Bump database version to 3, reject old version 2 due to timestamp changes

Merge branch 'rocket-cleanup'
Fill out test coverage
2013-03-15 18:37:38 -04:00 · 2013-03-15 18:08:36 -04:00 · 2013-03-15 18:08:21 -04:00 · 2013-03-15 18:08:21 -04:00 · 2013-03-15 18:08:21 -04:00 · 2013-03-15 18:08:21 -04:00
35 changed files with 15607 additions and 15467 deletions
--- a/docs/design.md
+++ b/docs/design.md
@@ -140,7 +140,7 @@ Speed

 - Next slowdown target is nilmdb.layout.Parser.parse().
    - Rewrote parsers using cython and sscanf
-    - Stats (rev 10831), with _add_interval disabled
+    - Stats (rev 10831), with `_add_interval` disabled

        layout.pyx.Parser.parse:128        6303 sec, 262k calls
         layout.pyx.parse:63               13913 sec, 5.1g calls
@@ -328,3 +328,51 @@ Current places where we use lines:
  - Finished.  Just a single insert() that takes any length string and
    does very little processing until it's time to send it to the
 	server.
+
+Timestamps
+----------
+
+Timestamps are currently double-precision floats (64 bit).  Since the
+mantissa is 53-bit, this can only represent about 15-17 significant
+figures, and microsecond Unix timestamps like 1222333444.000111 are
+already 16 significant figures.  Rounding is therefore an issue;
+it's hard to sure that converting from ASCII, then back to ASCII,
+will always give the same result.
+
+Also, if the client provides a floating point value like 1.9999999999,
+we need to be careful that we don't store it as 1.9999999999 but later
+print it as 2.000000, because then round-trips change the data.
+
+Possible solutions:
+
+- When the client provides a floating point value to the server,
+  always round to the 6th decimal digit before verifying & storing.
+  Good for compatibility and simplicity.  But still might have rounding
+  issues, and clients will also need to round when doing their own
+  verification.  Having every piece of code need to know which digit
+  to round at is not ideal.
+
+- Always store int64 timestamps on the server, representing
+  microseconds since epoch.  int64 timestamps are used in all HTTP
+  parameters, in insert/extract ASCII strings, client API, commandline
+  raw timestamps, etc.  Pretty big change.
+
+  This is what we'll go with...
+
+  - Client programs that interpret the timestamps as doubles instead
+    of ints will remain accurate until 2^53 microseconds, or year
+    2255.
+
+  - On insert, maybe it's OK to send floating point microsecond values
+    (1234567890123456.0), just to cope with clients that want to print
+    everything as a double.  Server could try parsing as int64, and if
+    that fails, parse as double and truncate to int64.  However, this
+    wouldn't catch imprecise inputs like "1.23456789012e+15".  But
+    maybe that can just be ignored; it's likely to cause a
+    non-monotonic error at the client.
+
+  - Timestamps like 1234567890.123456 never show up anywhere, except
+    for interfacing to datetime_tz etc.  Command line "raw timestamps"
+    are always printed as int64 values, and a new format
+    "@1234567890123456" is added to the parser for specifying them
+    exactly.
--- a/extras/nilmtool-bash-completion.sh
+++ b/extras/nilmtool-bash-completion.sh
@@ -0,0 +1,20 @@
+# To enable bash completion:
+#
+# 1. Ensure python-argcomplete is installed:
+#       pip install argcomplete
+# 2. Source this file:
+#       . nilmtool-bash-completion.sh
+
+_nilmtool_argcomplete() {
+    local IFS=$(printf "\013")
+    COMPREPLY=( $(IFS="$IFS" \
+                  COMP_LINE="$COMP_LINE" \
+	          COMP_WORDBREAKS="$COMP_WORDBREAKS" \
+                  COMP_POINT="$COMP_POINT" \
+                  _ARGCOMPLETE=1 \
+                  "$1" 8>&1 9>&2 1>/dev/null 2>/dev/null) )
+    if [[ $? != 0 ]]; then
+        unset COMPREPLY
+    fi
+}
+complete -o nospace -o default -F _nilmtool_argcomplete nilmtool
--- a/nilmdb/client/client.py
+++ b/nilmdb/client/client.py
@@ -10,11 +10,11 @@ import time
 import simplejson as json
 import contextlib

-from nilmdb.utils.time import float_time_to_string
+from nilmdb.utils.time import timestamp_to_string, string_to_timestamp

 def extract_timestamp(line):
    """Extract just the timestamp from a line of data text"""
-    return float(line.split()[0])
+    return string_to_timestamp(line.split()[0])

 class Client(object):
    """Main client interface to the Nilm database."""
@@ -101,15 +101,21 @@ class Client(object):
        params = { "path": path }
        return self.http.post("stream/destroy", params)

+    def stream_rename(self, oldpath, newpath):
+        """Rename a stream."""
+        params = { "oldpath": oldpath,
+                   "newpath": newpath }
+        return self.http.post("stream/rename", params)
+
    def stream_remove(self, path, start = None, end = None):
        """Remove data from the specified time range"""
        params = {
            "path": path
        }
        if start is not None:
-            params["start"] = float_time_to_string(start)
+            params["start"] = timestamp_to_string(start)
        if end is not None:
-            params["end"] = float_time_to_string(end)
+            params["end"] = timestamp_to_string(end)
        return self.http.post("stream/remove", params)

    @contextlib.contextmanager
@@ -160,9 +166,9 @@ class Client(object):
        if diffpath is not None:
            params["diffpath"] = diffpath
        if start is not None:
-            params["start"] = float_time_to_string(start)
+            params["start"] = timestamp_to_string(start)
        if end is not None:
-            params["end"] = float_time_to_string(end)
+            params["end"] = timestamp_to_string(end)
        return self.http.get_gen("stream/intervals", params)

    def stream_extract(self, path, start = None, end = None, count = False):
@@ -178,9 +184,9 @@ class Client(object):
            "path": path,
        }
        if start is not None:
-            params["start"] = float_time_to_string(start)
+            params["start"] = timestamp_to_string(start)
        if end is not None:
-            params["end"] = float_time_to_string(end)
+            params["end"] = timestamp_to_string(end)
        if count:
            params["count"] = 1
        return self.http.get_gen("stream/extract", params)
@@ -232,9 +238,6 @@ class StreamInserter(object):
    _max_data = 2 * 1024 * 1024
    _max_data_after_send = 64 * 1024

-    # Delta to add to the final timestamp, if "end" wasn't given
-    _end_epsilon = 1e-6
-
    def __init__(self, http, path, start = None, end = None):
        """'http' is the httpclient object.  'path' is the database
        path to insert to.  'start' and 'end' are used for the first
@@ -358,7 +361,7 @@ class StreamInserter(object):
                if end_ts is None:
                    (spos, epos) = self._get_last_noncomment(block)
                    end_ts = extract_timestamp(block[spos:epos])
-                    end_ts += self._end_epsilon
+                    end_ts += nilmdb.utils.time.epsilon
            except (ValueError, IndexError):
                pass # no timestamp is OK, if we have no data
            self._block_data = []
@@ -405,8 +408,8 @@ class StreamInserter(object):

        # Send it
        params = { "path": self._path,
-                   "start": float_time_to_string(start_ts),
-                   "end": float_time_to_string(end_ts) }
+                   "start": timestamp_to_string(start_ts),
+                   "end": timestamp_to_string(end_ts) }
        self.last_response = self._http.put("stream/insert", block, params)

        return
--- a/nilmdb/cmdline/cmdline.py
+++ b/nilmdb/cmdline/cmdline.py
@@ -11,11 +11,16 @@ import os
 import argparse
 from argparse import ArgumentDefaultsHelpFormatter as def_form

+try: # pragma: no cover
+    import argcomplete
+except ImportError: # pragma: no cover
+    argcomplete = None
+
 # Valid subcommands.  Defined in separate files just to break
 # things up -- they're still called with Cmdline as self.
 subcommands = [ "help", "info", "create", "list", "metadata",
                "insert", "extract", "remove", "destroy",
-                "intervals" ]
+                "intervals", "rename" ]

 # Import the subcommand modules
 subcmd_mods = {}
@@ -27,6 +32,50 @@ class JimArgumentParser(argparse.ArgumentParser):
        self.print_usage(sys.stderr)
        self.exit(2, sprintf("error: %s\n", message))

+class Complete(object): # pragma: no cover
+    # Completion helpers, for using argcomplete (see
+    # extras/nilmtool-bash-completion.sh)
+    def escape(self, s):
+        quote_chars = [ "\\", "\"", "'", " " ]
+        for char in quote_chars:
+            s = s.replace(char, "\\" + char)
+        return s
+
+    def none(self, prefix, parsed_args, **kwargs):
+        return []
+    rate = none
+    time = none
+    url = none
+
+    def path(self, prefix, parsed_args, **kwargs):
+        client = nilmdb.client.Client(parsed_args.url)
+        return ( self.escape(s[0])
+                 for s in client.stream_list()
+                 if s[0].startswith(prefix) )
+
+    def layout(self, prefix, parsed_args, **kwargs):
+        types = [ "int8", "int16", "int32", "int64",
+                  "uint8", "uint16", "uint32", "uint64",
+                  "float32", "float64" ]
+        layouts = []
+        for i in range(1,10):
+            layouts.extend([(t + "_" + str(i)) for t in types])
+        return ( l for l in layouts if l.startswith(prefix) )
+
+    def meta_key(self, prefix, parsed_args, **kwargs):
+        return (kv.split('=')[0] for kv
+                in self.meta_keyval(prefix, parsed_args, **kwargs))
+
+    def meta_keyval(self, prefix, parsed_args, **kwargs):
+        client = nilmdb.client.Client(parsed_args.url)
+        path = parsed_args.path
+        if not path:
+            return []
+        return ( self.escape(k + '=' + v)
+                 for (k,v) in client.stream_get_metadata(path).iteritems()
+                 if k.startswith(prefix) )
+
+
 class Cmdline(object):

    def __init__(self, argv = None):
@@ -34,15 +83,17 @@ class Cmdline(object):
        self.client = None
        self.def_url = os.environ.get("NILMDB_URL", "http://localhost:12380")
        self.subcmd = {}
+        self.complete = Complete()

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

+    # Set up the parser
    def parser_setup(self):
        self.parser = JimArgumentParser(add_help = False,
                                        formatter_class = def_form)
@@ -56,7 +107,8 @@ class Cmdline(object):
        group = self.parser.add_argument_group("Server")
        group.add_argument("-u", "--url", action="store",
                           default=self.def_url,
-                           help="NilmDB server URL (default: %(default)s)")
+                           help="NilmDB server URL (default: %(default)s)"
+                           ).completer = self.complete.url

        sub = self.parser.add_subparsers(
            title="Commands", dest="command",
@@ -80,6 +132,8 @@ class Cmdline(object):

        # Run parser
        self.parser_setup()
+        if argcomplete: # pragma: no cover
+            argcomplete.autocomplete(self.parser)
        self.args = self.parser.parse_args(self.argv)

        # Run arg verify handler if there is one
--- a/nilmdb/cmdline/create.py
+++ b/nilmdb/cmdline/create.py
@@ -22,9 +22,11 @@ Layout types are of the format: type_count
    cmd.set_defaults(handler = cmd_create)
    group = cmd.add_argument_group("Required arguments")
    group.add_argument("path",
-                       help="Path (in database) of new stream, e.g. /foo/bar")
+                       help="Path (in database) of new stream, e.g. /foo/bar",
+                       ).completer = self.complete.path
    group.add_argument("layout",
-                       help="Layout type for new stream, e.g. float32_8")
+                       help="Layout type for new stream, e.g. float32_8",
+                       ).completer = self.complete.layout
    return cmd

 def cmd_create(self):
--- a/nilmdb/cmdline/destroy.py
+++ b/nilmdb/cmdline/destroy.py
@@ -14,7 +14,8 @@ def setup(self, sub):
    cmd.set_defaults(handler = cmd_destroy)
    group = cmd.add_argument_group("Required arguments")
    group.add_argument("path",
-                       help="Path of the stream to delete, e.g. /foo/bar")
+                       help="Path of the stream to delete, e.g. /foo/bar",
+                       ).completer = self.complete.path
    return cmd

 def cmd_destroy(self):
--- a/nilmdb/cmdline/extract.py
+++ b/nilmdb/cmdline/extract.py
@@ -12,13 +12,16 @@ def setup(self, sub):

    group = cmd.add_argument_group("Data selection")
    group.add_argument("path",
-                       help="Path of stream, e.g. /foo/bar")
+                       help="Path of stream, e.g. /foo/bar",
+                       ).completer = self.complete.path
    group.add_argument("-s", "--start", required=True,
                       metavar="TIME", type=self.arg_time,
-                       help="Starting timestamp (free-form, inclusive)")
+                       help="Starting timestamp (free-form, inclusive)",
+                       ).completer = self.complete.time
    group.add_argument("-e", "--end", required=True,
                       metavar="TIME", type=self.arg_time,
-                       help="Ending timestamp (free-form, noninclusive)")
+                       help="Ending timestamp (free-form, noninclusive)",
+                       ).completer = self.complete.time

    group = cmd.add_argument_group("Output format")
    group.add_argument("-b", "--bare", action="store_true",
@@ -44,9 +47,9 @@ def cmd_extract(self):
    layout = streams[0][1]

    if self.args.timestamp_raw:
-        time_string = nilmdb.utils.time.float_time_to_string
+        time_string = nilmdb.utils.time.timestamp_to_string
    else:
-        time_string = nilmdb.utils.time.format_time
+        time_string = nilmdb.utils.time.timestamp_to_human

    if self.args.annotate:
        printf("# path: %s\n", self.args.path)
--- a/nilmdb/cmdline/insert.py
+++ b/nilmdb/cmdline/insert.py
@@ -25,7 +25,8 @@ def setup(self, sub):
    group.add_argument("-t", "--timestamp", action="store_true",
                       help="Add timestamps to each line")
    group.add_argument("-r", "--rate", type=float,
-                       help="Data rate, in Hz")
+                       help="Data rate, in Hz",
+                       ).completer = self.complete.rate

    group = cmd.add_argument_group("Start time",
                                   description="""
@@ -39,7 +40,8 @@ def setup(self, sub):
    exc = group.add_mutually_exclusive_group()
    exc.add_argument("-s", "--start",
                     metavar="TIME", type=self.arg_time,
-                     help="Starting timestamp (free-form)")
+                     help="Starting timestamp (free-form)",
+                     ).completer = self.complete.time
    exc.add_argument("-f", "--filename", action="store_true",
                     help="Use filename to determine start time")

@@ -52,11 +54,13 @@ def setup(self, sub):
                                   timezone.""")
    group.add_argument("-e", "--end",
                       metavar="TIME", type=self.arg_time,
-                       help="Ending timestamp (free-form)")
+                       help="Ending timestamp (free-form)",
+                       ).completer = self.complete.time

    group = cmd.add_argument_group("Required parameters")
    group.add_argument("path",
-                       help="Path of stream, e.g. /foo/bar")
+                       help="Path of stream, e.g. /foo/bar",
+                       ).completer = self.complete.path
    group.add_argument("file", nargs = '?', default='-',
                       help="File to insert (default: - (stdin))")
    return cmd
@@ -92,7 +96,7 @@ def cmd_insert(self):

        if arg.start is None:
            try:
-                arg.start = nilmdb.utils.time.parse_time(filename).totimestamp()
+                arg.start = nilmdb.utils.time.parse_time(filename)
            except ValueError:
                self.die("error extracting start time from filename '%s'",
                         filename)
@@ -106,10 +110,10 @@ def cmd_insert(self):
        if not arg.quiet:
            printf(" Input file: %s\n", filename)
            printf(" Start time: %s\n",
-                   nilmdb.utils.time.format_time(arg.start))
+                   nilmdb.utils.time.timestamp_to_human(arg.start))
            if arg.end:
                printf("   End time: %s\n",
-                       nilmdb.utils.time.format_time(arg.end))
+                       nilmdb.utils.time.timestamp_to_human(arg.end))
            if arg.timestamp:
                printf("Timestamper: %s\n", str(data))

--- a/nilmdb/cmdline/intervals.py
+++ b/nilmdb/cmdline/intervals.py
@@ -21,19 +21,23 @@ def setup(self, sub):

    group = cmd.add_argument_group("Stream selection")
    group.add_argument("path", metavar="PATH",
-                       help="List intervals for this path")
+                       help="List intervals for this path",
+                       ).completer = self.complete.path
    group.add_argument("-d", "--diff", metavar="PATH",
-                       help="Subtract intervals from this path")
+                       help="Subtract intervals from this path",
+                       ).completer = self.complete.path

    group = cmd.add_argument_group("Interval details")
    group.add_argument("-s", "--start",
                       metavar="TIME", type=self.arg_time,
                       help="Starting timestamp for intervals "
-                       "(free-form, inclusive)")
+                       "(free-form, inclusive)",
+                       ).completer = self.complete.time
    group.add_argument("-e", "--end",
                       metavar="TIME", type=self.arg_time,
                       help="Ending timestamp for intervals "
-                       "(free-form, noninclusive)")
+                       "(free-form, noninclusive)",
+                       ).completer = self.complete.time

    group = cmd.add_argument_group("Misc options")
    group.add_argument("-T", "--timestamp-raw", action="store_true",
@@ -49,9 +53,9 @@ def cmd_intervals_verify(self):
 def cmd_intervals(self):
    """List intervals in a stream"""
    if self.args.timestamp_raw:
-        time_string = nilmdb.utils.time.float_time_to_string
+        time_string = nilmdb.utils.time.timestamp_to_string
    else:
-        time_string = nilmdb.utils.time.format_time
+        time_string = nilmdb.utils.time.timestamp_to_human

    try:
           for (start, end) in self.client.stream_intervals(
--- a/nilmdb/cmdline/list.py
+++ b/nilmdb/cmdline/list.py
@@ -18,11 +18,14 @@ def setup(self, sub):

    group = cmd.add_argument_group("Stream filtering")
    group.add_argument("-p", "--path", metavar="PATH", default="*",
-                       help="Match only this path (-p can be omitted)")
+                       help="Match only this path (-p can be omitted)",
+                       ).completer = self.complete.path
    group.add_argument("path_positional", default="*",
-                       nargs="?", help=argparse.SUPPRESS)
+                       nargs="?", help=argparse.SUPPRESS,
+                       ).completer = self.complete.path
    group.add_argument("-l", "--layout", default="*",
-                       help="Match only this stream layout")
+                       help="Match only this stream layout",
+                       ).completer = self.complete.layout

    group = cmd.add_argument_group("Interval info")
    group.add_argument("-E", "--ext", action="store_true",
@@ -35,11 +38,13 @@ def setup(self, sub):
    group.add_argument("-s", "--start",
                       metavar="TIME", type=self.arg_time,
                       help="Starting timestamp for intervals "
-                       "(free-form, inclusive)")
+                       "(free-form, inclusive)",
+                       ).completer = self.complete.time
    group.add_argument("-e", "--end",
                       metavar="TIME", type=self.arg_time,
                       help="Ending timestamp for intervals "
-                       "(free-form, noninclusive)")
+                       "(free-form, noninclusive)",
+                       ).completer = self.complete.time

    group = cmd.add_argument_group("Misc options")
    group.add_argument("-T", "--timestamp-raw", action="store_true",
@@ -71,12 +76,12 @@ def cmd_list(self):
    streams = self.client.stream_list(extended = True)

    if self.args.timestamp_raw:
-        time_string = nilmdb.utils.time.float_time_to_string
+        time_string = nilmdb.utils.time.timestamp_to_string
    else:
-        time_string = nilmdb.utils.time.format_time
+        time_string = nilmdb.utils.time.timestamp_to_human

    for stream in streams:
-        (path, layout, int_min, int_max, rows, seconds) = stream[:6]
+        (path, layout, int_min, int_max, rows, time) = stream[:6]
        if not (fnmatch.fnmatch(path, self.args.path) and
                fnmatch.fnmatch(layout, self.args.layout)):
            continue
@@ -90,7 +95,8 @@ def cmd_list(self):
                printf("  interval extents: %s -> %s\n",
                       time_string(int_min), time_string(int_max))
            printf("        total data: %d rows, %.6f seconds\n",
-                   rows or 0, seconds or 0);
+                   rows or 0,
+                   nilmdb.utils.time.timestamp_to_seconds(time or 0))

        if self.args.detail:
            printed = False
--- a/nilmdb/cmdline/metadata.py
+++ b/nilmdb/cmdline/metadata.py
@@ -14,18 +14,22 @@ def setup(self, sub):

    group = cmd.add_argument_group("Required arguments")
    group.add_argument("path",
-                       help="Path of stream, e.g. /foo/bar")
+                       help="Path of stream, e.g. /foo/bar",
+                       ).completer = self.complete.path

    group = cmd.add_argument_group("Actions")
    exc = group.add_mutually_exclusive_group()
    exc.add_argument("-g", "--get", nargs="*", metavar="key",
-                     help="Get metadata for specified keys (default all)")
+                     help="Get metadata for specified keys (default all)",
+                     ).completer = self.complete.meta_key
    exc.add_argument("-s", "--set", nargs="+", metavar="key=value",
                     help="Replace all metadata with provided "
-                     "key=value pairs")
+                     "key=value pairs",
+                     ).completer = self.complete.meta_keyval
    exc.add_argument("-u", "--update", nargs="+", metavar="key=value",
                     help="Update metadata using provided "
-                     "key=value pairs")
+                     "key=value pairs",
+                     ).completer = self.complete.meta_keyval
    return cmd

 def cmd_metadata(self):
--- a/nilmdb/cmdline/remove.py
+++ b/nilmdb/cmdline/remove.py
@@ -11,13 +11,16 @@ def setup(self, sub):

    group = cmd.add_argument_group("Data selection")
    group.add_argument("path",
-                       help="Path of stream, e.g. /foo/bar")
+                       help="Path of stream, e.g. /foo/bar",
+                       ).completer = self.complete.path
    group.add_argument("-s", "--start", required=True,
                       metavar="TIME", type=self.arg_time,
-                       help="Starting timestamp (free-form, inclusive)")
+                       help="Starting timestamp (free-form, inclusive)",
+                       ).completer = self.complete.time
    group.add_argument("-e", "--end", required=True,
                       metavar="TIME", type=self.arg_time,
-                       help="Ending timestamp (free-form, noninclusive)")
+                       help="Ending timestamp (free-form, noninclusive)",
+                       ).completer = self.complete.time

    group = cmd.add_argument_group("Output format")
    group.add_argument("-c", "--count", action="store_true",
--- a/nilmdb/cmdline/rename.py
+++ b/nilmdb/cmdline/rename.py
@@ -0,0 +1,31 @@
+from nilmdb.utils.printf import *
+import nilmdb.client
+
+from argparse import ArgumentDefaultsHelpFormatter as def_form
+
+def setup(self, sub):
+    cmd = sub.add_parser("rename", help="Rename a stream",
+                         formatter_class = def_form,
+                         description="""
+                         Rename a stream.
+
+                         Only the stream's path is renamed; no
+                         metadata is changed.
+                         """)
+    cmd.set_defaults(handler = cmd_rename)
+    group = cmd.add_argument_group("Required arguments")
+    group.add_argument("oldpath",
+                       help="Old path, e.g. /foo/old",
+                       ).completer = self.complete.path
+    group.add_argument("newpath",
+                       help="New path, e.g. /foo/bar/new",
+                       ).completer = self.complete.path
+
+    return cmd
+
+def cmd_rename(self):
+    """Rename a stream"""
+    try:
+        self.client.stream_rename(self.args.oldpath, self.args.newpath)
+    except nilmdb.client.ClientError as e:
+        self.die("error renaming stream: %s", str(e))
--- a/nilmdb/server/bulkdata.py
+++ b/nilmdb/server/bulkdata.py
@@ -5,15 +5,15 @@
 from __future__ import absolute_import
 from __future__ import division
 from nilmdb.utils.printf import *
-from nilmdb.utils.time import float_time_to_string as ftts
+from nilmdb.utils.time import timestamp_to_string as timestamp_to_string
 import nilmdb.utils

 import os
 import cPickle as pickle
 import re
 import sys
+import tempfile

-#from . import pyrocket as rocket
 from . import rocket

 # Up to 256 open file descriptors at any given time.
@@ -56,6 +56,59 @@ class BulkData(object):
            return path.encode('utf-8')
        return path

+    def _create_check_ospath(self, ospath):
+        if ospath[-1] == '/':
+            raise ValueError("invalid path; should not end with a /")
+        if Table.exists(ospath):
+            raise ValueError("stream already exists at this path")
+        if os.path.isdir(ospath):
+            raise ValueError("subdirs of this path already exist")
+
+    def _create_parents(self, unicodepath):
+        """Verify the path name, and create parent directories if they
+        don't exist.  Returns a list of elements that got created."""
+        path = self._encode_filename(unicodepath)
+
+        if path[0] != '/':
+            raise ValueError("paths must start with /")
+        [ group, node ] = path.rsplit("/", 1)
+        if group == '':
+            raise ValueError("invalid path; path must contain at least one "
+                             "folder")
+        if node == '':
+            raise ValueError("invalid path; should not end with a /")
+        if not Table.valid_path(path):
+            raise ValueError("path name is invalid or contains reserved words")
+
+        # Create the table's base dir.  Note that we make a
+        # distinction here between NilmDB paths (always Unix style,
+        # split apart manually) and OS paths (built up with
+        # os.path.join)
+
+        # Make directories leading up to this one
+        elements = path.lstrip('/').split('/')
+        made_dirs = []
+        try:
+            # Make parent elements
+            for i in range(len(elements)):
+                ospath = os.path.join(self.root, *elements[0:i])
+                if Table.exists(ospath):
+                    raise ValueError("path is subdir of existing node")
+                if not os.path.isdir(ospath):
+                    os.mkdir(ospath)
+                    made_dirs.append(ospath)
+        except Exception as e:
+            # Try to remove paths that we created; ignore errors
+            exc_info = sys.exc_info()
+            for ospath in reversed(made_dirs): # pragma: no cover (hard to hit)
+                try:
+                    os.rmdir(ospath)
+                except OSError:
+                    pass
+            raise exc_info[1], None, exc_info[2]
+
+        return elements
+
    def create(self, unicodepath, layout_name):
        """
        unicodepath: path to the data (e.g. u'/newton/prep').
@@ -67,32 +120,11 @@ class BulkData(object):

        layout_name: string for nilmdb.layout.get_named(), e.g. 'float32_8'
        """
-        path = self._encode_filename(unicodepath)
-
-        if path[0] != '/':
-            raise ValueError("paths must start with /")
-        [ group, node ] = path.rsplit("/", 1)
-        if group == '':
-            raise ValueError("invalid path; path must contain at least one "
-                             "folder")
-
-        # Create the table.  Note that we make a distinction here
-        # between NilmDB paths (always Unix style, split apart
-        # manually) and OS paths (built up with os.path.join)
-
-        # Make directories leading up to this one
-        elements = path.lstrip('/').split('/')
-        for i in range(len(elements)):
-            ospath = os.path.join(self.root, *elements[0:i])
-            if Table.exists(ospath):
-                raise ValueError("path is subdir of existing node")
-            if not os.path.isdir(ospath):
-                os.mkdir(ospath)
+        elements = self._create_parents(unicodepath)

        # Make the final dir
        ospath = os.path.join(self.root, *elements)
-        if os.path.isdir(ospath):
-            raise ValueError("subdirs of this path already exist")
+        self._create_check_ospath(ospath)
        os.mkdir(ospath)

        try:
@@ -113,6 +145,55 @@ class BulkData(object):
        # Success
        return

+    def _remove_leaves(self, unicodepath):
+        """Remove empty directories starting at the leaves of unicodepath"""
+        path = self._encode_filename(unicodepath)
+        elements = path.lstrip('/').split('/')
+        for i in reversed(range(len(elements))):
+            ospath = os.path.join(self.root, *elements[0:i+1])
+            try:
+                os.rmdir(ospath)
+            except OSError:
+                pass
+
+    def rename(self, oldunicodepath, newunicodepath):
+        """Move entire tree from 'oldunicodepath' to
+        'newunicodepath'"""
+        oldpath = self._encode_filename(oldunicodepath)
+        newpath = self._encode_filename(newunicodepath)
+
+        # Get OS paths
+        oldelements = oldpath.lstrip('/').split('/')
+        oldospath = os.path.join(self.root, *oldelements)
+        newelements = newpath.lstrip('/').split('/')
+        newospath = os.path.join(self.root, *newelements)
+
+        # Basic checks
+        if oldospath == newospath:
+            raise ValueError("old and new paths are the same")
+        self._create_check_ospath(newospath)
+
+        # Move the table to a temporary location
+        tmpdir = tempfile.mkdtemp(prefix = "rename-", dir = self.root)
+        tmppath = os.path.join(tmpdir, "table")
+        os.rename(oldospath, tmppath)
+
+        try:
+            # Create parent dirs for new location
+            self._create_parents(newunicodepath)
+
+            # Move table into new location
+            os.rename(tmppath, newospath)
+        except Exception:
+            # On failure, move the table back to original path
+            os.rename(tmppath, oldospath)
+            os.rmdir(tmpdir)
+            raise
+
+        # Prune old dirs
+        self._remove_leaves(oldunicodepath)
+        os.rmdir(tmpdir)
+
    def destroy(self, unicodepath):
        """Fully remove all data at a particular path.  No way to undo
        it!  The group/path structure is removed, too."""
@@ -134,13 +215,8 @@ class BulkData(object):
            for name in dirs:
                os.rmdir(os.path.join(root, name))

-        # Remove empty parent directories
-        for i in reversed(range(len(elements))):
-            ospath = os.path.join(self.root, *elements[0:i+1])
-            try:
-                os.rmdir(ospath)
-            except OSError:
-                break
+        # Remove leftover empty directories
+        self._remove_leaves(unicodepath)

    # Cache open tables
    @nilmdb.utils.lru_cache(size = table_cache_size,
@@ -159,6 +235,11 @@ class Table(object):
    # See design.md for design details

    # Class methods, to help keep format details in this class.
+    @classmethod
+    def valid_path(cls, root):
+        """Return True if a root path is a valid name"""
+        return "_format" not in root.split("/")
+
    @classmethod
    def exists(cls, root):
        """Return True if a table appears to exist at this OS path"""
@@ -178,7 +259,7 @@ class Table(object):
        fmt = { "rows_per_file": rows_per_file,
                "files_per_dir": files_per_dir,
                "layout": layout,
-                "version": 2 }
+                "version": 3 }
        with open(os.path.join(root, "_format"), "wb") as f:
            pickle.dump(fmt, f, 2)

@@ -191,21 +272,11 @@ class Table(object):
        with open(os.path.join(self.root, "_format"), "rb") as f:
            fmt = pickle.load(f)

-        if fmt["version"] == 1: # pragma: no cover
-            # We can handle this old version by converting from
-            # struct_fmt back to layout name.
-            compat = { "<dHHHHHH": "uint16_6",
-                       "<dHHHHHHHHH": "uint16_9",
-                       "<dffffffff": "float32_8" }
-            if fmt["struct_fmt"] in compat:
-                fmt["version"] = 2
-                fmt["layout"] = compat[fmt["struct_fmt"]]
-            else:
-                raise NotImplementedError("old version 1 data with format "
-                                          + fmt["struct_fmt"] + " is no good")
-        elif fmt["version"] != 2: # pragma: no cover (just future proofing)
-            raise NotImplementedError("version " + str(fmt["version"]) +
-                                      " bulk data store not supported")
+        if fmt["version"] != 3: # pragma: no cover
+            # Old versions used floating point timestamps, which aren't
+            # valid anymore.
+            raise NotImplementedError("old version " + str(fmt["version"]) +
+                                      " bulk data store is not supported")

        self.rows_per_file = fmt["rows_per_file"]
        self.files_per_dir = fmt["files_per_dir"]
@@ -318,27 +389,6 @@ class Table(object):
        return rocket.Rocket(self.layout,
                             os.path.join(self.root, subdir, filename))

-    def append(self, data):
-        """Append the data and flush it to disk.
-        data is a nested Python list [[row],[row],[...]]"""
-        remaining = len(data)
-        dataiter = iter(data)
-        while remaining:
-            # See how many rows we can fit into the current file, and open it
-            (subdir, fname, offset, count) = self._offset_from_row(self.nrows)
-            if count > remaining:
-                count = remaining
-
-            f = self.file_open(subdir, fname)
-
-            # Write the data
-            written = f.append_iter(count, dataiter)
-            if written != count: # pragma: no cover
-                raise Exception("Didn't write the expected number of rows: "
-                                + str(written) + " != " + str(count))
-            remaining -= count
-            self.nrows += count
-
    def append_string(self, data, start, end):
        """Parse the formatted string in 'data', according to the
        current layout, and append it to the table.  If any timestamps
@@ -349,7 +399,7 @@ class Table(object):
        the table is reverted back to its original state by truncating
        or deleting files as necessary."""
        data_offset = 0
-        last_timestamp = -1e12
+        last_timestamp = nilmdb.utils.time.min_timestamp
        tot_rows = self.nrows
        count = 0
        linenum = 0
@@ -380,10 +430,12 @@ class Table(object):
                    elif errtype == rocket.ERR_OUT_OF_INTERVAL:
                        if obj < start:
                            err = sprintf("Data timestamp %s < start time %s",
-                                          ftts(obj), ftts(start))
+                                          timestamp_to_string(obj),
+                                          timestamp_to_string(start))
                        else:
                            err = sprintf("Data timestamp %s >= end time %s",
-                                          ftts(obj), ftts(end))
+                                          timestamp_to_string(obj),
+                                          timestamp_to_string(end))
                    else:
                        err = str(obj)
                    raise ValueError("error parsing input data: " +
@@ -403,10 +455,9 @@ class Table(object):
            # Success, so update self.nrows accordingly
            self.nrows = tot_rows

-    def _get_data(self, start, stop, as_string):
+    def get_data(self, start, stop):
        """Extract data corresponding to Python range [n:m],
-        and returns a numeric list or formatted string,
-        depending on as_string."""
+        and returns a formatted string"""
        if (start is None or
            stop is None or
            start > stop or
@@ -422,42 +473,18 @@ class Table(object):
            if count > remaining:
                count = remaining
            f = self.file_open(subdir, filename)
-            if as_string:
-                ret.append(f.extract_string(offset, count))
-            else:
-                ret.extend(f.extract_list(offset, count))
+            ret.append(f.extract_string(offset, count))
            remaining -= count
            row += count
-        if as_string:
-            return "".join(ret)
-        return ret
+        return "".join(ret)

-    def get_as_text(self, start, stop):
-        """Extract data corresponding to Python range [n:m],
-        and returns a formatted string"""
-        return self._get_data(start, stop, True)
-
-    def __getitem__(self, key):
-        """Extract data and return it.  Supports simple indexing
-        (table[n]) and range slices (table[n:m]).  Returns a nested
-        Python list [[row],[row],[...]]"""
-
-        # Handle simple slices
-        if isinstance(key, slice):
-            # Fall back to brute force if the slice isn't simple
-            try:
-                if (key.step is not None and key.step != 1):
-                    raise IndexError
-                return self._get_data(key.start, key.stop, False)
-            except IndexError:
-                return [ self[x] for x in xrange(*key.indices(self.nrows)) ]
-
-        # Handle single points (inefficiently!)
-        if key < 0 or key >= self.nrows:
+    def __getitem__(self, row):
+        """Extract timestamps from a row, with table[n] notation."""
+        if row < 0 or row >= self.nrows:
            raise IndexError("Index out of range")
-        (subdir, filename, offset, count) = self._offset_from_row(key)
+        (subdir, filename, offset, count) = self._offset_from_row(row)
        f = self.file_open(subdir, filename)
-        return f.extract_list(offset, 1)[0]
+        return f.extract_timestamp(offset)

    def _remove_rows(self, subdir, filename, start, stop):
        """Helper to mark specific rows as being removed from a
@@ -548,11 +575,3 @@ class Table(object):
            self._remove_rows(subdir, filename, row_offset, row_offset + count)
            remaining -= count
            row += count
-
-class TimestampOnlyTable(object):
-    """Helper that lets us pass a Tables object into bisect, by
-    returning only the timestamp when a particular row is requested."""
-    def __init__(self, table):
-        self.table = table
-    def __getitem__(self, index):
-        return self.table[index][0]
--- a/nilmdb/server/interval.pyx
+++ b/nilmdb/server/interval.pyx
@@ -19,13 +19,16 @@ Intervals are half-open, ie. they include data points with timestamps
 # Fourth version is an optimized rb-tree that stores interval starts
 # and ends directly in the tree, like bxinterval did.

-from ..utils.time import float_time_to_string as ftts
+from ..utils.time import min_timestamp as nilmdb_min_timestamp
+from ..utils.time import max_timestamp as nilmdb_max_timestamp
+from ..utils.time import timestamp_to_string
 from ..utils.iterator import imerge
 import itertools

 cimport rbtree
-cdef extern from "stdint.h":
-    ctypedef unsigned long long uint64_t
+from libc.stdint cimport uint64_t, int64_t
+
+ctypedef int64_t timestamp_t

 class IntervalError(Exception):
    """Error due to interval overlap, etc"""
@@ -34,24 +37,25 @@ class IntervalError(Exception):
 cdef class Interval:
    """Represents an interval of time."""

-    cdef public double start, end
+    cdef public timestamp_t start, end

-    def __init__(self, double start, double end):
+    def __init__(self, timestamp_t start, timestamp_t end):
        """
-        'start' and 'end' are arbitrary floats that represent time
+        'start' and 'end' are arbitrary numbers that represent time
        """
        if start >= end:
            # Explicitly disallow zero-width intervals (since they're half-open)
            raise IntervalError("start %s must precede end %s" % (start, end))
-        self.start = float(start)
-        self.end = float(end)
+        self.start = start
+        self.end = end

    def __repr__(self):
        s = repr(self.start) + ", " + repr(self.end)
        return self.__class__.__name__ + "(" + s + ")"

    def __str__(self):
-        return "[" + ftts(self.start) + " -> " + ftts(self.end) + ")"
+        return ("[" + timestamp_to_string(self.start) +
+                " -> " + timestamp_to_string(self.end) + ")")

    def __cmp__(self, Interval other):
        """Compare two intervals.  If non-equal, order by start then end"""
@@ -73,7 +77,7 @@ cdef class Interval:
            return False
        return True

-    cpdef subset(self, double start, double end):
+    cpdef subset(self, timestamp_t start, timestamp_t end):
        """Return a new Interval that is a subset of this one"""
        # A subclass that tracks additional data might override this.
        if start < self.start or end > self.end:
@@ -95,14 +99,14 @@ cdef class DBInterval(Interval):
        db_end = 200, db_endpos = 20000
    """

-    cpdef public double db_start, db_end
+    cpdef public timestamp_t db_start, db_end
    cpdef public uint64_t db_startpos, db_endpos

    def __init__(self, start, end,
                 db_start, db_end,
                 db_startpos, db_endpos):
        """
-        'db_start' and 'db_end' are arbitrary floats that represent
+        'db_start' and 'db_end' are arbitrary numbers that represent
        time.  They must be a strict superset of the time interval
        covered by 'start' and 'end'.  The 'db_startpos' and
        'db_endpos' are arbitrary database position indicators that
@@ -122,7 +126,7 @@ cdef class DBInterval(Interval):
        s += ", " + repr(self.db_startpos) + ", " + repr(self.db_endpos)
        return self.__class__.__name__ + "(" + s + ")"

-    cpdef subset(self, double start, double end):
+    cpdef subset(self, timestamp_t start, timestamp_t end):
        """
        Return a new DBInterval that is a subset of this one
        """
@@ -331,7 +335,8 @@ cdef class IntervalSet:
                yield i.start, key_start, i
                yield i.end, key_end, i
        if bounds is None:
-            bounds = Interval(-1e12, 1e12)
+            bounds = Interval(nilmdb_min_timestamp,
+                              nilmdb_max_timestamp)
        self_iter = decorate(self.intersection(bounds), 0, 2)
        other_iter = decorate(other.intersection(bounds), 1, 3)

@@ -372,7 +377,7 @@ cdef class IntervalSet:
                return True
        return False

-    def find_end(self, double t):
+    def find_end(self, timestamp_t t):
        """
        Return an Interval from this tree that ends at time t, or
        None if it doesn't exist.
--- a/nilmdb/server/layout.pyx
+++ b/nilmdb/server/layout.pyx
@@ -5,6 +5,8 @@ import sys
 import inspect
 import cStringIO

+from ..utils.time import min_timestamp as nilmdb_min_timestamp
+
 cdef enum:
    max_value_count = 64

@@ -146,7 +148,8 @@ class Parser(object):
        layout, into an internal data structure suitable for a
        pytables 'table.append(parser.data)'.
        """
-        cdef double last_ts = -1e12, ts
+        cdef double last_ts = nilmdb_min_timestamp
+        cdef double ts
        cdef int n = 0, i
        cdef char *line

--- a/nilmdb/server/nilmdb.py
+++ b/nilmdb/server/nilmdb.py
@@ -35,10 +35,10 @@ import bisect
 # seems that 'PRAGMA synchronous=NORMAL' and 'PRAGMA journal_mode=WAL'
 # give an equivalent speedup more safely.  That is what is used here.
 _sql_schema_updates = {
-    0: """
+    0: { "next": 1, "sql": """
    -- All streams
    CREATE TABLE streams(
-    	id INTEGER PRIMARY KEY,		-- stream ID
+        id INTEGER PRIMARY KEY,		-- stream ID
        path TEXT UNIQUE NOT NULL,	-- path, e.g. '/newton/prep'
        layout TEXT NOT NULL		-- layout name, e.g. float32_8
    );
@@ -59,16 +59,21 @@ _sql_schema_updates = {
        end_pos INTEGER NOT NULL
    );
    CREATE INDEX _ranges_index ON ranges (stream_id, start_time, end_time);
-    """,
+    """ },

-    1: """
+    1: { "next": 3, "sql": """
    -- Generic dictionary-type metadata that can be associated with a stream
    CREATE TABLE metadata(
    	stream_id INTEGER NOT NULL,
        key TEXT NOT NULL,
        value TEXT
    );
-    """,
+    """ },
+
+    2: { "error": "old format with floating-point timestamps requires "
+         "nilmdb 1.3.1 or older" },
+
+    3: { "next": None },
 }

@nilmdb.utils.must_close()
@@ -96,7 +101,10 @@ class NilmDB(object):
        # SQLite database too
        sqlfilename = os.path.join(self.basepath, "data.sql")
        self.con = sqlite3.connect(sqlfilename, check_same_thread = True)
-        self._sql_schema_update()
+        try:
+            self._sql_schema_update()
+        finally: # pragma: no cover
+            self.data.close()

        # See big comment at top about the performance implications of this
        self.con.execute("PRAGMA synchronous=NORMAL")
@@ -123,11 +131,20 @@ class NilmDB(object):
        version = cur.execute("PRAGMA user_version").fetchone()[0]
        oldversion = version

-        while version in _sql_schema_updates:
-            cur.executescript(_sql_schema_updates[version])
-            version = version + 1
+        while True:
+            if version not in _sql_schema_updates: # pragma: no cover
+                raise Exception(self.basepath + ": unknown database version "
+                                + str(version))
+            update = _sql_schema_updates[version]
+            if "error" in update: # pragma: no cover
+                raise Exception(self.basepath + ": can't use database version "
+                                + str(version) + ": " + update["error"])
+            if update["next"] is None:
+                break
+            cur.executescript(update["sql"])
+            version = update["next"]
            if self.verbose: # pragma: no cover
-                printf("Schema updated to %d\n", version)
+                printf("Database schema updated to %d\n", version)

        if version != oldversion:
            with self.con:
@@ -135,9 +152,9 @@ class NilmDB(object):

    def _check_user_times(self, start, end):
        if start is None:
-            start = -1e12
+            start = nilmdb.utils.time.min_timestamp
        if end is None:
-            end = 1e12
+            end = nilmdb.utils.time.max_timestamp
        if start >= end:
            raise NilmDBError("start must precede end")
        return (start, end)
@@ -284,7 +301,7 @@ class NilmDB(object):
           interval_min (earliest interval start)
           interval_max (latest interval end)
           rows         (total number of rows of data)
-           seconds      (total time covered by this stream)
+           time         (total time covered by this stream, in timestamp units)
        """
        params = ()
        query = "SELECT streams.path, streams.layout"
@@ -405,6 +422,18 @@ class NilmDB(object):
        data.update(newdata)
        self.stream_set_metadata(path, data)

+    def stream_rename(self, oldpath, newpath):
+        """Rename a stream."""
+        stream_id = self._stream_id(oldpath)
+
+        # Rename the data
+        self.data.rename(oldpath, newpath)
+
+        # Rename the stream in the database
+        with self.con as con:
+            con.execute("UPDATE streams SET path=? WHERE id=?",
+                        (newpath, stream_id))
+
    def stream_destroy(self, path):
        """Fully remove a table and all of its data from the database.
        No way to undo it!  Metadata is removed."""
@@ -460,7 +489,7 @@ class NilmDB(object):
        # Optimization for the common case where an interval wasn't truncated
        if dbinterval.start == dbinterval.db_start:
            return dbinterval.db_startpos
-        return bisect.bisect_left(bulkdata.TimestampOnlyTable(table),
+        return bisect.bisect_left(table,
                                  dbinterval.start,
                                  dbinterval.db_startpos,
                                  dbinterval.db_endpos)
@@ -479,7 +508,7 @@ class NilmDB(object):
        # want to include the given timestamp in the results.  This is
        # so a queries like 1:00 -> 2:00 and 2:00 -> 3:00 return
        # non-overlapping data.
-        return bisect.bisect_left(bulkdata.TimestampOnlyTable(table),
+        return bisect.bisect_left(table,
                                  dbinterval.end,
                                  dbinterval.db_startpos,
                                  dbinterval.db_endpos)
@@ -527,10 +556,10 @@ class NilmDB(object):
            row_max = row_start + remaining
            if row_max < row_end:
                row_end = row_max
-                restart = table[row_max][0]
+                restart = table[row_max]

            # Gather these results up
-            result.append(table.get_as_text(row_start, row_end))
+            result.append(table.get_data(row_start, row_end))

            # Count them
            remaining -= row_end - row_start
--- a/nilmdb/server/pyrocket.py
+++ b/nilmdb/server/pyrocket.py
@@ -1,143 +0,0 @@
-# Python implementation of the "rocket" data parsing interface.
-# This interface translates between the binary format on disk
-# and the ASCII format used when communicating with clients.
-
-# This is slow!  Use the C version instead.
-
-from __future__ import absolute_import
-import struct
-import cStringIO
-import itertools
-from . import layout as _layout
-import nilmdb.utils
-from nilmdb.utils.time import float_time_to_string as ftts
-
-ERR_UNKNOWN = 0
-ERR_NON_MONOTONIC = 1
-ERR_OUT_OF_INTERVAL = 2
-class ParseError(Exception):
-    pass
-
-@nilmdb.utils.must_close(wrap_verify = False)
-class Rocket(object):
-    def __init__(self, layout, filename):
-        self.layout = layout
-        if filename:
-            self.file = open(filename, "a+b")
-        else:
-            self.file = None
-
-        # For packing/unpacking into a binary file.
-        # This will change in the C version
-        try:
-            (self.ltype, lcount) = layout.split('_', 2)
-            self.lcount = int(lcount)
-        except:
-            raise ValueError("no such layout: badly formatted string")
-        if self.lcount < 1:
-            raise ValueError("no such layout: bad count")
-        try:
-            struct_fmt = '<d'  # Little endian, double timestamp
-            struct_mapping = {
-                "int8": 'b',
-                "uint8": 'B',
-                "int16": 'h',
-                "uint16": 'H',
-                "int32": 'i',
-                "uint32": 'I',
-                "int64": 'q',
-                "uint64": 'Q',
-                "float32": 'f',
-                "float64": 'd',
-                }
-            struct_fmt += struct_mapping[self.ltype] * self.lcount
-        except KeyError:
-            raise ValueError("no such layout: bad data type")
-        self.packer = struct.Struct(struct_fmt)
-
-        # For packing/unpacking from strings.
-        self.layoutparser = _layout.Layout(self.layout)
-        self.formatter = _layout.Formatter(self.layout)
-
-    def close(self):
-        if self.file:
-            self.file.close()
-
-    @property
-    def binary_size(self):
-        """Return size of one row of data in the binary file, in bytes"""
-        return self.packer.size
-
-    def append_iter(self, maxrows, data):
-        """Append the list data to the file"""
-        # We assume the file is opened in append mode,
-        # so all writes go to the end.
-        written = 0
-        for row in itertools.islice(data, maxrows):
-            self.file.write(self.packer.pack(*row))
-            written += 1
-        self.file.flush()
-        return written
-
-    def append_string(self, count, data, data_offset, linenum,
-                      start, end, last_timestamp):
-        """Parse string and append data.
-
-          count: maximum number of rows to add
-          data: string data
-          data_offset: byte offset into data to start parsing
-          linenum: current line number of data
-          start: starting timestamp for interval
-          end: end timestamp for interval
-          last_timestamp: last timestamp that was previously parsed
-
-        Raises ParseError if timestamps are non-monotonic, outside the
-        start/end interval, etc.
-
-        On success, return a tuple with three values:
-          added_rows: how many rows were added from the file
-          data_offset: current offset into the data string
-          last_timestamp: last timestamp we parsed
-        """
-        # Parse the input data
-        indata = cStringIO.StringIO(data)
-        indata.seek(data_offset)
-        written = 0
-        while written < count:
-            line = indata.readline()
-            linenum += 1
-            if line == "":
-                break
-            comment = line.find('#')
-            if comment >= 0:
-                line = line.split('#', 1)[0]
-            line = line.strip()
-            if line == "":
-                continue
-            try:
-                (ts, row) = self.layoutparser.parse(line)
-            except ValueError as e:
-                raise ParseError(linenum, 0, ERR_UNKNOWN, e)
-            if ts <= last_timestamp:
-                raise ParseError(linenum, 0, ERR_NON_MONOTONIC, ts)
-            last_timestamp = ts
-            if ts < start or ts >= end:
-                raise ParseError(linenum, 0, ERR_OUT_OF_INTERVAL, ts)
-            self.append_iter(1, [row])
-            written += 1
-        return (written, indata.tell(), last_timestamp, linenum)
-
-    def extract_list(self, offset, count):
-        """Extract count rows of data from the file at offset offset.
-        Return a list of lists [[row],[row],...]"""
-        ret = []
-        self.file.seek(offset)
-        for i in xrange(count):
-            data = self.file.read(self.binary_size)
-            ret.append(list(self.packer.unpack(data)))
-        return ret
-
-    def extract_string(self, offset, count):
-        """Extract count rows of data from the file at offset offset.
-        Return an ascii formatted string according to the layout"""
-        return self.formatter.format(self.extract_list(offset, count))
--- a/nilmdb/server/rocket.c
+++ b/nilmdb/server/rocket.c
@@ -17,6 +17,15 @@
 #define FLOAT64_MIN 0
 #define FLOAT64_MAX 0

+typedef int64_t timestamp_t;
+
+/* This code probably needs to be double-checked for the case where
+   sizeof(long) != 8, so enforce that here with something that will
+   fail at build time.  We assume that the python integer type can
+   hold an int64_t. */
+const static char __long_ok[1 - 2*!(sizeof(int64_t) ==
+				    sizeof(long int))] = { 0 };
+
 /* Somewhat arbitrary, just so we can use fixed sizes for strings
   etc. */
 static const int MAX_LAYOUT_COUNT = 128;
@@ -46,10 +55,10 @@ static PyObject *raise_str(int line, int col, int code, const char *string)
 	}
 	return NULL;
 }
-static PyObject *raise_num(int line, int col, int code, double num)
+static PyObject *raise_int(int line, int col, int code, int64_t num)
 {
 	PyObject *o;
-	o = Py_BuildValue("(iiid)", line, col, code, num);
+	o = Py_BuildValue("(iiil)", line, col, code, num);
 	if (o != NULL) {
 		PyErr_SetObject(ParseError, o);
 		Py_DECREF(o);
@@ -237,106 +246,6 @@ static PyObject *Rocket_get_file_size(Rocket *self)
 	return PyInt_FromLong(self->file_size);
 }

-/****
- * Append from iterator
- */
-
-/* Helper for writing Python objects to the file */
-static inline void append_pyobject(FILE *out, PyObject *val, layout_type_t type)
-{
-	union8_t t8;
-	union16_t t16;
-	union32_t t32;
-	union64_t t64;
-	int ret = 0;
-
-	switch (type) {
-#define CASE(type, pyconvert, pytype, disktype, htole, bytes)	     \
-		case LAYOUT_TYPE_##type:			     \
-			pytype = pyconvert(val);		     \
-			if (PyErr_Occurred())			     \
-				return;				     \
-			disktype = htole(disktype);		     \
-			ret = fwrite(&disktype, bytes, 1, out);	     \
-			break
-		CASE(INT8,   PyInt_AsLong, t8.i,  t8.u,         , 1);
-		CASE(UINT8,  PyInt_AsLong, t8.u,  t8.u,         , 1);
-		CASE(INT16,  PyInt_AsLong, t16.i, t16.u, htole16, 2);
-		CASE(UINT16, PyInt_AsLong, t16.u, t16.u, htole16, 2);
-		CASE(INT32,  PyInt_AsLong, t32.i, t32.u, htole32, 4);
-		CASE(UINT32, PyInt_AsLong, t32.u, t32.u, htole32, 4);
-		CASE(INT64,  PyInt_AsLong, t64.i, t64.u, htole64, 8);
-		CASE(UINT64, PyInt_AsLong, t64.u, t64.u, htole64, 8);
-		CASE(FLOAT32, PyFloat_AsDouble, t32.f, t32.u, htole32, 4);
-		CASE(FLOAT64, PyFloat_AsDouble, t64.d, t64.u, htole64, 8);
-#undef CASE
-	default:
-		PyErr_SetString(PyExc_TypeError, "unknown type");
-		return;
-	}
-	if (ret <= 0) {
-		PyErr_SetFromErrno(PyExc_OSError);
-	}
-}
-/* .append_iter(maxrows, dataiter) */
-static PyObject *Rocket_append_iter(Rocket *self, PyObject *args)
-{
-	int maxrows;
-	PyObject *iter;
-	PyObject *rowlist;
-	if (!PyArg_ParseTuple(args, "iO:append_iter", &maxrows, &iter))
-		return NULL;
-	if (!PyIter_Check(iter)) {
-		PyErr_SetString(PyExc_TypeError, "need an iterable");
-		return NULL;
-	}
-	if (!self->file) {
-		PyErr_SetString(PyExc_Exception, "no file");
-		return NULL;
-	}
-
-	/* Mark file size so that it will get updated next time it's read */
-	self->file_size = -1;
-
-	int row;
-	for (row = 0; row < maxrows; row++) {
-		rowlist = PyIter_Next(iter);
-		if (!rowlist)
-			break;
-		if (!PyList_Check(rowlist)) {
-			PyErr_SetString(PyExc_TypeError, "rows must be lists");
-			goto row_err;
-		}
-		if (PyList_Size(rowlist) != self->layout_count + 1) {
-			PyErr_SetString(PyExc_TypeError, "short row");
-			goto row_err;
-		}
-
-		/* Extract and write timestamp */
-		append_pyobject(self->file, PyList_GetItem(rowlist, 0),
-				LAYOUT_TYPE_FLOAT64);
-		if (PyErr_Occurred())
-			goto row_err;
-
-		/* Extract and write values */
-		int i;
-		for (i = 0; i < self->layout_count; i++) {
-			append_pyobject(self->file,
-					PyList_GetItem(rowlist, i+1),
-					self->layout_type);
-			if (PyErr_Occurred())
-				goto row_err;
-		}
-	}
-	fflush(self->file);
-	/* All done */
-	return PyLong_FromLong(row);
-row_err:
-	fflush(self->file);
-	Py_DECREF(rowlist);
-	return NULL;
-}
-
 /****
 * Append from string
 */
@@ -355,9 +264,9 @@ static PyObject *Rocket_append_string(Rocket *self, PyObject *args)
 	int offset;
 	const char *linestart;
 	int linenum;
-	double start;
-	double end;
-	double last_timestamp;
+	timestamp_t start;
+	timestamp_t end;
+	timestamp_t last_timestamp;

 	int written = 0;
 	char *endptr;
@@ -371,7 +280,7 @@ static PyObject *Rocket_append_string(Rocket *self, PyObject *args)
 	   but we need the null termination for strto*.  If we had
 	   strnto* that took a length, we could use t# and not require
 	   a copy. */
-	if (!PyArg_ParseTuple(args, "isiiddd:append_string", &count,
+	if (!PyArg_ParseTuple(args, "isiilll:append_string", &count,
 			      &data, &offset, &linenum,
 			      &start, &end, &last_timestamp))
 		return NULL;
@@ -401,17 +310,23 @@ static PyObject *Rocket_append_string(Rocket *self, PyObject *args)
 		}

 		/* Extract timestamp */
-		t64.d = strtod(buf, &endptr);
-		if (endptr == buf)
-			return raise_str(linenum, buf - linestart + 1,
-					 ERR_OTHER, "bad timestamp");
-		if (t64.d <= last_timestamp)
-			return raise_num(linenum, buf - linestart + 1,
-					 ERR_NON_MONOTONIC, t64.d);
-		last_timestamp = t64.d;
-		if (t64.d < start || t64.d >= end)
-			return raise_num(linenum, buf - linestart + 1,
-					 ERR_OUT_OF_INTERVAL, t64.d);
+		t64.i = strtoll(buf, &endptr, 10);
+		if (endptr == buf || !isspace(*endptr)) {
+			/* Try parsing as a double instead */
+			t64.d = strtod(buf, &endptr);
+			if (endptr == buf)
+				goto bad_timestamp;
+			if (!isspace(*endptr))
+				goto cant_parse_value;
+			t64.i = round(t64.d);
+		}
+		if (t64.i <= last_timestamp)
+			return raise_int(linenum, buf - linestart + 1,
+					 ERR_NON_MONOTONIC, t64.i);
+		last_timestamp = t64.i;
+		if (t64.i < start || t64.i >= end)
+			return raise_int(linenum, buf - linestart + 1,
+					 ERR_OUT_OF_INTERVAL, t64.i);
 		t64.u = le64toh(t64.u);
 		if (fwrite(&t64.u, 8, 1, self->file) != 1)
 			goto err;
@@ -479,14 +394,17 @@ static PyObject *Rocket_append_string(Rocket *self, PyObject *args)

 	fflush(self->file);

-	/* Build return value and return*/
+	/* Build return value and return */
 	offset = buf - data;
 	PyObject *o;
-	o = Py_BuildValue("(iidi)", written, offset, last_timestamp, linenum);
+	o = Py_BuildValue("(iili)", written, offset, last_timestamp, linenum);
 	return o;
 err:
 	PyErr_SetFromErrno(PyExc_OSError);
 	return NULL;
+bad_timestamp:
+	return raise_str(linenum, buf - linestart + 1,
+			 ERR_OTHER, "bad timestamp");
 cant_parse_value:
 	return raise_str(linenum, buf - linestart + 1,
 			 ERR_OTHER, "can't parse value");
@@ -501,116 +419,6 @@ extra_data_on_line:
 			 ERR_OTHER, "extra data on line");
 }

-/****
- * Extract to Python list
- */
-
-static int _extract_handle_params(Rocket *self, PyObject *args, long *count)
-{
-	long offset;
-	if (!PyArg_ParseTuple(args, "ll", &offset, count))
-		return -1;
-	if (!self->file) {
-		PyErr_SetString(PyExc_Exception, "no file");
-		return -1;
-	}
-	/* Seek to target location */
-	if (fseek(self->file, offset, SEEK_SET) < 0) {
-		PyErr_SetFromErrno(PyExc_OSError);
-		return -1;
-	}
-	return 0;
-}
-
-/* Helper for extracting data from a file as a Python object */
-static inline void *extract_pyobject(FILE *in, layout_type_t type)
-{
-	union8_t t8;
-	union16_t t16;
-	union32_t t32;
-	union64_t t64;
-
-	switch (type) {
-#define CASE(type, pyconvert, pytype, disktype, letoh, bytes)		\
-		case LAYOUT_TYPE_##type:				\
-			if (fread(&disktype, bytes, 1, in) <= 0)	\
-				break;					\
-			disktype = letoh(disktype);			\
-			return pyconvert(pytype);			\
-			break
-		CASE(INT8,   PyInt_FromLong, t8.i,  t8.u,         , 1);
-		CASE(UINT8,  PyInt_FromLong, t8.u,  t8.u,         , 1);
-		CASE(INT16,  PyInt_FromLong, t16.i, t16.u, le16toh, 2);
-		CASE(UINT16, PyInt_FromLong, t16.u, t16.u, le16toh, 2);
-		CASE(INT32,  PyInt_FromLong, t32.i, t32.u, le32toh, 4);
-		CASE(UINT32, PyInt_FromLong, t32.u, t32.u, le32toh, 4);
-		CASE(INT64,  PyInt_FromLong, t64.i, t64.u, le64toh, 8);
-		CASE(UINT64, PyInt_FromLong, t64.u, t64.u, le64toh, 8);
-		CASE(FLOAT32, PyFloat_FromDouble, t32.f, t32.u, le32toh, 4);
-		CASE(FLOAT64, PyFloat_FromDouble, t64.d, t64.u, le64toh, 8);
-#undef CASE
-	default:
-		PyErr_SetString(PyExc_TypeError, "unknown type");
-		return NULL;
-	}
-	PyErr_SetString(PyExc_OSError, "failed to read from file");
-	return NULL;
-}
-
-static PyObject *Rocket_extract_list(Rocket *self, PyObject *args)
-{
-	long count;
-	if (_extract_handle_params(self, args, &count) < 0)
-		return NULL;
-
-	/* Make a list to return */
-	PyObject *retlist = PyList_New(0);
-	if (!retlist)
-		return NULL;
-
-	/* Read data into new Python lists */
-	int row;
-	for (row = 0; row < count; row++)
-	{
-		PyObject *rowlist = PyList_New(self->layout_count + 1);
-		if (!rowlist) {
-			Py_DECREF(retlist);
-			return NULL;
-		}
-
-		/* Timestamp */
-		PyObject *entry = extract_pyobject(self->file,
-						   LAYOUT_TYPE_FLOAT64);
-		if (!entry || (PyList_SetItem(rowlist, 0, entry) < 0)) {
-			Py_DECREF(rowlist);
-			Py_DECREF(retlist);
-			return NULL;
-		}
-
-		/* Data */
-		int i;
-		for (i = 0; i < self->layout_count; i++) {
-			PyObject *ent = extract_pyobject(self->file,
-							 self->layout_type);
-			if (!ent || (PyList_SetItem(rowlist, i+1, ent) < 0)) {
-				Py_DECREF(rowlist);
-				Py_DECREF(retlist);
-				return NULL;
-			}
-		}
-
-		/* Add row to return value */
-		if (PyList_Append(retlist, rowlist) < 0) {
-			Py_DECREF(rowlist);
-			Py_DECREF(retlist);
-			return NULL;
-		}
-
-		Py_DECREF(rowlist);
-	}
-	return retlist;
-}
-
 /****
 * Extract to string
 */
@@ -618,8 +426,19 @@ static PyObject *Rocket_extract_list(Rocket *self, PyObject *args)
 static PyObject *Rocket_extract_string(Rocket *self, PyObject *args)
 {
 	long count;
-	if (_extract_handle_params(self, args, &count) < 0)
+	long offset;
+
+	if (!PyArg_ParseTuple(args, "ll", &offset, &count))
 		return NULL;
+	if (!self->file) {
+		PyErr_SetString(PyExc_Exception, "no file");
+		return NULL;
+	}
+	/* Seek to target location */
+	if (fseek(self->file, offset, SEEK_SET) < 0) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}

 	char *str = NULL, *new;
 	long len_alloc = 0;
@@ -653,8 +472,7 @@ static PyObject *Rocket_extract_string(Rocket *self, PyObject *args)
 		if (fread(&t64.u, 8, 1, self->file) != 1)
 			goto err;
 		t64.u = le64toh(t64.u);
-		/* Timestamps are always printed to the microsecond */
-		ret = sprintf(&str[len], "%.6f", t64.d);
+		ret = sprintf(&str[len], "%ld", t64.i);
 		if (ret <= 0)
 			goto err;
 		len += ret;
@@ -709,6 +527,33 @@ err:
 	return NULL;
 }

+
+/****
+ * Extract timestamp
+ */
+static PyObject *Rocket_extract_timestamp(Rocket *self, PyObject *args)
+{
+	long offset;
+	union64_t t64;
+	if (!PyArg_ParseTuple(args, "l", &offset))
+		return NULL;
+	if (!self->file) {
+		PyErr_SetString(PyExc_Exception, "no file");
+		return NULL;
+	}
+
+	/* Seek to target location and read timestamp */
+	if ((fseek(self->file, offset, SEEK_SET) < 0) ||
+	    (fread(&t64.u, 8, 1, self->file) != 1)) {
+		PyErr_SetFromErrno(PyExc_OSError);
+		return NULL;
+	}
+
+	/* Convert and return */
+	t64.u = le64toh(t64.u);
+	return Py_BuildValue("l", t64.i);
+}
+
 /****
 * Module and type setup
 */
@@ -730,10 +575,6 @@ static PyMethodDef Rocket_methods[] = {
 	  "close(self)\n\n"
 	  "Close file handle" },

-	{ "append_iter", (PyCFunction)Rocket_append_iter, METH_VARARGS,
-	  "append_iter(self, maxrows, iterable)\n\n"
-	  "Append up to maxrows of data from iter to the file" },
-
 	{ "append_string", (PyCFunction)Rocket_append_string, METH_VARARGS,
 	  "append_string(self, count, data, offset, line, start, end, ts)\n\n"
          "Parse string and append data.\n"
@@ -754,16 +595,16 @@ static PyMethodDef Rocket_methods[] = {
          "  data_offset: current offset into the data string\n"
          "  last_timestamp: last timestamp we parsed" },

-	{ "extract_list", (PyCFunction)Rocket_extract_list, METH_VARARGS,
-	  "extract_list(self, offset, count)\n\n"
-	  "Extract count rows of data from the file at offset offset.\n"
-	  "Return a list of lists [[row],[row],...]" },
-
 	{ "extract_string", (PyCFunction)Rocket_extract_string, METH_VARARGS,
 	  "extract_string(self, offset, count)\n\n"
 	  "Extract count rows of data from the file at offset offset.\n"
 	  "Return an ascii formatted string according to the layout" },

+	{ "extract_timestamp",
+	  (PyCFunction)Rocket_extract_timestamp, METH_VARARGS,
+	  "extract_timestamp(self, offset)\n\n"
+	  "Extract a single timestamp from the file" },
+
 	{ NULL },
 };

--- a/nilmdb/server/server.py
+++ b/nilmdb/server/server.py
@@ -6,6 +6,7 @@ from __future__ import absolute_import
 import nilmdb.server
 from nilmdb.utils.printf import *
 from nilmdb.server.errors import NilmDBError
+from nilmdb.utils.time import string_to_timestamp

 import cherrypy
 import sys
@@ -212,6 +213,16 @@ class Stream(NilmApp):
        """Delete a stream and its associated data."""
        return self.db.stream_destroy(path)

+    # /stream/rename?oldpath=/newton/prep&newpath=/newton/prep/1
+    @cherrypy.expose
+    @cherrypy.tools.json_in()
+    @cherrypy.tools.json_out()
+    @exception_to_httperror(NilmDBError, ValueError)
+    @cherrypy.tools.CORS_allow(methods = ["POST"])
+    def rename(self, oldpath, newpath):
+        """Rename a stream."""
+        return self.db.stream_rename(oldpath, newpath)
+
    # /stream/get_metadata?path=/newton/prep
    # /stream/get_metadata?path=/newton/prep&key=foo&key=bar
    @cherrypy.expose
@@ -294,8 +305,8 @@ class Stream(NilmApp):
            raise cherrypy.HTTPError("404 Not Found", "No such stream")

        # Check limits
-        start = float(start)
-        end = float(end)
+        start = string_to_timestamp(start)
+        end = string_to_timestamp(end)
        if start >= end:
            raise cherrypy.HTTPError("400 Bad Request",
                                     "start must precede end")
@@ -321,9 +332,9 @@ class Stream(NilmApp):
        removed.
        """
        if start is not None:
-            start = float(start)
+            start = string_to_timestamp(start)
        if end is not None:
-            end = float(end)
+            end = string_to_timestamp(end)
        if start is not None and end is not None:
            if start >= end:
                raise cherrypy.HTTPError("400 Bad Request",
@@ -352,9 +363,9 @@ class Stream(NilmApp):
        'application/x-json-stream' for lack of a better option.
        """
        if start is not None:
-            start = float(start)
+            start = string_to_timestamp(start)
        if end is not None:
-            end = float(end)
+            end = string_to_timestamp(end)

        if start is not None and end is not None:
            if start >= end:
@@ -394,9 +405,9 @@ class Stream(NilmApp):
        Add count=True to return a count rather than actual data.
        """
        if start is not None:
-            start = float(start)
+            start = string_to_timestamp(start)
        if end is not None:
-            end = float(end)
+            end = string_to_timestamp(end)

        # Check parameters
        if start is not None and end is not None:
--- a/nilmdb/utils/time.py
+++ b/nilmdb/utils/time.py
@@ -1,11 +1,60 @@
 from nilmdb.utils import datetime_tz
 import re

+# Range
+min_timestamp = (-2**63)
+max_timestamp = (2**62 - 1)
+
+# Smallest representable step
+epsilon = 1
+
+def string_to_timestamp(str):
+    """Convert a string that represents an integer number of microseconds
+    since epoch."""
+    try:
+        # Parse a string like "1234567890123456" and return an integer
+        return int(str)
+    except ValueError:
+        # Try parsing as a float, in case it's "1234567890123456.0"
+        return int(round(float(str)))
+
+def timestamp_to_string(timestamp):
+    """Convert a timestamp (integer microseconds since epoch) to a string"""
+    if isinstance(timestamp, float):
+        return str(int(round(timestamp)))
+    else:
+        return str(timestamp)
+
+def timestamp_to_human(timestamp):
+    """Convert a timestamp (integer microseconds since epoch) to a
+    human-readable string, using the local timezone for display
+    (e.g. from the TZ env var)."""
+    dt = datetime_tz.datetime_tz.fromtimestamp(timestamp_to_unix(timestamp))
+    return dt.strftime("%a, %d %b %Y %H:%M:%S.%f %z")
+
+def unix_to_timestamp(unix):
+    """Convert a Unix timestamp (floating point seconds since epoch)
+    into a NILM timestamp (integer microseconds since epoch)"""
+    return int(round(unix * 1e6))
+
+def timestamp_to_unix(timestamp):
+    """Convert a NILM timestamp (integer microseconds since epoch)
+    into a Unix timestamp (floating point seconds since epoch)"""
+    return timestamp / 1e6
+timestamp_to_seconds = timestamp_to_unix
+
+def rate_to_period(hz, cycles = 1):
+    """Convert a rate (in Hz) to a period (in timestamp units).
+    Returns an integer."""
+    period = unix_to_timestamp(cycles) / float(hz)
+    return int(round(period))
+
 def parse_time(toparse):
    """
-    Parse a free-form time string and return a datetime_tz object.
-    If the string doesn't contain a timestamp, the current local
-    timezone is assumed (e.g. from the TZ env var).
+    Parse a free-form time string and return a nilmdb timestamp
+    (integer seconds since epoch).  If the string doesn't contain a
+    timestamp, the current local timezone is assumed (e.g. from the TZ
+    env var).
    """
    # If string isn't "now" and doesn't contain at least 4 digits,
    # consider it invalid.  smartparse might otherwise accept
@@ -15,17 +64,28 @@ def parse_time(toparse):

    # Try to just parse the time as given
    try:
-        return datetime_tz.datetime_tz.smartparse(toparse)
-    except ValueError:
+        return unix_to_timestamp(datetime_tz.datetime_tz.
+                                 smartparse(toparse).totimestamp())
+    except (ValueError, OverflowError):
        pass

-    # Try to treat it as a single double
+    # If it starts with @, treat it as a NILM timestamp
+    # (integer microseconds since epoch)
    try:
-        timestamp = float(toparse)
-        # range is from about year 2001 - 2065
-        if timestamp < 1e9 or timestamp > 3e9:
-            raise ValueError
-        return datetime_tz.datetime_tz.fromtimestamp(timestamp)
+        if toparse[0] == '@':
+            return int(toparse[1:])
+    except (ValueError, KeyError):
+        pass
+
+    # If it's parseable as a float, treat it as a Unix or NILM
+    # timestamp based on its range.
+    try:
+        val = float(toparse)
+        # range is from about year 2001 - 2128
+        if val > 1e9 and val < 5e9:
+            return unix_to_timestamp(val)
+        if val > 1e15 and val < 5e15:
+            return val
    except ValueError:
        pass

@@ -47,7 +107,8 @@ def parse_time(toparse):
                    r")", toparse)
    if res is not None:
        try:
-            return datetime_tz.datetime_tz.smartparse(res.group(2))
+            return unix_to_timestamp(datetime_tz.datetime_tz.
+                                     smartparse(res.group(2)).totimestamp())
        except ValueError:
            pass

@@ -55,15 +116,6 @@ def parse_time(toparse):
    # just give up for now.
    raise ValueError("unable to parse timestamp")

-def format_time(timestamp):
-    """
-    Convert a Unix timestamp to a string for printing, using the
-    local timezone for display (e.g. from the TZ env var).
-    """
-    dt = datetime_tz.datetime_tz.fromtimestamp(timestamp)
-    return dt.strftime("%a, %d %b %Y %H:%M:%S.%f %z")
-
-def float_time_to_string(timestamp):
-    """Convert a floating-point Unix timestamp to a string,
-    like '1234567890.000000'"""
-    return "%.6f" % timestamp
+def now():
+    """Return current timestamp"""
+    return unix_to_timestamp(datetime_tz.datetime_tz.utcnow().totimestamp())
--- a/nilmdb/utils/timestamper.py
+++ b/nilmdb/utils/timestamper.py
@@ -1,7 +1,7 @@
 """File-like objects that add timestamps to the input lines"""

 from nilmdb.utils.printf import *
-from nilmdb.utils import datetime_tz
+import nilmdb.utils.time

 class Timestamper(object):
    """A file-like object that adds timestamps to lines of an input file."""
@@ -61,31 +61,33 @@ class TimestamperRate(Timestamper):

        end: If specified, raise StopIteration before outputting a value
             greater than this."""
+        timestamp_to_string = nilmdb.utils.time.timestamp_to_string
+        rate_to_period = nilmdb.utils.time.rate_to_period
        def iterator(start, rate, end):
            n = 0
            rate = float(rate)
            while True:
-                now = start + n / rate
+                now = start + rate_to_period(rate, n)
                if end and now >= end:
                    raise StopIteration
-                yield sprintf("%.6f ", start + n / rate)
+                yield timestamp_to_string(now) + " "
                n += 1
        Timestamper.__init__(self, infile, iterator(start, rate, end))
        self.start = start
        self.rate = rate
    def __str__(self):
-        start = datetime_tz.datetime_tz.fromtimestamp(self.start)
-        start = start.strftime("%a, %d %b %Y %H:%M:%S %Z")
        return sprintf("TimestamperRate(..., start=\"%s\", rate=%g)",
-                       str(start), self.rate)
+                       nilmdb.utils.time.timestamp_to_human(self.start),
+                       self.rate)

 class TimestamperNow(Timestamper):
    """Timestamper that uses current time"""
    def __init__(self, infile):
+        timestamp_to_string = nilmdb.utils.time.timestamp_to_string
+        get_now = nilmdb.utils.time.now
        def iterator():
            while True:
-                now = datetime_tz.datetime_tz.utcnow().totimestamp()
-                yield sprintf("%.6f ", now)
+                yield timestamp_to_string(get_now()) + " "
        Timestamper.__init__(self, infile, iterator())
    def __str__(self):
        return "TimestamperNow(...)"
--- a/setup.py
+++ b/setup.py
@@ -91,6 +91,9 @@ include tests/test.order

 # Docs
 recursive-include docs Makefile *.md
+
+# Extras
+recursive-include extras *
 """)

 # Run setup
--- a/tests/data/extract-1
+++ b/tests/data/extract-1
@@ -2,123 +2,123 @@
 # layout: float32_8
 # start: Fri, 23 Mar 2012 10:00:30.000000 +0000
 # end: Fri, 23 Mar 2012 10:00:31.000000 +0000
-1332496830.000000 2.517740e+05 2.242410e+05 5.688100e+03 1.915530e+03 9.329220e+03 4.183710e+03 1.212350e+03 2.641790e+03
-1332496830.008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
-1332496830.016667 2.630730e+05 2.233040e+05 4.961640e+03 2.197120e+03 7.687310e+03 4.861860e+03 2.732780e+03 3.008540e+03
-1332496830.025000 2.576140e+05 2.233230e+05 5.003660e+03 3.525140e+03 7.165310e+03 4.685620e+03 1.715380e+03 3.440480e+03
-1332496830.033333 2.557800e+05 2.219150e+05 6.357310e+03 2.145290e+03 8.426970e+03 3.775350e+03 1.475390e+03 3.797240e+03
-1332496830.041667 2.601660e+05 2.230080e+05 6.702590e+03 1.484960e+03 9.288100e+03 3.330830e+03 1.228500e+03 3.214320e+03
-1332496830.050000 2.612310e+05 2.264260e+05 4.980060e+03 2.982380e+03 8.499630e+03 4.267670e+03 9.940890e+02 2.292890e+03
-1332496830.058333 2.551170e+05 2.266420e+05 4.584410e+03 4.656440e+03 7.860150e+03 5.317310e+03 1.473600e+03 2.111690e+03
-1332496830.066667 2.533000e+05 2.235540e+05 6.455090e+03 3.036650e+03 8.869750e+03 4.986310e+03 2.607360e+03 2.839590e+03
-1332496830.075000 2.610610e+05 2.212630e+05 6.951980e+03 1.500240e+03 9.386100e+03 3.791680e+03 2.677010e+03 3.980630e+03
-1332496830.083333 2.665030e+05 2.231980e+05 5.189610e+03 2.594560e+03 8.571530e+03 3.175000e+03 9.198400e+02 3.792010e+03
-1332496830.091667 2.606920e+05 2.251840e+05 3.782480e+03 4.642880e+03 7.662960e+03 3.917790e+03 -2.510970e+02 2.907060e+03
-1332496830.100000 2.539630e+05 2.250810e+05 5.123530e+03 3.839550e+03 8.669030e+03 4.877820e+03 9.437240e+02 2.527450e+03
-1332496830.108333 2.565550e+05 2.241690e+05 5.930600e+03 2.298540e+03 8.906710e+03 5.331680e+03 2.549910e+03 3.053560e+03
-1332496830.116667 2.608890e+05 2.250100e+05 4.681130e+03 2.971870e+03 7.900040e+03 4.874080e+03 2.322430e+03 3.649120e+03
-1332496830.125000 2.579440e+05 2.249230e+05 3.291140e+03 4.357090e+03 7.131590e+03 4.385560e+03 1.077050e+03 3.664040e+03
-1332496830.133333 2.550090e+05 2.230180e+05 4.584820e+03 2.864000e+03 8.469490e+03 3.625580e+03 9.855570e+02 3.504230e+03
-1332496830.141667 2.601140e+05 2.219470e+05 5.676190e+03 1.210340e+03 9.393780e+03 3.390240e+03 1.654020e+03 3.018700e+03
-1332496830.150000 2.642770e+05 2.244380e+05 4.446620e+03 2.176720e+03 8.142090e+03 4.584880e+03 2.327830e+03 2.615800e+03
-1332496830.158333 2.592210e+05 2.264710e+05 2.734440e+03 4.182760e+03 6.389550e+03 5.540520e+03 1.958880e+03 2.720120e+03
-1332496830.166667 2.526500e+05 2.248310e+05 4.163640e+03 2.989990e+03 7.179200e+03 5.213060e+03 1.929550e+03 3.457660e+03
-1332496830.175000 2.570830e+05 2.220480e+05 5.759040e+03 7.024410e+02 8.566550e+03 3.552020e+03 1.832940e+03 3.956190e+03
-1332496830.183333 2.631300e+05 2.229670e+05 5.141140e+03 1.166120e+03 8.666960e+03 2.720370e+03 9.713740e+02 3.479730e+03
-1332496830.191667 2.602360e+05 2.252650e+05 3.425140e+03 3.339080e+03 7.853610e+03 3.674950e+03 5.259080e+02 2.443310e+03
-1332496830.200000 2.535030e+05 2.245270e+05 4.398130e+03 2.927430e+03 8.110280e+03 4.842470e+03 1.513870e+03 2.467100e+03
-1332496830.208333 2.561260e+05 2.226930e+05 6.043530e+03 6.562240e+02 8.797560e+03 4.832410e+03 2.832370e+03 3.426140e+03
-1332496830.216667 2.616770e+05 2.236080e+05 5.830460e+03 1.033910e+03 8.123940e+03 3.980690e+03 1.927960e+03 4.092720e+03
-1332496830.225000 2.594570e+05 2.255360e+05 4.015570e+03 2.995990e+03 7.135440e+03 3.713550e+03 3.072200e+02 3.849430e+03
-1332496830.233333 2.533520e+05 2.242160e+05 4.650560e+03 3.196620e+03 8.131280e+03 3.586160e+03 7.083230e+01 3.074180e+03
-1332496830.241667 2.561240e+05 2.215130e+05 6.100480e+03 8.219800e+02 9.757540e+03 3.474510e+03 1.647520e+03 2.559860e+03
-1332496830.250000 2.630240e+05 2.215590e+05 5.789960e+03 6.994170e+02 9.129740e+03 4.153080e+03 2.829250e+03 2.677270e+03
-1332496830.258333 2.617200e+05 2.240150e+05 4.358500e+03 2.645360e+03 7.414110e+03 4.810670e+03 2.225990e+03 3.185990e+03
-1332496830.266667 2.547560e+05 2.242400e+05 4.857380e+03 3.229680e+03 7.539310e+03 4.769140e+03 1.507130e+03 3.668260e+03
-1332496830.275000 2.568890e+05 2.226580e+05 6.473420e+03 1.214110e+03 9.010760e+03 3.848730e+03 1.303840e+03 3.778500e+03
-1332496830.283333 2.642080e+05 2.233160e+05 5.700450e+03 1.116560e+03 9.087610e+03 3.846680e+03 1.293590e+03 2.891560e+03
-1332496830.291667 2.633100e+05 2.257190e+05 3.936120e+03 3.252360e+03 7.552850e+03 4.897860e+03 1.156630e+03 2.037160e+03
-1332496830.300000 2.550790e+05 2.250860e+05 4.536450e+03 3.960110e+03 7.454590e+03 5.479070e+03 1.596360e+03 2.190800e+03
-1332496830.308333 2.544870e+05 2.225080e+05 6.635860e+03 1.758850e+03 8.732970e+03 4.466970e+03 2.650360e+03 3.139310e+03
-1332496830.316667 2.612410e+05 2.224320e+05 6.702270e+03 1.085130e+03 8.989230e+03 3.112990e+03 1.933560e+03 3.828410e+03
-1332496830.325000 2.621190e+05 2.255870e+05 4.714950e+03 2.892360e+03 8.107820e+03 2.961310e+03 2.399780e+02 3.273720e+03
-1332496830.333333 2.549990e+05 2.265140e+05 4.532090e+03 4.126900e+03 8.200130e+03 3.872590e+03 5.608900e+01 2.370580e+03
-1332496830.341667 2.542890e+05 2.240330e+05 6.538810e+03 2.251440e+03 9.419430e+03 4.564450e+03 2.077810e+03 2.508170e+03
-1332496830.350000 2.618900e+05 2.219600e+05 6.846090e+03 1.475270e+03 9.125590e+03 4.598290e+03 3.299220e+03 3.475420e+03
-1332496830.358333 2.645020e+05 2.230850e+05 5.066380e+03 3.270560e+03 7.933170e+03 4.173710e+03 1.908910e+03 3.867460e+03
-1332496830.366667 2.578890e+05 2.236560e+05 4.201660e+03 4.473640e+03 7.688340e+03 4.161580e+03 6.875790e+02 3.653690e+03
-1332496830.375000 2.542700e+05 2.231510e+05 5.715140e+03 2.752140e+03 9.273320e+03 3.772950e+03 8.964040e+02 3.256060e+03
-1332496830.383333 2.582570e+05 2.242170e+05 6.114310e+03 1.856860e+03 9.604320e+03 4.200490e+03 1.764380e+03 2.939220e+03
-1332496830.391667 2.600200e+05 2.268680e+05 4.237530e+03 3.605880e+03 8.066220e+03 5.430250e+03 2.138580e+03 2.696710e+03
-1332496830.400000 2.550830e+05 2.259240e+05 3.350310e+03 4.853070e+03 7.045820e+03 5.925200e+03 1.893610e+03 2.897340e+03
-1332496830.408333 2.544530e+05 2.221270e+05 5.271330e+03 2.491500e+03 8.436680e+03 5.032080e+03 2.436050e+03 3.724590e+03
-1332496830.416667 2.625880e+05 2.199500e+05 5.994620e+03 7.892740e+02 9.029650e+03 3.515740e+03 1.953570e+03 4.014520e+03
-1332496830.425000 2.656100e+05 2.233330e+05 4.391410e+03 2.400960e+03 8.146460e+03 3.536960e+03 5.302320e+02 3.133920e+03
-1332496830.433333 2.574700e+05 2.269770e+05 2.975320e+03 4.633530e+03 7.278560e+03 4.640100e+03 -5.015020e+01 2.024960e+03
-1332496830.441667 2.506870e+05 2.263310e+05 4.517860e+03 3.183800e+03 8.072600e+03 5.281660e+03 1.605140e+03 2.335140e+03
-1332496830.450000 2.555630e+05 2.244950e+05 5.551000e+03 1.101300e+03 8.461490e+03 4.725700e+03 2.726670e+03 3.480540e+03
-1332496830.458333 2.613350e+05 2.246450e+05 4.764680e+03 1.557020e+03 7.833350e+03 3.524810e+03 1.577410e+03 4.038620e+03
-1332496830.466667 2.602690e+05 2.240080e+05 3.558030e+03 2.987610e+03 7.362440e+03 3.279230e+03 5.624420e+02 3.786550e+03
-1332496830.475000 2.574350e+05 2.217770e+05 4.972600e+03 2.166880e+03 8.481440e+03 3.328720e+03 1.037130e+03 3.271370e+03
-1332496830.483333 2.610460e+05 2.215500e+05 5.816180e+03 5.902170e+02 9.120930e+03 3.895400e+03 2.382670e+03 2.824170e+03
-1332496830.491667 2.627660e+05 2.244730e+05 4.835050e+03 1.785770e+03 7.880760e+03 4.745620e+03 2.443660e+03 3.229550e+03
-1332496830.500000 2.565090e+05 2.264130e+05 3.758870e+03 3.461200e+03 6.743770e+03 4.928960e+03 1.536620e+03 3.546690e+03
-1332496830.508333 2.507930e+05 2.243720e+05 5.218490e+03 2.865260e+03 7.803960e+03 4.351090e+03 1.333820e+03 3.680490e+03
-1332496830.516667 2.563190e+05 2.220660e+05 6.403970e+03 7.323450e+02 9.627760e+03 3.089300e+03 1.516780e+03 3.653690e+03
-1332496830.525000 2.633430e+05 2.232350e+05 5.200430e+03 1.388580e+03 9.372850e+03 3.371230e+03 1.450390e+03 2.678910e+03
-1332496830.533333 2.609030e+05 2.251100e+05 3.722580e+03 3.246660e+03 7.876540e+03 4.716810e+03 1.498440e+03 2.116520e+03
-1332496830.541667 2.544160e+05 2.237690e+05 4.841650e+03 2.956400e+03 8.115920e+03 5.392360e+03 2.142810e+03 2.652320e+03
-1332496830.550000 2.566980e+05 2.221720e+05 6.471230e+03 9.703960e+02 8.834980e+03 4.816840e+03 2.376630e+03 3.605860e+03
-1332496830.558333 2.618410e+05 2.235370e+05 5.500740e+03 1.189660e+03 8.365730e+03 4.016470e+03 1.042270e+03 3.821200e+03
-1332496830.566667 2.595030e+05 2.258400e+05 3.827930e+03 3.088840e+03 7.676140e+03 3.978310e+03 -3.570070e+02 3.016420e+03
-1332496830.575000 2.534570e+05 2.246360e+05 4.914610e+03 3.097450e+03 8.224900e+03 4.321440e+03 1.713740e+02 2.412360e+03
-1332496830.583333 2.560290e+05 2.222210e+05 6.841800e+03 1.028500e+03 9.252300e+03 4.387570e+03 2.418140e+03 2.510100e+03
-1332496830.591667 2.628400e+05 2.225500e+05 6.210250e+03 1.410730e+03 8.538900e+03 4.152580e+03 3.009300e+03 3.219760e+03
-1332496830.600000 2.616330e+05 2.250650e+05 4.284530e+03 3.357210e+03 7.282170e+03 3.823590e+03 1.402840e+03 3.644670e+03
-1332496830.608333 2.545910e+05 2.251090e+05 4.693160e+03 3.647740e+03 7.745160e+03 3.686380e+03 4.901610e+02 3.448860e+03
-1332496830.616667 2.547800e+05 2.235990e+05 6.527380e+03 1.569870e+03 9.438430e+03 3.456580e+03 1.162520e+03 3.252010e+03
-1332496830.625000 2.606390e+05 2.241070e+05 6.531050e+03 1.633050e+03 9.283720e+03 4.174020e+03 2.089550e+03 2.775750e+03
-1332496830.633333 2.611080e+05 2.254720e+05 4.968260e+03 3.527850e+03 7.692870e+03 5.137100e+03 2.207390e+03 2.436660e+03
-1332496830.641667 2.557750e+05 2.237080e+05 4.963450e+03 4.017370e+03 7.701420e+03 5.269650e+03 2.284400e+03 2.842080e+03
-1332496830.650000 2.573980e+05 2.209470e+05 6.767500e+03 1.645710e+03 9.107070e+03 4.000180e+03 2.548860e+03 3.624770e+03
-1332496830.658333 2.649240e+05 2.215590e+05 6.471460e+03 1.110330e+03 9.459650e+03 3.108170e+03 1.696970e+03 3.893440e+03
-1332496830.666667 2.653390e+05 2.257330e+05 4.348800e+03 3.459510e+03 8.475300e+03 4.031240e+03 5.733470e+02 2.910270e+03
-1332496830.675000 2.568140e+05 2.269950e+05 3.479540e+03 4.949790e+03 7.499910e+03 5.624710e+03 7.516560e+02 2.347710e+03
-1332496830.683333 2.533160e+05 2.251610e+05 5.147060e+03 3.218430e+03 8.460160e+03 5.869300e+03 2.336320e+03 2.987960e+03
-1332496830.691667 2.593600e+05 2.231010e+05 5.549120e+03 1.869950e+03 8.740760e+03 4.668940e+03 2.457910e+03 3.758820e+03
-1332496830.700000 2.620120e+05 2.240160e+05 4.173610e+03 3.004130e+03 8.157040e+03 3.704730e+03 9.879640e+02 3.652750e+03
-1332496830.708333 2.571760e+05 2.244200e+05 3.517300e+03 4.118750e+03 7.822240e+03 3.718230e+03 3.726490e+01 2.953680e+03
-1332496830.716667 2.551460e+05 2.233220e+05 4.923980e+03 2.330680e+03 9.095910e+03 3.792400e+03 1.013070e+03 2.711240e+03
-1332496830.725000 2.605240e+05 2.236510e+05 5.413630e+03 1.146210e+03 8.817170e+03 4.419650e+03 2.446650e+03 2.832050e+03
-1332496830.733333 2.620980e+05 2.257520e+05 4.262980e+03 2.270970e+03 7.135480e+03 5.067120e+03 2.294680e+03 3.376620e+03
-1332496830.741667 2.568890e+05 2.253790e+05 3.606460e+03 3.568190e+03 6.552650e+03 4.970270e+03 1.516380e+03 3.662570e+03
-1332496830.750000 2.539480e+05 2.226310e+05 5.511700e+03 2.066300e+03 7.952660e+03 4.019910e+03 1.513140e+03 3.752630e+03
-1332496830.758333 2.597990e+05 2.220670e+05 5.873500e+03 6.085840e+02 9.253780e+03 2.870740e+03 1.348240e+03 3.344200e+03
-1332496830.766667 2.625470e+05 2.249010e+05 4.346080e+03 1.928100e+03 8.590970e+03 3.455460e+03 9.043910e+02 2.379270e+03
-1332496830.775000 2.561370e+05 2.267610e+05 3.423560e+03 3.379080e+03 7.471150e+03 4.894170e+03 1.153540e+03 2.031410e+03
-1332496830.783333 2.503260e+05 2.250130e+05 5.519980e+03 2.423970e+03 7.991760e+03 5.117950e+03 2.098790e+03 3.099240e+03
-1332496830.791667 2.554540e+05 2.229920e+05 6.547950e+03 4.964960e+02 8.751340e+03 3.900560e+03 2.132290e+03 4.076810e+03
-1332496830.800000 2.612860e+05 2.234890e+05 5.152850e+03 1.501510e+03 8.425610e+03 2.888030e+03 7.761140e+02 3.786360e+03
-1332496830.808333 2.589690e+05 2.240690e+05 3.832610e+03 3.001980e+03 7.979260e+03 3.182310e+03 5.271600e+01 2.874800e+03
-1332496830.816667 2.549460e+05 2.220350e+05 5.317880e+03 2.139800e+03 9.103140e+03 3.955610e+03 1.235170e+03 2.394150e+03
-1332496830.825000 2.586760e+05 2.212050e+05 6.594910e+03 5.053440e+02 9.423360e+03 4.562470e+03 2.913740e+03 2.892350e+03
-1332496830.833333 2.621250e+05 2.235660e+05 5.116750e+03 1.773600e+03 8.082200e+03 4.776370e+03 2.386390e+03 3.659730e+03
-1332496830.841667 2.578350e+05 2.259180e+05 3.714300e+03 3.477080e+03 7.205370e+03 4.554610e+03 7.115390e+02 3.878420e+03
-1332496830.850000 2.536600e+05 2.243710e+05 5.022450e+03 2.592430e+03 8.277200e+03 4.119370e+03 4.865080e+02 3.666740e+03
-1332496830.858333 2.595030e+05 2.220610e+05 6.589950e+03 6.599360e+02 9.596920e+03 3.598100e+03 1.702490e+03 3.036600e+03
-1332496830.866667 2.654950e+05 2.228430e+05 5.541850e+03 1.728430e+03 8.459960e+03 4.492000e+03 2.231970e+03 2.430620e+03
-1332496830.875000 2.609290e+05 2.249960e+05 4.000950e+03 3.745990e+03 6.983790e+03 5.430860e+03 1.855260e+03 2.533380e+03
-1332496830.883333 2.527160e+05 2.243350e+05 5.086560e+03 3.401150e+03 7.597970e+03 5.196120e+03 1.755720e+03 3.079760e+03
-1332496830.891667 2.541100e+05 2.231110e+05 6.822190e+03 1.229080e+03 9.164340e+03 3.761230e+03 1.679390e+03 3.584880e+03
-1332496830.900000 2.599690e+05 2.246930e+05 6.183950e+03 1.538500e+03 9.222080e+03 3.139170e+03 9.499020e+02 3.180800e+03
-1332496830.908333 2.590780e+05 2.269130e+05 4.388890e+03 3.694820e+03 8.195020e+03 3.933000e+03 4.260800e+02 2.388450e+03
-1332496830.916667 2.545630e+05 2.247600e+05 5.168440e+03 4.020940e+03 8.450270e+03 4.758910e+03 1.458900e+03 2.286430e+03
-1332496830.925000 2.580590e+05 2.212170e+05 6.883460e+03 1.649530e+03 9.232780e+03 4.457650e+03 3.057820e+03 3.031950e+03
-1332496830.933333 2.646670e+05 2.211770e+05 6.218510e+03 1.645730e+03 8.657180e+03 3.663500e+03 2.528280e+03 3.978340e+03
-1332496830.941667 2.629250e+05 2.243820e+05 4.627500e+03 3.635930e+03 7.892800e+03 3.431320e+03 6.045090e+02 3.901370e+03
-1332496830.950000 2.547080e+05 2.254480e+05 4.408250e+03 4.461040e+03 8.197170e+03 3.953750e+03 -4.453460e+01 3.154870e+03
-1332496830.958333 2.537020e+05 2.246350e+05 5.825770e+03 2.577050e+03 9.590050e+03 4.569250e+03 1.460270e+03 2.785170e+03
-1332496830.966667 2.602060e+05 2.241400e+05 5.387980e+03 1.951160e+03 8.789510e+03 5.131660e+03 2.706380e+03 2.972480e+03
-1332496830.975000 2.612400e+05 2.247370e+05 3.860810e+03 3.418310e+03 7.414530e+03 5.284520e+03 2.271380e+03 3.183150e+03
-1332496830.983333 2.561400e+05 2.232520e+05 3.850010e+03 3.957140e+03 7.262650e+03 4.964640e+03 1.499510e+03 3.453130e+03
-1332496830.991667 2.561160e+05 2.213490e+05 5.594480e+03 2.054400e+03 8.835130e+03 3.662010e+03 1.485510e+03 3.613010e+03
+1332496830000000 2.517740e+05 2.242410e+05 5.688100e+03 1.915530e+03 9.329220e+03 4.183710e+03 1.212350e+03 2.641790e+03
+1332496830008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
+1332496830016667 2.630730e+05 2.233040e+05 4.961640e+03 2.197120e+03 7.687310e+03 4.861860e+03 2.732780e+03 3.008540e+03
+1332496830025000 2.576140e+05 2.233230e+05 5.003660e+03 3.525140e+03 7.165310e+03 4.685620e+03 1.715380e+03 3.440480e+03
+1332496830033333 2.557800e+05 2.219150e+05 6.357310e+03 2.145290e+03 8.426970e+03 3.775350e+03 1.475390e+03 3.797240e+03
+1332496830041667 2.601660e+05 2.230080e+05 6.702590e+03 1.484960e+03 9.288100e+03 3.330830e+03 1.228500e+03 3.214320e+03
+1332496830050000 2.612310e+05 2.264260e+05 4.980060e+03 2.982380e+03 8.499630e+03 4.267670e+03 9.940890e+02 2.292890e+03
+1332496830058333 2.551170e+05 2.266420e+05 4.584410e+03 4.656440e+03 7.860150e+03 5.317310e+03 1.473600e+03 2.111690e+03
+1332496830066667 2.533000e+05 2.235540e+05 6.455090e+03 3.036650e+03 8.869750e+03 4.986310e+03 2.607360e+03 2.839590e+03
+1332496830075000 2.610610e+05 2.212630e+05 6.951980e+03 1.500240e+03 9.386100e+03 3.791680e+03 2.677010e+03 3.980630e+03
+1332496830083333 2.665030e+05 2.231980e+05 5.189610e+03 2.594560e+03 8.571530e+03 3.175000e+03 9.198400e+02 3.792010e+03
+1332496830091667 2.606920e+05 2.251840e+05 3.782480e+03 4.642880e+03 7.662960e+03 3.917790e+03 -2.510970e+02 2.907060e+03
+1332496830100000 2.539630e+05 2.250810e+05 5.123530e+03 3.839550e+03 8.669030e+03 4.877820e+03 9.437240e+02 2.527450e+03
+1332496830108333 2.565550e+05 2.241690e+05 5.930600e+03 2.298540e+03 8.906710e+03 5.331680e+03 2.549910e+03 3.053560e+03
+1332496830116667 2.608890e+05 2.250100e+05 4.681130e+03 2.971870e+03 7.900040e+03 4.874080e+03 2.322430e+03 3.649120e+03
+1332496830125000 2.579440e+05 2.249230e+05 3.291140e+03 4.357090e+03 7.131590e+03 4.385560e+03 1.077050e+03 3.664040e+03
+1332496830133333 2.550090e+05 2.230180e+05 4.584820e+03 2.864000e+03 8.469490e+03 3.625580e+03 9.855570e+02 3.504230e+03
+1332496830141667 2.601140e+05 2.219470e+05 5.676190e+03 1.210340e+03 9.393780e+03 3.390240e+03 1.654020e+03 3.018700e+03
+1332496830150000 2.642770e+05 2.244380e+05 4.446620e+03 2.176720e+03 8.142090e+03 4.584880e+03 2.327830e+03 2.615800e+03
+1332496830158333 2.592210e+05 2.264710e+05 2.734440e+03 4.182760e+03 6.389550e+03 5.540520e+03 1.958880e+03 2.720120e+03
+1332496830166667 2.526500e+05 2.248310e+05 4.163640e+03 2.989990e+03 7.179200e+03 5.213060e+03 1.929550e+03 3.457660e+03
+1332496830175000 2.570830e+05 2.220480e+05 5.759040e+03 7.024410e+02 8.566550e+03 3.552020e+03 1.832940e+03 3.956190e+03
+1332496830183333 2.631300e+05 2.229670e+05 5.141140e+03 1.166120e+03 8.666960e+03 2.720370e+03 9.713740e+02 3.479730e+03
+1332496830191667 2.602360e+05 2.252650e+05 3.425140e+03 3.339080e+03 7.853610e+03 3.674950e+03 5.259080e+02 2.443310e+03
+1332496830200000 2.535030e+05 2.245270e+05 4.398130e+03 2.927430e+03 8.110280e+03 4.842470e+03 1.513870e+03 2.467100e+03
+1332496830208333 2.561260e+05 2.226930e+05 6.043530e+03 6.562240e+02 8.797560e+03 4.832410e+03 2.832370e+03 3.426140e+03
+1332496830216667 2.616770e+05 2.236080e+05 5.830460e+03 1.033910e+03 8.123940e+03 3.980690e+03 1.927960e+03 4.092720e+03
+1332496830225000 2.594570e+05 2.255360e+05 4.015570e+03 2.995990e+03 7.135440e+03 3.713550e+03 3.072200e+02 3.849430e+03
+1332496830233333 2.533520e+05 2.242160e+05 4.650560e+03 3.196620e+03 8.131280e+03 3.586160e+03 7.083230e+01 3.074180e+03
+1332496830241667 2.561240e+05 2.215130e+05 6.100480e+03 8.219800e+02 9.757540e+03 3.474510e+03 1.647520e+03 2.559860e+03
+1332496830250000 2.630240e+05 2.215590e+05 5.789960e+03 6.994170e+02 9.129740e+03 4.153080e+03 2.829250e+03 2.677270e+03
+1332496830258333 2.617200e+05 2.240150e+05 4.358500e+03 2.645360e+03 7.414110e+03 4.810670e+03 2.225990e+03 3.185990e+03
+1332496830266667 2.547560e+05 2.242400e+05 4.857380e+03 3.229680e+03 7.539310e+03 4.769140e+03 1.507130e+03 3.668260e+03
+1332496830275000 2.568890e+05 2.226580e+05 6.473420e+03 1.214110e+03 9.010760e+03 3.848730e+03 1.303840e+03 3.778500e+03
+1332496830283333 2.642080e+05 2.233160e+05 5.700450e+03 1.116560e+03 9.087610e+03 3.846680e+03 1.293590e+03 2.891560e+03
+1332496830291667 2.633100e+05 2.257190e+05 3.936120e+03 3.252360e+03 7.552850e+03 4.897860e+03 1.156630e+03 2.037160e+03
+1332496830300000 2.550790e+05 2.250860e+05 4.536450e+03 3.960110e+03 7.454590e+03 5.479070e+03 1.596360e+03 2.190800e+03
+1332496830308333 2.544870e+05 2.225080e+05 6.635860e+03 1.758850e+03 8.732970e+03 4.466970e+03 2.650360e+03 3.139310e+03
+1332496830316667 2.612410e+05 2.224320e+05 6.702270e+03 1.085130e+03 8.989230e+03 3.112990e+03 1.933560e+03 3.828410e+03
+1332496830325000 2.621190e+05 2.255870e+05 4.714950e+03 2.892360e+03 8.107820e+03 2.961310e+03 2.399780e+02 3.273720e+03
+1332496830333333 2.549990e+05 2.265140e+05 4.532090e+03 4.126900e+03 8.200130e+03 3.872590e+03 5.608900e+01 2.370580e+03
+1332496830341667 2.542890e+05 2.240330e+05 6.538810e+03 2.251440e+03 9.419430e+03 4.564450e+03 2.077810e+03 2.508170e+03
+1332496830350000 2.618900e+05 2.219600e+05 6.846090e+03 1.475270e+03 9.125590e+03 4.598290e+03 3.299220e+03 3.475420e+03
+1332496830358333 2.645020e+05 2.230850e+05 5.066380e+03 3.270560e+03 7.933170e+03 4.173710e+03 1.908910e+03 3.867460e+03
+1332496830366667 2.578890e+05 2.236560e+05 4.201660e+03 4.473640e+03 7.688340e+03 4.161580e+03 6.875790e+02 3.653690e+03
+1332496830375000 2.542700e+05 2.231510e+05 5.715140e+03 2.752140e+03 9.273320e+03 3.772950e+03 8.964040e+02 3.256060e+03
+1332496830383333 2.582570e+05 2.242170e+05 6.114310e+03 1.856860e+03 9.604320e+03 4.200490e+03 1.764380e+03 2.939220e+03
+1332496830391667 2.600200e+05 2.268680e+05 4.237530e+03 3.605880e+03 8.066220e+03 5.430250e+03 2.138580e+03 2.696710e+03
+1332496830400000 2.550830e+05 2.259240e+05 3.350310e+03 4.853070e+03 7.045820e+03 5.925200e+03 1.893610e+03 2.897340e+03
+1332496830408333 2.544530e+05 2.221270e+05 5.271330e+03 2.491500e+03 8.436680e+03 5.032080e+03 2.436050e+03 3.724590e+03
+1332496830416667 2.625880e+05 2.199500e+05 5.994620e+03 7.892740e+02 9.029650e+03 3.515740e+03 1.953570e+03 4.014520e+03
+1332496830425000 2.656100e+05 2.233330e+05 4.391410e+03 2.400960e+03 8.146460e+03 3.536960e+03 5.302320e+02 3.133920e+03
+1332496830433333 2.574700e+05 2.269770e+05 2.975320e+03 4.633530e+03 7.278560e+03 4.640100e+03 -5.015020e+01 2.024960e+03
+1332496830441667 2.506870e+05 2.263310e+05 4.517860e+03 3.183800e+03 8.072600e+03 5.281660e+03 1.605140e+03 2.335140e+03
+1332496830450000 2.555630e+05 2.244950e+05 5.551000e+03 1.101300e+03 8.461490e+03 4.725700e+03 2.726670e+03 3.480540e+03
+1332496830458333 2.613350e+05 2.246450e+05 4.764680e+03 1.557020e+03 7.833350e+03 3.524810e+03 1.577410e+03 4.038620e+03
+1332496830466667 2.602690e+05 2.240080e+05 3.558030e+03 2.987610e+03 7.362440e+03 3.279230e+03 5.624420e+02 3.786550e+03
+1332496830475000 2.574350e+05 2.217770e+05 4.972600e+03 2.166880e+03 8.481440e+03 3.328720e+03 1.037130e+03 3.271370e+03
+1332496830483333 2.610460e+05 2.215500e+05 5.816180e+03 5.902170e+02 9.120930e+03 3.895400e+03 2.382670e+03 2.824170e+03
+1332496830491667 2.627660e+05 2.244730e+05 4.835050e+03 1.785770e+03 7.880760e+03 4.745620e+03 2.443660e+03 3.229550e+03
+1332496830500000 2.565090e+05 2.264130e+05 3.758870e+03 3.461200e+03 6.743770e+03 4.928960e+03 1.536620e+03 3.546690e+03
+1332496830508333 2.507930e+05 2.243720e+05 5.218490e+03 2.865260e+03 7.803960e+03 4.351090e+03 1.333820e+03 3.680490e+03
+1332496830516667 2.563190e+05 2.220660e+05 6.403970e+03 7.323450e+02 9.627760e+03 3.089300e+03 1.516780e+03 3.653690e+03
+1332496830525000 2.633430e+05 2.232350e+05 5.200430e+03 1.388580e+03 9.372850e+03 3.371230e+03 1.450390e+03 2.678910e+03
+1332496830533333 2.609030e+05 2.251100e+05 3.722580e+03 3.246660e+03 7.876540e+03 4.716810e+03 1.498440e+03 2.116520e+03
+1332496830541667 2.544160e+05 2.237690e+05 4.841650e+03 2.956400e+03 8.115920e+03 5.392360e+03 2.142810e+03 2.652320e+03
+1332496830550000 2.566980e+05 2.221720e+05 6.471230e+03 9.703960e+02 8.834980e+03 4.816840e+03 2.376630e+03 3.605860e+03
+1332496830558333 2.618410e+05 2.235370e+05 5.500740e+03 1.189660e+03 8.365730e+03 4.016470e+03 1.042270e+03 3.821200e+03
+1332496830566667 2.595030e+05 2.258400e+05 3.827930e+03 3.088840e+03 7.676140e+03 3.978310e+03 -3.570070e+02 3.016420e+03
+1332496830575000 2.534570e+05 2.246360e+05 4.914610e+03 3.097450e+03 8.224900e+03 4.321440e+03 1.713740e+02 2.412360e+03
+1332496830583333 2.560290e+05 2.222210e+05 6.841800e+03 1.028500e+03 9.252300e+03 4.387570e+03 2.418140e+03 2.510100e+03
+1332496830591667 2.628400e+05 2.225500e+05 6.210250e+03 1.410730e+03 8.538900e+03 4.152580e+03 3.009300e+03 3.219760e+03
+1332496830600000 2.616330e+05 2.250650e+05 4.284530e+03 3.357210e+03 7.282170e+03 3.823590e+03 1.402840e+03 3.644670e+03
+1332496830608333 2.545910e+05 2.251090e+05 4.693160e+03 3.647740e+03 7.745160e+03 3.686380e+03 4.901610e+02 3.448860e+03
+1332496830616667 2.547800e+05 2.235990e+05 6.527380e+03 1.569870e+03 9.438430e+03 3.456580e+03 1.162520e+03 3.252010e+03
+1332496830625000 2.606390e+05 2.241070e+05 6.531050e+03 1.633050e+03 9.283720e+03 4.174020e+03 2.089550e+03 2.775750e+03
+1332496830633333 2.611080e+05 2.254720e+05 4.968260e+03 3.527850e+03 7.692870e+03 5.137100e+03 2.207390e+03 2.436660e+03
+1332496830641667 2.557750e+05 2.237080e+05 4.963450e+03 4.017370e+03 7.701420e+03 5.269650e+03 2.284400e+03 2.842080e+03
+1332496830650000 2.573980e+05 2.209470e+05 6.767500e+03 1.645710e+03 9.107070e+03 4.000180e+03 2.548860e+03 3.624770e+03
+1332496830658333 2.649240e+05 2.215590e+05 6.471460e+03 1.110330e+03 9.459650e+03 3.108170e+03 1.696970e+03 3.893440e+03
+1332496830666667 2.653390e+05 2.257330e+05 4.348800e+03 3.459510e+03 8.475300e+03 4.031240e+03 5.733470e+02 2.910270e+03
+1332496830675000 2.568140e+05 2.269950e+05 3.479540e+03 4.949790e+03 7.499910e+03 5.624710e+03 7.516560e+02 2.347710e+03
+1332496830683333 2.533160e+05 2.251610e+05 5.147060e+03 3.218430e+03 8.460160e+03 5.869300e+03 2.336320e+03 2.987960e+03
+1332496830691667 2.593600e+05 2.231010e+05 5.549120e+03 1.869950e+03 8.740760e+03 4.668940e+03 2.457910e+03 3.758820e+03
+1332496830700000 2.620120e+05 2.240160e+05 4.173610e+03 3.004130e+03 8.157040e+03 3.704730e+03 9.879640e+02 3.652750e+03
+1332496830708333 2.571760e+05 2.244200e+05 3.517300e+03 4.118750e+03 7.822240e+03 3.718230e+03 3.726490e+01 2.953680e+03
+1332496830716667 2.551460e+05 2.233220e+05 4.923980e+03 2.330680e+03 9.095910e+03 3.792400e+03 1.013070e+03 2.711240e+03
+1332496830725000 2.605240e+05 2.236510e+05 5.413630e+03 1.146210e+03 8.817170e+03 4.419650e+03 2.446650e+03 2.832050e+03
+1332496830733333 2.620980e+05 2.257520e+05 4.262980e+03 2.270970e+03 7.135480e+03 5.067120e+03 2.294680e+03 3.376620e+03
+1332496830741667 2.568890e+05 2.253790e+05 3.606460e+03 3.568190e+03 6.552650e+03 4.970270e+03 1.516380e+03 3.662570e+03
+1332496830750000 2.539480e+05 2.226310e+05 5.511700e+03 2.066300e+03 7.952660e+03 4.019910e+03 1.513140e+03 3.752630e+03
+1332496830758333 2.597990e+05 2.220670e+05 5.873500e+03 6.085840e+02 9.253780e+03 2.870740e+03 1.348240e+03 3.344200e+03
+1332496830766667 2.625470e+05 2.249010e+05 4.346080e+03 1.928100e+03 8.590970e+03 3.455460e+03 9.043910e+02 2.379270e+03
+1332496830775000 2.561370e+05 2.267610e+05 3.423560e+03 3.379080e+03 7.471150e+03 4.894170e+03 1.153540e+03 2.031410e+03
+1332496830783333 2.503260e+05 2.250130e+05 5.519980e+03 2.423970e+03 7.991760e+03 5.117950e+03 2.098790e+03 3.099240e+03
+1332496830791667 2.554540e+05 2.229920e+05 6.547950e+03 4.964960e+02 8.751340e+03 3.900560e+03 2.132290e+03 4.076810e+03
+1332496830800000 2.612860e+05 2.234890e+05 5.152850e+03 1.501510e+03 8.425610e+03 2.888030e+03 7.761140e+02 3.786360e+03
+1332496830808333 2.589690e+05 2.240690e+05 3.832610e+03 3.001980e+03 7.979260e+03 3.182310e+03 5.271600e+01 2.874800e+03
+1332496830816667 2.549460e+05 2.220350e+05 5.317880e+03 2.139800e+03 9.103140e+03 3.955610e+03 1.235170e+03 2.394150e+03
+1332496830825000 2.586760e+05 2.212050e+05 6.594910e+03 5.053440e+02 9.423360e+03 4.562470e+03 2.913740e+03 2.892350e+03
+1332496830833333 2.621250e+05 2.235660e+05 5.116750e+03 1.773600e+03 8.082200e+03 4.776370e+03 2.386390e+03 3.659730e+03
+1332496830841667 2.578350e+05 2.259180e+05 3.714300e+03 3.477080e+03 7.205370e+03 4.554610e+03 7.115390e+02 3.878420e+03
+1332496830850000 2.536600e+05 2.243710e+05 5.022450e+03 2.592430e+03 8.277200e+03 4.119370e+03 4.865080e+02 3.666740e+03
+1332496830858333 2.595030e+05 2.220610e+05 6.589950e+03 6.599360e+02 9.596920e+03 3.598100e+03 1.702490e+03 3.036600e+03
+1332496830866667 2.654950e+05 2.228430e+05 5.541850e+03 1.728430e+03 8.459960e+03 4.492000e+03 2.231970e+03 2.430620e+03
+1332496830875000 2.609290e+05 2.249960e+05 4.000950e+03 3.745990e+03 6.983790e+03 5.430860e+03 1.855260e+03 2.533380e+03
+1332496830883333 2.527160e+05 2.243350e+05 5.086560e+03 3.401150e+03 7.597970e+03 5.196120e+03 1.755720e+03 3.079760e+03
+1332496830891667 2.541100e+05 2.231110e+05 6.822190e+03 1.229080e+03 9.164340e+03 3.761230e+03 1.679390e+03 3.584880e+03
+1332496830900000 2.599690e+05 2.246930e+05 6.183950e+03 1.538500e+03 9.222080e+03 3.139170e+03 9.499020e+02 3.180800e+03
+1332496830908333 2.590780e+05 2.269130e+05 4.388890e+03 3.694820e+03 8.195020e+03 3.933000e+03 4.260800e+02 2.388450e+03
+1332496830916667 2.545630e+05 2.247600e+05 5.168440e+03 4.020940e+03 8.450270e+03 4.758910e+03 1.458900e+03 2.286430e+03
+1332496830925000 2.580590e+05 2.212170e+05 6.883460e+03 1.649530e+03 9.232780e+03 4.457650e+03 3.057820e+03 3.031950e+03
+1332496830933333 2.646670e+05 2.211770e+05 6.218510e+03 1.645730e+03 8.657180e+03 3.663500e+03 2.528280e+03 3.978340e+03
+1332496830941667 2.629250e+05 2.243820e+05 4.627500e+03 3.635930e+03 7.892800e+03 3.431320e+03 6.045090e+02 3.901370e+03
+1332496830950000 2.547080e+05 2.254480e+05 4.408250e+03 4.461040e+03 8.197170e+03 3.953750e+03 -4.453460e+01 3.154870e+03
+1332496830958333 2.537020e+05 2.246350e+05 5.825770e+03 2.577050e+03 9.590050e+03 4.569250e+03 1.460270e+03 2.785170e+03
+1332496830966667 2.602060e+05 2.241400e+05 5.387980e+03 1.951160e+03 8.789510e+03 5.131660e+03 2.706380e+03 2.972480e+03
+1332496830975000 2.612400e+05 2.247370e+05 3.860810e+03 3.418310e+03 7.414530e+03 5.284520e+03 2.271380e+03 3.183150e+03
+1332496830983333 2.561400e+05 2.232520e+05 3.850010e+03 3.957140e+03 7.262650e+03 4.964640e+03 1.499510e+03 3.453130e+03
+1332496830991667 2.561160e+05 2.213490e+05 5.594480e+03 2.054400e+03 8.835130e+03 3.662010e+03 1.485510e+03 3.613010e+03
--- a/tests/data/extract-2
+++ b/tests/data/extract-2
@@ -1,119 +1,119 @@
-1332496830.008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
-1332496830.016667 2.630730e+05 2.233040e+05 4.961640e+03 2.197120e+03 7.687310e+03 4.861860e+03 2.732780e+03 3.008540e+03
-1332496830.025000 2.576140e+05 2.233230e+05 5.003660e+03 3.525140e+03 7.165310e+03 4.685620e+03 1.715380e+03 3.440480e+03
-1332496830.033333 2.557800e+05 2.219150e+05 6.357310e+03 2.145290e+03 8.426970e+03 3.775350e+03 1.475390e+03 3.797240e+03
-1332496830.041667 2.601660e+05 2.230080e+05 6.702590e+03 1.484960e+03 9.288100e+03 3.330830e+03 1.228500e+03 3.214320e+03
-1332496830.050000 2.612310e+05 2.264260e+05 4.980060e+03 2.982380e+03 8.499630e+03 4.267670e+03 9.940890e+02 2.292890e+03
-1332496830.058333 2.551170e+05 2.266420e+05 4.584410e+03 4.656440e+03 7.860150e+03 5.317310e+03 1.473600e+03 2.111690e+03
-1332496830.066667 2.533000e+05 2.235540e+05 6.455090e+03 3.036650e+03 8.869750e+03 4.986310e+03 2.607360e+03 2.839590e+03
-1332496830.075000 2.610610e+05 2.212630e+05 6.951980e+03 1.500240e+03 9.386100e+03 3.791680e+03 2.677010e+03 3.980630e+03
-1332496830.083333 2.665030e+05 2.231980e+05 5.189610e+03 2.594560e+03 8.571530e+03 3.175000e+03 9.198400e+02 3.792010e+03
-1332496830.091667 2.606920e+05 2.251840e+05 3.782480e+03 4.642880e+03 7.662960e+03 3.917790e+03 -2.510970e+02 2.907060e+03
-1332496830.100000 2.539630e+05 2.250810e+05 5.123530e+03 3.839550e+03 8.669030e+03 4.877820e+03 9.437240e+02 2.527450e+03
-1332496830.108333 2.565550e+05 2.241690e+05 5.930600e+03 2.298540e+03 8.906710e+03 5.331680e+03 2.549910e+03 3.053560e+03
-1332496830.116667 2.608890e+05 2.250100e+05 4.681130e+03 2.971870e+03 7.900040e+03 4.874080e+03 2.322430e+03 3.649120e+03
-1332496830.125000 2.579440e+05 2.249230e+05 3.291140e+03 4.357090e+03 7.131590e+03 4.385560e+03 1.077050e+03 3.664040e+03
-1332496830.133333 2.550090e+05 2.230180e+05 4.584820e+03 2.864000e+03 8.469490e+03 3.625580e+03 9.855570e+02 3.504230e+03
-1332496830.141667 2.601140e+05 2.219470e+05 5.676190e+03 1.210340e+03 9.393780e+03 3.390240e+03 1.654020e+03 3.018700e+03
-1332496830.150000 2.642770e+05 2.244380e+05 4.446620e+03 2.176720e+03 8.142090e+03 4.584880e+03 2.327830e+03 2.615800e+03
-1332496830.158333 2.592210e+05 2.264710e+05 2.734440e+03 4.182760e+03 6.389550e+03 5.540520e+03 1.958880e+03 2.720120e+03
-1332496830.166667 2.526500e+05 2.248310e+05 4.163640e+03 2.989990e+03 7.179200e+03 5.213060e+03 1.929550e+03 3.457660e+03
-1332496830.175000 2.570830e+05 2.220480e+05 5.759040e+03 7.024410e+02 8.566550e+03 3.552020e+03 1.832940e+03 3.956190e+03
-1332496830.183333 2.631300e+05 2.229670e+05 5.141140e+03 1.166120e+03 8.666960e+03 2.720370e+03 9.713740e+02 3.479730e+03
-1332496830.191667 2.602360e+05 2.252650e+05 3.425140e+03 3.339080e+03 7.853610e+03 3.674950e+03 5.259080e+02 2.443310e+03
-1332496830.200000 2.535030e+05 2.245270e+05 4.398130e+03 2.927430e+03 8.110280e+03 4.842470e+03 1.513870e+03 2.467100e+03
-1332496830.208333 2.561260e+05 2.226930e+05 6.043530e+03 6.562240e+02 8.797560e+03 4.832410e+03 2.832370e+03 3.426140e+03
-1332496830.216667 2.616770e+05 2.236080e+05 5.830460e+03 1.033910e+03 8.123940e+03 3.980690e+03 1.927960e+03 4.092720e+03
-1332496830.225000 2.594570e+05 2.255360e+05 4.015570e+03 2.995990e+03 7.135440e+03 3.713550e+03 3.072200e+02 3.849430e+03
-1332496830.233333 2.533520e+05 2.242160e+05 4.650560e+03 3.196620e+03 8.131280e+03 3.586160e+03 7.083230e+01 3.074180e+03
-1332496830.241667 2.561240e+05 2.215130e+05 6.100480e+03 8.219800e+02 9.757540e+03 3.474510e+03 1.647520e+03 2.559860e+03
-1332496830.250000 2.630240e+05 2.215590e+05 5.789960e+03 6.994170e+02 9.129740e+03 4.153080e+03 2.829250e+03 2.677270e+03
-1332496830.258333 2.617200e+05 2.240150e+05 4.358500e+03 2.645360e+03 7.414110e+03 4.810670e+03 2.225990e+03 3.185990e+03
-1332496830.266667 2.547560e+05 2.242400e+05 4.857380e+03 3.229680e+03 7.539310e+03 4.769140e+03 1.507130e+03 3.668260e+03
-1332496830.275000 2.568890e+05 2.226580e+05 6.473420e+03 1.214110e+03 9.010760e+03 3.848730e+03 1.303840e+03 3.778500e+03
-1332496830.283333 2.642080e+05 2.233160e+05 5.700450e+03 1.116560e+03 9.087610e+03 3.846680e+03 1.293590e+03 2.891560e+03
-1332496830.291667 2.633100e+05 2.257190e+05 3.936120e+03 3.252360e+03 7.552850e+03 4.897860e+03 1.156630e+03 2.037160e+03
-1332496830.300000 2.550790e+05 2.250860e+05 4.536450e+03 3.960110e+03 7.454590e+03 5.479070e+03 1.596360e+03 2.190800e+03
-1332496830.308333 2.544870e+05 2.225080e+05 6.635860e+03 1.758850e+03 8.732970e+03 4.466970e+03 2.650360e+03 3.139310e+03
-1332496830.316667 2.612410e+05 2.224320e+05 6.702270e+03 1.085130e+03 8.989230e+03 3.112990e+03 1.933560e+03 3.828410e+03
-1332496830.325000 2.621190e+05 2.255870e+05 4.714950e+03 2.892360e+03 8.107820e+03 2.961310e+03 2.399780e+02 3.273720e+03
-1332496830.333333 2.549990e+05 2.265140e+05 4.532090e+03 4.126900e+03 8.200130e+03 3.872590e+03 5.608900e+01 2.370580e+03
-1332496830.341667 2.542890e+05 2.240330e+05 6.538810e+03 2.251440e+03 9.419430e+03 4.564450e+03 2.077810e+03 2.508170e+03
-1332496830.350000 2.618900e+05 2.219600e+05 6.846090e+03 1.475270e+03 9.125590e+03 4.598290e+03 3.299220e+03 3.475420e+03
-1332496830.358333 2.645020e+05 2.230850e+05 5.066380e+03 3.270560e+03 7.933170e+03 4.173710e+03 1.908910e+03 3.867460e+03
-1332496830.366667 2.578890e+05 2.236560e+05 4.201660e+03 4.473640e+03 7.688340e+03 4.161580e+03 6.875790e+02 3.653690e+03
-1332496830.375000 2.542700e+05 2.231510e+05 5.715140e+03 2.752140e+03 9.273320e+03 3.772950e+03 8.964040e+02 3.256060e+03
-1332496830.383333 2.582570e+05 2.242170e+05 6.114310e+03 1.856860e+03 9.604320e+03 4.200490e+03 1.764380e+03 2.939220e+03
-1332496830.391667 2.600200e+05 2.268680e+05 4.237530e+03 3.605880e+03 8.066220e+03 5.430250e+03 2.138580e+03 2.696710e+03
-1332496830.400000 2.550830e+05 2.259240e+05 3.350310e+03 4.853070e+03 7.045820e+03 5.925200e+03 1.893610e+03 2.897340e+03
-1332496830.408333 2.544530e+05 2.221270e+05 5.271330e+03 2.491500e+03 8.436680e+03 5.032080e+03 2.436050e+03 3.724590e+03
-1332496830.416667 2.625880e+05 2.199500e+05 5.994620e+03 7.892740e+02 9.029650e+03 3.515740e+03 1.953570e+03 4.014520e+03
-1332496830.425000 2.656100e+05 2.233330e+05 4.391410e+03 2.400960e+03 8.146460e+03 3.536960e+03 5.302320e+02 3.133920e+03
-1332496830.433333 2.574700e+05 2.269770e+05 2.975320e+03 4.633530e+03 7.278560e+03 4.640100e+03 -5.015020e+01 2.024960e+03
-1332496830.441667 2.506870e+05 2.263310e+05 4.517860e+03 3.183800e+03 8.072600e+03 5.281660e+03 1.605140e+03 2.335140e+03
-1332496830.450000 2.555630e+05 2.244950e+05 5.551000e+03 1.101300e+03 8.461490e+03 4.725700e+03 2.726670e+03 3.480540e+03
-1332496830.458333 2.613350e+05 2.246450e+05 4.764680e+03 1.557020e+03 7.833350e+03 3.524810e+03 1.577410e+03 4.038620e+03
-1332496830.466667 2.602690e+05 2.240080e+05 3.558030e+03 2.987610e+03 7.362440e+03 3.279230e+03 5.624420e+02 3.786550e+03
-1332496830.475000 2.574350e+05 2.217770e+05 4.972600e+03 2.166880e+03 8.481440e+03 3.328720e+03 1.037130e+03 3.271370e+03
-1332496830.483333 2.610460e+05 2.215500e+05 5.816180e+03 5.902170e+02 9.120930e+03 3.895400e+03 2.382670e+03 2.824170e+03
-1332496830.491667 2.627660e+05 2.244730e+05 4.835050e+03 1.785770e+03 7.880760e+03 4.745620e+03 2.443660e+03 3.229550e+03
-1332496830.500000 2.565090e+05 2.264130e+05 3.758870e+03 3.461200e+03 6.743770e+03 4.928960e+03 1.536620e+03 3.546690e+03
-1332496830.508333 2.507930e+05 2.243720e+05 5.218490e+03 2.865260e+03 7.803960e+03 4.351090e+03 1.333820e+03 3.680490e+03
-1332496830.516667 2.563190e+05 2.220660e+05 6.403970e+03 7.323450e+02 9.627760e+03 3.089300e+03 1.516780e+03 3.653690e+03
-1332496830.525000 2.633430e+05 2.232350e+05 5.200430e+03 1.388580e+03 9.372850e+03 3.371230e+03 1.450390e+03 2.678910e+03
-1332496830.533333 2.609030e+05 2.251100e+05 3.722580e+03 3.246660e+03 7.876540e+03 4.716810e+03 1.498440e+03 2.116520e+03
-1332496830.541667 2.544160e+05 2.237690e+05 4.841650e+03 2.956400e+03 8.115920e+03 5.392360e+03 2.142810e+03 2.652320e+03
-1332496830.550000 2.566980e+05 2.221720e+05 6.471230e+03 9.703960e+02 8.834980e+03 4.816840e+03 2.376630e+03 3.605860e+03
-1332496830.558333 2.618410e+05 2.235370e+05 5.500740e+03 1.189660e+03 8.365730e+03 4.016470e+03 1.042270e+03 3.821200e+03
-1332496830.566667 2.595030e+05 2.258400e+05 3.827930e+03 3.088840e+03 7.676140e+03 3.978310e+03 -3.570070e+02 3.016420e+03
-1332496830.575000 2.534570e+05 2.246360e+05 4.914610e+03 3.097450e+03 8.224900e+03 4.321440e+03 1.713740e+02 2.412360e+03
-1332496830.583333 2.560290e+05 2.222210e+05 6.841800e+03 1.028500e+03 9.252300e+03 4.387570e+03 2.418140e+03 2.510100e+03
-1332496830.591667 2.628400e+05 2.225500e+05 6.210250e+03 1.410730e+03 8.538900e+03 4.152580e+03 3.009300e+03 3.219760e+03
-1332496830.600000 2.616330e+05 2.250650e+05 4.284530e+03 3.357210e+03 7.282170e+03 3.823590e+03 1.402840e+03 3.644670e+03
-1332496830.608333 2.545910e+05 2.251090e+05 4.693160e+03 3.647740e+03 7.745160e+03 3.686380e+03 4.901610e+02 3.448860e+03
-1332496830.616667 2.547800e+05 2.235990e+05 6.527380e+03 1.569870e+03 9.438430e+03 3.456580e+03 1.162520e+03 3.252010e+03
-1332496830.625000 2.606390e+05 2.241070e+05 6.531050e+03 1.633050e+03 9.283720e+03 4.174020e+03 2.089550e+03 2.775750e+03
-1332496830.633333 2.611080e+05 2.254720e+05 4.968260e+03 3.527850e+03 7.692870e+03 5.137100e+03 2.207390e+03 2.436660e+03
-1332496830.641667 2.557750e+05 2.237080e+05 4.963450e+03 4.017370e+03 7.701420e+03 5.269650e+03 2.284400e+03 2.842080e+03
-1332496830.650000 2.573980e+05 2.209470e+05 6.767500e+03 1.645710e+03 9.107070e+03 4.000180e+03 2.548860e+03 3.624770e+03
-1332496830.658333 2.649240e+05 2.215590e+05 6.471460e+03 1.110330e+03 9.459650e+03 3.108170e+03 1.696970e+03 3.893440e+03
-1332496830.666667 2.653390e+05 2.257330e+05 4.348800e+03 3.459510e+03 8.475300e+03 4.031240e+03 5.733470e+02 2.910270e+03
-1332496830.675000 2.568140e+05 2.269950e+05 3.479540e+03 4.949790e+03 7.499910e+03 5.624710e+03 7.516560e+02 2.347710e+03
-1332496830.683333 2.533160e+05 2.251610e+05 5.147060e+03 3.218430e+03 8.460160e+03 5.869300e+03 2.336320e+03 2.987960e+03
-1332496830.691667 2.593600e+05 2.231010e+05 5.549120e+03 1.869950e+03 8.740760e+03 4.668940e+03 2.457910e+03 3.758820e+03
-1332496830.700000 2.620120e+05 2.240160e+05 4.173610e+03 3.004130e+03 8.157040e+03 3.704730e+03 9.879640e+02 3.652750e+03
-1332496830.708333 2.571760e+05 2.244200e+05 3.517300e+03 4.118750e+03 7.822240e+03 3.718230e+03 3.726490e+01 2.953680e+03
-1332496830.716667 2.551460e+05 2.233220e+05 4.923980e+03 2.330680e+03 9.095910e+03 3.792400e+03 1.013070e+03 2.711240e+03
-1332496830.725000 2.605240e+05 2.236510e+05 5.413630e+03 1.146210e+03 8.817170e+03 4.419650e+03 2.446650e+03 2.832050e+03
-1332496830.733333 2.620980e+05 2.257520e+05 4.262980e+03 2.270970e+03 7.135480e+03 5.067120e+03 2.294680e+03 3.376620e+03
-1332496830.741667 2.568890e+05 2.253790e+05 3.606460e+03 3.568190e+03 6.552650e+03 4.970270e+03 1.516380e+03 3.662570e+03
-1332496830.750000 2.539480e+05 2.226310e+05 5.511700e+03 2.066300e+03 7.952660e+03 4.019910e+03 1.513140e+03 3.752630e+03
-1332496830.758333 2.597990e+05 2.220670e+05 5.873500e+03 6.085840e+02 9.253780e+03 2.870740e+03 1.348240e+03 3.344200e+03
-1332496830.766667 2.625470e+05 2.249010e+05 4.346080e+03 1.928100e+03 8.590970e+03 3.455460e+03 9.043910e+02 2.379270e+03
-1332496830.775000 2.561370e+05 2.267610e+05 3.423560e+03 3.379080e+03 7.471150e+03 4.894170e+03 1.153540e+03 2.031410e+03
-1332496830.783333 2.503260e+05 2.250130e+05 5.519980e+03 2.423970e+03 7.991760e+03 5.117950e+03 2.098790e+03 3.099240e+03
-1332496830.791667 2.554540e+05 2.229920e+05 6.547950e+03 4.964960e+02 8.751340e+03 3.900560e+03 2.132290e+03 4.076810e+03
-1332496830.800000 2.612860e+05 2.234890e+05 5.152850e+03 1.501510e+03 8.425610e+03 2.888030e+03 7.761140e+02 3.786360e+03
-1332496830.808333 2.589690e+05 2.240690e+05 3.832610e+03 3.001980e+03 7.979260e+03 3.182310e+03 5.271600e+01 2.874800e+03
-1332496830.816667 2.549460e+05 2.220350e+05 5.317880e+03 2.139800e+03 9.103140e+03 3.955610e+03 1.235170e+03 2.394150e+03
-1332496830.825000 2.586760e+05 2.212050e+05 6.594910e+03 5.053440e+02 9.423360e+03 4.562470e+03 2.913740e+03 2.892350e+03
-1332496830.833333 2.621250e+05 2.235660e+05 5.116750e+03 1.773600e+03 8.082200e+03 4.776370e+03 2.386390e+03 3.659730e+03
-1332496830.841667 2.578350e+05 2.259180e+05 3.714300e+03 3.477080e+03 7.205370e+03 4.554610e+03 7.115390e+02 3.878420e+03
-1332496830.850000 2.536600e+05 2.243710e+05 5.022450e+03 2.592430e+03 8.277200e+03 4.119370e+03 4.865080e+02 3.666740e+03
-1332496830.858333 2.595030e+05 2.220610e+05 6.589950e+03 6.599360e+02 9.596920e+03 3.598100e+03 1.702490e+03 3.036600e+03
-1332496830.866667 2.654950e+05 2.228430e+05 5.541850e+03 1.728430e+03 8.459960e+03 4.492000e+03 2.231970e+03 2.430620e+03
-1332496830.875000 2.609290e+05 2.249960e+05 4.000950e+03 3.745990e+03 6.983790e+03 5.430860e+03 1.855260e+03 2.533380e+03
-1332496830.883333 2.527160e+05 2.243350e+05 5.086560e+03 3.401150e+03 7.597970e+03 5.196120e+03 1.755720e+03 3.079760e+03
-1332496830.891667 2.541100e+05 2.231110e+05 6.822190e+03 1.229080e+03 9.164340e+03 3.761230e+03 1.679390e+03 3.584880e+03
-1332496830.900000 2.599690e+05 2.246930e+05 6.183950e+03 1.538500e+03 9.222080e+03 3.139170e+03 9.499020e+02 3.180800e+03
-1332496830.908333 2.590780e+05 2.269130e+05 4.388890e+03 3.694820e+03 8.195020e+03 3.933000e+03 4.260800e+02 2.388450e+03
-1332496830.916667 2.545630e+05 2.247600e+05 5.168440e+03 4.020940e+03 8.450270e+03 4.758910e+03 1.458900e+03 2.286430e+03
-1332496830.925000 2.580590e+05 2.212170e+05 6.883460e+03 1.649530e+03 9.232780e+03 4.457650e+03 3.057820e+03 3.031950e+03
-1332496830.933333 2.646670e+05 2.211770e+05 6.218510e+03 1.645730e+03 8.657180e+03 3.663500e+03 2.528280e+03 3.978340e+03
-1332496830.941667 2.629250e+05 2.243820e+05 4.627500e+03 3.635930e+03 7.892800e+03 3.431320e+03 6.045090e+02 3.901370e+03
-1332496830.950000 2.547080e+05 2.254480e+05 4.408250e+03 4.461040e+03 8.197170e+03 3.953750e+03 -4.453460e+01 3.154870e+03
-1332496830.958333 2.537020e+05 2.246350e+05 5.825770e+03 2.577050e+03 9.590050e+03 4.569250e+03 1.460270e+03 2.785170e+03
-1332496830.966667 2.602060e+05 2.241400e+05 5.387980e+03 1.951160e+03 8.789510e+03 5.131660e+03 2.706380e+03 2.972480e+03
-1332496830.975000 2.612400e+05 2.247370e+05 3.860810e+03 3.418310e+03 7.414530e+03 5.284520e+03 2.271380e+03 3.183150e+03
-1332496830.983333 2.561400e+05 2.232520e+05 3.850010e+03 3.957140e+03 7.262650e+03 4.964640e+03 1.499510e+03 3.453130e+03
-1332496830.991667 2.561160e+05 2.213490e+05 5.594480e+03 2.054400e+03 8.835130e+03 3.662010e+03 1.485510e+03 3.613010e+03
+1332496830008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
+1332496830016667 2.630730e+05 2.233040e+05 4.961640e+03 2.197120e+03 7.687310e+03 4.861860e+03 2.732780e+03 3.008540e+03
+1332496830025000 2.576140e+05 2.233230e+05 5.003660e+03 3.525140e+03 7.165310e+03 4.685620e+03 1.715380e+03 3.440480e+03
+1332496830033333 2.557800e+05 2.219150e+05 6.357310e+03 2.145290e+03 8.426970e+03 3.775350e+03 1.475390e+03 3.797240e+03
+1332496830041667 2.601660e+05 2.230080e+05 6.702590e+03 1.484960e+03 9.288100e+03 3.330830e+03 1.228500e+03 3.214320e+03
+1332496830050000 2.612310e+05 2.264260e+05 4.980060e+03 2.982380e+03 8.499630e+03 4.267670e+03 9.940890e+02 2.292890e+03
+1332496830058333 2.551170e+05 2.266420e+05 4.584410e+03 4.656440e+03 7.860150e+03 5.317310e+03 1.473600e+03 2.111690e+03
+1332496830066667 2.533000e+05 2.235540e+05 6.455090e+03 3.036650e+03 8.869750e+03 4.986310e+03 2.607360e+03 2.839590e+03
+1332496830075000 2.610610e+05 2.212630e+05 6.951980e+03 1.500240e+03 9.386100e+03 3.791680e+03 2.677010e+03 3.980630e+03
+1332496830083333 2.665030e+05 2.231980e+05 5.189610e+03 2.594560e+03 8.571530e+03 3.175000e+03 9.198400e+02 3.792010e+03
+1332496830091667 2.606920e+05 2.251840e+05 3.782480e+03 4.642880e+03 7.662960e+03 3.917790e+03 -2.510970e+02 2.907060e+03
+1332496830100000 2.539630e+05 2.250810e+05 5.123530e+03 3.839550e+03 8.669030e+03 4.877820e+03 9.437240e+02 2.527450e+03
+1332496830108333 2.565550e+05 2.241690e+05 5.930600e+03 2.298540e+03 8.906710e+03 5.331680e+03 2.549910e+03 3.053560e+03
+1332496830116667 2.608890e+05 2.250100e+05 4.681130e+03 2.971870e+03 7.900040e+03 4.874080e+03 2.322430e+03 3.649120e+03
+1332496830125000 2.579440e+05 2.249230e+05 3.291140e+03 4.357090e+03 7.131590e+03 4.385560e+03 1.077050e+03 3.664040e+03
+1332496830133333 2.550090e+05 2.230180e+05 4.584820e+03 2.864000e+03 8.469490e+03 3.625580e+03 9.855570e+02 3.504230e+03
+1332496830141667 2.601140e+05 2.219470e+05 5.676190e+03 1.210340e+03 9.393780e+03 3.390240e+03 1.654020e+03 3.018700e+03
+1332496830150000 2.642770e+05 2.244380e+05 4.446620e+03 2.176720e+03 8.142090e+03 4.584880e+03 2.327830e+03 2.615800e+03
+1332496830158333 2.592210e+05 2.264710e+05 2.734440e+03 4.182760e+03 6.389550e+03 5.540520e+03 1.958880e+03 2.720120e+03
+1332496830166667 2.526500e+05 2.248310e+05 4.163640e+03 2.989990e+03 7.179200e+03 5.213060e+03 1.929550e+03 3.457660e+03
+1332496830175000 2.570830e+05 2.220480e+05 5.759040e+03 7.024410e+02 8.566550e+03 3.552020e+03 1.832940e+03 3.956190e+03
+1332496830183333 2.631300e+05 2.229670e+05 5.141140e+03 1.166120e+03 8.666960e+03 2.720370e+03 9.713740e+02 3.479730e+03
+1332496830191667 2.602360e+05 2.252650e+05 3.425140e+03 3.339080e+03 7.853610e+03 3.674950e+03 5.259080e+02 2.443310e+03
+1332496830200000 2.535030e+05 2.245270e+05 4.398130e+03 2.927430e+03 8.110280e+03 4.842470e+03 1.513870e+03 2.467100e+03
+1332496830208333 2.561260e+05 2.226930e+05 6.043530e+03 6.562240e+02 8.797560e+03 4.832410e+03 2.832370e+03 3.426140e+03
+1332496830216667 2.616770e+05 2.236080e+05 5.830460e+03 1.033910e+03 8.123940e+03 3.980690e+03 1.927960e+03 4.092720e+03
+1332496830225000 2.594570e+05 2.255360e+05 4.015570e+03 2.995990e+03 7.135440e+03 3.713550e+03 3.072200e+02 3.849430e+03
+1332496830233333 2.533520e+05 2.242160e+05 4.650560e+03 3.196620e+03 8.131280e+03 3.586160e+03 7.083230e+01 3.074180e+03
+1332496830241667 2.561240e+05 2.215130e+05 6.100480e+03 8.219800e+02 9.757540e+03 3.474510e+03 1.647520e+03 2.559860e+03
+1332496830250000 2.630240e+05 2.215590e+05 5.789960e+03 6.994170e+02 9.129740e+03 4.153080e+03 2.829250e+03 2.677270e+03
+1332496830258333 2.617200e+05 2.240150e+05 4.358500e+03 2.645360e+03 7.414110e+03 4.810670e+03 2.225990e+03 3.185990e+03
+1332496830266667 2.547560e+05 2.242400e+05 4.857380e+03 3.229680e+03 7.539310e+03 4.769140e+03 1.507130e+03 3.668260e+03
+1332496830275000 2.568890e+05 2.226580e+05 6.473420e+03 1.214110e+03 9.010760e+03 3.848730e+03 1.303840e+03 3.778500e+03
+1332496830283333 2.642080e+05 2.233160e+05 5.700450e+03 1.116560e+03 9.087610e+03 3.846680e+03 1.293590e+03 2.891560e+03
+1332496830291667 2.633100e+05 2.257190e+05 3.936120e+03 3.252360e+03 7.552850e+03 4.897860e+03 1.156630e+03 2.037160e+03
+1332496830300000 2.550790e+05 2.250860e+05 4.536450e+03 3.960110e+03 7.454590e+03 5.479070e+03 1.596360e+03 2.190800e+03
+1332496830308333 2.544870e+05 2.225080e+05 6.635860e+03 1.758850e+03 8.732970e+03 4.466970e+03 2.650360e+03 3.139310e+03
+1332496830316667 2.612410e+05 2.224320e+05 6.702270e+03 1.085130e+03 8.989230e+03 3.112990e+03 1.933560e+03 3.828410e+03
+1332496830325000 2.621190e+05 2.255870e+05 4.714950e+03 2.892360e+03 8.107820e+03 2.961310e+03 2.399780e+02 3.273720e+03
+1332496830333333 2.549990e+05 2.265140e+05 4.532090e+03 4.126900e+03 8.200130e+03 3.872590e+03 5.608900e+01 2.370580e+03
+1332496830341667 2.542890e+05 2.240330e+05 6.538810e+03 2.251440e+03 9.419430e+03 4.564450e+03 2.077810e+03 2.508170e+03
+1332496830350000 2.618900e+05 2.219600e+05 6.846090e+03 1.475270e+03 9.125590e+03 4.598290e+03 3.299220e+03 3.475420e+03
+1332496830358333 2.645020e+05 2.230850e+05 5.066380e+03 3.270560e+03 7.933170e+03 4.173710e+03 1.908910e+03 3.867460e+03
+1332496830366667 2.578890e+05 2.236560e+05 4.201660e+03 4.473640e+03 7.688340e+03 4.161580e+03 6.875790e+02 3.653690e+03
+1332496830375000 2.542700e+05 2.231510e+05 5.715140e+03 2.752140e+03 9.273320e+03 3.772950e+03 8.964040e+02 3.256060e+03
+1332496830383333 2.582570e+05 2.242170e+05 6.114310e+03 1.856860e+03 9.604320e+03 4.200490e+03 1.764380e+03 2.939220e+03
+1332496830391667 2.600200e+05 2.268680e+05 4.237530e+03 3.605880e+03 8.066220e+03 5.430250e+03 2.138580e+03 2.696710e+03
+1332496830400000 2.550830e+05 2.259240e+05 3.350310e+03 4.853070e+03 7.045820e+03 5.925200e+03 1.893610e+03 2.897340e+03
+1332496830408333 2.544530e+05 2.221270e+05 5.271330e+03 2.491500e+03 8.436680e+03 5.032080e+03 2.436050e+03 3.724590e+03
+1332496830416667 2.625880e+05 2.199500e+05 5.994620e+03 7.892740e+02 9.029650e+03 3.515740e+03 1.953570e+03 4.014520e+03
+1332496830425000 2.656100e+05 2.233330e+05 4.391410e+03 2.400960e+03 8.146460e+03 3.536960e+03 5.302320e+02 3.133920e+03
+1332496830433333 2.574700e+05 2.269770e+05 2.975320e+03 4.633530e+03 7.278560e+03 4.640100e+03 -5.015020e+01 2.024960e+03
+1332496830441667 2.506870e+05 2.263310e+05 4.517860e+03 3.183800e+03 8.072600e+03 5.281660e+03 1.605140e+03 2.335140e+03
+1332496830450000 2.555630e+05 2.244950e+05 5.551000e+03 1.101300e+03 8.461490e+03 4.725700e+03 2.726670e+03 3.480540e+03
+1332496830458333 2.613350e+05 2.246450e+05 4.764680e+03 1.557020e+03 7.833350e+03 3.524810e+03 1.577410e+03 4.038620e+03
+1332496830466667 2.602690e+05 2.240080e+05 3.558030e+03 2.987610e+03 7.362440e+03 3.279230e+03 5.624420e+02 3.786550e+03
+1332496830475000 2.574350e+05 2.217770e+05 4.972600e+03 2.166880e+03 8.481440e+03 3.328720e+03 1.037130e+03 3.271370e+03
+1332496830483333 2.610460e+05 2.215500e+05 5.816180e+03 5.902170e+02 9.120930e+03 3.895400e+03 2.382670e+03 2.824170e+03
+1332496830491667 2.627660e+05 2.244730e+05 4.835050e+03 1.785770e+03 7.880760e+03 4.745620e+03 2.443660e+03 3.229550e+03
+1332496830500000 2.565090e+05 2.264130e+05 3.758870e+03 3.461200e+03 6.743770e+03 4.928960e+03 1.536620e+03 3.546690e+03
+1332496830508333 2.507930e+05 2.243720e+05 5.218490e+03 2.865260e+03 7.803960e+03 4.351090e+03 1.333820e+03 3.680490e+03
+1332496830516667 2.563190e+05 2.220660e+05 6.403970e+03 7.323450e+02 9.627760e+03 3.089300e+03 1.516780e+03 3.653690e+03
+1332496830525000 2.633430e+05 2.232350e+05 5.200430e+03 1.388580e+03 9.372850e+03 3.371230e+03 1.450390e+03 2.678910e+03
+1332496830533333 2.609030e+05 2.251100e+05 3.722580e+03 3.246660e+03 7.876540e+03 4.716810e+03 1.498440e+03 2.116520e+03
+1332496830541667 2.544160e+05 2.237690e+05 4.841650e+03 2.956400e+03 8.115920e+03 5.392360e+03 2.142810e+03 2.652320e+03
+1332496830550000 2.566980e+05 2.221720e+05 6.471230e+03 9.703960e+02 8.834980e+03 4.816840e+03 2.376630e+03 3.605860e+03
+1332496830558333 2.618410e+05 2.235370e+05 5.500740e+03 1.189660e+03 8.365730e+03 4.016470e+03 1.042270e+03 3.821200e+03
+1332496830566667 2.595030e+05 2.258400e+05 3.827930e+03 3.088840e+03 7.676140e+03 3.978310e+03 -3.570070e+02 3.016420e+03
+1332496830575000 2.534570e+05 2.246360e+05 4.914610e+03 3.097450e+03 8.224900e+03 4.321440e+03 1.713740e+02 2.412360e+03
+1332496830583333 2.560290e+05 2.222210e+05 6.841800e+03 1.028500e+03 9.252300e+03 4.387570e+03 2.418140e+03 2.510100e+03
+1332496830591667 2.628400e+05 2.225500e+05 6.210250e+03 1.410730e+03 8.538900e+03 4.152580e+03 3.009300e+03 3.219760e+03
+1332496830600000 2.616330e+05 2.250650e+05 4.284530e+03 3.357210e+03 7.282170e+03 3.823590e+03 1.402840e+03 3.644670e+03
+1332496830608333 2.545910e+05 2.251090e+05 4.693160e+03 3.647740e+03 7.745160e+03 3.686380e+03 4.901610e+02 3.448860e+03
+1332496830616667 2.547800e+05 2.235990e+05 6.527380e+03 1.569870e+03 9.438430e+03 3.456580e+03 1.162520e+03 3.252010e+03
+1332496830625000 2.606390e+05 2.241070e+05 6.531050e+03 1.633050e+03 9.283720e+03 4.174020e+03 2.089550e+03 2.775750e+03
+1332496830633333 2.611080e+05 2.254720e+05 4.968260e+03 3.527850e+03 7.692870e+03 5.137100e+03 2.207390e+03 2.436660e+03
+1332496830641667 2.557750e+05 2.237080e+05 4.963450e+03 4.017370e+03 7.701420e+03 5.269650e+03 2.284400e+03 2.842080e+03
+1332496830650000 2.573980e+05 2.209470e+05 6.767500e+03 1.645710e+03 9.107070e+03 4.000180e+03 2.548860e+03 3.624770e+03
+1332496830658333 2.649240e+05 2.215590e+05 6.471460e+03 1.110330e+03 9.459650e+03 3.108170e+03 1.696970e+03 3.893440e+03
+1332496830666667 2.653390e+05 2.257330e+05 4.348800e+03 3.459510e+03 8.475300e+03 4.031240e+03 5.733470e+02 2.910270e+03
+1332496830675000 2.568140e+05 2.269950e+05 3.479540e+03 4.949790e+03 7.499910e+03 5.624710e+03 7.516560e+02 2.347710e+03
+1332496830683333 2.533160e+05 2.251610e+05 5.147060e+03 3.218430e+03 8.460160e+03 5.869300e+03 2.336320e+03 2.987960e+03
+1332496830691667 2.593600e+05 2.231010e+05 5.549120e+03 1.869950e+03 8.740760e+03 4.668940e+03 2.457910e+03 3.758820e+03
+1332496830700000 2.620120e+05 2.240160e+05 4.173610e+03 3.004130e+03 8.157040e+03 3.704730e+03 9.879640e+02 3.652750e+03
+1332496830708333 2.571760e+05 2.244200e+05 3.517300e+03 4.118750e+03 7.822240e+03 3.718230e+03 3.726490e+01 2.953680e+03
+1332496830716667 2.551460e+05 2.233220e+05 4.923980e+03 2.330680e+03 9.095910e+03 3.792400e+03 1.013070e+03 2.711240e+03
+1332496830725000 2.605240e+05 2.236510e+05 5.413630e+03 1.146210e+03 8.817170e+03 4.419650e+03 2.446650e+03 2.832050e+03
+1332496830733333 2.620980e+05 2.257520e+05 4.262980e+03 2.270970e+03 7.135480e+03 5.067120e+03 2.294680e+03 3.376620e+03
+1332496830741667 2.568890e+05 2.253790e+05 3.606460e+03 3.568190e+03 6.552650e+03 4.970270e+03 1.516380e+03 3.662570e+03
+1332496830750000 2.539480e+05 2.226310e+05 5.511700e+03 2.066300e+03 7.952660e+03 4.019910e+03 1.513140e+03 3.752630e+03
+1332496830758333 2.597990e+05 2.220670e+05 5.873500e+03 6.085840e+02 9.253780e+03 2.870740e+03 1.348240e+03 3.344200e+03
+1332496830766667 2.625470e+05 2.249010e+05 4.346080e+03 1.928100e+03 8.590970e+03 3.455460e+03 9.043910e+02 2.379270e+03
+1332496830775000 2.561370e+05 2.267610e+05 3.423560e+03 3.379080e+03 7.471150e+03 4.894170e+03 1.153540e+03 2.031410e+03
+1332496830783333 2.503260e+05 2.250130e+05 5.519980e+03 2.423970e+03 7.991760e+03 5.117950e+03 2.098790e+03 3.099240e+03
+1332496830791667 2.554540e+05 2.229920e+05 6.547950e+03 4.964960e+02 8.751340e+03 3.900560e+03 2.132290e+03 4.076810e+03
+1332496830800000 2.612860e+05 2.234890e+05 5.152850e+03 1.501510e+03 8.425610e+03 2.888030e+03 7.761140e+02 3.786360e+03
+1332496830808333 2.589690e+05 2.240690e+05 3.832610e+03 3.001980e+03 7.979260e+03 3.182310e+03 5.271600e+01 2.874800e+03
+1332496830816667 2.549460e+05 2.220350e+05 5.317880e+03 2.139800e+03 9.103140e+03 3.955610e+03 1.235170e+03 2.394150e+03
+1332496830825000 2.586760e+05 2.212050e+05 6.594910e+03 5.053440e+02 9.423360e+03 4.562470e+03 2.913740e+03 2.892350e+03
+1332496830833333 2.621250e+05 2.235660e+05 5.116750e+03 1.773600e+03 8.082200e+03 4.776370e+03 2.386390e+03 3.659730e+03
+1332496830841667 2.578350e+05 2.259180e+05 3.714300e+03 3.477080e+03 7.205370e+03 4.554610e+03 7.115390e+02 3.878420e+03
+1332496830850000 2.536600e+05 2.243710e+05 5.022450e+03 2.592430e+03 8.277200e+03 4.119370e+03 4.865080e+02 3.666740e+03
+1332496830858333 2.595030e+05 2.220610e+05 6.589950e+03 6.599360e+02 9.596920e+03 3.598100e+03 1.702490e+03 3.036600e+03
+1332496830866667 2.654950e+05 2.228430e+05 5.541850e+03 1.728430e+03 8.459960e+03 4.492000e+03 2.231970e+03 2.430620e+03
+1332496830875000 2.609290e+05 2.249960e+05 4.000950e+03 3.745990e+03 6.983790e+03 5.430860e+03 1.855260e+03 2.533380e+03
+1332496830883333 2.527160e+05 2.243350e+05 5.086560e+03 3.401150e+03 7.597970e+03 5.196120e+03 1.755720e+03 3.079760e+03
+1332496830891667 2.541100e+05 2.231110e+05 6.822190e+03 1.229080e+03 9.164340e+03 3.761230e+03 1.679390e+03 3.584880e+03
+1332496830900000 2.599690e+05 2.246930e+05 6.183950e+03 1.538500e+03 9.222080e+03 3.139170e+03 9.499020e+02 3.180800e+03
+1332496830908333 2.590780e+05 2.269130e+05 4.388890e+03 3.694820e+03 8.195020e+03 3.933000e+03 4.260800e+02 2.388450e+03
+1332496830916667 2.545630e+05 2.247600e+05 5.168440e+03 4.020940e+03 8.450270e+03 4.758910e+03 1.458900e+03 2.286430e+03
+1332496830925000 2.580590e+05 2.212170e+05 6.883460e+03 1.649530e+03 9.232780e+03 4.457650e+03 3.057820e+03 3.031950e+03
+1332496830933333 2.646670e+05 2.211770e+05 6.218510e+03 1.645730e+03 8.657180e+03 3.663500e+03 2.528280e+03 3.978340e+03
+1332496830941667 2.629250e+05 2.243820e+05 4.627500e+03 3.635930e+03 7.892800e+03 3.431320e+03 6.045090e+02 3.901370e+03
+1332496830950000 2.547080e+05 2.254480e+05 4.408250e+03 4.461040e+03 8.197170e+03 3.953750e+03 -4.453460e+01 3.154870e+03
+1332496830958333 2.537020e+05 2.246350e+05 5.825770e+03 2.577050e+03 9.590050e+03 4.569250e+03 1.460270e+03 2.785170e+03
+1332496830966667 2.602060e+05 2.241400e+05 5.387980e+03 1.951160e+03 8.789510e+03 5.131660e+03 2.706380e+03 2.972480e+03
+1332496830975000 2.612400e+05 2.247370e+05 3.860810e+03 3.418310e+03 7.414530e+03 5.284520e+03 2.271380e+03 3.183150e+03
+1332496830983333 2.561400e+05 2.232520e+05 3.850010e+03 3.957140e+03 7.262650e+03 4.964640e+03 1.499510e+03 3.453130e+03
+1332496830991667 2.561160e+05 2.213490e+05 5.594480e+03 2.054400e+03 8.835130e+03 3.662010e+03 1.485510e+03 3.613010e+03
--- a/tests/data/extract-3
+++ b/tests/data/extract-3
@@ -1 +1 @@
-1332496830.008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
+1332496830008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
--- a/tests/data/extract-4
+++ b/tests/data/extract-4
@@ -1,2 +1,2 @@
-1332496830.008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
-1332496830.016667 2.630730e+05 2.233040e+05 4.961640e+03 2.197120e+03 7.687310e+03 4.861860e+03 2.732780e+03 3.008540e+03
+1332496830008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
+1332496830016667 2.630730e+05 2.233040e+05 4.961640e+03 2.197120e+03 7.687310e+03 4.861860e+03 2.732780e+03 3.008540e+03
--- a/tests/data/extract-7
+++ b/tests/data/extract-7
@@ -1,124 +1,124 @@
 # path: /newton/prep
 # layout: float32_8
-# start: 1332496830.000000
-# end: 1332496830.999000
-1332496830.000000 2.517740e+05 2.242410e+05 5.688100e+03 1.915530e+03 9.329220e+03 4.183710e+03 1.212350e+03 2.641790e+03
-1332496830.008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
-1332496830.016667 2.630730e+05 2.233040e+05 4.961640e+03 2.197120e+03 7.687310e+03 4.861860e+03 2.732780e+03 3.008540e+03
-1332496830.025000 2.576140e+05 2.233230e+05 5.003660e+03 3.525140e+03 7.165310e+03 4.685620e+03 1.715380e+03 3.440480e+03
-1332496830.033333 2.557800e+05 2.219150e+05 6.357310e+03 2.145290e+03 8.426970e+03 3.775350e+03 1.475390e+03 3.797240e+03
-1332496830.041667 2.601660e+05 2.230080e+05 6.702590e+03 1.484960e+03 9.288100e+03 3.330830e+03 1.228500e+03 3.214320e+03
-1332496830.050000 2.612310e+05 2.264260e+05 4.980060e+03 2.982380e+03 8.499630e+03 4.267670e+03 9.940890e+02 2.292890e+03
-1332496830.058333 2.551170e+05 2.266420e+05 4.584410e+03 4.656440e+03 7.860150e+03 5.317310e+03 1.473600e+03 2.111690e+03
-1332496830.066667 2.533000e+05 2.235540e+05 6.455090e+03 3.036650e+03 8.869750e+03 4.986310e+03 2.607360e+03 2.839590e+03
-1332496830.075000 2.610610e+05 2.212630e+05 6.951980e+03 1.500240e+03 9.386100e+03 3.791680e+03 2.677010e+03 3.980630e+03
-1332496830.083333 2.665030e+05 2.231980e+05 5.189610e+03 2.594560e+03 8.571530e+03 3.175000e+03 9.198400e+02 3.792010e+03
-1332496830.091667 2.606920e+05 2.251840e+05 3.782480e+03 4.642880e+03 7.662960e+03 3.917790e+03 -2.510970e+02 2.907060e+03
-1332496830.100000 2.539630e+05 2.250810e+05 5.123530e+03 3.839550e+03 8.669030e+03 4.877820e+03 9.437240e+02 2.527450e+03
-1332496830.108333 2.565550e+05 2.241690e+05 5.930600e+03 2.298540e+03 8.906710e+03 5.331680e+03 2.549910e+03 3.053560e+03
-1332496830.116667 2.608890e+05 2.250100e+05 4.681130e+03 2.971870e+03 7.900040e+03 4.874080e+03 2.322430e+03 3.649120e+03
-1332496830.125000 2.579440e+05 2.249230e+05 3.291140e+03 4.357090e+03 7.131590e+03 4.385560e+03 1.077050e+03 3.664040e+03
-1332496830.133333 2.550090e+05 2.230180e+05 4.584820e+03 2.864000e+03 8.469490e+03 3.625580e+03 9.855570e+02 3.504230e+03
-1332496830.141667 2.601140e+05 2.219470e+05 5.676190e+03 1.210340e+03 9.393780e+03 3.390240e+03 1.654020e+03 3.018700e+03
-1332496830.150000 2.642770e+05 2.244380e+05 4.446620e+03 2.176720e+03 8.142090e+03 4.584880e+03 2.327830e+03 2.615800e+03
-1332496830.158333 2.592210e+05 2.264710e+05 2.734440e+03 4.182760e+03 6.389550e+03 5.540520e+03 1.958880e+03 2.720120e+03
-1332496830.166667 2.526500e+05 2.248310e+05 4.163640e+03 2.989990e+03 7.179200e+03 5.213060e+03 1.929550e+03 3.457660e+03
-1332496830.175000 2.570830e+05 2.220480e+05 5.759040e+03 7.024410e+02 8.566550e+03 3.552020e+03 1.832940e+03 3.956190e+03
-1332496830.183333 2.631300e+05 2.229670e+05 5.141140e+03 1.166120e+03 8.666960e+03 2.720370e+03 9.713740e+02 3.479730e+03
-1332496830.191667 2.602360e+05 2.252650e+05 3.425140e+03 3.339080e+03 7.853610e+03 3.674950e+03 5.259080e+02 2.443310e+03
-1332496830.200000 2.535030e+05 2.245270e+05 4.398130e+03 2.927430e+03 8.110280e+03 4.842470e+03 1.513870e+03 2.467100e+03
-1332496830.208333 2.561260e+05 2.226930e+05 6.043530e+03 6.562240e+02 8.797560e+03 4.832410e+03 2.832370e+03 3.426140e+03
-1332496830.216667 2.616770e+05 2.236080e+05 5.830460e+03 1.033910e+03 8.123940e+03 3.980690e+03 1.927960e+03 4.092720e+03
-1332496830.225000 2.594570e+05 2.255360e+05 4.015570e+03 2.995990e+03 7.135440e+03 3.713550e+03 3.072200e+02 3.849430e+03
-1332496830.233333 2.533520e+05 2.242160e+05 4.650560e+03 3.196620e+03 8.131280e+03 3.586160e+03 7.083230e+01 3.074180e+03
-1332496830.241667 2.561240e+05 2.215130e+05 6.100480e+03 8.219800e+02 9.757540e+03 3.474510e+03 1.647520e+03 2.559860e+03
-1332496830.250000 2.630240e+05 2.215590e+05 5.789960e+03 6.994170e+02 9.129740e+03 4.153080e+03 2.829250e+03 2.677270e+03
-1332496830.258333 2.617200e+05 2.240150e+05 4.358500e+03 2.645360e+03 7.414110e+03 4.810670e+03 2.225990e+03 3.185990e+03
-1332496830.266667 2.547560e+05 2.242400e+05 4.857380e+03 3.229680e+03 7.539310e+03 4.769140e+03 1.507130e+03 3.668260e+03
-1332496830.275000 2.568890e+05 2.226580e+05 6.473420e+03 1.214110e+03 9.010760e+03 3.848730e+03 1.303840e+03 3.778500e+03
-1332496830.283333 2.642080e+05 2.233160e+05 5.700450e+03 1.116560e+03 9.087610e+03 3.846680e+03 1.293590e+03 2.891560e+03
-1332496830.291667 2.633100e+05 2.257190e+05 3.936120e+03 3.252360e+03 7.552850e+03 4.897860e+03 1.156630e+03 2.037160e+03
-1332496830.300000 2.550790e+05 2.250860e+05 4.536450e+03 3.960110e+03 7.454590e+03 5.479070e+03 1.596360e+03 2.190800e+03
-1332496830.308333 2.544870e+05 2.225080e+05 6.635860e+03 1.758850e+03 8.732970e+03 4.466970e+03 2.650360e+03 3.139310e+03
-1332496830.316667 2.612410e+05 2.224320e+05 6.702270e+03 1.085130e+03 8.989230e+03 3.112990e+03 1.933560e+03 3.828410e+03
-1332496830.325000 2.621190e+05 2.255870e+05 4.714950e+03 2.892360e+03 8.107820e+03 2.961310e+03 2.399780e+02 3.273720e+03
-1332496830.333333 2.549990e+05 2.265140e+05 4.532090e+03 4.126900e+03 8.200130e+03 3.872590e+03 5.608900e+01 2.370580e+03
-1332496830.341667 2.542890e+05 2.240330e+05 6.538810e+03 2.251440e+03 9.419430e+03 4.564450e+03 2.077810e+03 2.508170e+03
-1332496830.350000 2.618900e+05 2.219600e+05 6.846090e+03 1.475270e+03 9.125590e+03 4.598290e+03 3.299220e+03 3.475420e+03
-1332496830.358333 2.645020e+05 2.230850e+05 5.066380e+03 3.270560e+03 7.933170e+03 4.173710e+03 1.908910e+03 3.867460e+03
-1332496830.366667 2.578890e+05 2.236560e+05 4.201660e+03 4.473640e+03 7.688340e+03 4.161580e+03 6.875790e+02 3.653690e+03
-1332496830.375000 2.542700e+05 2.231510e+05 5.715140e+03 2.752140e+03 9.273320e+03 3.772950e+03 8.964040e+02 3.256060e+03
-1332496830.383333 2.582570e+05 2.242170e+05 6.114310e+03 1.856860e+03 9.604320e+03 4.200490e+03 1.764380e+03 2.939220e+03
-1332496830.391667 2.600200e+05 2.268680e+05 4.237530e+03 3.605880e+03 8.066220e+03 5.430250e+03 2.138580e+03 2.696710e+03
-1332496830.400000 2.550830e+05 2.259240e+05 3.350310e+03 4.853070e+03 7.045820e+03 5.925200e+03 1.893610e+03 2.897340e+03
-1332496830.408333 2.544530e+05 2.221270e+05 5.271330e+03 2.491500e+03 8.436680e+03 5.032080e+03 2.436050e+03 3.724590e+03
-1332496830.416667 2.625880e+05 2.199500e+05 5.994620e+03 7.892740e+02 9.029650e+03 3.515740e+03 1.953570e+03 4.014520e+03
-1332496830.425000 2.656100e+05 2.233330e+05 4.391410e+03 2.400960e+03 8.146460e+03 3.536960e+03 5.302320e+02 3.133920e+03
-1332496830.433333 2.574700e+05 2.269770e+05 2.975320e+03 4.633530e+03 7.278560e+03 4.640100e+03 -5.015020e+01 2.024960e+03
-1332496830.441667 2.506870e+05 2.263310e+05 4.517860e+03 3.183800e+03 8.072600e+03 5.281660e+03 1.605140e+03 2.335140e+03
-1332496830.450000 2.555630e+05 2.244950e+05 5.551000e+03 1.101300e+03 8.461490e+03 4.725700e+03 2.726670e+03 3.480540e+03
-1332496830.458333 2.613350e+05 2.246450e+05 4.764680e+03 1.557020e+03 7.833350e+03 3.524810e+03 1.577410e+03 4.038620e+03
-1332496830.466667 2.602690e+05 2.240080e+05 3.558030e+03 2.987610e+03 7.362440e+03 3.279230e+03 5.624420e+02 3.786550e+03
-1332496830.475000 2.574350e+05 2.217770e+05 4.972600e+03 2.166880e+03 8.481440e+03 3.328720e+03 1.037130e+03 3.271370e+03
-1332496830.483333 2.610460e+05 2.215500e+05 5.816180e+03 5.902170e+02 9.120930e+03 3.895400e+03 2.382670e+03 2.824170e+03
-1332496830.491667 2.627660e+05 2.244730e+05 4.835050e+03 1.785770e+03 7.880760e+03 4.745620e+03 2.443660e+03 3.229550e+03
-1332496830.500000 2.565090e+05 2.264130e+05 3.758870e+03 3.461200e+03 6.743770e+03 4.928960e+03 1.536620e+03 3.546690e+03
-1332496830.508333 2.507930e+05 2.243720e+05 5.218490e+03 2.865260e+03 7.803960e+03 4.351090e+03 1.333820e+03 3.680490e+03
-1332496830.516667 2.563190e+05 2.220660e+05 6.403970e+03 7.323450e+02 9.627760e+03 3.089300e+03 1.516780e+03 3.653690e+03
-1332496830.525000 2.633430e+05 2.232350e+05 5.200430e+03 1.388580e+03 9.372850e+03 3.371230e+03 1.450390e+03 2.678910e+03
-1332496830.533333 2.609030e+05 2.251100e+05 3.722580e+03 3.246660e+03 7.876540e+03 4.716810e+03 1.498440e+03 2.116520e+03
-1332496830.541667 2.544160e+05 2.237690e+05 4.841650e+03 2.956400e+03 8.115920e+03 5.392360e+03 2.142810e+03 2.652320e+03
-1332496830.550000 2.566980e+05 2.221720e+05 6.471230e+03 9.703960e+02 8.834980e+03 4.816840e+03 2.376630e+03 3.605860e+03
-1332496830.558333 2.618410e+05 2.235370e+05 5.500740e+03 1.189660e+03 8.365730e+03 4.016470e+03 1.042270e+03 3.821200e+03
-1332496830.566667 2.595030e+05 2.258400e+05 3.827930e+03 3.088840e+03 7.676140e+03 3.978310e+03 -3.570070e+02 3.016420e+03
-1332496830.575000 2.534570e+05 2.246360e+05 4.914610e+03 3.097450e+03 8.224900e+03 4.321440e+03 1.713740e+02 2.412360e+03
-1332496830.583333 2.560290e+05 2.222210e+05 6.841800e+03 1.028500e+03 9.252300e+03 4.387570e+03 2.418140e+03 2.510100e+03
-1332496830.591667 2.628400e+05 2.225500e+05 6.210250e+03 1.410730e+03 8.538900e+03 4.152580e+03 3.009300e+03 3.219760e+03
-1332496830.600000 2.616330e+05 2.250650e+05 4.284530e+03 3.357210e+03 7.282170e+03 3.823590e+03 1.402840e+03 3.644670e+03
-1332496830.608333 2.545910e+05 2.251090e+05 4.693160e+03 3.647740e+03 7.745160e+03 3.686380e+03 4.901610e+02 3.448860e+03
-1332496830.616667 2.547800e+05 2.235990e+05 6.527380e+03 1.569870e+03 9.438430e+03 3.456580e+03 1.162520e+03 3.252010e+03
-1332496830.625000 2.606390e+05 2.241070e+05 6.531050e+03 1.633050e+03 9.283720e+03 4.174020e+03 2.089550e+03 2.775750e+03
-1332496830.633333 2.611080e+05 2.254720e+05 4.968260e+03 3.527850e+03 7.692870e+03 5.137100e+03 2.207390e+03 2.436660e+03
-1332496830.641667 2.557750e+05 2.237080e+05 4.963450e+03 4.017370e+03 7.701420e+03 5.269650e+03 2.284400e+03 2.842080e+03
-1332496830.650000 2.573980e+05 2.209470e+05 6.767500e+03 1.645710e+03 9.107070e+03 4.000180e+03 2.548860e+03 3.624770e+03
-1332496830.658333 2.649240e+05 2.215590e+05 6.471460e+03 1.110330e+03 9.459650e+03 3.108170e+03 1.696970e+03 3.893440e+03
-1332496830.666667 2.653390e+05 2.257330e+05 4.348800e+03 3.459510e+03 8.475300e+03 4.031240e+03 5.733470e+02 2.910270e+03
-1332496830.675000 2.568140e+05 2.269950e+05 3.479540e+03 4.949790e+03 7.499910e+03 5.624710e+03 7.516560e+02 2.347710e+03
-1332496830.683333 2.533160e+05 2.251610e+05 5.147060e+03 3.218430e+03 8.460160e+03 5.869300e+03 2.336320e+03 2.987960e+03
-1332496830.691667 2.593600e+05 2.231010e+05 5.549120e+03 1.869950e+03 8.740760e+03 4.668940e+03 2.457910e+03 3.758820e+03
-1332496830.700000 2.620120e+05 2.240160e+05 4.173610e+03 3.004130e+03 8.157040e+03 3.704730e+03 9.879640e+02 3.652750e+03
-1332496830.708333 2.571760e+05 2.244200e+05 3.517300e+03 4.118750e+03 7.822240e+03 3.718230e+03 3.726490e+01 2.953680e+03
-1332496830.716667 2.551460e+05 2.233220e+05 4.923980e+03 2.330680e+03 9.095910e+03 3.792400e+03 1.013070e+03 2.711240e+03
-1332496830.725000 2.605240e+05 2.236510e+05 5.413630e+03 1.146210e+03 8.817170e+03 4.419650e+03 2.446650e+03 2.832050e+03
-1332496830.733333 2.620980e+05 2.257520e+05 4.262980e+03 2.270970e+03 7.135480e+03 5.067120e+03 2.294680e+03 3.376620e+03
-1332496830.741667 2.568890e+05 2.253790e+05 3.606460e+03 3.568190e+03 6.552650e+03 4.970270e+03 1.516380e+03 3.662570e+03
-1332496830.750000 2.539480e+05 2.226310e+05 5.511700e+03 2.066300e+03 7.952660e+03 4.019910e+03 1.513140e+03 3.752630e+03
-1332496830.758333 2.597990e+05 2.220670e+05 5.873500e+03 6.085840e+02 9.253780e+03 2.870740e+03 1.348240e+03 3.344200e+03
-1332496830.766667 2.625470e+05 2.249010e+05 4.346080e+03 1.928100e+03 8.590970e+03 3.455460e+03 9.043910e+02 2.379270e+03
-1332496830.775000 2.561370e+05 2.267610e+05 3.423560e+03 3.379080e+03 7.471150e+03 4.894170e+03 1.153540e+03 2.031410e+03
-1332496830.783333 2.503260e+05 2.250130e+05 5.519980e+03 2.423970e+03 7.991760e+03 5.117950e+03 2.098790e+03 3.099240e+03
-1332496830.791667 2.554540e+05 2.229920e+05 6.547950e+03 4.964960e+02 8.751340e+03 3.900560e+03 2.132290e+03 4.076810e+03
-1332496830.800000 2.612860e+05 2.234890e+05 5.152850e+03 1.501510e+03 8.425610e+03 2.888030e+03 7.761140e+02 3.786360e+03
-1332496830.808333 2.589690e+05 2.240690e+05 3.832610e+03 3.001980e+03 7.979260e+03 3.182310e+03 5.271600e+01 2.874800e+03
-1332496830.816667 2.549460e+05 2.220350e+05 5.317880e+03 2.139800e+03 9.103140e+03 3.955610e+03 1.235170e+03 2.394150e+03
-1332496830.825000 2.586760e+05 2.212050e+05 6.594910e+03 5.053440e+02 9.423360e+03 4.562470e+03 2.913740e+03 2.892350e+03
-1332496830.833333 2.621250e+05 2.235660e+05 5.116750e+03 1.773600e+03 8.082200e+03 4.776370e+03 2.386390e+03 3.659730e+03
-1332496830.841667 2.578350e+05 2.259180e+05 3.714300e+03 3.477080e+03 7.205370e+03 4.554610e+03 7.115390e+02 3.878420e+03
-1332496830.850000 2.536600e+05 2.243710e+05 5.022450e+03 2.592430e+03 8.277200e+03 4.119370e+03 4.865080e+02 3.666740e+03
-1332496830.858333 2.595030e+05 2.220610e+05 6.589950e+03 6.599360e+02 9.596920e+03 3.598100e+03 1.702490e+03 3.036600e+03
-1332496830.866667 2.654950e+05 2.228430e+05 5.541850e+03 1.728430e+03 8.459960e+03 4.492000e+03 2.231970e+03 2.430620e+03
-1332496830.875000 2.609290e+05 2.249960e+05 4.000950e+03 3.745990e+03 6.983790e+03 5.430860e+03 1.855260e+03 2.533380e+03
-1332496830.883333 2.527160e+05 2.243350e+05 5.086560e+03 3.401150e+03 7.597970e+03 5.196120e+03 1.755720e+03 3.079760e+03
-1332496830.891667 2.541100e+05 2.231110e+05 6.822190e+03 1.229080e+03 9.164340e+03 3.761230e+03 1.679390e+03 3.584880e+03
-1332496830.900000 2.599690e+05 2.246930e+05 6.183950e+03 1.538500e+03 9.222080e+03 3.139170e+03 9.499020e+02 3.180800e+03
-1332496830.908333 2.590780e+05 2.269130e+05 4.388890e+03 3.694820e+03 8.195020e+03 3.933000e+03 4.260800e+02 2.388450e+03
-1332496830.916667 2.545630e+05 2.247600e+05 5.168440e+03 4.020940e+03 8.450270e+03 4.758910e+03 1.458900e+03 2.286430e+03
-1332496830.925000 2.580590e+05 2.212170e+05 6.883460e+03 1.649530e+03 9.232780e+03 4.457650e+03 3.057820e+03 3.031950e+03
-1332496830.933333 2.646670e+05 2.211770e+05 6.218510e+03 1.645730e+03 8.657180e+03 3.663500e+03 2.528280e+03 3.978340e+03
-1332496830.941667 2.629250e+05 2.243820e+05 4.627500e+03 3.635930e+03 7.892800e+03 3.431320e+03 6.045090e+02 3.901370e+03
-1332496830.950000 2.547080e+05 2.254480e+05 4.408250e+03 4.461040e+03 8.197170e+03 3.953750e+03 -4.453460e+01 3.154870e+03
-1332496830.958333 2.537020e+05 2.246350e+05 5.825770e+03 2.577050e+03 9.590050e+03 4.569250e+03 1.460270e+03 2.785170e+03
-1332496830.966667 2.602060e+05 2.241400e+05 5.387980e+03 1.951160e+03 8.789510e+03 5.131660e+03 2.706380e+03 2.972480e+03
-1332496830.975000 2.612400e+05 2.247370e+05 3.860810e+03 3.418310e+03 7.414530e+03 5.284520e+03 2.271380e+03 3.183150e+03
-1332496830.983333 2.561400e+05 2.232520e+05 3.850010e+03 3.957140e+03 7.262650e+03 4.964640e+03 1.499510e+03 3.453130e+03
-1332496830.991667 2.561160e+05 2.213490e+05 5.594480e+03 2.054400e+03 8.835130e+03 3.662010e+03 1.485510e+03 3.613010e+03
+# start: 1332496830000000
+# end: 1332496830999000
+1332496830000000 2.517740e+05 2.242410e+05 5.688100e+03 1.915530e+03 9.329220e+03 4.183710e+03 1.212350e+03 2.641790e+03
+1332496830008333 2.595670e+05 2.226980e+05 6.207600e+03 6.786720e+02 9.380230e+03 4.575580e+03 2.830610e+03 2.688630e+03
+1332496830016667 2.630730e+05 2.233040e+05 4.961640e+03 2.197120e+03 7.687310e+03 4.861860e+03 2.732780e+03 3.008540e+03
+1332496830025000 2.576140e+05 2.233230e+05 5.003660e+03 3.525140e+03 7.165310e+03 4.685620e+03 1.715380e+03 3.440480e+03
+1332496830033333 2.557800e+05 2.219150e+05 6.357310e+03 2.145290e+03 8.426970e+03 3.775350e+03 1.475390e+03 3.797240e+03
+1332496830041667 2.601660e+05 2.230080e+05 6.702590e+03 1.484960e+03 9.288100e+03 3.330830e+03 1.228500e+03 3.214320e+03
+1332496830050000 2.612310e+05 2.264260e+05 4.980060e+03 2.982380e+03 8.499630e+03 4.267670e+03 9.940890e+02 2.292890e+03
+1332496830058333 2.551170e+05 2.266420e+05 4.584410e+03 4.656440e+03 7.860150e+03 5.317310e+03 1.473600e+03 2.111690e+03
+1332496830066667 2.533000e+05 2.235540e+05 6.455090e+03 3.036650e+03 8.869750e+03 4.986310e+03 2.607360e+03 2.839590e+03
+1332496830075000 2.610610e+05 2.212630e+05 6.951980e+03 1.500240e+03 9.386100e+03 3.791680e+03 2.677010e+03 3.980630e+03
+1332496830083333 2.665030e+05 2.231980e+05 5.189610e+03 2.594560e+03 8.571530e+03 3.175000e+03 9.198400e+02 3.792010e+03
+1332496830091667 2.606920e+05 2.251840e+05 3.782480e+03 4.642880e+03 7.662960e+03 3.917790e+03 -2.510970e+02 2.907060e+03
+1332496830100000 2.539630e+05 2.250810e+05 5.123530e+03 3.839550e+03 8.669030e+03 4.877820e+03 9.437240e+02 2.527450e+03
+1332496830108333 2.565550e+05 2.241690e+05 5.930600e+03 2.298540e+03 8.906710e+03 5.331680e+03 2.549910e+03 3.053560e+03
+1332496830116667 2.608890e+05 2.250100e+05 4.681130e+03 2.971870e+03 7.900040e+03 4.874080e+03 2.322430e+03 3.649120e+03
+1332496830125000 2.579440e+05 2.249230e+05 3.291140e+03 4.357090e+03 7.131590e+03 4.385560e+03 1.077050e+03 3.664040e+03
+1332496830133333 2.550090e+05 2.230180e+05 4.584820e+03 2.864000e+03 8.469490e+03 3.625580e+03 9.855570e+02 3.504230e+03
+1332496830141667 2.601140e+05 2.219470e+05 5.676190e+03 1.210340e+03 9.393780e+03 3.390240e+03 1.654020e+03 3.018700e+03
+1332496830150000 2.642770e+05 2.244380e+05 4.446620e+03 2.176720e+03 8.142090e+03 4.584880e+03 2.327830e+03 2.615800e+03
+1332496830158333 2.592210e+05 2.264710e+05 2.734440e+03 4.182760e+03 6.389550e+03 5.540520e+03 1.958880e+03 2.720120e+03
+1332496830166667 2.526500e+05 2.248310e+05 4.163640e+03 2.989990e+03 7.179200e+03 5.213060e+03 1.929550e+03 3.457660e+03
+1332496830175000 2.570830e+05 2.220480e+05 5.759040e+03 7.024410e+02 8.566550e+03 3.552020e+03 1.832940e+03 3.956190e+03
+1332496830183333 2.631300e+05 2.229670e+05 5.141140e+03 1.166120e+03 8.666960e+03 2.720370e+03 9.713740e+02 3.479730e+03
+1332496830191667 2.602360e+05 2.252650e+05 3.425140e+03 3.339080e+03 7.853610e+03 3.674950e+03 5.259080e+02 2.443310e+03
+1332496830200000 2.535030e+05 2.245270e+05 4.398130e+03 2.927430e+03 8.110280e+03 4.842470e+03 1.513870e+03 2.467100e+03
+1332496830208333 2.561260e+05 2.226930e+05 6.043530e+03 6.562240e+02 8.797560e+03 4.832410e+03 2.832370e+03 3.426140e+03
+1332496830216667 2.616770e+05 2.236080e+05 5.830460e+03 1.033910e+03 8.123940e+03 3.980690e+03 1.927960e+03 4.092720e+03
+1332496830225000 2.594570e+05 2.255360e+05 4.015570e+03 2.995990e+03 7.135440e+03 3.713550e+03 3.072200e+02 3.849430e+03
+1332496830233333 2.533520e+05 2.242160e+05 4.650560e+03 3.196620e+03 8.131280e+03 3.586160e+03 7.083230e+01 3.074180e+03
+1332496830241667 2.561240e+05 2.215130e+05 6.100480e+03 8.219800e+02 9.757540e+03 3.474510e+03 1.647520e+03 2.559860e+03
+1332496830250000 2.630240e+05 2.215590e+05 5.789960e+03 6.994170e+02 9.129740e+03 4.153080e+03 2.829250e+03 2.677270e+03
+1332496830258333 2.617200e+05 2.240150e+05 4.358500e+03 2.645360e+03 7.414110e+03 4.810670e+03 2.225990e+03 3.185990e+03
+1332496830266667 2.547560e+05 2.242400e+05 4.857380e+03 3.229680e+03 7.539310e+03 4.769140e+03 1.507130e+03 3.668260e+03
+1332496830275000 2.568890e+05 2.226580e+05 6.473420e+03 1.214110e+03 9.010760e+03 3.848730e+03 1.303840e+03 3.778500e+03
+1332496830283333 2.642080e+05 2.233160e+05 5.700450e+03 1.116560e+03 9.087610e+03 3.846680e+03 1.293590e+03 2.891560e+03
+1332496830291667 2.633100e+05 2.257190e+05 3.936120e+03 3.252360e+03 7.552850e+03 4.897860e+03 1.156630e+03 2.037160e+03
+1332496830300000 2.550790e+05 2.250860e+05 4.536450e+03 3.960110e+03 7.454590e+03 5.479070e+03 1.596360e+03 2.190800e+03
+1332496830308333 2.544870e+05 2.225080e+05 6.635860e+03 1.758850e+03 8.732970e+03 4.466970e+03 2.650360e+03 3.139310e+03
+1332496830316667 2.612410e+05 2.224320e+05 6.702270e+03 1.085130e+03 8.989230e+03 3.112990e+03 1.933560e+03 3.828410e+03
+1332496830325000 2.621190e+05 2.255870e+05 4.714950e+03 2.892360e+03 8.107820e+03 2.961310e+03 2.399780e+02 3.273720e+03
+1332496830333333 2.549990e+05 2.265140e+05 4.532090e+03 4.126900e+03 8.200130e+03 3.872590e+03 5.608900e+01 2.370580e+03
+1332496830341667 2.542890e+05 2.240330e+05 6.538810e+03 2.251440e+03 9.419430e+03 4.564450e+03 2.077810e+03 2.508170e+03
+1332496830350000 2.618900e+05 2.219600e+05 6.846090e+03 1.475270e+03 9.125590e+03 4.598290e+03 3.299220e+03 3.475420e+03
+1332496830358333 2.645020e+05 2.230850e+05 5.066380e+03 3.270560e+03 7.933170e+03 4.173710e+03 1.908910e+03 3.867460e+03
+1332496830366667 2.578890e+05 2.236560e+05 4.201660e+03 4.473640e+03 7.688340e+03 4.161580e+03 6.875790e+02 3.653690e+03
+1332496830375000 2.542700e+05 2.231510e+05 5.715140e+03 2.752140e+03 9.273320e+03 3.772950e+03 8.964040e+02 3.256060e+03
+1332496830383333 2.582570e+05 2.242170e+05 6.114310e+03 1.856860e+03 9.604320e+03 4.200490e+03 1.764380e+03 2.939220e+03
+1332496830391667 2.600200e+05 2.268680e+05 4.237530e+03 3.605880e+03 8.066220e+03 5.430250e+03 2.138580e+03 2.696710e+03
+1332496830400000 2.550830e+05 2.259240e+05 3.350310e+03 4.853070e+03 7.045820e+03 5.925200e+03 1.893610e+03 2.897340e+03
+1332496830408333 2.544530e+05 2.221270e+05 5.271330e+03 2.491500e+03 8.436680e+03 5.032080e+03 2.436050e+03 3.724590e+03
+1332496830416667 2.625880e+05 2.199500e+05 5.994620e+03 7.892740e+02 9.029650e+03 3.515740e+03 1.953570e+03 4.014520e+03
+1332496830425000 2.656100e+05 2.233330e+05 4.391410e+03 2.400960e+03 8.146460e+03 3.536960e+03 5.302320e+02 3.133920e+03
+1332496830433333 2.574700e+05 2.269770e+05 2.975320e+03 4.633530e+03 7.278560e+03 4.640100e+03 -5.015020e+01 2.024960e+03
+1332496830441667 2.506870e+05 2.263310e+05 4.517860e+03 3.183800e+03 8.072600e+03 5.281660e+03 1.605140e+03 2.335140e+03
+1332496830450000 2.555630e+05 2.244950e+05 5.551000e+03 1.101300e+03 8.461490e+03 4.725700e+03 2.726670e+03 3.480540e+03
+1332496830458333 2.613350e+05 2.246450e+05 4.764680e+03 1.557020e+03 7.833350e+03 3.524810e+03 1.577410e+03 4.038620e+03
+1332496830466667 2.602690e+05 2.240080e+05 3.558030e+03 2.987610e+03 7.362440e+03 3.279230e+03 5.624420e+02 3.786550e+03
+1332496830475000 2.574350e+05 2.217770e+05 4.972600e+03 2.166880e+03 8.481440e+03 3.328720e+03 1.037130e+03 3.271370e+03
+1332496830483333 2.610460e+05 2.215500e+05 5.816180e+03 5.902170e+02 9.120930e+03 3.895400e+03 2.382670e+03 2.824170e+03
+1332496830491667 2.627660e+05 2.244730e+05 4.835050e+03 1.785770e+03 7.880760e+03 4.745620e+03 2.443660e+03 3.229550e+03
+1332496830500000 2.565090e+05 2.264130e+05 3.758870e+03 3.461200e+03 6.743770e+03 4.928960e+03 1.536620e+03 3.546690e+03
+1332496830508333 2.507930e+05 2.243720e+05 5.218490e+03 2.865260e+03 7.803960e+03 4.351090e+03 1.333820e+03 3.680490e+03
+1332496830516667 2.563190e+05 2.220660e+05 6.403970e+03 7.323450e+02 9.627760e+03 3.089300e+03 1.516780e+03 3.653690e+03
+1332496830525000 2.633430e+05 2.232350e+05 5.200430e+03 1.388580e+03 9.372850e+03 3.371230e+03 1.450390e+03 2.678910e+03
+1332496830533333 2.609030e+05 2.251100e+05 3.722580e+03 3.246660e+03 7.876540e+03 4.716810e+03 1.498440e+03 2.116520e+03
+1332496830541667 2.544160e+05 2.237690e+05 4.841650e+03 2.956400e+03 8.115920e+03 5.392360e+03 2.142810e+03 2.652320e+03
+1332496830550000 2.566980e+05 2.221720e+05 6.471230e+03 9.703960e+02 8.834980e+03 4.816840e+03 2.376630e+03 3.605860e+03
+1332496830558333 2.618410e+05 2.235370e+05 5.500740e+03 1.189660e+03 8.365730e+03 4.016470e+03 1.042270e+03 3.821200e+03
+1332496830566667 2.595030e+05 2.258400e+05 3.827930e+03 3.088840e+03 7.676140e+03 3.978310e+03 -3.570070e+02 3.016420e+03
+1332496830575000 2.534570e+05 2.246360e+05 4.914610e+03 3.097450e+03 8.224900e+03 4.321440e+03 1.713740e+02 2.412360e+03
+1332496830583333 2.560290e+05 2.222210e+05 6.841800e+03 1.028500e+03 9.252300e+03 4.387570e+03 2.418140e+03 2.510100e+03
+1332496830591667 2.628400e+05 2.225500e+05 6.210250e+03 1.410730e+03 8.538900e+03 4.152580e+03 3.009300e+03 3.219760e+03
+1332496830600000 2.616330e+05 2.250650e+05 4.284530e+03 3.357210e+03 7.282170e+03 3.823590e+03 1.402840e+03 3.644670e+03
+1332496830608333 2.545910e+05 2.251090e+05 4.693160e+03 3.647740e+03 7.745160e+03 3.686380e+03 4.901610e+02 3.448860e+03
+1332496830616667 2.547800e+05 2.235990e+05 6.527380e+03 1.569870e+03 9.438430e+03 3.456580e+03 1.162520e+03 3.252010e+03
+1332496830625000 2.606390e+05 2.241070e+05 6.531050e+03 1.633050e+03 9.283720e+03 4.174020e+03 2.089550e+03 2.775750e+03
+1332496830633333 2.611080e+05 2.254720e+05 4.968260e+03 3.527850e+03 7.692870e+03 5.137100e+03 2.207390e+03 2.436660e+03
+1332496830641667 2.557750e+05 2.237080e+05 4.963450e+03 4.017370e+03 7.701420e+03 5.269650e+03 2.284400e+03 2.842080e+03
+1332496830650000 2.573980e+05 2.209470e+05 6.767500e+03 1.645710e+03 9.107070e+03 4.000180e+03 2.548860e+03 3.624770e+03
+1332496830658333 2.649240e+05 2.215590e+05 6.471460e+03 1.110330e+03 9.459650e+03 3.108170e+03 1.696970e+03 3.893440e+03
+1332496830666667 2.653390e+05 2.257330e+05 4.348800e+03 3.459510e+03 8.475300e+03 4.031240e+03 5.733470e+02 2.910270e+03
+1332496830675000 2.568140e+05 2.269950e+05 3.479540e+03 4.949790e+03 7.499910e+03 5.624710e+03 7.516560e+02 2.347710e+03
+1332496830683333 2.533160e+05 2.251610e+05 5.147060e+03 3.218430e+03 8.460160e+03 5.869300e+03 2.336320e+03 2.987960e+03
+1332496830691667 2.593600e+05 2.231010e+05 5.549120e+03 1.869950e+03 8.740760e+03 4.668940e+03 2.457910e+03 3.758820e+03
+1332496830700000 2.620120e+05 2.240160e+05 4.173610e+03 3.004130e+03 8.157040e+03 3.704730e+03 9.879640e+02 3.652750e+03
+1332496830708333 2.571760e+05 2.244200e+05 3.517300e+03 4.118750e+03 7.822240e+03 3.718230e+03 3.726490e+01 2.953680e+03
+1332496830716667 2.551460e+05 2.233220e+05 4.923980e+03 2.330680e+03 9.095910e+03 3.792400e+03 1.013070e+03 2.711240e+03
+1332496830725000 2.605240e+05 2.236510e+05 5.413630e+03 1.146210e+03 8.817170e+03 4.419650e+03 2.446650e+03 2.832050e+03
+1332496830733333 2.620980e+05 2.257520e+05 4.262980e+03 2.270970e+03 7.135480e+03 5.067120e+03 2.294680e+03 3.376620e+03
+1332496830741667 2.568890e+05 2.253790e+05 3.606460e+03 3.568190e+03 6.552650e+03 4.970270e+03 1.516380e+03 3.662570e+03
+1332496830750000 2.539480e+05 2.226310e+05 5.511700e+03 2.066300e+03 7.952660e+03 4.019910e+03 1.513140e+03 3.752630e+03
+1332496830758333 2.597990e+05 2.220670e+05 5.873500e+03 6.085840e+02 9.253780e+03 2.870740e+03 1.348240e+03 3.344200e+03
+1332496830766667 2.625470e+05 2.249010e+05 4.346080e+03 1.928100e+03 8.590970e+03 3.455460e+03 9.043910e+02 2.379270e+03
+1332496830775000 2.561370e+05 2.267610e+05 3.423560e+03 3.379080e+03 7.471150e+03 4.894170e+03 1.153540e+03 2.031410e+03
+1332496830783333 2.503260e+05 2.250130e+05 5.519980e+03 2.423970e+03 7.991760e+03 5.117950e+03 2.098790e+03 3.099240e+03
+1332496830791667 2.554540e+05 2.229920e+05 6.547950e+03 4.964960e+02 8.751340e+03 3.900560e+03 2.132290e+03 4.076810e+03
+1332496830800000 2.612860e+05 2.234890e+05 5.152850e+03 1.501510e+03 8.425610e+03 2.888030e+03 7.761140e+02 3.786360e+03
+1332496830808333 2.589690e+05 2.240690e+05 3.832610e+03 3.001980e+03 7.979260e+03 3.182310e+03 5.271600e+01 2.874800e+03
+1332496830816667 2.549460e+05 2.220350e+05 5.317880e+03 2.139800e+03 9.103140e+03 3.955610e+03 1.235170e+03 2.394150e+03
+1332496830825000 2.586760e+05 2.212050e+05 6.594910e+03 5.053440e+02 9.423360e+03 4.562470e+03 2.913740e+03 2.892350e+03
+1332496830833333 2.621250e+05 2.235660e+05 5.116750e+03 1.773600e+03 8.082200e+03 4.776370e+03 2.386390e+03 3.659730e+03
+1332496830841667 2.578350e+05 2.259180e+05 3.714300e+03 3.477080e+03 7.205370e+03 4.554610e+03 7.115390e+02 3.878420e+03
+1332496830850000 2.536600e+05 2.243710e+05 5.022450e+03 2.592430e+03 8.277200e+03 4.119370e+03 4.865080e+02 3.666740e+03
+1332496830858333 2.595030e+05 2.220610e+05 6.589950e+03 6.599360e+02 9.596920e+03 3.598100e+03 1.702490e+03 3.036600e+03
+1332496830866667 2.654950e+05 2.228430e+05 5.541850e+03 1.728430e+03 8.459960e+03 4.492000e+03 2.231970e+03 2.430620e+03
+1332496830875000 2.609290e+05 2.249960e+05 4.000950e+03 3.745990e+03 6.983790e+03 5.430860e+03 1.855260e+03 2.533380e+03
+1332496830883333 2.527160e+05 2.243350e+05 5.086560e+03 3.401150e+03 7.597970e+03 5.196120e+03 1.755720e+03 3.079760e+03
+1332496830891667 2.541100e+05 2.231110e+05 6.822190e+03 1.229080e+03 9.164340e+03 3.761230e+03 1.679390e+03 3.584880e+03
+1332496830900000 2.599690e+05 2.246930e+05 6.183950e+03 1.538500e+03 9.222080e+03 3.139170e+03 9.499020e+02 3.180800e+03
+1332496830908333 2.590780e+05 2.269130e+05 4.388890e+03 3.694820e+03 8.195020e+03 3.933000e+03 4.260800e+02 2.388450e+03
+1332496830916667 2.545630e+05 2.247600e+05 5.168440e+03 4.020940e+03 8.450270e+03 4.758910e+03 1.458900e+03 2.286430e+03
+1332496830925000 2.580590e+05 2.212170e+05 6.883460e+03 1.649530e+03 9.232780e+03 4.457650e+03 3.057820e+03 3.031950e+03
+1332496830933333 2.646670e+05 2.211770e+05 6.218510e+03 1.645730e+03 8.657180e+03 3.663500e+03 2.528280e+03 3.978340e+03
+1332496830941667 2.629250e+05 2.243820e+05 4.627500e+03 3.635930e+03 7.892800e+03 3.431320e+03 6.045090e+02 3.901370e+03
+1332496830950000 2.547080e+05 2.254480e+05 4.408250e+03 4.461040e+03 8.197170e+03 3.953750e+03 -4.453460e+01 3.154870e+03
+1332496830958333 2.537020e+05 2.246350e+05 5.825770e+03 2.577050e+03 9.590050e+03 4.569250e+03 1.460270e+03 2.785170e+03
+1332496830966667 2.602060e+05 2.241400e+05 5.387980e+03 1.951160e+03 8.789510e+03 5.131660e+03 2.706380e+03 2.972480e+03
+1332496830975000 2.612400e+05 2.247370e+05 3.860810e+03 3.418310e+03 7.414530e+03 5.284520e+03 2.271380e+03 3.183150e+03
+1332496830983333 2.561400e+05 2.232520e+05 3.850010e+03 3.957140e+03 7.262650e+03 4.964640e+03 1.499510e+03 3.453130e+03
+1332496830991667 2.561160e+05 2.213490e+05 5.594480e+03 2.054400e+03 8.835130e+03 3.662010e+03 1.485510e+03 3.613010e+03
--- a/tests/data/prep-20120323T1004-badtimes
+++ b/tests/data/prep-20120323T1004-badtimes
@@ -1,11 +1,11 @@
-1332497040.000000 2.56439e+05  2.24775e+05  2.92897e+03  4.66646e+03  7.58491e+03  3.57351e+03  -4.34171e+02  2.98819e+03  
-1332497040.010000 2.51903e+05  2.23202e+05  4.23696e+03  3.49363e+03  8.53493e+03  4.29416e+03  8.49573e+02  2.38189e+03  
-1332497040.020000 2.57625e+05  2.20247e+05  5.47017e+03  1.35872e+03  9.18903e+03  4.56136e+03  2.65599e+03  2.60912e+03  
-1332497040.030000 2.63375e+05  2.20706e+05  4.51842e+03  1.80758e+03  8.17208e+03  4.17463e+03  2.57884e+03  3.32848e+03  
-1332497040.040000 2.59221e+05  2.22346e+05  2.98879e+03  3.66264e+03  6.87274e+03  3.94223e+03  1.25928e+03  3.51786e+03  
-1332497040.050000 2.51918e+05  2.22281e+05  4.22677e+03  2.84764e+03  7.78323e+03  3.81659e+03  8.04944e+02  3.46314e+03  
-1332497040.050000 2.54478e+05  2.21701e+05  5.61366e+03  1.02262e+03  9.26581e+03  3.50152e+03  1.29331e+03  3.07271e+03  
-1332497040.060000 2.59568e+05  2.22945e+05  4.97190e+03  1.28250e+03  8.62081e+03  4.06316e+03  1.85717e+03  2.61990e+03  
-1332497040.070000 2.57269e+05  2.23697e+05  3.60527e+03  3.05749e+03  7.22363e+03  4.90330e+03  1.93736e+03  2.35357e+03  
-1332497040.080000 2.52274e+05  2.21438e+05  5.01228e+03  2.86309e+03  7.87115e+03  4.80448e+03  2.18291e+03  2.93397e+03  
-1332497040.090000 2.56468e+05  2.19205e+05  6.29804e+03  8.09467e+02  9.12895e+03  3.52055e+03  2.16980e+03  3.88739e+03  
+1332497040000000 2.56439e+05  2.24775e+05  2.92897e+03  4.66646e+03  7.58491e+03  3.57351e+03  -4.34171e+02  2.98819e+03  
+1332497040010000 2.51903e+05  2.23202e+05  4.23696e+03  3.49363e+03  8.53493e+03  4.29416e+03  8.49573e+02  2.38189e+03  
+1332497040020000 2.57625e+05  2.20247e+05  5.47017e+03  1.35872e+03  9.18903e+03  4.56136e+03  2.65599e+03  2.60912e+03  
+1332497040030000 2.63375e+05  2.20706e+05  4.51842e+03  1.80758e+03  8.17208e+03  4.17463e+03  2.57884e+03  3.32848e+03  
+1332497040040000 2.59221e+05  2.22346e+05  2.98879e+03  3.66264e+03  6.87274e+03  3.94223e+03  1.25928e+03  3.51786e+03  
+1332497040050000 2.51918e+05  2.22281e+05  4.22677e+03  2.84764e+03  7.78323e+03  3.81659e+03  8.04944e+02  3.46314e+03  
+1332497040050000 2.54478e+05  2.21701e+05  5.61366e+03  1.02262e+03  9.26581e+03  3.50152e+03  1.29331e+03  3.07271e+03  
+1332497040060000 2.59568e+05  2.22945e+05  4.97190e+03  1.28250e+03  8.62081e+03  4.06316e+03  1.85717e+03  2.61990e+03  
+1332497040070000 2.57269e+05  2.23697e+05  3.60527e+03  3.05749e+03  7.22363e+03  4.90330e+03  1.93736e+03  2.35357e+03  
+1332497040080000 2.52274e+05  2.21438e+05  5.01228e+03  2.86309e+03  7.87115e+03  4.80448e+03  2.18291e+03  2.93397e+03  
+1332497040090000 2.56468e+05  2.19205e+05  6.29804e+03  8.09467e+02  9.12895e+03  3.52055e+03  2.16980e+03  3.88739e+03  
--- a/tests/data/prep-20120323T1004-timestamped
+++ b/tests/data/prep-20120323T1004-timestamped
--- a/tests/test_bulkdata.py
+++ b/tests/test_bulkdata.py
@@ -49,31 +49,40 @@ class TestBulkData(object):
            nodes.append(data.getnode("/foo/baz/quux"))
        del nodes

+        def get_node_slice(key):
+            if isinstance(key, slice):
+                return [ node.get_data(x, x+1) for x in
+                         xrange(*key.indices(node.nrows)) ]
+            return node.get_data(key, key+1)
+
        # Test node
        node = data.getnode("/foo/bar")
        with assert_raises(IndexError):
-            x = node[0]
+            x = get_node_slice(0)
+        with assert_raises(IndexError):
+            x = node[0] # timestamp
        raw = []
        for i in range(1000):
-            raw.append([10000+i, 1, 2, 3, 4, 5, 6, 7, 8 ])
-        node.append(raw[0:1])
-        node.append(raw[1:100])
-        node.append(raw[100:])
+            raw.append("%d 1 2 3 4 5 6 7 8\n" % (10000 + i))
+        node.append_string("".join(raw[0:1]), 0, 50000)
+        node.append_string("".join(raw[1:100]), 0, 50000)
+        node.append_string("".join(raw[100:]), 0, 50000)

        misc_slices = [ 0, 100, slice(None), slice(0), slice(10),
                        slice(5,10), slice(3,None), slice(3,-3),
                        slice(20,10), slice(200,100,-1), slice(None,0,-1),
                        slice(100,500,5) ]
+
        # Extract slices
        for s in misc_slices:
-            eq_(node[s], raw[s])
+            eq_(get_node_slice(s), raw[s])

        # Extract misc slices while appending, to make sure the
        # data isn't being added in the middle of the file
        for s in [2, slice(1,5), 2, slice(1,5)]:
-            node.append([[0,0,0,0,0,0,0,0,0]])
-            raw.append([0,0,0,0,0,0,0,0,0])
-            eq_(node[s], raw[s])
+            node.append_string("0 0 0 0 0 0 0 0 0\n", 0, 50000)
+            raw.append("0 0 0 0 0 0 0 0 0\n")
+            eq_(get_node_slice(s), raw[s])

        # Get some coverage of remove; remove is more fully tested
        # in cmdline
@@ -91,7 +100,7 @@ class TestBulkData(object):

        # Extract slices
        for s in misc_slices:
-            eq_(node[s], raw[s])
+            eq_(get_node_slice(s), raw[s])

        # destroy
        with assert_raises(ValueError):
--- a/tests/test_client.py
+++ b/tests/test_client.py
@@ -186,8 +186,7 @@ class TestClient(object):
        datetime_tz.localtz_set("America/New_York")

        testfile = "tests/data/prep-20120323T1000"
-        start = datetime_tz.datetime_tz.smartparse("20120323T1000")
-        start = start.totimestamp()
+        start = nilmdb.utils.time.parse_time("20120323T1000")
        rate = 120

        # First try a nonexistent path
@@ -243,32 +242,32 @@ class TestClient(object):
        data = timestamper.TimestamperRate(testfile, start, 120)
        with assert_raises(ClientError) as e:
            result = client.stream_insert("/newton/prep", data,
-                                          start + 5, start + 120)
+                                          start + 5000000, start + 120000000)
        in_("400 Bad Request", str(e.exception))
-        in_("Data timestamp 1332511200.000000 < start time 1332511205.000000",
+        in_("Data timestamp 1332511200000000 < start time 1332511205000000",
            str(e.exception))

        # Specify start/end (ends too early)
        data = timestamper.TimestamperRate(testfile, start, 120)
        with assert_raises(ClientError) as e:
            result = client.stream_insert("/newton/prep", data,
-                                          start, start + 1)
+                                          start, start + 1000000)
        in_("400 Bad Request", str(e.exception))
        # Client chunks the input, so the exact timestamp here might change
        # if the chunk positions change.
-        assert(re.search("Data timestamp 13325[0-9]+\.[0-9]+ "
-                         ">= end time 1332511201.000000", str(e.exception))
+        assert(re.search("Data timestamp 13325[0-9]+ "
+                         ">= end time 1332511201000000", str(e.exception))
               is not None)

        # Now do the real load
        data = timestamper.TimestamperRate(testfile, start, 120)
        result = client.stream_insert("/newton/prep", data,
-                                      start, start + 119.999777)
+                                      start, start + 119999777)

        # Verify the intervals.  Should be just one, even if the data
        # was inserted in chunks, due to nilmdb interval concatenation.
        intervals = list(client.stream_intervals("/newton/prep"))
-        eq_(intervals, [[start, start + 119.999777]])
+        eq_(intervals, [[start, start + 119999777]])

        # Try some overlapping data -- just insert it again
        data = timestamper.TimestamperRate(testfile, start, 120)
@@ -284,11 +283,12 @@ class TestClient(object):
        # Misc tests for extract and remove.  Most of them are in test_cmdline.
        client = nilmdb.client.Client(url = testurl)

-        for x in client.stream_extract("/newton/prep", 999123, 999124):
+        for x in client.stream_extract("/newton/prep",
+                                       999123000000, 999124000000):
            raise AssertionError("shouldn't be any data for this request")

        with assert_raises(ClientError) as e:
-            client.stream_remove("/newton/prep", 123, 120)
+            client.stream_remove("/newton/prep", 123000000, 120000000)

        # Test count
        eq_(client.stream_count("/newton/prep"), 14400)
@@ -301,13 +301,11 @@ class TestClient(object):
        client = nilmdb.client.Client(url = testurl)

        # Trigger a client error in generator
-        start = datetime_tz.datetime_tz.smartparse("20120323T2000")
-        end = datetime_tz.datetime_tz.smartparse("20120323T1000")
+        start = nilmdb.utils.time.parse_time("20120323T2000")
+        end = nilmdb.utils.time.parse_time("20120323T1000")
        for function in [ client.stream_intervals, client.stream_extract ]:
            with assert_raises(ClientError) as e:
-                function("/newton/prep",
-                         start.totimestamp(),
-                         end.totimestamp()).next()
+                function("/newton/prep", start, end).next()
            in_("400 Bad Request", str(e.exception))
            in_("start must precede end", str(e.exception))

@@ -455,21 +453,21 @@ class TestClient(object):
            ctx.insert("  105 1\n")
            ctx.finalize()

-            ctx.insert("106 1\n")
-            ctx.update_end(106.5)
-            ctx.finalize()
-            ctx.update_start(106.8)
            ctx.insert("107 1\n")
-            ctx.insert("108 1\n")
-            ctx.insert("109 1\n")
+            ctx.update_end(108)
+            ctx.finalize()
+            ctx.update_start(109)
            ctx.insert("110 1\n")
            ctx.insert("111 1\n")
-            ctx.update_end(113)
            ctx.insert("112 1\n")
-            ctx.update_end(114)
            ctx.insert("113 1\n")
-            ctx.update_end(115)
-            ctx.insert("114 1" +
+            ctx.insert("114 1\n")
+            ctx.update_end(116)
+            ctx.insert("115 1\n")
+            ctx.update_end(117)
+            ctx.insert("116 1\n")
+            ctx.update_end(118)
+            ctx.insert("117 1" +
                       " # this is super long" * 100 +
                       "\n")
            ctx.finalize()
@@ -477,11 +475,11 @@ class TestClient(object):

        with assert_raises(ClientError):
            with client.stream_insert_context("/context/test", 100, 200) as ctx:
-                ctx.insert("115 1\n")
+                ctx.insert("118 1\n")

        with assert_raises(ClientError):
            with client.stream_insert_context("/context/test", 200, 300) as ctx:
-                ctx.insert("115 1\n")
+                ctx.insert("118 1\n")

        with assert_raises(ClientError):
            with client.stream_insert_context("/context/test") as ctx:
@@ -503,9 +501,9 @@ class TestClient(object):
                ctx.finalize()

        eq_(list(client.stream_intervals("/context/test")),
-            [ [ 100, 105.000001 ],
-              [ 106, 106.5 ],
-              [ 106.8, 115 ],
+            [ [ 100, 106 ],
+              [ 107, 108 ],
+              [ 109, 118 ],
              [ 200, 300 ] ])

        client.stream_destroy("/context/test")
@@ -567,7 +565,7 @@ class TestClient(object):
            ctx.finalize() # nothing
            ctx.finalize() # nothing
            ctx.insert("1100 1\n")
-            ctx.finalize() # inserts [1100, 1100.000001]
+            ctx.finalize() # inserts [1100, 1101]
            ctx.update_start(1199)
            ctx.insert("1200 1\n")
            ctx.update_end(1250)
@@ -595,7 +593,7 @@ class TestClient(object):
                     (0, [400, 450]),
                     (0, [500, 550]),
                     (0, [1000, 1050]),
-                     (1, [1100, 1100.000001]),
+                     (1, [1100, 1101]),
                     (1, [1199, 1250]),
                     (0, [1400, 1450]),
                     (0, [1500, 1550]),
@@ -639,3 +637,29 @@ class TestClient(object):
            # Clean up
            c.stream_destroy("/persist/test")
            eq_(connections(), (1, 6))
+
+    def test_client_13_timestamp_rounding(self):
+        # Test potentially bad timestamps (due to floating point
+        # roundoff etc).  The server will round floating point values
+        # to the nearest int.
+        client = nilmdb.client.Client(testurl)
+
+        client.stream_create("/rounding/test", "uint16_1")
+        with client.stream_insert_context("/rounding/test",
+                                          100000000, 200000000.1) as ctx:
+            ctx.insert("100000000.1 1\n")
+            ctx.insert("150000000.00003 1\n")
+            ctx.insert("199999999.4 1\n")
+        eq_(list(client.stream_intervals("/rounding/test")),
+            [ [ 100000000, 200000000 ] ])
+
+        with assert_raises(ClientError):
+            with client.stream_insert_context("/rounding/test",
+                                              200000000, 300000000) as ctx:
+                ctx.insert("200000000 1\n")
+                ctx.insert("250000000 1\n")
+                # Server will round this and give an error on finalize()
+                ctx.insert("299999999.99 1\n")
+
+        client.stream_destroy("/rounding/test")
+        client.close()
--- a/tests/test_cmdline.py
+++ b/tests/test_cmdline.py
@@ -15,6 +15,7 @@ import re
 import sys
 import StringIO
 import shlex
+import warnings

 from testutil.helpers import *

@@ -215,9 +216,11 @@ class TestCmdline(object):
    def test_02_parsetime(self):
        os.environ['TZ'] = "America/New_York"
        test = datetime_tz.datetime_tz.now()
+        u2ts = nilmdb.utils.time.unix_to_timestamp
        parse_time = nilmdb.utils.time.parse_time
-        eq_(parse_time(str(test)), test)
-        test = datetime_tz.datetime_tz.smartparse("20120405 1400-0400")
+        eq_(parse_time(str(test)), u2ts(test.totimestamp()))
+        test = u2ts(datetime_tz.datetime_tz.smartparse("20120405 1400-0400").
+                    totimestamp())
        eq_(parse_time("hi there 20120405 1400-0400 testing! 123"), test)
        eq_(parse_time("20120405 1800 UTC"), test)
        eq_(parse_time("20120405 1400-0400 UTC"), test)
@@ -227,6 +230,11 @@ class TestCmdline(object):
        x = parse_time("now")
        eq_(parse_time("snapshot-20120405-140000.raw.gz"), test)
        eq_(parse_time("prep-20120405T1400"), test)
+        eq_(parse_time("1333648800.0"), test)
+        eq_(parse_time("1333648800000000"), test)
+        eq_(parse_time("@1333648800000000"), test)
+        with assert_raises(ValueError):
+            parse_time("@hashtag12345")

    def test_03_info(self):
        self.ok("info")
@@ -250,6 +258,15 @@ class TestCmdline(object):

        self.fail("create /foo float32_8")
        self.contain("invalid path")
+        self.fail("create /newton/prep/ float32_8")
+        self.contain("invalid path")
+
+        self.fail("create /newton/_format/prep float32_8")
+        self.contain("path name is invalid")
+        self.fail("create /_format/newton/prep float32_8")
+        self.contain("path name is invalid")
+        self.fail("create /newton/prep/_format float32_8")
+        self.contain("path name is invalid")

        # Bad layout type
        self.fail("create /newton/prep NoSuchLayout")
@@ -264,6 +281,10 @@ class TestCmdline(object):
        self.ok("create /newton/prep float32_8")
        self.ok("create /newton/raw uint16_6")

+        # Create a stream that already exists
+        self.fail("create /newton/raw uint16_6")
+        self.contain("stream already exists at this path")
+
        # Should not be able to create a stream with another stream as
        # its parent
        self.fail("create /newton/prep/blah float32_8")
@@ -491,14 +512,14 @@ class TestCmdline(object):
        self.ok("list --detail --path *prep --timestamp-raw "
                "--start='23 Mar 2012 10:05:15.50'")
        lines_(self.captured, 2)
-        self.contain("[ 1332497115.500000 -> 1332497160.000000 ]")
+        self.contain("[ 1332497115500000 -> 1332497160000000 ]")

        # bad time
        self.fail("list --detail --path *prep -T --start='9332497115.612'")
        # good time
        self.ok("list --detail --path *prep -T --start='1332497115.612'")
        lines_(self.captured, 2)
-        self.contain("[ 1332497115.612000 -> 1332497160.000000 ]")
+        self.contain("[ 1332497115612000 -> 1332497160000000 ]")

        # Check --ext output
        self.ok("list --ext")
@@ -506,7 +527,7 @@ class TestCmdline(object):

        self.ok("list -E -T")
        c = self.contain
-        c("\n  interval extents: 1332496800.000000 -> 1332497160.000000\n")
+        c("\n  interval extents: 1332496800000000 -> 1332497160000000\n")
        c("\n        total data: 43200 rows, 359.983336 seconds\n")
        c("\n  interval extents: (no data)\n")
        c("\n        total data: 0 rows, 0.000000 seconds\n")
@@ -955,3 +976,85 @@ class TestCmdline(object):

        self.ok("destroy /diff/1")
        self.ok("destroy /diff/2")
+
+    def test_16_rename(self):
+        # Test renaming.  Force file size smaller so we get more files
+        server_stop()
+        recursive_unlink(testdb)
+        server_start(bulkdata_args = { "file_size" : 920, # 23 rows per file
+                                       "files_per_dir" : 3 })
+
+
+        # Fill data
+        self.ok("create /newton/prep float32_8")
+        os.environ['TZ'] = "UTC"
+        with open("tests/data/prep-20120323T1004-timestamped") as input:
+            self.ok("insert -s 20120323T1004 -e 20120323T1006 /newton/prep",
+                    input)
+
+        # Extract it
+        self.ok("extract /newton/prep --start '2000-01-01' " +
+                "--end '2012-03-23 10:04:01'")
+        extract_before = self.captured
+
+        def check_path(*components):
+            # Verify the paths look right on disk
+            seek = os.path.join(testdb, "data", *components)
+            for (dirpath, dirnames, filenames) in os.walk(testdb):
+                if "_format" in filenames:
+                    if dirpath == seek:
+                        break
+                    raise AssertionError("data also found at " + dirpath)
+            else:
+                raise AssertionError("data not found at " + seek)
+            # Verify "list" output
+            self.ok("list")
+            self.match("/" + "/".join(components) + " float32_8\n")
+
+        # Lots of renames
+        check_path("newton", "prep")
+
+        self.fail("rename /newton/prep /newton/prep")
+        self.contain("old and new paths are the same")
+        check_path("newton", "prep")
+        self.fail("rename /newton/prep /newton")
+        self.contain("subdirs of this path already exist")
+        self.fail("rename /newton/prep /newton/prep/")
+        self.contain("invalid path")
+        self.ok("rename /newton/prep /newton/foo")
+        check_path("newton", "foo")
+        self.ok("rename /newton/foo /totally/different/thing")
+        check_path("totally", "different", "thing")
+        self.ok("rename /totally/different/thing /totally/something")
+        check_path("totally", "something")
+        self.ok("rename /totally/something /totally/something/cool")
+        check_path("totally", "something", "cool")
+        self.ok("rename /totally/something/cool /foo/bar")
+        check_path("foo", "bar")
+        self.ok("create /xxx/yyy/zzz float32_8")
+        self.fail("rename /foo/bar /xxx/yyy")
+        self.contain("subdirs of this path already exist")
+        self.fail("rename /foo/bar /xxx/yyy/zzz")
+        self.contain("stream already exists at this path")
+        self.fail("rename /foo/bar /xxx/yyy/zzz/www")
+        self.contain("path is subdir of existing node")
+        self.ok("rename /foo/bar /xxx/yyy/mmm")
+        self.ok("destroy /xxx/yyy/zzz")
+        check_path("xxx", "yyy", "mmm")
+
+        # Extract it at the final path
+        self.ok("extract /xxx/yyy/mmm --start '2000-01-01' " +
+                "--end '2012-03-23 10:04:01'")
+        eq_(self.captured, extract_before)
+
+        self.ok("destroy /xxx/yyy/mmm")
+
+        # Make sure temporary rename dirs weren't left around
+        for (dirpath, dirnames, filenames) in os.walk(testdb):
+            if "rename-" in dirpath:
+                raise AssertionError("temporary directories not cleaned up")
+            if "totally" in dirpath or "newton" in dirpath:
+                raise AssertionError("old directories not cleaned up")
+
+        server_stop()
+        server_start()
--- a/tests/test_interval.py
+++ b/tests/test_interval.py
@@ -51,7 +51,7 @@ class TestInterval:
        # Test Interval class
        os.environ['TZ'] = "America/New_York"
        datetime_tz._localtz = None
-        (d1, d2, d3) = [ datetime_tz.datetime_tz.smartparse(x).totimestamp()
+        (d1, d2, d3) = [ nilmdb.utils.time.parse_time(x)
                         for x in [ "03/24/2012", "03/25/2012", "03/26/2012" ] ]

        # basic construction
@@ -77,8 +77,8 @@ class TestInterval:
        assert(Interval(d1, d3) > Interval(d1, d2))
        assert(Interval(d1, d2) < Interval(d2, d3))
        assert(Interval(d1, d3) < Interval(d2, d3))
-        assert(Interval(d2, d2+0.01) > Interval(d1, d3))
-        assert(Interval(d3, d3+0.01) == Interval(d3, d3+0.01))
+        assert(Interval(d2, d2+1) > Interval(d1, d3))
+        assert(Interval(d3, d3+1) == Interval(d3, d3+1))
        #with assert_raises(TypeError): # was AttributeError, that's wrong
        #    x = (i == 123)

@@ -87,16 +87,16 @@ class TestInterval:
        with assert_raises(IntervalError):
            x = Interval(d2, d3).subset(d1, d2)

-        # big integers and floats
-        x = Interval(5000111222, 6000111222)
-        eq_(str(x), "[5000111222.000000 -> 6000111222.000000)")
-        x = Interval(123.45, 234.56)
-        eq_(str(x), "[123.450000 -> 234.560000)")
+        # big integers, negative integers
+        x = Interval(5000111222000000, 6000111222000000)
+        eq_(str(x), "[5000111222000000 -> 6000111222000000)")
+        x = Interval(-5000111222000000, -4000111222000000)
+        eq_(str(x), "[-5000111222000000 -> -4000111222000000)")

        # misc
        i = Interval(d1, d2)
        eq_(repr(i), repr(eval(repr(i))))
-        eq_(str(i), "[1332561600.000000 -> 1332648000.000000)")
+        eq_(str(i), "[1332561600000000 -> 1332648000000000)")

    def test_interval_intersect(self):
        # Test Interval intersections
@@ -193,7 +193,7 @@ class TestInterval:
        # misc
        eq_(repr(iset), repr(eval(repr(iset))))
        eq_(str(iset),
-            "[[100.000000 -> 200.000000), [200.000000 -> 300.000000)]")
+            "[[100 -> 200), [200 -> 300)]")

    def test_intervalset_geniset(self):
        # Test basic iset construction
--- a/tests/test_timestamper.py
+++ b/tests/test_timestamper.py
@@ -1,6 +1,5 @@
 import nilmdb
 from nilmdb.utils.printf import *
-from nilmdb.utils import datetime_tz

 from nose.tools import *
 from nose.tools import assert_raises
@@ -20,11 +19,11 @@ class TestTimestamper(object):
        def join(list):
            return "\n".join(list) + "\n"

-        start = datetime_tz.datetime_tz.smartparse("03/24/2012").totimestamp()
+        start = nilmdb.utils.time.parse_time("03/24/2012")
        lines_in  = [ "hello", "world", "hello world", "# commented out" ]
-        lines_out = [ "1332561600.000000 hello",
-                      "1332561600.000125 world",
-                      "1332561600.000250 hello world" ]
+        lines_out = [ "1332561600000000 hello",
+                      "1332561600000125 world",
+                      "1332561600000250 hello world" ]

        # full
        input = cStringIO.StringIO(join(lines_in))
@@ -42,7 +41,7 @@ class TestTimestamper(object):
        # stop iteration early
        input = cStringIO.StringIO(join(lines_in))
        ts = timestamper.TimestamperRate(input, start, 8000,
-                                                1332561600.000200)
+                                                1332561600000200)
        foo = ""
        for line in ts:
            foo += line
@@ -51,14 +50,14 @@ class TestTimestamper(object):
        # stop iteration early (readlines)
        input = cStringIO.StringIO(join(lines_in))
        ts = timestamper.TimestamperRate(input, start, 8000,
-                                                1332561600.000200)
+                                                1332561600000200)
        foo = ts.readlines()
        eq_(foo, join(lines_out[0:2]))

        # stop iteration really early
        input = cStringIO.StringIO(join(lines_in))
        ts = timestamper.TimestamperRate(input, start, 8000,
-                                                1332561600.000000)
+                                                1332561600000000)
        foo = ts.readlines()
        eq_(foo, "")
Author	SHA1	Message	Date
Jim Paris	52ae397d7d	Bump database version to 3, reject old version 2 due to timestamp changes	2013-03-15 18:37:38 -04:00
Jim Paris	d05b6f6348	Merge branch 'rocket-cleanup'	2013-03-15 18:08:36 -04:00
Jim Paris	049375d30e	Fill out test coverage	2013-03-15 18:08:21 -04:00
Jim Paris	88eb0123f5	Add test for Table.__getitem__ indexing	2013-03-15 18:08:21 -04:00
Jim Paris	a547ddbbba	Change table.get_timestamp to table.__getitem__ This lets us use simple indexing to get timestamps from the table, which allows us to use 'bisect' directly without needing a proxy class.	2013-03-15 18:08:21 -04:00
Jim Paris	28e72fd53e	Remove Table.__getitem__; used only by tests	2013-03-15 18:08:21 -04:00
Jim Paris	f63107b334	Add rocket.extract_timestamp to speed up bisections	2013-03-15 18:08:21 -04:00
Jim Paris	955d7aa871	Remove floating port time support from nilmdb.utils.time	2013-03-15 18:08:21 -04:00
Jim Paris	b8d2cf1b78	Consolidate rocket._extract_handle.params with extract_string	2013-03-15 18:08:21 -04:00
Jim Paris	7c465730de	Remove rocket.extract_pyobject	2013-03-15 18:08:21 -04:00
Jim Paris	aca130272d	Remove rocket.extract_list	2013-03-15 18:08:21 -04:00
Jim Paris	76e5e9883f	Remove Table.append, rocket.append_iter	2013-03-15 18:08:20 -04:00
Jim Paris	fb4f4519ff	Clean up and simplify Table.get_*, including __getitem__	2013-03-15 18:08:20 -04:00
Jim Paris	30328714a7	Remove python implementation of rocket	2013-03-15 18:08:20 -04:00
Jim Paris	759466de4a	Merge branch 'timestamp-integers'	2013-03-15 18:07:51 -04:00
Jim Paris	d3efb829b5	Try to parse timestamps as double, if int64 parse fails	2013-03-15 15:19:41 -04:00
Jim Paris	90b96799ac	Bulk of the switch to int64 microsecond timestamps, including test data.	2013-03-15 15:08:58 -04:00
Jim Paris	56679ad770	Move more datetime_tz calls into common code	2013-03-15 15:08:58 -04:00
Jim Paris	b5541722c2	Continue moving time-handling code into nilmdb.utils.time	2013-03-15 15:08:58 -04:00
Jim Paris	aaea105861	Consolidate most timestamp <-> string conversions (outside of rocket)	2013-03-15 15:08:57 -04:00
Jim Paris	e6a081d639	Consolidate timestamp constants into nilmdb.utils.time	2013-03-15 15:08:57 -04:00
Jim Paris	1835d03412	Bump bulkdata database version to 3	2013-03-15 15:08:57 -04:00
Jim Paris	c7a712d8d8	Partial test for rounding issues	2013-03-15 15:08:57 -04:00
Jim Paris	20d315b4f7	Add documentation about upcoming timestamp changes	2013-03-15 15:08:57 -04:00
Jim Paris	a44a5e3135	Merge branch 'argcomplete'	2013-03-15 15:08:42 -04:00
Jim Paris	039b2a0557	Include nilmtool-bash-completion.sh script in .tar.gz	2013-03-15 15:08:28 -04:00
Jim Paris	cd1dfe7dcd	Add completion functions to most commandline arguments	2013-03-15 14:26:38 -04:00
Jim Paris	fb35517dfa	Add basic argument completion	2013-03-15 13:57:35 -04:00
Jim Paris	b9f0b35bbe	Stream renaming support, and comprehensive tests Implemented in command line, client, server, nilmdb, bulkdata	2013-03-14 11:02:30 -04:00
Jim Paris	b1b09f8cd0	Strengthen checks when creating paths, fix some bugs, and add tests	2013-03-13 17:45:47 -04:00
Jim Paris	d467df7980	Add specific error for creating a path that already exists	2013-03-13 10:14:28 -04:00