view hgext/blackbox.py @ 42619:20d0e59be79b

tests: show the files fields of changelogs for many merges

I don't think there's coverage for many of the subtle cases, and I
found it hard to understand what the code is doing by reading it.  The
test takes 40s to run on a laptop, or 9s with --chg.

I have yet to find a description of what the files field is supposed
to be for merges. I thought it could be one of:
1. the files added/modified/removed relative to p1 (wouldn't seem
  useful, but `hg diff -c -r mergerev` has this behavior)
2. the files with filelog nodes not in either parent (i.e., what is
needed to create a bundle out of a commit)
3. the files added/removed/modified files by merge itself [1]

It's clearly not 1, because file contents merges are symmetric. It's
clearly not 2 because removed files and exec bit changes are
listed. It's also not 3 but I think it's intended to be 3 and the
differences are bugs.

Assuming 3, the test shows that, for merges, the list of files both
overapproximates and underapproximates. All the cases involve file
changes not in the filelog but in the manifest (existence of file
at revision, exec bit and file vs symlink).

I didn't look at all underapproximations, but they looked minor. The
two overapproximations are problematic though because they both cause
potentially long lists of files when merging cleanly.


[1] even what it means for the merge commit itself to change a file is
not completely trivial. A file in the merge being the same as in one
of the parent is too lax as it would consider that merges change
nothing when they revert all the changes done on one side. The
criteria used in the test and in the next commit for "merge didn't
touch a file" is:
- the parents and the merge all have the same file
- or, one parent didn't touch the file and the other parent contains
  the same file as the merge

Differential Revision: https://phab.mercurial-scm.org/D6612

# blackbox.py - log repository events to a file for post-mortem debugging
#
# Copyright 2010 Nicolas Dumazet
# Copyright 2013 Facebook, Inc.
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.

"""log repository events to a blackbox for debugging

Logs event information to .hg/blackbox.log to help debug and diagnose problems.
The events that get logged can be configured via the blackbox.track and
blackbox.ignore config keys.

Examples::

  [blackbox]
  track = *
  ignore = pythonhook
  # dirty is *EXPENSIVE* (slow);
  # each log entry indicates `+` if the repository is dirty, like :hg:`id`.
  dirty = True
  # record the source of log messages
  logsource = True

  [blackbox]
  track = command, commandfinish, commandexception, exthook, pythonhook

  [blackbox]
  track = incoming

  [blackbox]
  # limit the size of a log file
  maxsize = 1.5 MB
  # rotate up to N log files when the current one gets too big
  maxfiles = 3

  [blackbox]
  # Include nanoseconds in log entries with %f (see Python function
  # datetime.datetime.strftime)
  date-format = '%Y-%m-%d @ %H:%M:%S.%f'

"""

from __future__ import absolute_import

import re

from mercurial.i18n import _
from mercurial.node import hex

from mercurial import (
    encoding,
    loggingutil,
    registrar,
)
from mercurial.utils import (
    dateutil,
    procutil,
)

# Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
# extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
# be specifying the version(s) of Mercurial they are tested with, or
# leave the attribute unspecified.
testedwith = 'ships-with-hg-core'

cmdtable = {}
command = registrar.command(cmdtable)

configtable = {}
configitem = registrar.configitem(configtable)

configitem('blackbox', 'dirty',
    default=False,
)
configitem('blackbox', 'maxsize',
    default='1 MB',
)
configitem('blackbox', 'logsource',
    default=False,
)
configitem('blackbox', 'maxfiles',
    default=7,
)
configitem('blackbox', 'track',
    default=lambda: ['*'],
)
configitem('blackbox', 'ignore',
    default=lambda: ['chgserver', 'cmdserver', 'extension'],
)
configitem('blackbox', 'date-format',
    default='%Y/%m/%d %H:%M:%S',
)

_lastlogger = loggingutil.proxylogger()

class blackboxlogger(object):
    def __init__(self, ui, repo):
        self._repo = repo
        self._trackedevents = set(ui.configlist('blackbox', 'track'))
        self._ignoredevents = set(ui.configlist('blackbox', 'ignore'))
        self._maxfiles = ui.configint('blackbox', 'maxfiles')
        self._maxsize = ui.configbytes('blackbox', 'maxsize')
        self._inlog = False

    def tracked(self, event):
        return ((b'*' in self._trackedevents
                 and event not in self._ignoredevents)
                or event in self._trackedevents)

    def log(self, ui, event, msg, opts):
        # self._log() -> ctx.dirty() may create new subrepo instance, which
        # ui is derived from baseui. So the recursion guard in ui.log()
        # doesn't work as it's local to the ui instance.
        if self._inlog:
            return
        self._inlog = True
        try:
            self._log(ui, event, msg, opts)
        finally:
            self._inlog = False

    def _log(self, ui, event, msg, opts):
        default = ui.configdate('devel', 'default-date')
        date = dateutil.datestr(default, ui.config('blackbox', 'date-format'))
        user = procutil.getuser()
        pid = '%d' % procutil.getpid()
        changed = ''
        ctx = self._repo[None]
        parents = ctx.parents()
        rev = ('+'.join([hex(p.node()) for p in parents]))
        if (ui.configbool('blackbox', 'dirty') and
            ctx.dirty(missing=True, merge=False, branch=False)):
            changed = '+'
        if ui.configbool('blackbox', 'logsource'):
            src = ' [%s]' % event
        else:
            src = ''
        try:
            fmt = '%s %s @%s%s (%s)%s> %s'
            args = (date, user, rev, changed, pid, src, msg)
            with loggingutil.openlogfile(
                    ui, self._repo.vfs, name='blackbox.log',
                    maxfiles=self._maxfiles, maxsize=self._maxsize) as fp:
                fp.write(fmt % args)
        except (IOError, OSError) as err:
            # deactivate this to avoid failed logging again
            self._trackedevents.clear()
            ui.debug('warning: cannot write to blackbox.log: %s\n' %
                     encoding.strtolocal(err.strerror))
            return
        _lastlogger.logger = self

def uipopulate(ui):
    ui.setlogger(b'blackbox', _lastlogger)

def reposetup(ui, repo):
    # During 'hg pull' a httppeer repo is created to represent the remote repo.
    # It doesn't have a .hg directory to put a blackbox in, so we don't do
    # the blackbox setup for it.
    if not repo.local():
        return

    # Since blackbox.log is stored in the repo directory, the logger should be
    # instantiated per repository.
    logger = blackboxlogger(ui, repo)
    ui.setlogger(b'blackbox', logger)

    # Set _lastlogger even if ui.log is not called. This gives blackbox a
    # fallback place to log
    if _lastlogger.logger is None:
        _lastlogger.logger = logger

    repo._wlockfreeprefix.add('blackbox.log')

@command('blackbox',
    [('l', 'limit', 10, _('the number of events to show')),
    ],
    _('hg blackbox [OPTION]...'),
    helpcategory=command.CATEGORY_MAINTENANCE,
    helpbasic=True)
def blackbox(ui, repo, *revs, **opts):
    '''view the recent repository events
    '''

    if not repo.vfs.exists('blackbox.log'):
        return

    limit = opts.get(r'limit')
    fp = repo.vfs('blackbox.log', 'r')
    lines = fp.read().split('\n')

    count = 0
    output = []
    for line in reversed(lines):
        if count >= limit:
            break

        # count the commands by matching lines like: 2013/01/23 19:13:36 root>
        if re.match(br'^\d{4}/\d{2}/\d{2} \d{2}:\d{2}:\d{2} .*> .*', line):
            count += 1
        output.append(line)

    ui.status('\n'.join(reversed(output)))