch, an obsolete revision can no
  longer be tiprev. And depending on what way we use for finding tiprev (input
  data vs computed heads) we'll get a different result. This is relevant when
  recomputing cache key from cache contents, and may lead to updating cache for
  obsolete revisions multiple times (not from scratch, because it still would
  be considered valid for a subset of revisions in the repo).

- If all commits on a branch are obsolete, the branchcache will include that
  branch, but the list of heads will be empty (that's why there's now `if not
  heads` when recomputing tiprev/tipnode from cache contents). Having an entry
  for every branch is currently required for notify extension (and
  test-notify.t to pass), because notify doesn't handle revsets in its
  subscription config very well and will throw an error if e.g. a branch
  doesn't exist.

- Cloning static HTTP repos may try to stat() a non-existent obsstore file. The
  issue is that we now care about obsolescence during clone, but statichttpvfs
  doesn't implement a stat method, so a regular vfs.stat() is used, and it
  assumes that file is local and calls os.stat().  During a clone, we're trying
  to stat() .hg/store/obsstore, but in static HTTP case we provide a literal
  URL to the obsstore file on the remote as if it were a local file path. On
  windows it actually results in a failure in test-static-http.t.

The first issue is going to be addressed in a series dedicated to making sure
branchcache is properly and timely written on disk (it wasn't perfect even
before this patch, but there aren't enough tests to demonstrate that). The
second issue will be addressed in a future patch for notify extension that will
make it not raise an exception if a branch doesn't exist. And the third one was
partially addressed in the previous patch in this series and will be properly
fixed in a future patch when this series is accepted.

filteredhash() grows a keyword argument to make sure that branchcache is also
invalidated when there are new obsolete revisions in its repo view. This way
the on-disk cache format is unchanged and compatible between versions (although
it will obviously be recomputed when switching versions before/after this patch
and the repo has obsolete revisions).

There's one test that uses plain `hg up` without arguments while updated to a
pruned commit. To make this test pass, simply return current working directory
parent. Later in this series this code will be replaced by what prune command
does: updating to the closest non-obsolete ancestor.

Test changes:

test-branch-change.t: update branch head and cache update message. The head of
default listed in hg heads is changed because revision 2 was rewritten as 7,
and 1 is the closest ancestor on the same branch, so it's the head of default
now.

The cache invalidation message appears now because of the cache hash change,
since we're now accounting for obsolete revisions. Here's some context:

"served.hidden" repo filter means everything is visible (no filtered
revisions), so before this series branch2-served.hidden file would not contain
any cache hash, only revnum and node. Now it also has a hash when there are
obsolete changesets in the repo. The command that the message appears for is
changing branch of 5 and 6, which are now obsolete, so the cache hash changes.

In general, when cache is simply out-of-date, it can be updated using the old
version as a base. But if cache hash differs, then the cache for that
particular repo filter is recomputed (at least with the current
implementation). This is what happens here.

test-obsmarker-template.t: the pull reports 2 heads changed, but after that the
repo correctly sees only 1. The new message could be better, but it's still an
improvement over the previous one where hg pull suggested merging with an
obsolete revision.

test-obsolete.t: we can see these revisions in hg log --hidden, but they
shouldn't be considered heads even with --hidden.

test-rebase-obsolete{,2}.t: there were new heads created previously after
making new orphan changesets, but they weren't detected. Now we are properly
detecting and reporting them.

test-rebase-obsolete4.t: there's only one head now because the other head is
pruned and was falsely reported before.

test-static-http.t: add obsstore to the list of requested files. This file
doesn't exist on the remotes, but clients want it anyway (they get 404). This
is fine, because there are other nonexistent files that clients request, like
.hg/bookmarks or .hg/cache/tags2-served.

Differential Revision: https://phab.mercurial-scm.org/D12097

# rewriteutil.py - utility functions for rewriting changesets
#
# Copyright 2017 Octobus <[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 re

from .i18n import _
from .node import (
    hex,
    nullrev,
)

from . import (
    error,
    node,
    obsolete,
    obsutil,
    revset,
    scmutil,
    util,
)


NODE_RE = re.compile(br'\b[0-9a-f]{6,64}\b')


def _formatrevs(repo, revs, maxrevs=4):
    """returns a string summarizing revisions in a decent size

    If there are few enough revisions, we list them all. Otherwise we display a
    summary of the form:

        1ea73414a91b and 5 others
    """
    tonode = repo.changelog.node
    numrevs = len(revs)
    if numrevs < maxrevs:
        shorts = [node.short(tonode(r)) for r in revs]
        summary = b', '.join(shorts)
    else:
        first = revs.first()
        summary = _(b'%s and %d others')
        summary %= (node.short(tonode(first)), numrevs - 1)
    return summary


def precheck(repo, revs, action=b'rewrite'):
    """check if revs can be rewritten
    action is used to control the error message.

    Make sure this function is called after taking the lock.
    """
    if nullrev in revs:
        msg = _(b"cannot %s the null revision") % action
        hint = _(b"no changeset checked out")
        raise error.InputError(msg, hint=hint)

    if any(util.safehasattr(r, 'rev') for r in revs):
        repo.ui.develwarn(b"rewriteutil.precheck called with ctx not revs")
        revs = (r.rev() for r in revs)

    if len(repo[None].parents()) > 1:
        raise error.StateError(
            _(b"cannot %s changesets while merging") % action
        )

    publicrevs = repo.revs(b'%ld and public()', revs)
    if publicrevs:
        summary = _formatrevs(repo, publicrevs)
        msg = _(b"cannot %s public changesets: %s") % (action, summary)
        hint = _(b"see 'hg help phases' for details")
        raise error.InputError(msg, hint=hint)

    newunstable = disallowednewunstable(repo, revs)
    if newunstable:
        hint = _(b"see 'hg help evolution.instability'")
        raise error.InputError(
            _(b"cannot %s changeset, as that will orphan %d descendants")
            % (action, len(newunstable)),
            hint=hint,
        )

    if not obsolete.isenabled(repo, obsolete.allowdivergenceopt):
        new_divergence = _find_new_divergence(repo, revs)
        if new_divergence:
            local_ctx, other_ctx, base_ctx = new_divergence
            msg = _(
                b'cannot %s %s, as that creates content-divergence with %s'
            ) % (
                action,
                local_ctx,
                other_ctx,
            )
            if local_ctx.rev() != base_ctx.rev():
                msg += _(b', from %s') % base_ctx
            if repo.ui.verbose:
                if local_ctx.rev() != base_ctx.rev():
                    msg += _(
                        b'\n    changeset %s is a successor of ' b'changeset %s'
                    ) % (local_ctx, base_ctx)
                msg += _(
                    b'\n    changeset %s already has a successor in '
                    b'changeset %s\n'
                    b'    rewriting changeset %s would create '
                    b'"content-divergence"\n'
                    b'    set experimental.evolution.allowdivergence=True to '
                    b'skip this check'
                ) % (base_ctx, other_ctx, local_ctx)
                raise error.InputError(
                    msg,
                    hint=_(
                        b"see 'hg help evolution.instability' for details on content-divergence"
                    ),
                )
            else:
                raise error.InputError(
                    msg,
                    hint=_(
                        b"add --verbose for details or see "
                        b"'hg help evolution.instability'"
                    ),
                )


def disallowednewunstable(repo, revs):
    """Checks whether editing the revs will create new unstable changesets and
    are we allowed to create them.

    To allow new unstable changesets, set the config:
        `experimental.evolution.allowunstable=True`
    """
    allowunstable = obsolete.isenabled(repo, obsolete.allowunstableopt)
    if allowunstable:
        return revset.baseset()
    return repo.revs(b"(%ld::) - %ld", revs, revs)


def _find_new_divergence(repo, revs):
    obsrevs = repo.revs(b'%ld and obsolete()', revs)
    for r in obsrevs:
        div = find_new_divergence_from(repo, repo[r])
        if div:
            return (repo[r], repo[div[0]], repo.unfiltered()[div[1]])
    return None


def find_new_divergence_from(repo, ctx):
    """return divergent revision if rewriting an obsolete cset (ctx) will
    create divergence

    Returns (<other node>, <common ancestor node>) or None
    """
    if not ctx.obsolete():
        return None
    # We need to check two cases that can cause divergence:
    # case 1: the rev being rewritten has a non-obsolete successor (easily
    #     detected by successorssets)
    sset = obsutil.successorssets(repo, ctx.node())
    if sset:
        return (sset[0][0], ctx.node())
    else:
        # case 2: one of the precursors of the rev being revived has a
        #     non-obsolete successor (we need divergentsets for this)
        divsets = obsutil.divergentsets(repo, ctx)
        if divsets:
            nsuccset = divsets[0][b'divergentnodes']
            prec = divsets[0][b'commonpredecessor']
            return (nsuccset[0], prec)
        return None


def skip_empty_successor(ui, command):
    empty_successor = ui.config(b'rewrite', b'empty-successor')
    if empty_successor == b'skip':
        return True
    elif empty_successor == b'keep':
        return False
    else:
        raise error.ConfigError(
            _(
                b"%s doesn't know how to handle config "
                b"rewrite.empty-successor=%s (only 'skip' and 'keep' are "
                b"supported)"
            )
            % (command, empty_successor)
        )


def update_hash_refs(repo, commitmsg, pending=None):
    """Replace all obsolete commit hashes in the message with the current hash.

    If the obsolete commit was split or is divergent, the hash is not replaced
    as there's no way to know which successor to choose.

    For commands that update a series of commits in the current transaction, the
    new obsolete markers can be considered by setting ``pending`` to a mapping
    of ``pending[oldnode] = [successor_node1, successor_node2,..]``.
    """
    if not pending:
        pending = {}
    cache = {}
    hashes = re.findall(NODE_RE, commitmsg)
    unfi = repo.unfiltered()
    for h in hashes:
        try:
            fullnode = scmutil.resolvehexnodeidprefix(unfi, h)
        except error.WdirUnsupported:
            # Someone has an fffff... in a commit message we're
            # rewriting. Don't try rewriting that.
            continue
        if fullnode is None:
            continue
        ctx = unfi[fullnode]
        if not ctx.obsolete():
            successors = pending.get(fullnode)
            if successors is None:
                continue
            # obsutil.successorssets() returns a list of list of nodes
            successors = [successors]
        else:
            successors = obsutil.successorssets(repo, ctx.node(), cache=cache)

        # We can't make any assumptions about how to update the hash if the
        # cset in question was split or diverged.
        if len(successors) == 1 and len(successors[0]) == 1:
            successor = successors[0][0]
            if successor is not None:
                newhash = hex(successor)
                commitmsg = commitmsg.replace(h, newhash[: len(h)])
            else:
                repo.ui.note(
                    _(
                        b'The stale commit message reference to %s could '
                        b'not be updated\n(The referenced commit was dropped)\n'
                    )
                    % h
                )
        else:
            repo.ui.note(
                _(
                    b'The stale commit message reference to %s could '
                    b'not be updated\n'
                )
                % h
            )

    return commitmsg