nilmtools/src/filter.py

#!/usr/bin/python

from __future__ import absolute_import

import nilmdb.client
from nilmdb.client import Client
from nilmdb.client.numpyclient import NumpyClient
from nilmdb.utils.printf import *
from nilmdb.utils.time import (parse_time, timestamp_to_human,
                               timestamp_to_seconds)
from nilmdb.utils.interval import Interval

import nilmtools

import itertools
import time
import sys
import re
import argparse
import numpy as np
import cStringIO

class MissingDestination(Exception):
    def __init__(self, args, src, dest):
        self.parsed_args = args
        self.src = src
        self.dest = dest
        Exception.__init__(self, "destination path " + dest.path + " not found")

class StreamInfo(object):
    def __init__(self, url, info):
        self.url = url
        self.info = info
        try:
            self.path = info[0]
            self.layout = info[1]
            self.layout_type = self.layout.split('_')[0]
            self.layout_count = int(self.layout.split('_')[1])
            self.total_count = self.layout_count + 1
            self.timestamp_min = info[2]
            self.timestamp_max = info[3]
            self.rows = info[4]
            self.seconds = nilmdb.utils.time.timestamp_to_seconds(info[5])
        except IndexError, TypeError:
            pass

    def string(self, interhost):
        """Return stream info as a string.  If interhost is true,
        include the host URL."""
        if interhost:
            return sprintf("[%s] ", self.url) + str(self)
        return str(self)

    def __str__(self):
        """Return stream info as a string."""
        return sprintf("%s (%s), %.2fM rows, %.2f hours",
                       self.path, self.layout, self.rows / 1e6,
                       self.seconds / 3600.0)

def get_stream_info(client, path):
    """Return a StreamInfo object about the given path, or None if it
    doesn't exist"""
    streams = client.stream_list(path, extended = True)
    if len(streams) != 1:
        return None
    return StreamInfo(client.geturl(), streams[0])

class Filter(object):

    def __init__(self, parser_description = None):
        self._parser = None
        self._client_src = None
        self._client_dest = None
        self._using_client = False
        self.src = None
        self.dest = None
        self.start = None
        self.end = None
        self.interhost = False
        self.force_metadata = False
        if parser_description is not None:
            self.setup_parser(parser_description)
            self.parse_args()

    @property
    def client_src(self):
        if self._using_client:
            raise Exception("Filter client is in use; make another")
        return self._client_src

    @property
    def client_dest(self):
        if self._using_client:
            raise Exception("Filter client is in use; make another")
        return self._client_dest

    def setup_parser(self, description = "Filter data", skip_paths = False):
        parser = argparse.ArgumentParser(
            formatter_class = argparse.RawDescriptionHelpFormatter,
            version = nilmtools.__version__,
            description = description)
        group = parser.add_argument_group("General filter arguments")
        group.add_argument("-u", "--url", action="store",
                           default="http://localhost/nilmdb/",
                           help="Server URL (default: %(default)s)")
        group.add_argument("-U", "--dest-url", action="store",
                           help="Destination server URL "
                           "(default: same as source)")
        group.add_argument("-D", "--dry-run", action="store_true",
                           default = False,
                           help="Just print intervals that would be "
                           "processed")
        group.add_argument("--force-metadata", action="store_true",
                           default = False,
                           help="Force metadata changes if the dest "
                           "doesn't match")
        group.add_argument("-s", "--start",
                           metavar="TIME", type=self.arg_time,
                           help="Starting timestamp for intervals "
                           "(free-form, inclusive)")
        group.add_argument("-e", "--end",
                           metavar="TIME", type=self.arg_time,
                           help="Ending timestamp for intervals "
                           "(free-form, noninclusive)")
        if not skip_paths:
            # Individual filter scripts might want to add these arguments
            # themselves, to include multiple sources in a different order
            # (for example).  "srcpath" and "destpath" arguments must exist,
            # though.
            group.add_argument("srcpath", action="store",
                               help="Path of source stream, e.g. /foo/bar")
            group.add_argument("destpath", action="store",
                               help="Path of destination stream, e.g. /foo/bar")
        self._parser = parser
        return parser

    def interval_string(self, interval):
        return sprintf("[ %s -> %s ]",
                       timestamp_to_human(interval.start),
                       timestamp_to_human(interval.end))

    def parse_args(self, argv = None):
        args = self._parser.parse_args(argv)

        if args.dest_url is None:
            args.dest_url = args.url
        if args.url != args.dest_url:
            self.interhost = True

        self._client_src = Client(args.url)
        self._client_dest = Client(args.dest_url)

        if (not self.interhost) and (args.srcpath == args.destpath):
            self._parser.error("source and destination path must be different")

        # Open and print info about the streams
        self.src = get_stream_info(self._client_src, args.srcpath)
        if not self.src:
            self._parser.error("source path " + args.srcpath + " not found")

        self.dest = get_stream_info(self._client_dest, args.destpath)
        if not self.dest:
            raise MissingDestination(args, self.src,
                                     StreamInfo(args.dest_url, [args.destpath]))

        print "Source:", self.src.string(self.interhost)
        print "  Dest:", self.dest.string(self.interhost)

        if args.dry_run:
            for interval in self.intervals():
                print self.interval_string(interval)
            raise SystemExit(0)

        self.force_metadata = args.force_metadata

        self.start = args.start
        self.end = args.end

        return args

    def _optimize_int(self, it):
        """Join and yield adjacent intervals from the iterator 'it'"""
        saved_int = None
        for interval in it:
            if saved_int is not None:
                if saved_int.end == interval.start:
                    interval.start = saved_int.start
                else:
                    yield saved_int
            saved_int = interval
        if saved_int is not None:
            yield saved_int

    def intervals(self):
        """Generate all the intervals that this filter should process"""
        self._using_client = True

        if self.interhost:
            # Do the difference ourselves
            s_intervals = ( Interval(start, end)
                            for (start, end) in
                            self._client_src.stream_intervals(
                                self.src.path,
                                start = self.start, end = self.end) )
            d_intervals = ( Interval(start, end)
                            for (start, end) in
                            self._client_dest.stream_intervals(
                                self.dest.path,
                                start = self.start, end = self.end) )
            intervals = nilmdb.utils.interval.set_difference(s_intervals,
                                                             d_intervals)
        else:
            # Let the server do the difference for us
            intervals = ( Interval(start, end)
                          for (start, end) in
                          self._client_src.stream_intervals(
                              self.src.path, diffpath = self.dest.path,
                              start = self.start, end = self.end) )
        # Optimize intervals: join intervals that are adjacent
        for interval in self._optimize_int(intervals):
            yield interval
        self._using_client = False

    # Misc helpers
    def arg_time(self, toparse):
        """Parse a time string argument"""
        try:
            return nilmdb.utils.time.parse_time(toparse)
        except ValueError as e:
            raise argparse.ArgumentTypeError(sprintf("%s \"%s\"",
                                                     str(e), toparse))

    def check_dest_metadata(self, data):
        """See if the metadata jives, and complain if it doesn't.  If
        there's no conflict, update the metadata to match 'data'."""
        metadata = self._client_dest.stream_get_metadata(self.dest.path)
        if not self.force_metadata:
            for key in data:
                wanted = str(data[key])
                val = metadata.get(key, wanted)
                if val != wanted and self.dest.rows > 0:
                    m =  "Metadata in destination stream:\n"
                    m += "  %s = %s\n" % (key, val)
                    m += "doesn't match desired data:\n"
                    m += "  %s = %s\n" % (key, wanted)
                    m += "Refusing to change it.  To prevent this error, "
                    m += "change or delete the metadata with nilmtool,\n"
                    m += "remove existing data from the stream, or "
                    m += "retry with --force-metadata."
                    raise Exception(m)
        # All good -- write the metadata in case it's not already there
        self._client_dest.stream_update_metadata(self.dest.path, data)

    # The main filter processing method.
    def process_numpy(self, function, args = None, rows = 100000):
        """For all intervals that exist in self.src but don't exist in
        self.dest, call 'function' with a Numpy array corresponding to
        the data.  The data is converted to a Numpy array in chunks of
        'rows' rows at a time.

        'function' should be defined as:
           def function(data, interval, args, insert_func, final)

        'data': array of data to process -- may be empty

        'interval': overall interval we're processing (but not necessarily
        the interval of this particular chunk of data)

        'args': opaque arguments passed to process_numpy

        'insert_func': function to call in order to insert array of data.
        Should be passed a 2-dimensional array of data to insert.
        Data timestamps must be within the provided interval.

        'final': True if this is the last bit of data for this
        contiguous interval, False otherwise.

        Return value of 'function' is the number of data rows processed.
        Unprocessed data will be provided again in a subsequent call
        (unless 'final' is True).

        If unprocessed data remains after 'final' is True, the interval
        being inserted will be ended at the timestamp of the first
        unprocessed data point.
        """
        if args is None:
            args = []
        extractor = NumpyClient(self.src.url).stream_extract_numpy
        inserter = NumpyClient(self.dest.url).stream_insert_numpy_context

        for interval in self.intervals():
            print "Processing", self.interval_string(interval)
            with inserter(self.dest.path,
                          interval.start, interval.end) as insert_ctx:
                insert_function = insert_ctx.insert
                old_array = np.array([])
                for new_array in extractor(self.src.path,
                                           interval.start, interval.end,
                                           layout = self.src.layout,
                                           maxrows = rows):
                    # If we still had old data left, combine it
                    if old_array.shape[0] != 0:
                        array = np.vstack((old_array, new_array))
                    else:
                        array = new_array

                    # Pass it to the process function
                    processed = function(array, interval, args,
                                         insert_function, False)

                    # Send any pending data
                    insert_ctx.send()

                    # Save the unprocessed parts
                    if processed >= 0:
                        old_array = array[processed:]
                    else:
                        raise Exception(
                            sprintf("%s return value %s must be >= 0",
                                    str(function), str(processed)))

                    # Warn if there's too much data remaining
                    if old_array.shape[0] > 3 * rows:
                        printf("warning: %d unprocessed rows in buffer\n",
                               old_array.shape[0])

                # Last call for this contiguous interval
                if old_array.shape[0] != 0:
                    processed = function(old_array, interval, args,
                                         insert_function, True)
                    if processed != old_array.shape[0]:
                        # Truncate the interval we're inserting at the first
                        # unprocessed data point.  This ensures that
                        # we'll not miss any data when we run again later.
                        insert_ctx.update_end(old_array[processed][0])

def main(argv = None):
    # This is just a dummy function; actual filters can use the other
    # functions to prepare stuff, and then do something with the data.
    f = Filter()
    parser = f.setup_parser()
    args = f.parse_args(argv)
    for i in f.intervals():
        print "Generic filter: need to handle", f.interval_string(i)

if __name__ == "__main__":
    main()