| 1 | import re
|
|---|
| 2 |
|
|---|
| 3 | import time
|
|---|
| 4 | import datetime
|
|---|
| 5 | import pytz
|
|---|
| 6 |
|
|---|
| 7 | from ....builtins import *
|
|---|
| 8 |
|
|---|
| 9 | from ...exceptions import *
|
|---|
| 10 | from ...expressions import *
|
|---|
| 11 | from ...autotype import *
|
|---|
| 12 | from ...streams import *
|
|---|
| 13 |
|
|---|
| 14 | ##
|
|---|
| 15 | ## nbstlutil -U records begin with the string "Image:" on a line by itself
|
|---|
| 16 | ##
|
|---|
| 17 | def stream(stream, format='nbstlutil list -U'):
|
|---|
| 18 |
|
|---|
| 19 | if format in ['nbstlutil list -U']:
|
|---|
| 20 | return LineSeparatorStream(stream, separator='Image:', header=1)
|
|---|
| 21 | else:
|
|---|
| 22 | raise ParseError, 'Unknown format %s' % (format)
|
|---|
| 23 |
|
|---|
| 24 |
|
|---|
| 25 | ##
|
|---|
| 26 | ## Parse a nbstlutil record
|
|---|
| 27 | ##
|
|---|
| 28 | def parse(record, format='nbstlutil list -U', version=None, tz=None):
|
|---|
| 29 |
|
|---|
| 30 | re_pair = re.compile('^\s*([^:]+):\s*(.*)\s*$')
|
|---|
| 31 | re_type_id = re.compile('^([0-9a-zA-Z\-_ ]+)\s+\((\d+)\)$')
|
|---|
| 32 | re_id_type = re.compile('^(\d+)\s+\(([0-9a-zA-Z\-_ ]+)\)$')
|
|---|
| 33 | re_in_parens = re.compile('^\((.*)\)$')
|
|---|
| 34 | re_starts_with_time = re.compile('^time_.*$')
|
|---|
| 35 | re_ends_with_time = re.compile('^.*_time$')
|
|---|
| 36 | re_time_plus_epoch_seconds_in_parens = re.compile('^.*\d:\d\d:\d\d.*\((\d+)\)$')
|
|---|
| 37 | re_epoch_seconds_plus_time_in_parens = re.compile('^(\d+)\s+\(.*\d:\d\d:\d\d.*\)$')
|
|---|
| 38 |
|
|---|
| 39 | image = ExtendedDict()
|
|---|
| 40 |
|
|---|
| 41 | if format == 'nbstlutil list -U':
|
|---|
| 42 |
|
|---|
| 43 | header = True
|
|---|
| 44 |
|
|---|
| 45 | try:
|
|---|
| 46 |
|
|---|
| 47 | i = 0
|
|---|
| 48 |
|
|---|
| 49 | #
|
|---|
| 50 | # process image header
|
|---|
| 51 | #
|
|---|
| 52 | while header and i < len(record):
|
|---|
| 53 |
|
|---|
| 54 | line = record[i]
|
|---|
| 55 |
|
|---|
| 56 | i += 1
|
|---|
| 57 |
|
|---|
| 58 | #
|
|---|
| 59 | # skip blank lines
|
|---|
| 60 | #
|
|---|
| 61 | if line == '':
|
|---|
| 62 | continue
|
|---|
| 63 |
|
|---|
| 64 | match = re_pair.match(line)
|
|---|
| 65 |
|
|---|
| 66 | key = match.group(1)
|
|---|
| 67 | value = match.group(2)
|
|---|
| 68 | key = key.lower()
|
|---|
| 69 | key = key.rstrip()
|
|---|
| 70 | key = key.replace('-', '_')
|
|---|
| 71 | key = key.replace(' ', '_')
|
|---|
| 72 |
|
|---|
| 73 | if key == 'copy':
|
|---|
| 74 | header = False
|
|---|
| 75 | else:
|
|---|
| 76 |
|
|---|
| 77 | #
|
|---|
| 78 | # convert text values to python datatypes
|
|---|
| 79 | #
|
|---|
| 80 | while True:
|
|---|
| 81 |
|
|---|
| 82 | #
|
|---|
| 83 | # key name starts with time_ or ends with _time
|
|---|
| 84 | #
|
|---|
| 85 | if re_ends_with_time.match(key) or re_starts_with_time.match(key):
|
|---|
| 86 |
|
|---|
| 87 | match = re_epoch_seconds_plus_time_in_parens.match(value)
|
|---|
| 88 | if match:
|
|---|
| 89 | value = datetime.datetime.fromtimestamp(float(match.group(1)), tz)
|
|---|
| 90 | break
|
|---|
| 91 |
|
|---|
| 92 | match = re_time_plus_epoch_seconds_in_parens.match(value)
|
|---|
| 93 | if match:
|
|---|
| 94 | value = datetime.datetime.fromtimestamp(float(match.group(1)), tz)
|
|---|
| 95 | break
|
|---|
| 96 |
|
|---|
| 97 | #
|
|---|
| 98 | # separate id and type
|
|---|
| 99 | #
|
|---|
| 100 | match = re_id_type.match(value)
|
|---|
| 101 | if match:
|
|---|
| 102 | idkey = '%s_id' % (key)
|
|---|
| 103 | value = match.group(2)
|
|---|
| 104 | image[idkey] = int(match.group(1))
|
|---|
| 105 | break
|
|---|
| 106 |
|
|---|
| 107 | #
|
|---|
| 108 | # strip enclosing parens
|
|---|
| 109 | #
|
|---|
| 110 | match = re_in_parens.match(value)
|
|---|
| 111 | if match:
|
|---|
| 112 | value = match.group(1)
|
|---|
| 113 | break
|
|---|
| 114 |
|
|---|
| 115 | #
|
|---|
| 116 | # separate type and id
|
|---|
| 117 | #
|
|---|
| 118 | match = re_type_id.match(value)
|
|---|
| 119 | if match:
|
|---|
| 120 | idkey = '%s_id' % (key)
|
|---|
| 121 | value = match.group(1)
|
|---|
| 122 | image[idkey] = int(match.group(2))
|
|---|
| 123 | break
|
|---|
| 124 |
|
|---|
| 125 | #
|
|---|
| 126 | # ..otherwise autodetect and break if no datatype conversion was possible
|
|---|
| 127 | #
|
|---|
| 128 | value = autotype(value)
|
|---|
| 129 | break
|
|---|
| 130 |
|
|---|
| 131 | #
|
|---|
| 132 | # store value
|
|---|
| 133 | #
|
|---|
| 134 | image[key] = value
|
|---|
| 135 |
|
|---|
| 136 | #
|
|---|
| 137 | # parse image copies
|
|---|
| 138 | #
|
|---|
| 139 | copies = ExtendedDict()
|
|---|
| 140 | fragment = False
|
|---|
| 141 |
|
|---|
| 142 | while i < len(record):
|
|---|
| 143 |
|
|---|
| 144 | line = record[i]
|
|---|
| 145 |
|
|---|
| 146 | i += 1
|
|---|
| 147 |
|
|---|
| 148 | #
|
|---|
| 149 | # skip blank lines
|
|---|
| 150 | #
|
|---|
| 151 | if line == '':
|
|---|
| 152 | continue
|
|---|
| 153 |
|
|---|
| 154 | match = re_pair.match(line)
|
|---|
| 155 |
|
|---|
| 156 | key = match.group(1)
|
|---|
| 157 | value = match.group(2)
|
|---|
| 158 | key = key.lower()
|
|---|
| 159 | key = key.rstrip()
|
|---|
| 160 | key = key.replace('-', '_')
|
|---|
| 161 | key = key.replace(' ', '_')
|
|---|
| 162 |
|
|---|
| 163 | if key == 'copy':
|
|---|
| 164 | fragment = False
|
|---|
| 165 |
|
|---|
| 166 | if key == 'fragment':
|
|---|
| 167 | fragment = True
|
|---|
| 168 | #
|
|---|
| 169 | # #
|
|---|
| 170 | # # identify current copy
|
|---|
| 171 | # #
|
|---|
| 172 | # if key == 'copy_number':
|
|---|
| 173 | #
|
|---|
| 174 | # copy_number = int(value)
|
|---|
| 175 | #
|
|---|
| 176 | # if copy_number not in copies:
|
|---|
| 177 | # copies[copy_number] = ExtendedDict()
|
|---|
| 178 | # copies[copy_number]['copy_number'] = copy_number
|
|---|
| 179 | # copies[copy_number]['fragments'] = ExtendedDict()
|
|---|
| 180 | #
|
|---|
| 181 | # #
|
|---|
| 182 | # # identify current fragment
|
|---|
| 183 | # #
|
|---|
| 184 | # elif key == 'fragment':
|
|---|
| 185 | #
|
|---|
| 186 | # if value == 'TIR (-1)':
|
|---|
| 187 | # fragment = -1
|
|---|
| 188 | # elif value == 'TIR (-2)':
|
|---|
| 189 | # fragment = -2
|
|---|
| 190 | # else:
|
|---|
| 191 | # fragment = int(value)
|
|---|
| 192 | #
|
|---|
| 193 | # if fragment not in copies[copy_number].fragments:
|
|---|
| 194 | # copies[copy_number].fragments[fragment] = ExtendedDict()
|
|---|
| 195 | # copies[copy_number].fragments[fragment]['fragment'] = fragment
|
|---|
| 196 | #
|
|---|
| 197 | # #
|
|---|
| 198 | # # save the key and value using the copy and fragment
|
|---|
| 199 | # #
|
|---|
| 200 | # else:
|
|---|
| 201 | #
|
|---|
| 202 | # #
|
|---|
| 203 | # # match timestamp with Unix time in parentheses
|
|---|
| 204 | # #
|
|---|
| 205 | # match = re_nbu_datetime.match(line)
|
|---|
| 206 | # if match:
|
|---|
| 207 | # value = datetime.datetime.fromtimestamp(float(match.group(1)), tz)
|
|---|
| 208 | #
|
|---|
| 209 | # #
|
|---|
| 210 | # # separate type and id
|
|---|
| 211 | # #
|
|---|
| 212 | # elif re_type_id.match(value):
|
|---|
| 213 | # match = re_type_id.match(value)
|
|---|
| 214 | # idkey = '%s_id' % (key)
|
|---|
| 215 | # value = match.group(1)
|
|---|
| 216 | # copies[copy_number].fragments[fragment][idkey] = int(match.group(2))
|
|---|
| 217 | #
|
|---|
| 218 | # #
|
|---|
| 219 | # # ..otherwise autodetect
|
|---|
| 220 | # #
|
|---|
| 221 | # else:
|
|---|
| 222 | # value = autotype(value)
|
|---|
| 223 | #
|
|---|
| 224 | # copies[copy_number].fragments[fragment][key] = value
|
|---|
| 225 | #
|
|---|
| 226 | # image['copies'] = copies
|
|---|
| 227 | #
|
|---|
| 228 |
|
|---|
| 229 | return image
|
|---|
| 230 |
|
|---|
| 231 | except Exception, e:
|
|---|
| 232 |
|
|---|
| 233 | # for line in record:
|
|---|
| 234 | # print line
|
|---|
| 235 |
|
|---|
| 236 | raise ParseError, e
|
|---|
| 237 |
|
|---|
| 238 | else:
|
|---|
| 239 |
|
|---|
| 240 | raise ParseError, 'Unknown format %s' % (format)
|
|---|
| 241 |
|
|---|
