##// END OF EJS Templates
localrepo: iteratively derive local repository type...
localrepo: iteratively derive local repository type This commit implements the dynamic local repository type derivation that was explained in the recent commit bfeab472e3c0 "localrepo: create new function for instantiating a local repo object." Instead of a static localrepository class/type which must be customized after construction, we now dynamically construct a type by building up base classes/types to represent specific repository interfaces. Conceptually, the end state is similar to what was happening when various extensions would monkeypatch the __class__ of newly-constructed repo instances. However, the approach is inverted. Instead of making the instance then customizing it, we do the customization up front by influencing the behavior of the type then we instantiate that custom type. This approach gives us much more flexibility. For example, we can use completely separate classes for implementing different aspects of the repository. For example, we could have one class representing revlog-based file storage and another representing non-revlog based file storage. When then choose which implementation to use based on the presence of repo requirements. A concern with this approach is that it creates a lot more types and complexity and that complexity adds overhead. Yes, it is true that this approach will result in more types being created. Yes, this is more complicated than traditional "instantiate a static type." However, I believe the alternatives to supporting alternate storage backends are just as complicated. (Before I arrived at this solution, I had patches storing factory functions on local repo instances for e.g. constructing a file storage instance. We ended up having a handful of these. And this was logically identical to assigning custom methods. Since we were logically changing the type of the instance, I figured it would be better to just use specialized types instead of introducing levels of abstraction at run-time.) On the performance front, I don't believe that having N base classes has any significant performance overhead compared to just a single base class. Intuition says that Python will need to iterate the base classes to find an attribute. However, CPython caches method lookups: as long as the __class__ or MRO isn't changing, method attribute lookup should be constant time after first access. And non-method attributes are stored in __dict__, of which there is only 1 per object, so the number of base classes for __dict__ is irrelevant. Anyway, this commit splits up the monolithic completelocalrepository interface into sub-interfaces: 1 for file storage and 1 representing everything else. We've taught ``makelocalrepository()`` to call a series of factory functions which will produce types implementing specific interfaces. It then calls type() to create a new type from the built-up list of base types. This commit should be considered a start and not the end state. I suspect we'll hit a number of problems as we start to implement alternate storage backends: * Passing custom arguments to __init__ and setting custom attributes on __dict__. * Customizing the set of interfaces that are needed. e.g. the "readonly" intent could translate to not requesting an interface providing methods related to writing. * More ergonomic way for extensions to insert themselves so their callbacks aren't unconditionally called. * Wanting to modify vfs instances, other arguments passed to __init__. That being said, this code is usable in its current state and I'm convinced future commits will demonstrate the value in this approach. Differential Revision: https://phab.mercurial-scm.org/D4642

File last commit:

r39655:2aa12d7f default
r39800:e4e88157 default
Show More
monotone.py
374 lines | 13.1 KiB | text/x-python | PythonLexer
# monotone.py - monotone support for the convert extension
#
# Copyright 2008, 2009 Mikkel Fahnoe Jorgensen <mikkel@dvide.com> 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 os
import re
from mercurial.i18n import _
from mercurial import (
error,
pycompat,
)
from mercurial.utils import dateutil
from . import common
class monotone_source(common.converter_source, common.commandline):
def __init__(self, ui, repotype, path=None, revs=None):
common.converter_source.__init__(self, ui, repotype, path, revs)
if revs and len(revs) > 1:
raise error.Abort(_('monotone source does not support specifying '
'multiple revs'))
common.commandline.__init__(self, ui, 'mtn')
self.ui = ui
self.path = path
self.automatestdio = False
self.revs = revs
norepo = common.NoRepo(_("%s does not look like a monotone repository")
% path)
if not os.path.exists(os.path.join(path, '_MTN')):
# Could be a monotone repository (SQLite db file)
try:
f = open(path, 'rb')
header = f.read(16)
f.close()
except IOError:
header = ''
if header != 'SQLite format 3\x00':
raise norepo
# regular expressions for parsing monotone output
space = br'\s*'
name = br'\s+"((?:\\"|[^"])*)"\s*'
value = name
revision = br'\s+\[(\w+)\]\s*'
lines = br'(?:.|\n)+'
self.dir_re = re.compile(space + "dir" + name)
self.file_re = re.compile(space + "file" + name +
"content" + revision)
self.add_file_re = re.compile(space + "add_file" + name +
"content" + revision)
self.patch_re = re.compile(space + "patch" + name +
"from" + revision + "to" + revision)
self.rename_re = re.compile(space + "rename" + name + "to" + name)
self.delete_re = re.compile(space + "delete" + name)
self.tag_re = re.compile(space + "tag" + name + "revision" +
revision)
self.cert_re = re.compile(lines + space + "name" + name +
"value" + value)
attr = space + "file" + lines + space + "attr" + space
self.attr_execute_re = re.compile(attr + '"mtn:execute"' +
space + '"true"')
# cached data
self.manifest_rev = None
self.manifest = None
self.files = None
self.dirs = None
common.checktool('mtn', abort=False)
def mtnrun(self, *args, **kwargs):
if self.automatestdio:
return self.mtnrunstdio(*args, **kwargs)
else:
return self.mtnrunsingle(*args, **kwargs)
def mtnrunsingle(self, *args, **kwargs):
kwargs[r'd'] = self.path
return self.run0('automate', *args, **kwargs)
def mtnrunstdio(self, *args, **kwargs):
# Prepare the command in automate stdio format
kwargs = pycompat.byteskwargs(kwargs)
command = []
for k, v in kwargs.iteritems():
command.append("%s:%s" % (len(k), k))
if v:
command.append("%s:%s" % (len(v), v))
if command:
command.insert(0, 'o')
command.append('e')
command.append('l')
for arg in args:
command += "%s:%s" % (len(arg), arg)
command.append('e')
command = ''.join(command)
self.ui.debug("mtn: sending '%s'\n" % command)
self.mtnwritefp.write(command)
self.mtnwritefp.flush()
return self.mtnstdioreadcommandoutput(command)
def mtnstdioreadpacket(self):
read = None
commandnbr = ''
while read != ':':
read = self.mtnreadfp.read(1)
if not read:
raise error.Abort(_('bad mtn packet - no end of commandnbr'))
commandnbr += read
commandnbr = commandnbr[:-1]
stream = self.mtnreadfp.read(1)
if stream not in 'mewptl':
raise error.Abort(_('bad mtn packet - bad stream type %s') % stream)
read = self.mtnreadfp.read(1)
if read != ':':
raise error.Abort(_('bad mtn packet - no divider before size'))
read = None
lengthstr = ''
while read != ':':
read = self.mtnreadfp.read(1)
if not read:
raise error.Abort(_('bad mtn packet - no end of packet size'))
lengthstr += read
try:
length = long(lengthstr[:-1])
except TypeError:
raise error.Abort(_('bad mtn packet - bad packet size %s')
% lengthstr)
read = self.mtnreadfp.read(length)
if len(read) != length:
raise error.Abort(_("bad mtn packet - unable to read full packet "
"read %s of %s") % (len(read), length))
return (commandnbr, stream, length, read)
def mtnstdioreadcommandoutput(self, command):
retval = []
while True:
commandnbr, stream, length, output = self.mtnstdioreadpacket()
self.ui.debug('mtn: read packet %s:%s:%s\n' %
(commandnbr, stream, length))
if stream == 'l':
# End of command
if output != '0':
raise error.Abort(_("mtn command '%s' returned %s") %
(command, output))
break
elif stream in 'ew':
# Error, warning output
self.ui.warn(_('%s error:\n') % self.command)
self.ui.warn(output)
elif stream == 'p':
# Progress messages
self.ui.debug('mtn: ' + output)
elif stream == 'm':
# Main stream - command output
retval.append(output)
return ''.join(retval)
def mtnloadmanifest(self, rev):
if self.manifest_rev == rev:
return
self.manifest = self.mtnrun("get_manifest_of", rev).split("\n\n")
self.manifest_rev = rev
self.files = {}
self.dirs = {}
for e in self.manifest:
m = self.file_re.match(e)
if m:
attr = ""
name = m.group(1)
node = m.group(2)
if self.attr_execute_re.match(e):
attr += "x"
self.files[name] = (node, attr)
m = self.dir_re.match(e)
if m:
self.dirs[m.group(1)] = True
def mtnisfile(self, name, rev):
# a non-file could be a directory or a deleted or renamed file
self.mtnloadmanifest(rev)
return name in self.files
def mtnisdir(self, name, rev):
self.mtnloadmanifest(rev)
return name in self.dirs
def mtngetcerts(self, rev):
certs = {"author":"<missing>", "date":"<missing>",
"changelog":"<missing>", "branch":"<missing>"}
certlist = self.mtnrun("certs", rev)
# mtn < 0.45:
# key "test@selenic.com"
# mtn >= 0.45:
# key [ff58a7ffb771907c4ff68995eada1c4da068d328]
certlist = re.split('\n\n key ["\[]', certlist)
for e in certlist:
m = self.cert_re.match(e)
if m:
name, value = m.groups()
value = value.replace(r'\"', '"')
value = value.replace(r'\\', '\\')
certs[name] = value
# Monotone may have subsecond dates: 2005-02-05T09:39:12.364306
# and all times are stored in UTC
certs["date"] = certs["date"].split('.')[0] + " UTC"
return certs
# implement the converter_source interface:
def getheads(self):
if not self.revs:
return self.mtnrun("leaves").splitlines()
else:
return self.revs
def getchanges(self, rev, full):
if full:
raise error.Abort(_("convert from monotone does not support "
"--full"))
revision = self.mtnrun("get_revision", rev).split("\n\n")
files = {}
ignoremove = {}
renameddirs = []
copies = {}
for e in revision:
m = self.add_file_re.match(e)
if m:
files[m.group(1)] = rev
ignoremove[m.group(1)] = rev
m = self.patch_re.match(e)
if m:
files[m.group(1)] = rev
# Delete/rename is handled later when the convert engine
# discovers an IOError exception from getfile,
# but only if we add the "from" file to the list of changes.
m = self.delete_re.match(e)
if m:
files[m.group(1)] = rev
m = self.rename_re.match(e)
if m:
toname = m.group(2)
fromname = m.group(1)
if self.mtnisfile(toname, rev):
ignoremove[toname] = 1
copies[toname] = fromname
files[toname] = rev
files[fromname] = rev
elif self.mtnisdir(toname, rev):
renameddirs.append((fromname, toname))
# Directory renames can be handled only once we have recorded
# all new files
for fromdir, todir in renameddirs:
renamed = {}
for tofile in self.files:
if tofile in ignoremove:
continue
if tofile.startswith(todir + '/'):
renamed[tofile] = fromdir + tofile[len(todir):]
# Avoid chained moves like:
# d1(/a) => d3/d1(/a)
# d2 => d3
ignoremove[tofile] = 1
for tofile, fromfile in renamed.items():
self.ui.debug (_("copying file in renamed directory "
"from '%s' to '%s'")
% (fromfile, tofile), '\n')
files[tofile] = rev
copies[tofile] = fromfile
for fromfile in renamed.values():
files[fromfile] = rev
return (files.items(), copies, set())
def getfile(self, name, rev):
if not self.mtnisfile(name, rev):
return None, None
try:
data = self.mtnrun("get_file_of", name, r=rev)
except Exception:
return None, None
self.mtnloadmanifest(rev)
node, attr = self.files.get(name, (None, ""))
return data, attr
def getcommit(self, rev):
extra = {}
certs = self.mtngetcerts(rev)
if certs.get('suspend') == certs["branch"]:
extra['close'] = 1
dateformat = "%Y-%m-%dT%H:%M:%S"
return common.commit(
author=certs["author"],
date=dateutil.datestr(dateutil.strdate(certs["date"], dateformat)),
desc=certs["changelog"],
rev=rev,
parents=self.mtnrun("parents", rev).splitlines(),
branch=certs["branch"],
extra=extra)
def gettags(self):
tags = {}
for e in self.mtnrun("tags").split("\n\n"):
m = self.tag_re.match(e)
if m:
tags[m.group(1)] = m.group(2)
return tags
def getchangedfiles(self, rev, i):
# This function is only needed to support --filemap
# ... and we don't support that
raise NotImplementedError
def before(self):
# Check if we have a new enough version to use automate stdio
version = 0.0
try:
versionstr = self.mtnrunsingle("interface_version")
version = float(versionstr)
except Exception:
raise error.Abort(_("unable to determine mtn automate interface "
"version"))
if version >= 12.0:
self.automatestdio = True
self.ui.debug("mtn automate version %f - using automate stdio\n" %
version)
# launch the long-running automate stdio process
self.mtnwritefp, self.mtnreadfp = self._run2('automate', 'stdio',
'-d', self.path)
# read the headers
read = self.mtnreadfp.readline()
if read != 'format-version: 2\n':
raise error.Abort(_('mtn automate stdio header unexpected: %s')
% read)
while read != '\n':
read = self.mtnreadfp.readline()
if not read:
raise error.Abort(_("failed to reach end of mtn automate "
"stdio headers"))
else:
self.ui.debug("mtn automate version %s - not using automate stdio "
"(automate >= 12.0 - mtn >= 0.46 is needed)\n" % version)
def after(self):
if self.automatestdio:
self.mtnwritefp.close()
self.mtnwritefp = None
self.mtnreadfp.close()
self.mtnreadfp = None