add Client.resubmit for re-running tasks

closes gh-411

* allow `content` in session.serialize to be a unicode object, because mongo+JSON cannot be relied upon to produce encoded bytes.

IPython/parallel/client/client.py

@@ -1,1292 +1,1354 b''
 """A semi-synchronous Client for the ZMQ cluster"""
 #-----------------------------------------------------------------------------
 #  Copyright (C) 2010  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-----------------------------------------------------------------------------
 
 #-----------------------------------------------------------------------------
 # Imports
 #-----------------------------------------------------------------------------
 
 import os
 import json
 import time
 import warnings
 from datetime import datetime
 from getpass import getpass
 from pprint import pprint
 
 pjoin = os.path.join
 
 import zmq
 # from zmq.eventloop import ioloop, zmqstream
 
 from IPython.utils.path import get_ipython_dir
 from IPython.utils.traitlets import (HasTraits, Int, Instance, CUnicode, 
                                     Dict, List, Bool, Str, Set)
 from IPython.external.decorator import decorator
 from IPython.external.ssh import tunnel
 
 from IPython.parallel import error
 from IPython.parallel import streamsession as ss
 from IPython.parallel import util
 
 from .asyncresult import AsyncResult, AsyncHubResult
 from IPython.parallel.apps.clusterdir import ClusterDir, ClusterDirError
 from .view import DirectView, LoadBalancedView
 
 #--------------------------------------------------------------------------
 # Decorators for Client methods
 #--------------------------------------------------------------------------
 
 @decorator
 def spin_first(f, self, *args, **kwargs):
     """Call spin() to sync state prior to calling the method."""
     self.spin()
     return f(self, *args, **kwargs)
 
 
 #--------------------------------------------------------------------------
 # Classes
 #--------------------------------------------------------------------------
 
 class Metadata(dict):
     """Subclass of dict for initializing metadata values.
     
     Attribute access works on keys.
     
     These objects have a strict set of keys - errors will raise if you try
     to add new keys.
     """
     def __init__(self, *args, **kwargs):
         dict.__init__(self)
         md = {'msg_id' : None,
               'submitted' : None,
               'started' : None,
               'completed' : None,
               'received' : None,
               'engine_uuid' : None,
               'engine_id' : None,
               'follow' : None,
               'after' : None,
               'status' : None,
 
               'pyin' : None,
               'pyout' : None,
               'pyerr' : None,
               'stdout' : '',
               'stderr' : '',
             }
         self.update(md)
         self.update(dict(*args, **kwargs))
     
     def __getattr__(self, key):
         """getattr aliased to getitem"""
         if key in self.iterkeys():
             return self[key]
         else:
             raise AttributeError(key)
 
     def __setattr__(self, key, value):
         """setattr aliased to setitem, with strict"""
         if key in self.iterkeys():
             self[key] = value
         else:
             raise AttributeError(key)
     
     def __setitem__(self, key, value):
         """strict static key enforcement"""
         if key in self.iterkeys():
             dict.__setitem__(self, key, value)
         else:
             raise KeyError(key)
         
 
 class Client(HasTraits):
     """A semi-synchronous client to the IPython ZMQ cluster
     
     Parameters
     ----------
     
     url_or_file : bytes; zmq url or path to ipcontroller-client.json
         Connection information for the Hub's registration.  If a json connector
         file is given, then likely no further configuration is necessary.
         [Default: use profile]
     profile : bytes
         The name of the Cluster profile to be used to find connector information.
         [Default: 'default']
     context : zmq.Context
         Pass an existing zmq.Context instance, otherwise the client will create its own.
     username : bytes
         set username to be passed to the Session object
     debug : bool
         flag for lots of message printing for debug purposes
  
     #-------------- ssh related args ----------------
     # These are args for configuring the ssh tunnel to be used
     # credentials are used to forward connections over ssh to the Controller
     # Note that the ip given in `addr` needs to be relative to sshserver
     # The most basic case is to leave addr as pointing to localhost (127.0.0.1),
     # and set sshserver as the same machine the Controller is on. However, 
     # the only requirement is that sshserver is able to see the Controller
     # (i.e. is within the same trusted network).
     
     sshserver : str
         A string of the form passed to ssh, i.e. 'server.tld' or '[e-mail]:port'
         If keyfile or password is specified, and this is not, it will default to
         the ip given in addr.
     sshkey : str; path to public ssh key file
         This specifies a key to be used in ssh login, default None.
         Regular default ssh keys will be used without specifying this argument.
     password : str 
         Your ssh password to sshserver. Note that if this is left None,
         you will be prompted for it if passwordless key based login is unavailable.
     paramiko : bool
         flag for whether to use paramiko instead of shell ssh for tunneling.
         [default: True on win32, False else]
     
     ------- exec authentication args -------
     If even localhost is untrusted, you can have some protection against
     unauthorized execution by using a key.  Messages are still sent
     as cleartext, so if someone can snoop your loopback traffic this will
     not help against malicious attacks.
     
     exec_key : str
         an authentication key or file containing a key
         default: None
     
     
     Attributes
     ----------
     
     ids : list of int engine IDs
         requesting the ids attribute always synchronizes
         the registration state. To request ids without synchronization,
         use semi-private _ids attributes.
     
     history : list of msg_ids
         a list of msg_ids, keeping track of all the execution
         messages you have submitted in order.
     
     outstanding : set of msg_ids
         a set of msg_ids that have been submitted, but whose
         results have not yet been received.
     
     results : dict
         a dict of all our results, keyed by msg_id
     
     block : bool
         determines default behavior when block not specified
         in execution methods
     
     Methods
     -------
     
     spin
         flushes incoming results and registration state changes
         control methods spin, and requesting `ids` also ensures up to date
     
     wait
         wait on one or more msg_ids
     
     execution methods
         apply
         legacy: execute, run
     
     data movement
         push, pull, scatter, gather
     
     query methods
         queue_status, get_result, purge, result_status
     
     control methods
         abort, shutdown
     
     """
     
     
     block = Bool(False)
     outstanding = Set()
     results = Instance('collections.defaultdict', (dict,))
     metadata = Instance('collections.defaultdict', (Metadata,))
     history = List()
     debug = Bool(False)
     profile=CUnicode('default')
     
     _outstanding_dict = Instance('collections.defaultdict', (set,))
     _ids = List()
     _connected=Bool(False)
     _ssh=Bool(False)
     _context = Instance('zmq.Context')
     _config = Dict()
     _engines=Instance(util.ReverseDict, (), {})
     # _hub_socket=Instance('zmq.Socket')
     _query_socket=Instance('zmq.Socket')
     _control_socket=Instance('zmq.Socket')
     _iopub_socket=Instance('zmq.Socket')
     _notification_socket=Instance('zmq.Socket')
     _mux_socket=Instance('zmq.Socket')
     _task_socket=Instance('zmq.Socket')
     _task_scheme=Str()
     _closed = False
     _ignored_control_replies=Int(0)
     _ignored_hub_replies=Int(0)
     
     def __init__(self, url_or_file=None, profile='default', cluster_dir=None, ipython_dir=None,
             context=None, username=None, debug=False, exec_key=None,
             sshserver=None, sshkey=None, password=None, paramiko=None,
             timeout=10
             ):
         super(Client, self).__init__(debug=debug, profile=profile)
         if context is None:
             context = zmq.Context.instance()
         self._context = context
             
         
         self._setup_cluster_dir(profile, cluster_dir, ipython_dir)
         if self._cd is not None:
             if url_or_file is None:
                 url_or_file = pjoin(self._cd.security_dir, 'ipcontroller-client.json')
         assert url_or_file is not None, "I can't find enough information to connect to a hub!"\
             " Please specify at least one of url_or_file or profile."
         
         try:
             util.validate_url(url_or_file)
         except AssertionError:
             if not os.path.exists(url_or_file):
                 if self._cd:
                     url_or_file = os.path.join(self._cd.security_dir, url_or_file)
                 assert os.path.exists(url_or_file), "Not a valid connection file or url: %r"%url_or_file
             with open(url_or_file) as f:
                 cfg = json.loads(f.read())
         else:
             cfg = {'url':url_or_file}
         
         # sync defaults from args, json:
         if sshserver:
             cfg['ssh'] = sshserver
         if exec_key:
             cfg['exec_key'] = exec_key
         exec_key = cfg['exec_key']
         sshserver=cfg['ssh']
         url = cfg['url']
         location = cfg.setdefault('location', None)
         cfg['url'] = util.disambiguate_url(cfg['url'], location)
         url = cfg['url']
         
         self._config = cfg
         
         self._ssh = bool(sshserver or sshkey or password)
         if self._ssh and sshserver is None:
             # default to ssh via localhost
             sshserver = url.split('://')[1].split(':')[0]
         if self._ssh and password is None:
             if tunnel.try_passwordless_ssh(sshserver, sshkey, paramiko):
                 password=False
             else:
                 password = getpass("SSH Password for %s: "%sshserver)
         ssh_kwargs = dict(keyfile=sshkey, password=password, paramiko=paramiko)
         if exec_key is not None and os.path.isfile(exec_key):
             arg = 'keyfile'
         else:
             arg = 'key'
         key_arg = {arg:exec_key}
         if username is None:
             self.session = ss.StreamSession(**key_arg)
         else:
             self.session = ss.StreamSession(username, **key_arg)
         self._query_socket = self._context.socket(zmq.XREQ)
         self._query_socket.setsockopt(zmq.IDENTITY, self.session.session)
         if self._ssh:
             tunnel.tunnel_connection(self._query_socket, url, sshserver, **ssh_kwargs)
         else:
             self._query_socket.connect(url)
         
         self.session.debug = self.debug
         
         self._notification_handlers = {'registration_notification' : self._register_engine,
                                     'unregistration_notification' : self._unregister_engine,
                                     'shutdown_notification' : lambda msg: self.close(),
                                     }
         self._queue_handlers = {'execute_reply' : self._handle_execute_reply,
                                 'apply_reply' : self._handle_apply_reply}
         self._connect(sshserver, ssh_kwargs, timeout)
         
     def __del__(self):
         """cleanup sockets, but _not_ context."""
         self.close()
     
     def _setup_cluster_dir(self, profile, cluster_dir, ipython_dir):
         if ipython_dir is None:
             ipython_dir = get_ipython_dir()
         if cluster_dir is not None:
             try:
                 self._cd = ClusterDir.find_cluster_dir(cluster_dir)
                 return
             except ClusterDirError:
                 pass
         elif profile is not None:
             try:
                 self._cd = ClusterDir.find_cluster_dir_by_profile(
                     ipython_dir, profile)
                 return
             except ClusterDirError:
                 pass
         self._cd = None
     
     def _update_engines(self, engines):
         """Update our engines dict and _ids from a dict of the form: {id:uuid}."""
         for k,v in engines.iteritems():
             eid = int(k)
             self._engines[eid] = bytes(v) # force not unicode
             self._ids.append(eid)
         self._ids = sorted(self._ids)
         if sorted(self._engines.keys()) != range(len(self._engines)) and \
                         self._task_scheme == 'pure' and self._task_socket:
             self._stop_scheduling_tasks()
     
     def _stop_scheduling_tasks(self):
         """Stop scheduling tasks because an engine has been unregistered
         from a pure ZMQ scheduler.
         """
         self._task_socket.close()
         self._task_socket = None
         msg = "An engine has been unregistered, and we are using pure " +\
               "ZMQ task scheduling.  Task farming will be disabled."
         if self.outstanding:
             msg += " If you were running tasks when this happened, " +\
                    "some `outstanding` msg_ids may never resolve."
         warnings.warn(msg, RuntimeWarning)
     
     def _build_targets(self, targets):
         """Turn valid target IDs or 'all' into two lists:
         (int_ids, uuids).
         """
         if not self._ids:
             # flush notification socket if no engines yet, just in case
             if not self.ids:
                 raise error.NoEnginesRegistered("Can't build targets without any engines")
         
         if targets is None:
             targets = self._ids
         elif isinstance(targets, str):
             if targets.lower() == 'all':
                 targets = self._ids
             else:
                 raise TypeError("%r not valid str target, must be 'all'"%(targets))
         elif isinstance(targets, int):
             if targets < 0:
                 targets = self.ids[targets]
             if targets not in self._ids:
                 raise IndexError("No such engine: %i"%targets)
             targets = [targets]
         
         if isinstance(targets, slice):
             indices = range(len(self._ids))[targets]
             ids = self.ids
             targets = [ ids[i] for i in indices ]
         
         if not isinstance(targets, (tuple, list, xrange)):
             raise TypeError("targets by int/slice/collection of ints only, not %s"%(type(targets)))
             
         return [self._engines[t] for t in targets], list(targets)
     
     def _connect(self, sshserver, ssh_kwargs, timeout):
         """setup all our socket connections to the cluster. This is called from
         __init__."""
         
         # Maybe allow reconnecting?
         if self._connected:
             return
         self._connected=True
 
         def connect_socket(s, url):
             url = util.disambiguate_url(url, self._config['location'])
             if self._ssh:
                 return tunnel.tunnel_connection(s, url, sshserver, **ssh_kwargs)
             else:
                 return s.connect(url)
             
         self.session.send(self._query_socket, 'connection_request')
         r,w,x = zmq.select([self._query_socket],[],[], timeout)
         if not r:
             raise error.TimeoutError("Hub connection request timed out")
         idents,msg = self.session.recv(self._query_socket,mode=0)
         if self.debug:
             pprint(msg)
         msg = ss.Message(msg)
         content = msg.content
         self._config['registration'] = dict(content)
         if content.status == 'ok':
             if content.mux:
                 self._mux_socket = self._context.socket(zmq.XREQ)
                 self._mux_socket.setsockopt(zmq.IDENTITY, self.session.session)
                 connect_socket(self._mux_socket, content.mux)
             if content.task:
                 self._task_scheme, task_addr = content.task
                 self._task_socket = self._context.socket(zmq.XREQ)
                 self._task_socket.setsockopt(zmq.IDENTITY, self.session.session)
                 connect_socket(self._task_socket, task_addr)
             if content.notification:
                 self._notification_socket = self._context.socket(zmq.SUB)
                 connect_socket(self._notification_socket, content.notification)
                 self._notification_socket.setsockopt(zmq.SUBSCRIBE, b'')
             # if content.query:
             #     self._query_socket = self._context.socket(zmq.XREQ)
             #     self._query_socket.setsockopt(zmq.IDENTITY, self.session.session)
             #     connect_socket(self._query_socket, content.query)
             if content.control:
                 self._control_socket = self._context.socket(zmq.XREQ)
                 self._control_socket.setsockopt(zmq.IDENTITY, self.session.session)
                 connect_socket(self._control_socket, content.control)
             if content.iopub:
                 self._iopub_socket = self._context.socket(zmq.SUB)
                 self._iopub_socket.setsockopt(zmq.SUBSCRIBE, b'')
                 self._iopub_socket.setsockopt(zmq.IDENTITY, self.session.session)
                 connect_socket(self._iopub_socket, content.iopub)
             self._update_engines(dict(content.engines))
         else:
             self._connected = False
             raise Exception("Failed to connect!")
     
     #--------------------------------------------------------------------------
     # handlers and callbacks for incoming messages
     #--------------------------------------------------------------------------
     
     def _unwrap_exception(self, content):
         """unwrap exception, and remap engine_id to int."""
         e = error.unwrap_exception(content)
         # print e.traceback
         if e.engine_info:
             e_uuid = e.engine_info['engine_uuid']
             eid = self._engines[e_uuid]
             e.engine_info['engine_id'] = eid
         return e
     
     def _extract_metadata(self, header, parent, content):
         md = {'msg_id' : parent['msg_id'],
               'received' : datetime.now(),
               'engine_uuid' : header.get('engine', None),
               'follow' : parent.get('follow', []),
               'after' : parent.get('after', []),
               'status' : content['status'],
             }
         
         if md['engine_uuid'] is not None:
             md['engine_id'] = self._engines.get(md['engine_uuid'], None)
         
         if 'date' in parent:
             md['submitted'] = datetime.strptime(parent['date'], util.ISO8601)
         if 'started' in header:
             md['started'] = datetime.strptime(header['started'], util.ISO8601)
         if 'date' in header:
             md['completed'] = datetime.strptime(header['date'], util.ISO8601)
         return md
     
     def _register_engine(self, msg):
         """Register a new engine, and update our connection info."""
         content = msg['content']
         eid = content['id']
         d = {eid : content['queue']}
         self._update_engines(d)
 
     def _unregister_engine(self, msg):
         """Unregister an engine that has died."""
         content = msg['content']
         eid = int(content['id'])
         if eid in self._ids:
             self._ids.remove(eid)
             uuid = self._engines.pop(eid)
             
             self._handle_stranded_msgs(eid, uuid)
                 
         if self._task_socket and self._task_scheme == 'pure':
             self._stop_scheduling_tasks()
     
     def _handle_stranded_msgs(self, eid, uuid):
         """Handle messages known to be on an engine when the engine unregisters.
         
         It is possible that this will fire prematurely - that is, an engine will
         go down after completing a result, and the client will be notified
         of the unregistration and later receive the successful result.
         """
         
         outstanding = self._outstanding_dict[uuid]
         
         for msg_id in list(outstanding):
             if msg_id in self.results:
                 # we already 
                 continue
             try:
                 raise error.EngineError("Engine %r died while running task %r"%(eid, msg_id))
             except:
                 content = error.wrap_exception()
             # build a fake message:
             parent = {}
             header = {}
             parent['msg_id'] = msg_id
             header['engine'] = uuid
             header['date'] = datetime.now().strftime(util.ISO8601)
             msg = dict(parent_header=parent, header=header, content=content)
             self._handle_apply_reply(msg)
     
     def _handle_execute_reply(self, msg):
         """Save the reply to an execute_request into our results.
         
         execute messages are never actually used. apply is used instead.
         """
         
         parent = msg['parent_header']
         msg_id = parent['msg_id']
         if msg_id not in self.outstanding:
             if msg_id in self.history:
                 print ("got stale result: %s"%msg_id)
             else:
                 print ("got unknown result: %s"%msg_id)
         else:
             self.outstanding.remove(msg_id)
         self.results[msg_id] = self._unwrap_exception(msg['content'])
     
     def _handle_apply_reply(self, msg):
         """Save the reply to an apply_request into our results."""
         parent = msg['parent_header']
         msg_id = parent['msg_id']
         if msg_id not in self.outstanding:
             if msg_id in self.history:
                 print ("got stale result: %s"%msg_id)
                 print self.results[msg_id]
                 print msg
             else:
                 print ("got unknown result: %s"%msg_id)
         else:
             self.outstanding.remove(msg_id)
         content = msg['content']
         header = msg['header']
         
         # construct metadata:
         md = self.metadata[msg_id]
         md.update(self._extract_metadata(header, parent, content))
         # is this redundant?
         self.metadata[msg_id] = md
         
         e_outstanding = self._outstanding_dict[md['engine_uuid']]
         if msg_id in e_outstanding:
             e_outstanding.remove(msg_id)
         
         # construct result:
         if content['status'] == 'ok':
             self.results[msg_id] = util.unserialize_object(msg['buffers'])[0]
         elif content['status'] == 'aborted':
             self.results[msg_id] = error.TaskAborted(msg_id)
         elif content['status'] == 'resubmitted':
             # TODO: handle resubmission
             pass
         else:
             self.results[msg_id] = self._unwrap_exception(content)
     
     def _flush_notifications(self):
         """Flush notifications of engine registrations waiting
         in ZMQ queue."""
         msg = self.session.recv(self._notification_socket, mode=zmq.NOBLOCK)
         while msg is not None:
             if self.debug:
                 pprint(msg)
             msg = msg[-1]
             msg_type = msg['msg_type']
             handler = self._notification_handlers.get(msg_type, None)
             if handler is None:
                 raise Exception("Unhandled message type: %s"%msg.msg_type)
             else:
                 handler(msg)
             msg = self.session.recv(self._notification_socket, mode=zmq.NOBLOCK)
     
     def _flush_results(self, sock):
         """Flush task or queue results waiting in ZMQ queue."""
         msg = self.session.recv(sock, mode=zmq.NOBLOCK)
         while msg is not None:
             if self.debug:
                 pprint(msg)
             msg = msg[-1]
             msg_type = msg['msg_type']
             handler = self._queue_handlers.get(msg_type, None)
             if handler is None:
                 raise Exception("Unhandled message type: %s"%msg.msg_type)
             else:
                 handler(msg)
             msg = self.session.recv(sock, mode=zmq.NOBLOCK)
     
     def _flush_control(self, sock):
         """Flush replies from the control channel waiting
         in the ZMQ queue.
         
         Currently: ignore them."""
         if self._ignored_control_replies <= 0:
             return
         msg = self.session.recv(sock, mode=zmq.NOBLOCK)
         while msg is not None:
             self._ignored_control_replies -= 1
             if self.debug:
                 pprint(msg)
             msg = self.session.recv(sock, mode=zmq.NOBLOCK)
     
     def _flush_ignored_control(self):
         """flush ignored control replies"""
         while self._ignored_control_replies > 0:
             self.session.recv(self._control_socket)
             self._ignored_control_replies -= 1
     
     def _flush_ignored_hub_replies(self):
         msg = self.session.recv(self._query_socket, mode=zmq.NOBLOCK)
         while msg is not None:
             msg = self.session.recv(self._query_socket, mode=zmq.NOBLOCK)
     
     def _flush_iopub(self, sock):
         """Flush replies from the iopub channel waiting
         in the ZMQ queue.
         """
         msg = self.session.recv(sock, mode=zmq.NOBLOCK)
         while msg is not None:
             if self.debug:
                 pprint(msg)
             msg = msg[-1]
             parent = msg['parent_header']
             msg_id = parent['msg_id']
             content = msg['content']
             header = msg['header']
             msg_type = msg['msg_type']
             
             # init metadata:
             md = self.metadata[msg_id]
             
             if msg_type == 'stream':
                 name = content['name']
                 s = md[name] or ''
                 md[name] = s + content['data']
             elif msg_type == 'pyerr':
                 md.update({'pyerr' : self._unwrap_exception(content)})
             elif msg_type == 'pyin':
                 md.update({'pyin' : content['code']})
             else:
                 md.update({msg_type : content.get('data', '')})
             
             # reduntant?
             self.metadata[msg_id] = md
             
             msg = self.session.recv(sock, mode=zmq.NOBLOCK)
     
     #--------------------------------------------------------------------------
     # len, getitem
     #--------------------------------------------------------------------------
     
     def __len__(self):
         """len(client) returns # of engines."""
         return len(self.ids)
     
     def __getitem__(self, key):
         """index access returns DirectView multiplexer objects
         
         Must be int, slice, or list/tuple/xrange of ints"""
         if not isinstance(key, (int, slice, tuple, list, xrange)):
             raise TypeError("key by int/slice/iterable of ints only, not %s"%(type(key)))
         else:
             return self.direct_view(key)
     
     #--------------------------------------------------------------------------
     # Begin public methods
     #--------------------------------------------------------------------------
     
     @property
     def ids(self):
         """Always up-to-date ids property."""
         self._flush_notifications()
         # always copy:
         return list(self._ids)
         
     def close(self):
         if self._closed:
             return
         snames = filter(lambda n: n.endswith('socket'), dir(self))
         for socket in map(lambda name: getattr(self, name), snames):
             if isinstance(socket, zmq.Socket) and not socket.closed:
                 socket.close()
         self._closed = True
     
     def spin(self):
         """Flush any registration notifications and execution results
         waiting in the ZMQ queue.
         """
         if self._notification_socket:
             self._flush_notifications()
         if self._mux_socket:
             self._flush_results(self._mux_socket)
         if self._task_socket:
             self._flush_results(self._task_socket)
         if self._control_socket:
             self._flush_control(self._control_socket)
         if self._iopub_socket:
             self._flush_iopub(self._iopub_socket)
         if self._query_socket:
             self._flush_ignored_hub_replies()
     
     def wait(self, jobs=None, timeout=-1):
         """waits on one or more `jobs`, for up to `timeout` seconds.
         
         Parameters
         ----------
         
         jobs : int, str, or list of ints and/or strs, or one or more AsyncResult objects
                 ints are indices to self.history
                 strs are msg_ids
                 default: wait on all outstanding messages
         timeout : float
                 a time in seconds, after which to give up.
                 default is -1, which means no timeout
         
         Returns
         -------
         
         True : when all msg_ids are done
         False : timeout reached, some msg_ids still outstanding
         """
         tic = time.time()
         if jobs is None:
             theids = self.outstanding
         else:
             if isinstance(jobs, (int, str, AsyncResult)):
                 jobs = [jobs]
             theids = set()
             for job in jobs:
                 if isinstance(job, int):
                     # index access
                     job = self.history[job]
                 elif isinstance(job, AsyncResult):
                     map(theids.add, job.msg_ids)
                     continue
                 theids.add(job)
         if not theids.intersection(self.outstanding):
             return True
         self.spin()
         while theids.intersection(self.outstanding):
             if timeout >= 0 and ( time.time()-tic ) > timeout:
                 break
             time.sleep(1e-3)
             self.spin()
         return len(theids.intersection(self.outstanding)) == 0
     
     #--------------------------------------------------------------------------
     # Control methods
     #--------------------------------------------------------------------------
     
     @spin_first
     def clear(self, targets=None, block=None):
         """Clear the namespace in target(s)."""
         block = self.block if block is None else block
         targets = self._build_targets(targets)[0]
         for t in targets:
             self.session.send(self._control_socket, 'clear_request', content={}, ident=t)
         error = False
         if block:
             self._flush_ignored_control()
             for i in range(len(targets)):
                 idents,msg = self.session.recv(self._control_socket,0)
                 if self.debug:
                     pprint(msg)
                 if msg['content']['status'] != 'ok':
                     error = self._unwrap_exception(msg['content'])
         else:
             self._ignored_control_replies += len(targets)
         if error:
             raise error
         
     
     @spin_first
     def abort(self, jobs=None, targets=None, block=None):
         """Abort specific jobs from the execution queues of target(s).
         
         This is a mechanism to prevent jobs that have already been submitted
         from executing.
         
         Parameters
         ----------
         
         jobs : msg_id, list of msg_ids, or AsyncResult
             The jobs to be aborted
         
         
         """
         block = self.block if block is None else block
         targets = self._build_targets(targets)[0]
         msg_ids = []
         if isinstance(jobs, (basestring,AsyncResult)):
             jobs = [jobs]
         bad_ids = filter(lambda obj: not isinstance(obj, (basestring, AsyncResult)), jobs)
         if bad_ids:
             raise TypeError("Invalid msg_id type %r, expected str or AsyncResult"%bad_ids[0])
         for j in jobs:
             if isinstance(j, AsyncResult):
                 msg_ids.extend(j.msg_ids)
             else:
                 msg_ids.append(j)
         content = dict(msg_ids=msg_ids)
         for t in targets:
             self.session.send(self._control_socket, 'abort_request', 
                     content=content, ident=t)
         error = False
         if block:
             self._flush_ignored_control()
             for i in range(len(targets)):
                 idents,msg = self.session.recv(self._control_socket,0)
                 if self.debug:
                     pprint(msg)
                 if msg['content']['status'] != 'ok':
                     error = self._unwrap_exception(msg['content'])
         else:
             self._ignored_control_replies += len(targets)
         if error:
             raise error
     
     @spin_first
     def shutdown(self, targets=None, restart=False, hub=False, block=None):
         """Terminates one or more engine processes, optionally including the hub."""
         block = self.block if block is None else block
         if hub:
             targets = 'all'
         targets = self._build_targets(targets)[0]
         for t in targets:
             self.session.send(self._control_socket, 'shutdown_request', 
                         content={'restart':restart},ident=t)
         error = False
         if block or hub:
             self._flush_ignored_control()
             for i in range(len(targets)):
                 idents,msg = self.session.recv(self._control_socket, 0)
                 if self.debug:
                     pprint(msg)
                 if msg['content']['status'] != 'ok':
                     error = self._unwrap_exception(msg['content'])
         else:
             self._ignored_control_replies += len(targets)
         
         if hub:
             time.sleep(0.25)
             self.session.send(self._query_socket, 'shutdown_request')
             idents,msg = self.session.recv(self._query_socket, 0)
             if self.debug:
                 pprint(msg)
             if msg['content']['status'] != 'ok':
                 error = self._unwrap_exception(msg['content'])
         
         if error:
             raise error
     
     #--------------------------------------------------------------------------
     # Execution related methods
     #--------------------------------------------------------------------------
     
     def _maybe_raise(self, result):
         """wrapper for maybe raising an exception if apply failed."""
         if isinstance(result, error.RemoteError):
             raise result
         
         return result
     
     def send_apply_message(self, socket, f, args=None, kwargs=None, subheader=None, track=False,
                             ident=None):
         """construct and send an apply message via a socket.
         
         This is the principal method with which all engine execution is performed by views.
         """
                             
         assert not self._closed, "cannot use me anymore, I'm closed!"
         # defaults:
         args = args if args is not None else []
         kwargs = kwargs if kwargs is not None else {}
         subheader = subheader if subheader is not None else {}
         
         # validate arguments
         if not callable(f):
             raise TypeError("f must be callable, not %s"%type(f))
         if not isinstance(args, (tuple, list)):
             raise TypeError("args must be tuple or list, not %s"%type(args))
         if not isinstance(kwargs, dict):
             raise TypeError("kwargs must be dict, not %s"%type(kwargs))
         if not isinstance(subheader, dict):
             raise TypeError("subheader must be dict, not %s"%type(subheader))
         
         bufs = util.pack_apply_message(f,args,kwargs)
         
         msg = self.session.send(socket, "apply_request", buffers=bufs, ident=ident,
                             subheader=subheader, track=track)
         
         msg_id = msg['msg_id']
         self.outstanding.add(msg_id)
         if ident:
             # possibly routed to a specific engine
             if isinstance(ident, list):
                 ident = ident[-1]
             if ident in self._engines.values():
                 # save for later, in case of engine death
                 self._outstanding_dict[ident].add(msg_id)
         self.history.append(msg_id)
         self.metadata[msg_id]['submitted'] = datetime.now()
         
         return msg
 
     #--------------------------------------------------------------------------
     # construct a View object
     #--------------------------------------------------------------------------
     
     def load_balanced_view(self, targets=None):
         """construct a DirectView object.
         
         If no arguments are specified, create a LoadBalancedView
         using all engines.
         
         Parameters
         ----------
         
         targets: list,slice,int,etc. [default: use all engines]
             The subset of engines across which to load-balance
         """
         if targets is not None:
             targets = self._build_targets(targets)[1]
         return LoadBalancedView(client=self, socket=self._task_socket, targets=targets)
     
     def direct_view(self, targets='all'):
         """construct a DirectView object.
         
         If no targets are specified, create a DirectView
         using all engines.
         
         Parameters
         ----------
         
         targets: list,slice,int,etc. [default: use all engines]
             The engines to use for the View
         """
         single = isinstance(targets, int)
         targets = self._build_targets(targets)[1]
         if single:
             targets = targets[0]
         return DirectView(client=self, socket=self._mux_socket, targets=targets)
     
     #--------------------------------------------------------------------------
     # Query methods
     #--------------------------------------------------------------------------
     
     @spin_first
     def get_result(self, indices_or_msg_ids=None, block=None):
         """Retrieve a result by msg_id or history index, wrapped in an AsyncResult object.
         
         If the client already has the results, no request to the Hub will be made.
         
         This is a convenient way to construct AsyncResult objects, which are wrappers
         that include metadata about execution, and allow for awaiting results that
         were not submitted by this Client.
         
         It can also be a convenient way to retrieve the metadata associated with
         blocking execution, since it always retrieves
         
         Examples
         --------
         ::
         
             In [10]: r = client.apply()
         
         Parameters
         ----------
         
         indices_or_msg_ids : integer history index, str msg_id, or list of either
             The indices or msg_ids of indices to be retrieved
         
         block : bool
             Whether to wait for the result to be done
         
         Returns
         -------
         
         AsyncResult
             A single AsyncResult object will always be returned.
         
         AsyncHubResult
             A subclass of AsyncResult that retrieves results from the Hub
         
         """
         block = self.block if block is None else block
         if indices_or_msg_ids is None:
             indices_or_msg_ids = -1
         
         if not isinstance(indices_or_msg_ids, (list,tuple)):
             indices_or_msg_ids = [indices_or_msg_ids]
         
         theids = []
         for id in indices_or_msg_ids:
             if isinstance(id, int):
                 id = self.history[id]
             if not isinstance(id, str):
                 raise TypeError("indices must be str or int, not %r"%id)
             theids.append(id)
         
         local_ids = filter(lambda msg_id: msg_id in self.history or msg_id in self.results, theids)
         remote_ids = filter(lambda msg_id: msg_id not in local_ids, theids)
         
         if remote_ids:
             ar = AsyncHubResult(self, msg_ids=theids)
         else:
             ar = AsyncResult(self, msg_ids=theids)
         
         if block:
             ar.wait()
         
         return ar
     
     @spin_first
+    def resubmit(self, indices_or_msg_ids=None, subheader=None, block=None):
+        """Resubmit one or more tasks.
+
+        in-flight tasks may not be resubmitted.
+
+        Parameters
+        ----------
+
+        indices_or_msg_ids : integer history index, str msg_id, or list of either
+            The indices or msg_ids of indices to be retrieved
+
+        block : bool
+            Whether to wait for the result to be done
+
+        Returns
+        -------
+
+        AsyncHubResult
+            A subclass of AsyncResult that retrieves results from the Hub
+
+        """
+        block = self.block if block is None else block
+        if indices_or_msg_ids is None:
+            indices_or_msg_ids = -1
+
+        if not isinstance(indices_or_msg_ids, (list,tuple)):
+            indices_or_msg_ids = [indices_or_msg_ids]
+
+        theids = []
+        for id in indices_or_msg_ids:
+            if isinstance(id, int):
+                id = self.history[id]
+            if not isinstance(id, str):
+                raise TypeError("indices must be str or int, not %r"%id)
+            theids.append(id)
+
+        for msg_id in theids:
+            self.outstanding.discard(msg_id)
+            if msg_id in self.history:
+                self.history.remove(msg_id)
+            self.results.pop(msg_id, None)
+            self.metadata.pop(msg_id, None)
+        content = dict(msg_ids = theids)
+
+        self.session.send(self._query_socket, 'resubmit_request', content)
+
+        zmq.select([self._query_socket], [], [])
+        idents,msg = self.session.recv(self._query_socket, zmq.NOBLOCK)
+        if self.debug:
+            pprint(msg)
+        content = msg['content']
+        if content['status'] != 'ok':
+            raise self._unwrap_exception(content)
+
+        ar = AsyncHubResult(self, msg_ids=theids)
+
+        if block:
+            ar.wait()
+
+        return ar
+    
+    @spin_first
     def result_status(self, msg_ids, status_only=True):
         """Check on the status of the result(s) of the apply request with `msg_ids`.
         
         If status_only is False, then the actual results will be retrieved, else
         only the status of the results will be checked.
         
         Parameters
         ----------
         
         msg_ids : list of msg_ids
             if int:
                 Passed as index to self.history for convenience.
         status_only : bool (default: True)
             if False:
                 Retrieve the actual results of completed tasks.
         
         Returns
         -------
         
         results : dict
             There will always be the keys 'pending' and 'completed', which will
             be lists of msg_ids that are incomplete or complete. If `status_only`
             is False, then completed results will be keyed by their `msg_id`.
         """
         if not isinstance(msg_ids, (list,tuple)):
             msg_ids = [msg_ids]
             
         theids = []
         for msg_id in msg_ids:
             if isinstance(msg_id, int):
                 msg_id = self.history[msg_id]
             if not isinstance(msg_id, basestring):
                 raise TypeError("msg_ids must be str, not %r"%msg_id)
             theids.append(msg_id)
         
         completed = []
         local_results = {}
         
         # comment this block out to temporarily disable local shortcut:
         for msg_id in theids:
             if msg_id in self.results:
                 completed.append(msg_id)
                 local_results[msg_id] = self.results[msg_id]
                 theids.remove(msg_id)
         
         if theids: # some not locally cached
             content = dict(msg_ids=theids, status_only=status_only)
             msg = self.session.send(self._query_socket, "result_request", content=content)
             zmq.select([self._query_socket], [], [])
             idents,msg = self.session.recv(self._query_socket, zmq.NOBLOCK)
             if self.debug:
                 pprint(msg)
             content = msg['content']
             if content['status'] != 'ok':
                 raise self._unwrap_exception(content)
             buffers = msg['buffers']
         else:
             content = dict(completed=[],pending=[])
         
         content['completed'].extend(completed)
         
         if status_only:
             return content
         
         failures = []
         # load cached results into result:
         content.update(local_results)
         # update cache with results:
         for msg_id in sorted(theids):
             if msg_id in content['completed']:
                 rec = content[msg_id]
                 parent = rec['header']
                 header = rec['result_header']
                 rcontent = rec['result_content']
                 iodict = rec['io']
                 if isinstance(rcontent, str):
                     rcontent = self.session.unpack(rcontent)
                 
                 md = self.metadata[msg_id]
                 md.update(self._extract_metadata(header, parent, rcontent))
                 md.update(iodict)
                 
                 if rcontent['status'] == 'ok':
                     res,buffers = util.unserialize_object(buffers)
                 else:
                     print rcontent
                     res = self._unwrap_exception(rcontent)
                     failures.append(res)
                 
                 self.results[msg_id] = res
                 content[msg_id] = res
         
         if len(theids) == 1 and failures:
                 raise failures[0]
         
         error.collect_exceptions(failures, "result_status")
         return content
 
     @spin_first
     def queue_status(self, targets='all', verbose=False):
         """Fetch the status of engine queues.
         
         Parameters
         ----------
         
         targets : int/str/list of ints/strs
                 the engines whose states are to be queried.
                 default : all
         verbose : bool
                 Whether to return lengths only, or lists of ids for each element
         """
         engine_ids = self._build_targets(targets)[1]
         content = dict(targets=engine_ids, verbose=verbose)
         self.session.send(self._query_socket, "queue_request", content=content)
         idents,msg = self.session.recv(self._query_socket, 0)
         if self.debug:
             pprint(msg)
         content = msg['content']
         status = content.pop('status')
         if status != 'ok':
             raise self._unwrap_exception(content)
         content = util.rekey(content)
         if isinstance(targets, int):
             return content[targets]
         else:
             return content
         
     @spin_first
     def purge_results(self, jobs=[], targets=[]):
         """Tell the Hub to forget results.
         
         Individual results can be purged by msg_id, or the entire
         history of specific targets can be purged.
         
         Parameters
         ----------
         
         jobs : str or list of str or AsyncResult objects
                 the msg_ids whose results should be forgotten.
         targets : int/str/list of ints/strs
                 The targets, by uuid or int_id, whose entire history is to be purged.
                 Use `targets='all'` to scrub everything from the Hub's memory.
                 
                 default : None
         """
         if not targets and not jobs:
             raise ValueError("Must specify at least one of `targets` and `jobs`")
         if targets:
             targets = self._build_targets(targets)[1]
         
         # construct msg_ids from jobs
         msg_ids = []
         if isinstance(jobs, (basestring,AsyncResult)):
             jobs = [jobs]
         bad_ids = filter(lambda obj: not isinstance(obj, (basestring, AsyncResult)), jobs)
         if bad_ids:
             raise TypeError("Invalid msg_id type %r, expected str or AsyncResult"%bad_ids[0])
         for j in jobs:
             if isinstance(j, AsyncResult):
                 msg_ids.extend(j.msg_ids)
             else:
                 msg_ids.append(j)
         
         content = dict(targets=targets, msg_ids=msg_ids)
         self.session.send(self._query_socket, "purge_request", content=content)
         idents, msg = self.session.recv(self._query_socket, 0)
         if self.debug:
             pprint(msg)
         content = msg['content']
         if content['status'] != 'ok':
             raise self._unwrap_exception(content)
 
     @spin_first
     def hub_history(self):
         """Get the Hub's history
         
         Just like the Client, the Hub has a history, which is a list of msg_ids.
         This will contain the history of all clients, and, depending on configuration,
         may contain history across multiple cluster sessions.
         
         Any msg_id returned here is a valid argument to `get_result`.
         
         Returns
         -------
         
         msg_ids : list of strs
                 list of all msg_ids, ordered by task submission time.
         """
         
         self.session.send(self._query_socket, "history_request", content={})
         idents, msg = self.session.recv(self._query_socket, 0)
         
         if self.debug:
             pprint(msg)
         content = msg['content']
         if content['status'] != 'ok':
             raise self._unwrap_exception(content)
         else:
             return content['history']
 
     @spin_first
     def db_query(self, query, keys=None):
         """Query the Hub's TaskRecord database
         
         This will return a list of task record dicts that match `query`
         
         Parameters
         ----------
         
         query : mongodb query dict
             The search dict. See mongodb query docs for details.
         keys : list of strs [optional]
             THe subset of keys to be returned.  The default is to fetch everything.
             'msg_id' will *always* be included.
         """
         content = dict(query=query, keys=keys)
         self.session.send(self._query_socket, "db_request", content=content)
         idents, msg = self.session.recv(self._query_socket, 0)
         if self.debug:
             pprint(msg)
         content = msg['content']
         if content['status'] != 'ok':
             raise self._unwrap_exception(content)
         
         records = content['records']
         buffer_lens = content['buffer_lens']
         result_buffer_lens = content['result_buffer_lens']
         buffers = msg['buffers']
         has_bufs = buffer_lens is not None
         has_rbufs = result_buffer_lens is not None
         for i,rec in enumerate(records):
             # relink buffers
             if has_bufs:
                 blen = buffer_lens[i]
                 rec['buffers'], buffers = buffers[:blen],buffers[blen:]
             if has_rbufs:
                 blen = result_buffer_lens[i]
                 rec['result_buffers'], buffers = buffers[:blen],buffers[blen:]
             # turn timestamps back into times
             for key in 'submitted started completed resubmitted'.split():
                 maybedate = rec.get(key, None)
                 if maybedate and util.ISO8601_RE.match(maybedate):
                     rec[key] = datetime.strptime(maybedate, util.ISO8601)
             
         return records
 
 __all__ = [ 'Client' ]

IPython/parallel/controller/hub.py

@@ -1,1193 +1,1284 b''
 #!/usr/bin/env python
 """The IPython Controller Hub with 0MQ
 This is the master object that handles connections from engines and clients,
 and monitors traffic through the various queues.
 """
 #-----------------------------------------------------------------------------
 #  Copyright (C) 2010  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-----------------------------------------------------------------------------
 
 #-----------------------------------------------------------------------------
 # Imports
 #-----------------------------------------------------------------------------
 from __future__ import print_function
 
 import sys
 import time
 from datetime import datetime
 
 import zmq
 from zmq.eventloop import ioloop
 from zmq.eventloop.zmqstream import ZMQStream
 
 # internal:
 from IPython.utils.importstring import import_item
 from IPython.utils.traitlets import HasTraits, Instance, Int, CStr, Str, Dict, Set, List, Bool
 
 from IPython.parallel import error, util
 from IPython.parallel.factory import RegistrationFactory, LoggingFactory
 
 from .heartmonitor import HeartMonitor
 
 #-----------------------------------------------------------------------------
 # Code
 #-----------------------------------------------------------------------------
 
 def _passer(*args, **kwargs):
     return
 
 def _printer(*args, **kwargs):
     print (args)
     print (kwargs)
 
 def empty_record():
     """Return an empty dict with all record keys."""
     return {
         'msg_id' : None,
         'header' : None,
         'content': None,
         'buffers': None,
         'submitted': None,
         'client_uuid' : None,
         'engine_uuid' : None,
         'started': None,
         'completed': None,
         'resubmitted': None,
         'result_header' : None,
         'result_content' : None,
         'result_buffers' : None,
         'queue' : None,
         'pyin' : None,
         'pyout': None,
         'pyerr': None,
         'stdout': '',
         'stderr': '',
     }
     
 def init_record(msg):
     """Initialize a TaskRecord based on a request."""
     header = msg['header']
     return {
         'msg_id' : header['msg_id'],
         'header' : header,
         'content': msg['content'],
         'buffers': msg['buffers'],
         'submitted': datetime.strptime(header['date'], util.ISO8601),
         'client_uuid' : None,
         'engine_uuid' : None,
         'started': None,
         'completed': None,
         'resubmitted': None,
         'result_header' : None,
         'result_content' : None,
         'result_buffers' : None,
         'queue' : None,
         'pyin' : None,
         'pyout': None,
         'pyerr': None,
         'stdout': '',
         'stderr': '',
     }
 
 
 class EngineConnector(HasTraits):
     """A simple object for accessing the various zmq connections of an object.
     Attributes are:
     id (int): engine ID
     uuid (str): uuid (unused?)
     queue (str): identity of queue's XREQ socket
     registration (str): identity of registration XREQ socket
     heartbeat (str): identity of heartbeat XREQ socket
     """
     id=Int(0)
     queue=Str()
     control=Str()
     registration=Str()
     heartbeat=Str()
     pending=Set()
 
 class HubFactory(RegistrationFactory):
     """The Configurable for setting up a Hub."""
     
     # name of a scheduler scheme
     scheme = Str('leastload', config=True)
     
     # port-pairs for monitoredqueues:
     hb = Instance(list, config=True)
     def _hb_default(self):
         return util.select_random_ports(2)
     
     mux = Instance(list, config=True)
     def _mux_default(self):
         return util.select_random_ports(2)
     
     task = Instance(list, config=True)
     def _task_default(self):
         return util.select_random_ports(2)
     
     control = Instance(list, config=True)
     def _control_default(self):
         return util.select_random_ports(2)
     
     iopub = Instance(list, config=True)
     def _iopub_default(self):
         return util.select_random_ports(2)
     
     # single ports:
     mon_port = Instance(int, config=True)
     def _mon_port_default(self):
         return util.select_random_ports(1)[0]
     
     notifier_port = Instance(int, config=True)
     def _notifier_port_default(self):
         return util.select_random_ports(1)[0]
     
     ping = Int(1000, config=True) # ping frequency
     
     engine_ip = CStr('127.0.0.1', config=True)
     engine_transport = CStr('tcp', config=True)
     
     client_ip = CStr('127.0.0.1', config=True)
     client_transport = CStr('tcp', config=True)
     
     monitor_ip = CStr('127.0.0.1', config=True)
     monitor_transport = CStr('tcp', config=True)
     
     monitor_url = CStr('')
     
     db_class = CStr('IPython.parallel.controller.dictdb.DictDB', config=True)
     
     # not configurable
     db = Instance('IPython.parallel.controller.dictdb.BaseDB')
     heartmonitor = Instance('IPython.parallel.controller.heartmonitor.HeartMonitor')
     subconstructors = List()
     _constructed = Bool(False)
     
     def _ip_changed(self, name, old, new):
         self.engine_ip = new
         self.client_ip = new
         self.monitor_ip = new
         self._update_monitor_url()
     
     def _update_monitor_url(self):
         self.monitor_url = "%s://%s:%i"%(self.monitor_transport, self.monitor_ip, self.mon_port)
     
     def _transport_changed(self, name, old, new):
         self.engine_transport = new
         self.client_transport = new
         self.monitor_transport = new
         self._update_monitor_url()
         
     def __init__(self, **kwargs):
         super(HubFactory, self).__init__(**kwargs)
         self._update_monitor_url()
         # self.on_trait_change(self._sync_ips, 'ip')
         # self.on_trait_change(self._sync_transports, 'transport')
         self.subconstructors.append(self.construct_hub)
     
     
     def construct(self):
         assert not self._constructed, "already constructed!"
         
         for subc in self.subconstructors:
             subc()
         
         self._constructed = True
         
     
     def start(self):
         assert self._constructed, "must be constructed by self.construct() first!"
         self.heartmonitor.start()
         self.log.info("Heartmonitor started")
     
     def construct_hub(self):
         """construct"""
         client_iface = "%s://%s:"%(self.client_transport, self.client_ip) + "%i"
         engine_iface = "%s://%s:"%(self.engine_transport, self.engine_ip) + "%i"
         
         ctx = self.context
         loop = self.loop
         
         # Registrar socket
         q = ZMQStream(ctx.socket(zmq.XREP), loop)
         q.bind(client_iface % self.regport)
         self.log.info("Hub listening on %s for registration."%(client_iface%self.regport))
         if self.client_ip != self.engine_ip:
             q.bind(engine_iface % self.regport)
             self.log.info("Hub listening on %s for registration."%(engine_iface%self.regport))
         
         ### Engine connections ###
 
         # heartbeat
         hpub = ctx.socket(zmq.PUB)
         hpub.bind(engine_iface % self.hb[0])
         hrep = ctx.socket(zmq.XREP)
         hrep.bind(engine_iface % self.hb[1])
         self.heartmonitor = HeartMonitor(loop=loop, pingstream=ZMQStream(hpub,loop), pongstream=ZMQStream(hrep,loop), 
                                 period=self.ping, logname=self.log.name)
 
         ### Client connections ###
         # Notifier socket
         n = ZMQStream(ctx.socket(zmq.PUB), loop)
         n.bind(client_iface%self.notifier_port)
 
         ### build and launch the queues ###
 
         # monitor socket
         sub = ctx.socket(zmq.SUB)
         sub.setsockopt(zmq.SUBSCRIBE, "")
         sub.bind(self.monitor_url)
         sub.bind('inproc://monitor')
         sub = ZMQStream(sub, loop)
         
         # connect the db
         self.log.info('Hub using DB backend: %r'%(self.db_class.split()[-1]))
         # cdir = self.config.Global.cluster_dir
         self.db = import_item(self.db_class)(session=self.session.session, config=self.config)
         time.sleep(.25)
 
         # build connection dicts
         self.engine_info = {
             'control' : engine_iface%self.control[1],
             'mux': engine_iface%self.mux[1],
             'heartbeat': (engine_iface%self.hb[0], engine_iface%self.hb[1]),
             'task' : engine_iface%self.task[1],
             'iopub' : engine_iface%self.iopub[1],
             # 'monitor' : engine_iface%self.mon_port,
             }
 
         self.client_info = {
             'control' : client_iface%self.control[0],
             'mux': client_iface%self.mux[0],
             'task' : (self.scheme, client_iface%self.task[0]),
             'iopub' : client_iface%self.iopub[0],
             'notification': client_iface%self.notifier_port
             }
         self.log.debug("Hub engine addrs: %s"%self.engine_info)
         self.log.debug("Hub client addrs: %s"%self.client_info)
+
+        # resubmit stream
+        r = ZMQStream(ctx.socket(zmq.XREQ), loop)
+        url = util.disambiguate_url(self.client_info['task'][-1])
+        r.setsockopt(zmq.IDENTITY, self.session.session)
+        r.connect(url)
+
         self.hub = Hub(loop=loop, session=self.session, monitor=sub, heartmonitor=self.heartmonitor,
-                query=q, notifier=n, db=self.db,
+                query=q, notifier=n, resubmit=r, db=self.db,
                 engine_info=self.engine_info, client_info=self.client_info,
                 logname=self.log.name)
     
 
 class Hub(LoggingFactory):
     """The IPython Controller Hub with 0MQ connections
     
     Parameters
     ==========
     loop: zmq IOLoop instance
     session: StreamSession object
     <removed> context: zmq context for creating new connections (?)
     queue: ZMQStream for monitoring the command queue (SUB)
     query: ZMQStream for engine registration and client queries requests (XREP)
     heartbeat: HeartMonitor object checking the pulse of the engines
     notifier: ZMQStream for broadcasting engine registration changes (PUB)
     db: connection to db for out of memory logging of commands
                 NotImplemented
     engine_info: dict of zmq connection information for engines to connect
                 to the queues.
     client_info: dict of zmq connection information for engines to connect
                 to the queues.
     """
     # internal data structures:
     ids=Set() # engine IDs
     keytable=Dict()
     by_ident=Dict()
     engines=Dict()
     clients=Dict()
     hearts=Dict()
     pending=Set()
     queues=Dict()  # pending msg_ids keyed by engine_id
     tasks=Dict() # pending msg_ids submitted as tasks, keyed by client_id
     completed=Dict() # completed msg_ids keyed by engine_id
     all_completed=Set() # completed msg_ids keyed by engine_id
     dead_engines=Set() # completed msg_ids keyed by engine_id
     unassigned=Set() # set of task msg_ds not yet assigned a destination
     incoming_registrations=Dict()
     registration_timeout=Int()
     _idcounter=Int(0)
     
     # objects from constructor:
     loop=Instance(ioloop.IOLoop)
     query=Instance(ZMQStream)
     monitor=Instance(ZMQStream)
-    heartmonitor=Instance(HeartMonitor)
     notifier=Instance(ZMQStream)
+    resubmit=Instance(ZMQStream)
+    heartmonitor=Instance(HeartMonitor)
     db=Instance(object)
     client_info=Dict()
     engine_info=Dict()
     
     
     def __init__(self, **kwargs):
         """
         # universal:
         loop: IOLoop for creating future connections
         session: streamsession for sending serialized data
         # engine:
         queue: ZMQStream for monitoring queue messages
         query: ZMQStream for engine+client registration and client requests
         heartbeat: HeartMonitor object for tracking engines
         # extra:
         db: ZMQStream for db connection (NotImplemented)
         engine_info: zmq address/protocol dict for engine connections
         client_info: zmq address/protocol dict for client connections
         """
         
         super(Hub, self).__init__(**kwargs)
         self.registration_timeout = max(5000, 2*self.heartmonitor.period)
         
         # validate connection dicts:
         for k,v in self.client_info.iteritems():
             if k == 'task':
                 util.validate_url_container(v[1])
             else:
                 util.validate_url_container(v)
         # util.validate_url_container(self.client_info)
         util.validate_url_container(self.engine_info)
         
         # register our callbacks
         self.query.on_recv(self.dispatch_query)
         self.monitor.on_recv(self.dispatch_monitor_traffic)
         
         self.heartmonitor.add_heart_failure_handler(self.handle_heart_failure)
         self.heartmonitor.add_new_heart_handler(self.handle_new_heart)
         
         self.monitor_handlers = { 'in' : self.save_queue_request,
                                 'out': self.save_queue_result,
                                 'intask': self.save_task_request,
                                 'outtask': self.save_task_result,
                                 'tracktask': self.save_task_destination,
                                 'incontrol': _passer,
                                 'outcontrol': _passer,
                                 'iopub': self.save_iopub_message,
         }
         
         self.query_handlers = {'queue_request': self.queue_status,
                                 'result_request': self.get_results,
                                 'history_request': self.get_history,
                                 'db_request': self.db_query,
                                 'purge_request': self.purge_results,
                                 'load_request': self.check_load,
                                 'resubmit_request': self.resubmit_task,
                                 'shutdown_request': self.shutdown_request,
                                 'registration_request' : self.register_engine,
                                 'unregistration_request' : self.unregister_engine,
                                 'connection_request': self.connection_request,
         }
         
+        # ignore resubmit replies
+        self.resubmit.on_recv(lambda msg: None, copy=False)
+
         self.log.info("hub::created hub")
     
     @property
     def _next_id(self):
         """gemerate a new ID.
         
         No longer reuse old ids, just count from 0."""
         newid = self._idcounter
         self._idcounter += 1
         return newid
         # newid = 0
         # incoming = [id[0] for id in self.incoming_registrations.itervalues()]
         # # print newid, self.ids, self.incoming_registrations
         # while newid in self.ids or newid in incoming:
         #     newid += 1
         # return newid
     
     #-----------------------------------------------------------------------------
     # message validation
     #-----------------------------------------------------------------------------
     
     def _validate_targets(self, targets):
         """turn any valid targets argument into a list of integer ids"""
         if targets is None:
             # default to all
             targets = self.ids
             
         if isinstance(targets, (int,str,unicode)):
             # only one target specified
             targets = [targets]
         _targets = []
         for t in targets:
             # map raw identities to ids
             if isinstance(t, (str,unicode)):
                 t = self.by_ident.get(t, t)
             _targets.append(t)
         targets = _targets
         bad_targets = [ t for t in targets if t not in self.ids ]
         if bad_targets:
             raise IndexError("No Such Engine: %r"%bad_targets)
         if not targets:
             raise IndexError("No Engines Registered")
         return targets
     
     #-----------------------------------------------------------------------------
     # dispatch methods (1 per stream)
     #-----------------------------------------------------------------------------
     
     # def dispatch_registration_request(self, msg):
     #     """"""
     #     self.log.debug("registration::dispatch_register_request(%s)"%msg)
     #     idents,msg = self.session.feed_identities(msg)
     #     if not idents:
     #         self.log.error("Bad Query Message: %s"%msg, exc_info=True)
     #         return
     #     try:
     #         msg = self.session.unpack_message(msg,content=True)
     #     except:
     #         self.log.error("registration::got bad registration message: %s"%msg, exc_info=True)
     #         return
     #     
     #     msg_type = msg['msg_type']
     #     content = msg['content']
     #     
     #     handler = self.query_handlers.get(msg_type, None)
     #     if handler is None:
     #         self.log.error("registration::got bad registration message: %s"%msg)
     #     else:
     #         handler(idents, msg)
     
     def dispatch_monitor_traffic(self, msg):
         """all ME and Task queue messages come through here, as well as
         IOPub traffic."""
-        self.log.debug("monitor traffic: %s"%msg[:2])
+        self.log.debug("monitor traffic: %r"%msg[:2])
         switch = msg[0]
         idents, msg = self.session.feed_identities(msg[1:])
         if not idents:
-            self.log.error("Bad Monitor Message: %s"%msg)
+            self.log.error("Bad Monitor Message: %r"%msg)
             return
         handler = self.monitor_handlers.get(switch, None)
         if handler is not None:
             handler(idents, msg)
         else:
-            self.log.error("Invalid monitor topic: %s"%switch)
+            self.log.error("Invalid monitor topic: %r"%switch)
         
     
     def dispatch_query(self, msg):
         """Route registration requests and queries from clients."""
         idents, msg = self.session.feed_identities(msg)
         if not idents:
-            self.log.error("Bad Query Message: %s"%msg)
+            self.log.error("Bad Query Message: %r"%msg)
             return
         client_id = idents[0]
         try:
             msg = self.session.unpack_message(msg, content=True)
         except:
             content = error.wrap_exception()
-            self.log.error("Bad Query Message: %s"%msg, exc_info=True)
+            self.log.error("Bad Query Message: %r"%msg, exc_info=True)
             self.session.send(self.query, "hub_error", ident=client_id, 
                     content=content)
             return
         
         # print client_id, header, parent, content
         #switch on message type:
         msg_type = msg['msg_type']
-        self.log.info("client::client %s requested %s"%(client_id, msg_type))
+        self.log.info("client::client %r requested %r"%(client_id, msg_type))
         handler = self.query_handlers.get(msg_type, None)
         try:
-            assert handler is not None, "Bad Message Type: %s"%msg_type
+            assert handler is not None, "Bad Message Type: %r"%msg_type
         except:
             content = error.wrap_exception()
-            self.log.error("Bad Message Type: %s"%msg_type, exc_info=True)
+            self.log.error("Bad Message Type: %r"%msg_type, exc_info=True)
             self.session.send(self.query, "hub_error", ident=client_id, 
                     content=content)
             return
+
         else:
             handler(idents, msg)
             
     def dispatch_db(self, msg):
         """"""
         raise NotImplementedError
     
     #---------------------------------------------------------------------------
     # handler methods (1 per event)
     #---------------------------------------------------------------------------
     
     #----------------------- Heartbeat --------------------------------------
     
     def handle_new_heart(self, heart):
         """handler to attach to heartbeater.
         Called when a new heart starts to beat.
         Triggers completion of registration."""
         self.log.debug("heartbeat::handle_new_heart(%r)"%heart)
         if heart not in self.incoming_registrations:
             self.log.info("heartbeat::ignoring new heart: %r"%heart)
         else:
             self.finish_registration(heart)
         
     
     def handle_heart_failure(self, heart):
         """handler to attach to heartbeater.
         called when a previously registered heart fails to respond to beat request.
         triggers unregistration"""
         self.log.debug("heartbeat::handle_heart_failure(%r)"%heart)
         eid = self.hearts.get(heart, None)
         queue = self.engines[eid].queue
         if eid is None:
             self.log.info("heartbeat::ignoring heart failure %r"%heart)
         else:
             self.unregister_engine(heart, dict(content=dict(id=eid, queue=queue)))
     
     #----------------------- MUX Queue Traffic ------------------------------
     
     def save_queue_request(self, idents, msg):
         if len(idents) < 2:
             self.log.error("invalid identity prefix: %s"%idents)
             return
         queue_id, client_id = idents[:2]
         try:
             msg = self.session.unpack_message(msg, content=False)
         except:
             self.log.error("queue::client %r sent invalid message to %r: %s"%(client_id, queue_id, msg), exc_info=True)
             return
         
         eid = self.by_ident.get(queue_id, None)
         if eid is None:
             self.log.error("queue::target %r not registered"%queue_id)
             self.log.debug("queue::    valid are: %s"%(self.by_ident.keys()))
             return
             
         header = msg['header']
         msg_id = header['msg_id']
         record = init_record(msg)
         record['engine_uuid'] = queue_id
         record['client_uuid'] = client_id
         record['queue'] = 'mux'
 
         try:
             # it's posible iopub arrived first:
             existing = self.db.get_record(msg_id)
             for key,evalue in existing.iteritems():
-                rvalue = record[key]
+                rvalue = record.get(key, None)
                 if evalue and rvalue and evalue != rvalue:
-                    self.log.error("conflicting initial state for record: %s:%s <> %s"%(msg_id, rvalue, evalue))
+                    self.log.warn("conflicting initial state for record: %r:%r <%r> %r"%(msg_id, rvalue, key, evalue))
                 elif evalue and not rvalue:
                     record[key] = evalue
             self.db.update_record(msg_id, record)
         except KeyError:
             self.db.add_record(msg_id, record)
         
         self.pending.add(msg_id)
         self.queues[eid].append(msg_id)
     
     def save_queue_result(self, idents, msg):
         if len(idents) < 2:
             self.log.error("invalid identity prefix: %s"%idents)
             return
             
         client_id, queue_id = idents[:2]
         try:
             msg = self.session.unpack_message(msg, content=False)
         except:
             self.log.error("queue::engine %r sent invalid message to %r: %s"%(
                     queue_id,client_id, msg), exc_info=True)
             return
         
         eid = self.by_ident.get(queue_id, None)
         if eid is None:
             self.log.error("queue::unknown engine %r is sending a reply: "%queue_id)
             # self.log.debug("queue::       %s"%msg[2:])
             return
         
         parent = msg['parent_header']
         if not parent:
             return
         msg_id = parent['msg_id']
         if msg_id in self.pending:
             self.pending.remove(msg_id)
             self.all_completed.add(msg_id)
             self.queues[eid].remove(msg_id)
             self.completed[eid].append(msg_id)
         elif msg_id not in self.all_completed:
             # it could be a result from a dead engine that died before delivering the
             # result
             self.log.warn("queue:: unknown msg finished %s"%msg_id)
             return
         # update record anyway, because the unregistration could have been premature
         rheader = msg['header']
         completed = datetime.strptime(rheader['date'], util.ISO8601)
         started = rheader.get('started', None)
         if started is not None:
             started = datetime.strptime(started, util.ISO8601)
         result = {
             'result_header' : rheader,
             'result_content': msg['content'],
             'started' : started,
             'completed' : completed
         }
 
         result['result_buffers'] = msg['buffers']
         try:
             self.db.update_record(msg_id, result)
         except Exception:
             self.log.error("DB Error updating record %r"%msg_id, exc_info=True)
         
             
     #--------------------- Task Queue Traffic ------------------------------
     
     def save_task_request(self, idents, msg):
         """Save the submission of a task."""
         client_id = idents[0]
         
         try:
             msg = self.session.unpack_message(msg, content=False)
         except:
             self.log.error("task::client %r sent invalid task message: %s"%(
                     client_id, msg), exc_info=True)
             return
         record = init_record(msg)
 
         record['client_uuid'] = client_id
         record['queue'] = 'task'
         header = msg['header']
         msg_id = header['msg_id']
         self.pending.add(msg_id)
         self.unassigned.add(msg_id)
         try:
             # it's posible iopub arrived first:
             existing = self.db.get_record(msg_id)
+            if existing['resubmitted']:
+                for key in ('submitted', 'client_uuid', 'buffers'):
+                    # don't clobber these keys on resubmit
+                    # submitted and client_uuid should be different
+                    # and buffers might be big, and shouldn't have changed
+                    record.pop(key)
+                    # still check content,header which should not change
+                    # but are not expensive to compare as buffers
+
             for key,evalue in existing.iteritems():
-                rvalue = record[key]
+                if key.endswith('buffers'):
+                    # don't compare buffers
+                    continue
+                rvalue = record.get(key, None)
                 if evalue and rvalue and evalue != rvalue:
-                    self.log.error("conflicting initial state for record: %s:%s <> %s"%(msg_id, rvalue, evalue))
+                    self.log.warn("conflicting initial state for record: %r:%r <%r> %r"%(msg_id, rvalue, key, evalue))
                 elif evalue and not rvalue:
                     record[key] = evalue
             self.db.update_record(msg_id, record)
         except KeyError:
             self.db.add_record(msg_id, record)
         except Exception:
             self.log.error("DB Error saving task request %r"%msg_id, exc_info=True)
     
     def save_task_result(self, idents, msg):
         """save the result of a completed task."""
         client_id = idents[0]
         try:
             msg = self.session.unpack_message(msg, content=False)
         except:
             self.log.error("task::invalid task result message send to %r: %s"%(
                     client_id, msg), exc_info=True)
             raise
             return
         
         parent = msg['parent_header']
         if not parent:
             # print msg
             self.log.warn("Task %r had no parent!"%msg)
             return
         msg_id = parent['msg_id']
         if msg_id in self.unassigned:
             self.unassigned.remove(msg_id)
         
         header = msg['header']
         engine_uuid = header.get('engine', None)
         eid = self.by_ident.get(engine_uuid, None)
         
         if msg_id in self.pending:
             self.pending.remove(msg_id)
             self.all_completed.add(msg_id)
             if eid is not None:
                 self.completed[eid].append(msg_id)
                 if msg_id in self.tasks[eid]:
                     self.tasks[eid].remove(msg_id)
             completed = datetime.strptime(header['date'], util.ISO8601)
             started = header.get('started', None)
             if started is not None:
                 started = datetime.strptime(started, util.ISO8601)
             result = {
                 'result_header' : header,
                 'result_content': msg['content'],
                 'started' : started,
                 'completed' : completed,
                 'engine_uuid': engine_uuid
             }
 
             result['result_buffers'] = msg['buffers']
             try:
                 self.db.update_record(msg_id, result)
             except Exception:
                 self.log.error("DB Error saving task request %r"%msg_id, exc_info=True)
             
         else:
             self.log.debug("task::unknown task %s finished"%msg_id)
     
     def save_task_destination(self, idents, msg):
         try:
             msg = self.session.unpack_message(msg, content=True)
         except:
             self.log.error("task::invalid task tracking message", exc_info=True)
             return
         content = msg['content']
         # print (content)
         msg_id = content['msg_id']
         engine_uuid = content['engine_id']
         eid = self.by_ident[engine_uuid]
         
         self.log.info("task::task %s arrived on %s"%(msg_id, eid))
         if msg_id in self.unassigned:
             self.unassigned.remove(msg_id)
         # else:
         #     self.log.debug("task::task %s not listed as MIA?!"%(msg_id))
         
         self.tasks[eid].append(msg_id)
         # self.pending[msg_id][1].update(received=datetime.now(),engine=(eid,engine_uuid))
         try:
             self.db.update_record(msg_id, dict(engine_uuid=engine_uuid))
         except Exception:
             self.log.error("DB Error saving task destination %r"%msg_id, exc_info=True)
             
     
     def mia_task_request(self, idents, msg):
         raise NotImplementedError
         client_id = idents[0]
         # content = dict(mia=self.mia,status='ok')
         # self.session.send('mia_reply', content=content, idents=client_id)
         
         
     #--------------------- IOPub Traffic ------------------------------
     
     def save_iopub_message(self, topics, msg):
         """save an iopub message into the db"""
         # print (topics)
         try:
             msg = self.session.unpack_message(msg, content=True)
         except:
             self.log.error("iopub::invalid IOPub message", exc_info=True)
             return
         
         parent = msg['parent_header']
         if not parent:
             self.log.error("iopub::invalid IOPub message: %s"%msg)
             return
         msg_id = parent['msg_id']
         msg_type = msg['msg_type']
         content = msg['content']
         
         # ensure msg_id is in db
         try:
             rec = self.db.get_record(msg_id)
         except KeyError:
             rec = empty_record()
             rec['msg_id'] = msg_id
             self.db.add_record(msg_id, rec)
         # stream
         d = {}
         if msg_type == 'stream':
             name = content['name']
             s = rec[name] or ''
             d[name] = s + content['data']
             
         elif msg_type == 'pyerr':
             d['pyerr'] = content
         elif msg_type == 'pyin':
             d['pyin'] = content['code']
         else:
             d[msg_type] = content.get('data', '')
         
         try:
             self.db.update_record(msg_id, d)
         except Exception:
             self.log.error("DB Error saving iopub message %r"%msg_id, exc_info=True)
         
     
             
     #-------------------------------------------------------------------------
     # Registration requests
     #-------------------------------------------------------------------------
         
     def connection_request(self, client_id, msg):
         """Reply with connection addresses for clients."""
         self.log.info("client::client %s connected"%client_id)
         content = dict(status='ok')
         content.update(self.client_info)
         jsonable = {}
         for k,v in self.keytable.iteritems():
             if v not in self.dead_engines:
                 jsonable[str(k)] = v
         content['engines'] = jsonable
         self.session.send(self.query, 'connection_reply', content, parent=msg, ident=client_id)
     
     def register_engine(self, reg, msg):
         """Register a new engine."""
         content = msg['content']
         try:
             queue = content['queue']
         except KeyError:
             self.log.error("registration::queue not specified", exc_info=True)
             return
         heart = content.get('heartbeat', None)
         """register a new engine, and create the socket(s) necessary"""
         eid = self._next_id
         # print (eid, queue, reg, heart)
         
         self.log.debug("registration::register_engine(%i, %r, %r, %r)"%(eid, queue, reg, heart))
         
         content = dict(id=eid,status='ok')
         content.update(self.engine_info)
         # check if requesting available IDs:
         if queue in self.by_ident:
             try:
                 raise KeyError("queue_id %r in use"%queue)
             except:
                 content = error.wrap_exception()
                 self.log.error("queue_id %r in use"%queue, exc_info=True)
         elif heart in self.hearts: # need to check unique hearts?
             try:
                 raise KeyError("heart_id %r in use"%heart)
             except:
                 self.log.error("heart_id %r in use"%heart, exc_info=True)
                 content = error.wrap_exception()
         else:
             for h, pack in self.incoming_registrations.iteritems():
                 if heart == h:
                     try:
                         raise KeyError("heart_id %r in use"%heart)
                     except:
                         self.log.error("heart_id %r in use"%heart, exc_info=True)
                         content = error.wrap_exception()
                     break
                 elif queue == pack[1]:
                     try:
                         raise KeyError("queue_id %r in use"%queue)
                     except:
                         self.log.error("queue_id %r in use"%queue, exc_info=True)
                         content = error.wrap_exception()
                     break
         
         msg = self.session.send(self.query, "registration_reply", 
                 content=content, 
                 ident=reg)
         
         if content['status'] == 'ok':
             if heart in self.heartmonitor.hearts:
                 # already beating
                 self.incoming_registrations[heart] = (eid,queue,reg[0],None)
                 self.finish_registration(heart)
             else:
                 purge = lambda : self._purge_stalled_registration(heart)
                 dc = ioloop.DelayedCallback(purge, self.registration_timeout, self.loop)
                 dc.start()
                 self.incoming_registrations[heart] = (eid,queue,reg[0],dc)
         else:
             self.log.error("registration::registration %i failed: %s"%(eid, content['evalue']))
         return eid
     
     def unregister_engine(self, ident, msg):
         """Unregister an engine that explicitly requested to leave."""
         try:
             eid = msg['content']['id']
         except:
             self.log.error("registration::bad engine id for unregistration: %s"%ident, exc_info=True)
             return
         self.log.info("registration::unregister_engine(%s)"%eid)
         # print (eid)
         uuid = self.keytable[eid]
         content=dict(id=eid, queue=uuid)
         self.dead_engines.add(uuid)
         # self.ids.remove(eid)
         # uuid = self.keytable.pop(eid)
         # 
         # ec = self.engines.pop(eid)
         # self.hearts.pop(ec.heartbeat)
         # self.by_ident.pop(ec.queue)
         # self.completed.pop(eid)
         handleit = lambda : self._handle_stranded_msgs(eid, uuid)
         dc = ioloop.DelayedCallback(handleit, self.registration_timeout, self.loop)
         dc.start()
         ############## TODO: HANDLE IT ################
         
         if self.notifier:
             self.session.send(self.notifier, "unregistration_notification", content=content)
     
     def _handle_stranded_msgs(self, eid, uuid):
         """Handle messages known to be on an engine when the engine unregisters.
         
         It is possible that this will fire prematurely - that is, an engine will
         go down after completing a result, and the client will be notified
         that the result failed and later receive the actual result.
         """
         
         outstanding = self.queues[eid]
         
         for msg_id in outstanding:
             self.pending.remove(msg_id)
             self.all_completed.add(msg_id)
             try:
                 raise error.EngineError("Engine %r died while running task %r"%(eid, msg_id))
             except:
                 content = error.wrap_exception()
             # build a fake header:
             header = {}
             header['engine'] = uuid
             header['date'] = datetime.now()
             rec = dict(result_content=content, result_header=header, result_buffers=[])
             rec['completed'] = header['date']
             rec['engine_uuid'] = uuid
             try:
                 self.db.update_record(msg_id, rec)
             except Exception:
                 self.log.error("DB Error handling stranded msg %r"%msg_id, exc_info=True)
                 
     
     def finish_registration(self, heart):
         """Second half of engine registration, called after our HeartMonitor
         has received a beat from the Engine's Heart."""
         try: 
             (eid,queue,reg,purge) = self.incoming_registrations.pop(heart)
         except KeyError:
             self.log.error("registration::tried to finish nonexistant registration", exc_info=True)
             return
         self.log.info("registration::finished registering engine %i:%r"%(eid,queue))
         if purge is not None:
             purge.stop()
         control = queue
         self.ids.add(eid)
         self.keytable[eid] = queue
         self.engines[eid] = EngineConnector(id=eid, queue=queue, registration=reg, 
                                     control=control, heartbeat=heart)
         self.by_ident[queue] = eid
         self.queues[eid] = list()
         self.tasks[eid] = list()
         self.completed[eid] = list()
         self.hearts[heart] = eid
         content = dict(id=eid, queue=self.engines[eid].queue)
         if self.notifier:
             self.session.send(self.notifier, "registration_notification", content=content)
         self.log.info("engine::Engine Connected: %i"%eid)
     
     def _purge_stalled_registration(self, heart):
         if heart in self.incoming_registrations:
             eid = self.incoming_registrations.pop(heart)[0]
             self.log.info("registration::purging stalled registration: %i"%eid)
         else:
             pass
             
     #-------------------------------------------------------------------------
     # Client Requests
     #-------------------------------------------------------------------------
     
     def shutdown_request(self, client_id, msg):
         """handle shutdown request."""
         self.session.send(self.query, 'shutdown_reply', content={'status': 'ok'}, ident=client_id)
         # also notify other clients of shutdown
         self.session.send(self.notifier, 'shutdown_notice', content={'status': 'ok'})
         dc = ioloop.DelayedCallback(lambda : self._shutdown(), 1000, self.loop)
         dc.start()
     
     def _shutdown(self):
         self.log.info("hub::hub shutting down.")
         time.sleep(0.1)
         sys.exit(0)
         
     
     def check_load(self, client_id, msg):
         content = msg['content']
         try:
             targets = content['targets']
             targets = self._validate_targets(targets)
         except:
             content = error.wrap_exception()
             self.session.send(self.query, "hub_error", 
                     content=content, ident=client_id)
             return
         
         content = dict(status='ok')
         # loads = {}
         for t in targets:
             content[bytes(t)] = len(self.queues[t])+len(self.tasks[t])
         self.session.send(self.query, "load_reply", content=content, ident=client_id)
             
     
     def queue_status(self, client_id, msg):
         """Return the Queue status of one or more targets.
         if verbose: return the msg_ids
         else: return len of each type.
         keys: queue (pending MUX jobs)
             tasks (pending Task jobs)
             completed (finished jobs from both queues)"""
         content = msg['content']
         targets = content['targets']
         try:
             targets = self._validate_targets(targets)
         except:
             content = error.wrap_exception()
             self.session.send(self.query, "hub_error", 
                     content=content, ident=client_id)
             return
         verbose = content.get('verbose', False)
         content = dict(status='ok')
         for t in targets:
             queue = self.queues[t]
             completed = self.completed[t]
             tasks = self.tasks[t]
             if not verbose:
                 queue = len(queue)
                 completed = len(completed)
                 tasks = len(tasks)
             content[bytes(t)] = {'queue': queue, 'completed': completed , 'tasks': tasks}
         content['unassigned'] = list(self.unassigned) if verbose else len(self.unassigned)
         
         self.session.send(self.query, "queue_reply", content=content, ident=client_id)
     
     def purge_results(self, client_id, msg):
         """Purge results from memory. This method is more valuable before we move
         to a DB based message storage mechanism."""
         content = msg['content']
         msg_ids = content.get('msg_ids', [])
         reply = dict(status='ok')
         if msg_ids == 'all':
             try:
                 self.db.drop_matching_records(dict(completed={'$ne':None}))
             except Exception:
                 reply = error.wrap_exception()
         else:
             for msg_id in msg_ids:
                 if msg_id in self.all_completed:
                     self.db.drop_record(msg_id)
                 else:
                     if msg_id in self.pending:
                         try:
                             raise IndexError("msg pending: %r"%msg_id)
                         except:
                             reply = error.wrap_exception()
                     else:
                         try:
                             raise IndexError("No such msg: %r"%msg_id)
                         except:
                             reply = error.wrap_exception()
                     break
             eids = content.get('engine_ids', [])
             for eid in eids:
                 if eid not in self.engines:
                     try:
                         raise IndexError("No such engine: %i"%eid)
                     except:
                         reply = error.wrap_exception()
                     break
                 msg_ids = self.completed.pop(eid)
                 uid = self.engines[eid].queue
                 try:
                     self.db.drop_matching_records(dict(engine_uuid=uid, completed={'$ne':None}))
                 except Exception:
                     reply = error.wrap_exception()
                     break
         
         self.session.send(self.query, 'purge_reply', content=reply, ident=client_id)
     
-    def resubmit_task(self, client_id, msg, buffers):
-        """Resubmit a task."""
-        raise NotImplementedError
+    def resubmit_task(self, client_id, msg):
+        """Resubmit one or more tasks."""
+        def finish(reply):
+            self.session.send(self.query, 'resubmit_reply', content=reply, ident=client_id)
+
+        content = msg['content']
+        msg_ids = content['msg_ids']
+        reply = dict(status='ok')
+        try:
+            records = self.db.find_records({'msg_id' : {'$in' : msg_ids}}, keys=[
+                'header', 'content', 'buffers'])
+        except Exception:
+            self.log.error('db::db error finding tasks to resubmit', exc_info=True)
+            return finish(error.wrap_exception())
+
+        # validate msg_ids
+        found_ids = [ rec['msg_id'] for rec in records ]
+        invalid_ids = filter(lambda m: m in self.pending, found_ids)
+        if len(records) > len(msg_ids):
+            try:
+                raise RuntimeError("DB appears to be in an inconsistent state."
+                    "More matching records were found than should exist")
+            except Exception:
+                return finish(error.wrap_exception())
+        elif len(records) < len(msg_ids):
+            missing = [ m for m in msg_ids if m not in found_ids ]
+            try:
+                raise KeyError("No such msg(s): %s"%missing)
+            except KeyError:
+                return finish(error.wrap_exception())
+        elif invalid_ids:
+            msg_id = invalid_ids[0]
+            try:
+                raise ValueError("Task %r appears to be inflight"%(msg_id))
+            except Exception:
+                return finish(error.wrap_exception())
+
+        # clear the existing records
+        rec = empty_record()
+        map(rec.pop, ['msg_id', 'header', 'content', 'buffers', 'submitted'])
+        rec['resubmitted'] = datetime.now()
+        rec['queue'] = 'task'
+        rec['client_uuid'] = client_id[0]
+        try:
+            for msg_id in msg_ids:
+                self.all_completed.discard(msg_id)
+                self.db.update_record(msg_id, rec)
+        except Exception:
+            self.log.error('db::db error upating record', exc_info=True)
+            reply = error.wrap_exception()
+        else:
+            # send the messages
+            for rec in records:
+                header = rec['header']
+                msg = self.session.msg(header['msg_type'])
+                msg['content'] = rec['content']
+                msg['header'] = header
+                msg['msg_id'] = rec['msg_id']
+                self.session.send(self.resubmit, msg, buffers=rec['buffers'])
+
+        finish(dict(status='ok'))
+
     
     def _extract_record(self, rec):
         """decompose a TaskRecord dict into subsection of reply for get_result"""
         io_dict = {}
         for key in 'pyin pyout pyerr stdout stderr'.split():
                 io_dict[key] = rec[key]
         content = { 'result_content': rec['result_content'],
                             'header': rec['header'],
                             'result_header' : rec['result_header'],
                             'io' : io_dict,
                           }
         if rec['result_buffers']:
             buffers = map(str, rec['result_buffers'])
         else:
             buffers = []
         
         return content, buffers
     
     def get_results(self, client_id, msg):
         """Get the result of 1 or more messages."""
         content = msg['content']
         msg_ids = sorted(set(content['msg_ids']))
         statusonly = content.get('status_only', False)
         pending = []
         completed = []
         content = dict(status='ok')
         content['pending'] = pending
         content['completed'] = completed
         buffers = []
         if not statusonly:
             try:
                 matches = self.db.find_records(dict(msg_id={'$in':msg_ids}))
                 # turn match list into dict, for faster lookup
                 records = {}
                 for rec in matches:
                     records[rec['msg_id']] = rec
             except Exception:
                 content = error.wrap_exception()
                 self.session.send(self.query, "result_reply", content=content, 
                                                     parent=msg, ident=client_id)
                 return
         else:
             records = {}
         for msg_id in msg_ids:
             if msg_id in self.pending:
                 pending.append(msg_id)
-            elif msg_id in self.all_completed or msg_id in records:
+            elif msg_id in self.all_completed:
                 completed.append(msg_id)
                 if not statusonly:
                     c,bufs = self._extract_record(records[msg_id])
                     content[msg_id] = c
                     buffers.extend(bufs)
+            elif msg_id in records:
+                if rec['completed']:
+                    completed.append(msg_id)
+                    c,bufs = self._extract_record(records[msg_id])
+                    content[msg_id] = c
+                    buffers.extend(bufs)
+                else:
+                    pending.append(msg_id)
             else:
                 try:
                     raise KeyError('No such message: '+msg_id)
                 except:
                     content = error.wrap_exception()
                 break
         self.session.send(self.query, "result_reply", content=content, 
                                             parent=msg, ident=client_id,
                                             buffers=buffers)
 
     def get_history(self, client_id, msg):
         """Get a list of all msg_ids in our DB records"""
         try:
             msg_ids = self.db.get_history()
         except Exception as e:
             content = error.wrap_exception()
         else:
             content = dict(status='ok', history=msg_ids)
         
         self.session.send(self.query, "history_reply", content=content, 
                                             parent=msg, ident=client_id)
 
     def db_query(self, client_id, msg):
         """Perform a raw query on the task record database."""
         content = msg['content']
         query = content.get('query', {})
         keys = content.get('keys', None)
         query = util.extract_dates(query)
         buffers = []
         empty = list()
         
         try:
             records = self.db.find_records(query, keys)
         except Exception as e:
             content = error.wrap_exception()
         else:
             # extract buffers from reply content:
             if keys is not None:
                 buffer_lens = [] if 'buffers' in keys else None
                 result_buffer_lens = [] if 'result_buffers' in keys else None
             else:
                 buffer_lens = []
                 result_buffer_lens = []
             
             for rec in records:
                 # buffers may be None, so double check
                 if buffer_lens is not None:
                     b = rec.pop('buffers', empty) or empty
                     buffer_lens.append(len(b))
                     buffers.extend(b)
                 if result_buffer_lens is not None:
                     rb = rec.pop('result_buffers', empty) or empty
                     result_buffer_lens.append(len(rb))
                     buffers.extend(rb)
             content = dict(status='ok', records=records, buffer_lens=buffer_lens,
                                     result_buffer_lens=result_buffer_lens)
         
         self.session.send(self.query, "db_reply", content=content, 
                                             parent=msg, ident=client_id,
                                             buffers=buffers)
 

IPython/parallel/streamsession.py

@@ -1,416 +1,419 b''
 #!/usr/bin/env python
 """edited session.py to work with streams, and move msg_type to the header
 """
 #-----------------------------------------------------------------------------
 #  Copyright (C) 2010-2011  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-----------------------------------------------------------------------------
 
 
 import os
 import pprint
 import uuid
 from datetime import datetime
 
 try:
     import cPickle
     pickle = cPickle
 except:
     cPickle = None
     import pickle
 
 import zmq
 from zmq.utils import jsonapi
 from zmq.eventloop.zmqstream import ZMQStream
 
 from .util import ISO8601
 
 def squash_unicode(obj):
     """coerce unicode back to bytestrings."""
     if isinstance(obj,dict):
         for key in obj.keys():
             obj[key] = squash_unicode(obj[key])
             if isinstance(key, unicode):
                 obj[squash_unicode(key)] = obj.pop(key)
     elif isinstance(obj, list):
         for i,v in enumerate(obj):
             obj[i] = squash_unicode(v)
     elif isinstance(obj, unicode):
         obj = obj.encode('utf8')
     return obj
 
 def _date_default(obj):
     if isinstance(obj, datetime):
         return obj.strftime(ISO8601)
     else:
         raise TypeError("%r is not JSON serializable"%obj)
 
 _default_key = 'on_unknown' if jsonapi.jsonmod.__name__ == 'jsonlib' else 'default'
 json_packer = lambda obj: jsonapi.dumps(obj, **{_default_key:_date_default})
 json_unpacker = lambda s: squash_unicode(jsonapi.loads(s))
 
 pickle_packer = lambda o: pickle.dumps(o,-1)
 pickle_unpacker = pickle.loads
 
 default_packer = json_packer
 default_unpacker = json_unpacker
 
 
 DELIM="<IDS|MSG>"
 
 class Message(object):
     """A simple message object that maps dict keys to attributes.
 
     A Message can be created from a dict and a dict from a Message instance
     simply by calling dict(msg_obj)."""
     
     def __init__(self, msg_dict):
         dct = self.__dict__
         for k, v in dict(msg_dict).iteritems():
             if isinstance(v, dict):
                 v = Message(v)
             dct[k] = v
 
     # Having this iterator lets dict(msg_obj) work out of the box.
     def __iter__(self):
         return iter(self.__dict__.iteritems())
     
     def __repr__(self):
         return repr(self.__dict__)
 
     def __str__(self):
         return pprint.pformat(self.__dict__)
 
     def __contains__(self, k):
         return k in self.__dict__
 
     def __getitem__(self, k):
         return self.__dict__[k]
 
 
 def msg_header(msg_id, msg_type, username, session):
     date=datetime.now().strftime(ISO8601)
     return locals()
 
 def extract_header(msg_or_header):
     """Given a message or header, return the header."""
     if not msg_or_header:
         return {}
     try:
         # See if msg_or_header is the entire message.
         h = msg_or_header['header']
     except KeyError:
         try:
             # See if msg_or_header is just the header
             h = msg_or_header['msg_id']
         except KeyError:
             raise
         else:
             h = msg_or_header
     if not isinstance(h, dict):
         h = dict(h)
     return h
 
 class StreamSession(object):
     """tweaked version of IPython.zmq.session.Session, for development in Parallel"""
     debug=False
     key=None
     
     def __init__(self, username=None, session=None, packer=None, unpacker=None, key=None, keyfile=None):
         if username is None:
             username = os.environ.get('USER','username')
         self.username = username
         if session is None:
             self.session = str(uuid.uuid4())
         else:
             self.session = session
         self.msg_id = str(uuid.uuid4())
         if packer is None:
             self.pack = default_packer
         else:
             if not callable(packer):
                 raise TypeError("packer must be callable, not %s"%type(packer))
             self.pack = packer
         
         if unpacker is None:
             self.unpack = default_unpacker
         else:
             if not callable(unpacker):
                 raise TypeError("unpacker must be callable, not %s"%type(unpacker))
             self.unpack = unpacker
         
         if key is not None and keyfile is not None:
             raise TypeError("Must specify key OR keyfile, not both")
         if keyfile is not None:
             with open(keyfile) as f:
                 self.key = f.read().strip()
         else:
             self.key = key
         if isinstance(self.key, unicode):
             self.key = self.key.encode('utf8')
         # print key, keyfile, self.key
         self.none = self.pack({})
             
     def msg_header(self, msg_type):
         h = msg_header(self.msg_id, msg_type, self.username, self.session)
         self.msg_id = str(uuid.uuid4())
         return h
 
     def msg(self, msg_type, content=None, parent=None, subheader=None):
         msg = {}
         msg['header'] = self.msg_header(msg_type)
         msg['msg_id'] = msg['header']['msg_id']
         msg['parent_header'] = {} if parent is None else extract_header(parent)
         msg['msg_type'] = msg_type
         msg['content'] = {} if content is None else content
         sub = {} if subheader is None else subheader
         msg['header'].update(sub)
         return msg
 
     def check_key(self, msg_or_header):
         """Check that a message's header has the right key"""
         if self.key is None:
             return True
         header = extract_header(msg_or_header)
         return header.get('key', None) == self.key
             
 
     def serialize(self, msg, ident=None):
         content = msg.get('content', {})
         if content is None:
             content = self.none
         elif isinstance(content, dict):
             content = self.pack(content)
         elif isinstance(content, bytes):
             # content is already packed, as in a relayed message
             pass
+        elif isinstance(content, unicode):
+            # should be bytes, but JSON often spits out unicode
+            content = content.encode('utf8')
         else:
             raise TypeError("Content incorrect type: %s"%type(content))
 
         to_send = []
 
         if isinstance(ident, list):
             # accept list of idents
             to_send.extend(ident)
         elif ident is not None:
             to_send.append(ident)
         to_send.append(DELIM)
         if self.key is not None:
             to_send.append(self.key)
         to_send.append(self.pack(msg['header']))
         to_send.append(self.pack(msg['parent_header']))
         to_send.append(content)
 
         return to_send
         
     def send(self, stream, msg_or_type, content=None, buffers=None, parent=None, subheader=None, ident=None, track=False):
         """Build and send a message via stream or socket.
         
         Parameters
         ----------
         
         stream : zmq.Socket or ZMQStream
             the socket-like object used to send the data
         msg_or_type : str or Message/dict
             Normally, msg_or_type will be a msg_type unless a message is being sent more
             than once.
         
         content : dict or None
             the content of the message (ignored if msg_or_type is a message)
         buffers : list or None
             the already-serialized buffers to be appended to the message
         parent : Message or dict or None
             the parent or parent header describing the parent of this message
         subheader : dict or None
             extra header keys for this message's header
         ident : bytes or list of bytes
             the zmq.IDENTITY routing path
         track : bool
             whether to track.  Only for use with Sockets, because ZMQStream objects cannot track messages.
         
         Returns
         -------
         msg : message dict
             the constructed message
         (msg,tracker) : (message dict, MessageTracker)
             if track=True, then a 2-tuple will be returned, the first element being the constructed
             message, and the second being the MessageTracker
             
         """
 
         if not isinstance(stream, (zmq.Socket, ZMQStream)):
             raise TypeError("stream must be Socket or ZMQStream, not %r"%type(stream))
         elif track and isinstance(stream, ZMQStream):
             raise TypeError("ZMQStream cannot track messages")
         
         if isinstance(msg_or_type, (Message, dict)):
             # we got a Message, not a msg_type
             # don't build a new Message
             msg = msg_or_type
         else:
             msg = self.msg(msg_or_type, content, parent, subheader)
         
         buffers = [] if buffers is None else buffers
         to_send = self.serialize(msg, ident)
         flag = 0
         if buffers:
             flag = zmq.SNDMORE
             _track = False
         else:
             _track=track
         if track:
             tracker = stream.send_multipart(to_send, flag, copy=False, track=_track)
         else:
             tracker = stream.send_multipart(to_send, flag, copy=False)
         for b in buffers[:-1]:
             stream.send(b, flag, copy=False)
         if buffers:
             if track:
                 tracker = stream.send(buffers[-1], copy=False, track=track)
             else:
                 tracker = stream.send(buffers[-1], copy=False)
                 
         # omsg = Message(msg)
         if self.debug:
             pprint.pprint(msg)
             pprint.pprint(to_send)
             pprint.pprint(buffers)
         
         msg['tracker'] = tracker
         
         return msg
     
     def send_raw(self, stream, msg, flags=0, copy=True, ident=None):
         """Send a raw message via ident path.
         
         Parameters
         ----------
         msg : list of sendable buffers"""
         to_send = []
         if isinstance(ident, bytes):
             ident = [ident]
         if ident is not None:
             to_send.extend(ident)
         to_send.append(DELIM)
         if self.key is not None:
             to_send.append(self.key)
         to_send.extend(msg)
         stream.send_multipart(msg, flags, copy=copy)
     
     def recv(self, socket, mode=zmq.NOBLOCK, content=True, copy=True):
         """receives and unpacks a message
         returns [idents], msg"""
         if isinstance(socket, ZMQStream):
             socket = socket.socket
         try:
             msg = socket.recv_multipart(mode)
         except zmq.ZMQError as e:
             if e.errno == zmq.EAGAIN:
                 # We can convert EAGAIN to None as we know in this case
                 # recv_multipart won't return None.
                 return None
             else:
                 raise
         # return an actual Message object
         # determine the number of idents by trying to unpack them.
         # this is terrible:
         idents, msg = self.feed_identities(msg, copy)
         try:
             return idents, self.unpack_message(msg, content=content, copy=copy)
         except Exception as e:
             print (idents, msg)
             # TODO: handle it
             raise e
     
     def feed_identities(self, msg, copy=True):
         """feed until DELIM is reached, then return the prefix as idents and remainder as
         msg. This is easily broken by setting an IDENT to DELIM, but that would be silly.
         
         Parameters
         ----------
         msg : a list of Message or bytes objects
             the message to be split
         copy : bool
             flag determining whether the arguments are bytes or Messages
         
         Returns
         -------
         (idents,msg) : two lists
             idents will always be a list of bytes - the indentity prefix
             msg will be a list of bytes or Messages, unchanged from input
             msg should be unpackable via self.unpack_message at this point.
         """
         ikey = int(self.key is not None)
         minlen = 3 + ikey
         msg = list(msg)
         idents = []
         while len(msg) > minlen:
             if copy:
                 s = msg[0]
             else:
                 s = msg[0].bytes
             if s == DELIM:
                 msg.pop(0)
                 break
             else:
                 idents.append(s)
                 msg.pop(0)
                 
         return idents, msg
     
     def unpack_message(self, msg, content=True, copy=True):
         """Return a message object from the format
         sent by self.send.
         
         Parameters:
         -----------
         
         content : bool (True)
             whether to unpack the content dict (True), 
             or leave it serialized (False)
         
         copy : bool (True)
             whether to return the bytes (True), 
             or the non-copying Message object in each place (False)
         
         """
         ikey = int(self.key is not None)
         minlen = 3 + ikey
         message = {}
         if not copy:
             for i in range(minlen):
                 msg[i] = msg[i].bytes
         if ikey:
             if not self.key == msg[0]:
                 raise KeyError("Invalid Session Key: %s"%msg[0])
         if not len(msg) >= minlen:
             raise TypeError("malformed message, must have at least %i elements"%minlen)
         message['header'] = self.unpack(msg[ikey+0])
         message['msg_type'] = message['header']['msg_type']
         message['parent_header'] = self.unpack(msg[ikey+1])
         if content:
             message['content'] = self.unpack(msg[ikey+2])
         else:
             message['content'] = msg[ikey+2]
         
         message['buffers'] = msg[ikey+3:]# [ m.buffer for m in msg[3:] ]
         return message
         
 
 def test_msg2obj():
     am = dict(x=1)
     ao = Message(am)
     assert ao.x == am['x']
 
     am['y'] = dict(z=1)
     ao = Message(am)
     assert ao.y.z == am['y']['z']
     
     k1, k2 = 'y', 'z'
     assert ao[k1][k2] == am[k1][k2]
     
     am2 = dict(ao)
     assert am['x'] == am2['x']
     assert am['y']['z'] == am2['y']['z']

IPython/parallel/tests/test_client.py

@@ -1,214 +1,237 b''
 """Tests for parallel client.py"""
 
 #-------------------------------------------------------------------------------
 #  Copyright (C) 2011  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-------------------------------------------------------------------------------
 
 #-------------------------------------------------------------------------------
 # Imports
 #-------------------------------------------------------------------------------
 
 import time
 from datetime import datetime
 from tempfile import mktemp
 
 import zmq
 
 from IPython.parallel.client import client as clientmod
 from IPython.parallel import error
 from IPython.parallel import AsyncResult, AsyncHubResult
 from IPython.parallel import LoadBalancedView, DirectView
 
 from clienttest import ClusterTestCase, segfault, wait, add_engines
 
 def setup():
     add_engines(4)
 
 class TestClient(ClusterTestCase):
     
     def test_ids(self):
         n = len(self.client.ids)
         self.add_engines(3)
         self.assertEquals(len(self.client.ids), n+3)
     
     def test_view_indexing(self):
         """test index access for views"""
         self.add_engines(2)
         targets = self.client._build_targets('all')[-1]
         v = self.client[:]
         self.assertEquals(v.targets, targets)
         t = self.client.ids[2]
         v = self.client[t]
         self.assert_(isinstance(v, DirectView))
         self.assertEquals(v.targets, t)
         t = self.client.ids[2:4]
         v = self.client[t]
         self.assert_(isinstance(v, DirectView))
         self.assertEquals(v.targets, t)
         v = self.client[::2]
         self.assert_(isinstance(v, DirectView))
         self.assertEquals(v.targets, targets[::2])
         v = self.client[1::3]
         self.assert_(isinstance(v, DirectView))
         self.assertEquals(v.targets, targets[1::3])
         v = self.client[:-3]
         self.assert_(isinstance(v, DirectView))
         self.assertEquals(v.targets, targets[:-3])
         v = self.client[-1]
         self.assert_(isinstance(v, DirectView))
         self.assertEquals(v.targets, targets[-1])
         self.assertRaises(TypeError, lambda : self.client[None])
     
     def test_lbview_targets(self):
         """test load_balanced_view targets"""
         v = self.client.load_balanced_view()
         self.assertEquals(v.targets, None)
         v = self.client.load_balanced_view(-1)
         self.assertEquals(v.targets, [self.client.ids[-1]])
         v = self.client.load_balanced_view('all')
         self.assertEquals(v.targets, self.client.ids)
     
     def test_targets(self):
         """test various valid targets arguments"""
         build = self.client._build_targets
         ids = self.client.ids
         idents,targets = build(None)
         self.assertEquals(ids, targets)
     
     def test_clear(self):
         """test clear behavior"""
         # self.add_engines(2)
         v = self.client[:]
         v.block=True
         v.push(dict(a=5))
         v.pull('a')
         id0 = self.client.ids[-1]
         self.client.clear(targets=id0, block=True)
         a = self.client[:-1].get('a')
         self.assertRaisesRemote(NameError, self.client[id0].get, 'a')
         self.client.clear(block=True)
         for i in self.client.ids:
             # print i
             self.assertRaisesRemote(NameError, self.client[i].get, 'a')
     
     def test_get_result(self):
         """test getting results from the Hub."""
         c = clientmod.Client(profile='iptest')
         # self.add_engines(1)
         t = c.ids[-1]
         ar = c[t].apply_async(wait, 1)
         # give the monitor time to notice the message
         time.sleep(.25)
         ahr = self.client.get_result(ar.msg_ids)
         self.assertTrue(isinstance(ahr, AsyncHubResult))
         self.assertEquals(ahr.get(), ar.get())
         ar2 = self.client.get_result(ar.msg_ids)
         self.assertFalse(isinstance(ar2, AsyncHubResult))
         c.close()
     
     def test_ids_list(self):
         """test client.ids"""
         # self.add_engines(2)
         ids = self.client.ids
         self.assertEquals(ids, self.client._ids)
         self.assertFalse(ids is self.client._ids)
         ids.remove(ids[-1])
         self.assertNotEquals(ids, self.client._ids)
     
     def test_queue_status(self):
         # self.addEngine(4)
         ids = self.client.ids
         id0 = ids[0]
         qs = self.client.queue_status(targets=id0)
         self.assertTrue(isinstance(qs, dict))
         self.assertEquals(sorted(qs.keys()), ['completed', 'queue', 'tasks'])
         allqs = self.client.queue_status()
         self.assertTrue(isinstance(allqs, dict))
         self.assertEquals(sorted(allqs.keys()), sorted(self.client.ids + ['unassigned']))
         unassigned = allqs.pop('unassigned')
         for eid,qs in allqs.items():
             self.assertTrue(isinstance(qs, dict))
             self.assertEquals(sorted(qs.keys()), ['completed', 'queue', 'tasks'])
 
     def test_shutdown(self):
         # self.addEngine(4)
         ids = self.client.ids
         id0 = ids[0]
         self.client.shutdown(id0, block=True)
         while id0 in self.client.ids:
             time.sleep(0.1)
             self.client.spin()
         
         self.assertRaises(IndexError, lambda : self.client[id0])
         
     def test_result_status(self):
         pass
         # to be written
     
     def test_db_query_dt(self):
         """test db query by date"""
         hist = self.client.hub_history()
         middle = self.client.db_query({'msg_id' : hist[len(hist)/2]})[0]
         tic = middle['submitted']
         before = self.client.db_query({'submitted' : {'$lt' : tic}})
         after = self.client.db_query({'submitted' : {'$gte' : tic}})
         self.assertEquals(len(before)+len(after),len(hist))
         for b in before:
             self.assertTrue(b['submitted'] < tic)
         for a in after:
             self.assertTrue(a['submitted'] >= tic)
         same = self.client.db_query({'submitted' : tic})
         for s in same:
             self.assertTrue(s['submitted'] == tic)
     
     def test_db_query_keys(self):
         """test extracting subset of record keys"""
         found = self.client.db_query({'msg_id': {'$ne' : ''}},keys=['submitted', 'completed'])
         for rec in found:
             self.assertEquals(set(rec.keys()), set(['msg_id', 'submitted', 'completed']))
     
     def test_db_query_msg_id(self):
         """ensure msg_id is always in db queries"""
         found = self.client.db_query({'msg_id': {'$ne' : ''}},keys=['submitted', 'completed'])
         for rec in found:
             self.assertTrue('msg_id' in rec.keys())
         found = self.client.db_query({'msg_id': {'$ne' : ''}},keys=['submitted'])
         for rec in found:
             self.assertTrue('msg_id' in rec.keys())
         found = self.client.db_query({'msg_id': {'$ne' : ''}},keys=['msg_id'])
         for rec in found:
             self.assertTrue('msg_id' in rec.keys())
     
     def test_db_query_in(self):
         """test db query with '$in','$nin' operators"""
         hist = self.client.hub_history()
         even = hist[::2]
         odd = hist[1::2]
         recs = self.client.db_query({ 'msg_id' : {'$in' : even}})
         found = [ r['msg_id'] for r in recs ]
         self.assertEquals(set(even), set(found))
         recs = self.client.db_query({ 'msg_id' : {'$nin' : even}})
         found = [ r['msg_id'] for r in recs ]
         self.assertEquals(set(odd), set(found))
     
     def test_hub_history(self):
         hist = self.client.hub_history()
         recs = self.client.db_query({ 'msg_id' : {"$ne":''}})
         recdict = {}
         for rec in recs:
             recdict[rec['msg_id']] = rec
         
         latest = datetime(1984,1,1)
         for msg_id in hist:
             rec = recdict[msg_id]
             newt = rec['submitted']
             self.assertTrue(newt >= latest)
             latest = newt
         ar = self.client[-1].apply_async(lambda : 1)
         ar.get()
         time.sleep(0.25)
         self.assertEquals(self.client.hub_history()[-1:],ar.msg_ids)
     
+    def test_resubmit(self):
+        def f():
+            import random
+            return random.random()
+        v = self.client.load_balanced_view()
+        ar = v.apply_async(f)
+        r1 = ar.get(1)
+        ahr = self.client.resubmit(ar.msg_ids)
+        r2 = ahr.get(1)
+        self.assertFalse(r1 == r2)
+
+    def test_resubmit_inflight(self):
+        """ensure ValueError on resubmit of inflight task"""
+        v = self.client.load_balanced_view()
+        ar = v.apply_async(time.sleep,1)
+        # give the message a chance to arrive
+        time.sleep(0.2)
+        self.assertRaisesRemote(ValueError, self.client.resubmit, ar.msg_ids)
+        ar.get(2)
+
+    def test_resubmit_badkey(self):
+        """ensure KeyError on resubmit of nonexistant task"""
+        self.assertRaisesRemote(KeyError, self.client.resubmit, ['invalid'])

IPython/parallel/tests/test_lbview.py

@@ -1,120 +1,120 b''
 """test LoadBalancedView objects"""
 # -*- coding: utf-8 -*-
 #-------------------------------------------------------------------------------
 #  Copyright (C) 2011  The IPython Development Team
 #
 #  Distributed under the terms of the BSD License.  The full license is in
 #  the file COPYING, distributed as part of this software.
 #-------------------------------------------------------------------------------
 
 #-------------------------------------------------------------------------------
 # Imports
 #-------------------------------------------------------------------------------
 
 import sys
 import time
 
 import zmq
 
 from IPython import parallel  as pmod
 from IPython.parallel import error
 
 from IPython.parallel.tests import add_engines
 
 from .clienttest import ClusterTestCase, crash, wait, skip_without
 
 def setup():
     add_engines(3)
 
 class TestLoadBalancedView(ClusterTestCase):
 
     def setUp(self):
         ClusterTestCase.setUp(self)
         self.view = self.client.load_balanced_view()
 
     def test_z_crash_task(self):
         """test graceful handling of engine death (balanced)"""
         # self.add_engines(1)
         ar = self.view.apply_async(crash)
-        self.assertRaisesRemote(error.EngineError, ar.get)
+        self.assertRaisesRemote(error.EngineError, ar.get, 10)
         eid = ar.engine_id
         tic = time.time()
         while eid in self.client.ids and time.time()-tic < 5:
             time.sleep(.01)
             self.client.spin()
         self.assertFalse(eid in self.client.ids, "Engine should have died")
 
     def test_map(self):
         def f(x):
             return x**2
         data = range(16)
         r = self.view.map_sync(f, data)
         self.assertEquals(r, map(f, data))
 
     def test_abort(self):
         view = self.view
         ar = self.client[:].apply_async(time.sleep, .5)
         ar2 = view.apply_async(lambda : 2)
         ar3 = view.apply_async(lambda : 3)
         view.abort(ar2)
         view.abort(ar3.msg_ids)
         self.assertRaises(error.TaskAborted, ar2.get)
         self.assertRaises(error.TaskAborted, ar3.get)
 
     def test_retries(self):
         add_engines(3)
         view = self.view
         view.timeout = 1 # prevent hang if this doesn't behave
         def fail():
             assert False
         for r in range(len(self.client)-1):
             with view.temp_flags(retries=r):
                 self.assertRaisesRemote(AssertionError, view.apply_sync, fail)
 
         with view.temp_flags(retries=len(self.client), timeout=0.25):
             self.assertRaisesRemote(error.TaskTimeout, view.apply_sync, fail)
 
     def test_invalid_dependency(self):
         view = self.view
         with view.temp_flags(after='12345'):
             self.assertRaisesRemote(error.InvalidDependency, view.apply_sync, lambda : 1)
 
     def test_impossible_dependency(self):
         if len(self.client) < 2:
             add_engines(2)
         view = self.client.load_balanced_view()
         ar1 = view.apply_async(lambda : 1)
         ar1.get()
         e1 = ar1.engine_id
         e2 = e1
         while e2 == e1:
             ar2 = view.apply_async(lambda : 1)
             ar2.get()
             e2 = ar2.engine_id
 
         with view.temp_flags(follow=[ar1, ar2]):
             self.assertRaisesRemote(error.ImpossibleDependency, view.apply_sync, lambda : 1)
 
 
     def test_follow(self):
         ar = self.view.apply_async(lambda : 1)
         ar.get()
         ars = []
         first_id = ar.engine_id
 
         self.view.follow = ar
         for i in range(5):
             ars.append(self.view.apply_async(lambda : 1))
         self.view.wait(ars)
         for ar in ars:
             self.assertEquals(ar.engine_id, first_id)
 
     def test_after(self):
         view = self.view
         ar = view.apply_async(time.sleep, 0.5)
         with view.temp_flags(after=ar):
             ar2 = view.apply_async(lambda : 1)
 
         ar.wait()
         ar2.wait()
         self.assertTrue(ar2.started > ar.completed)