Mercurial > babel > old > mirror

comparison scripts/import_cldr.py @ 235:d0cd235ede46

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Upgraded to CLDR 1.5 and improved timezone formatting.
author cmlenz
date Wed, 01 Aug 2007 12:32:20 +0000
parents a72de8971819
children c22f292731be
comparison
equal deleted inserted replaced
234:541b6d630575 235:d0cd235ede46
49 if len(args) != 1: 49 if len(args) != 1:
50 parser.error('incorrect number of arguments') 50 parser.error('incorrect number of arguments')
51 51
52 srcdir = args[0] 52 srcdir = args[0]
53 destdir = os.path.join(os.path.dirname(os.path.abspath(sys.argv[0])), 53 destdir = os.path.join(os.path.dirname(os.path.abspath(sys.argv[0])),
54 '..', 'babel', 'localedata') 54 '..', 'babel')
55 55
56 sup = parse(os.path.join(srcdir, 'supplemental', 'supplementalData.xml')) 56 sup = parse(os.path.join(srcdir, 'supplemental', 'supplementalData.xml'))
57
58 # import global data from the supplemental files
59 global_data = {}
60
61 territory_zones = global_data.setdefault('territory_zones', {})
62 zone_aliases = global_data.setdefault('zone_aliases', {})
63 zone_territories = global_data.setdefault('zone_territories', {})
64 for elem in sup.findall('//timezoneData/zoneFormatting/zoneItem'):
65 tzid = elem.attrib['type']
66 territory_zones.setdefault(elem.attrib['territory'], []).append(tzid)
67 zone_territories[tzid] = elem.attrib['territory']
68 if 'aliases' in elem.attrib:
69 for alias in elem.attrib['aliases'].split():
70 zone_aliases[alias] = tzid
71
72 outfile = open(os.path.join(destdir, 'global.dat'), 'wb')
73 try:
74 pickle.dump(global_data, outfile, 2)
75 finally:
76 outfile.close()
57 77
58 # build a territory containment mapping for inheritance 78 # build a territory containment mapping for inheritance
59 regions = {} 79 regions = {}
60 for elem in sup.findall('//territoryContainment/group'): 80 for elem in sup.findall('//territoryContainment/group'):
61 regions[elem.attrib['type']] = elem.attrib['contains'].split() 81 regions[elem.attrib['type']] = elem.attrib['contains'].split()
74 filenames = os.listdir(os.path.join(srcdir, 'main')) 94 filenames = os.listdir(os.path.join(srcdir, 'main'))
75 filenames.remove('root.xml') 95 filenames.remove('root.xml')
76 filenames.sort(lambda a,b: len(a)-len(b)) 96 filenames.sort(lambda a,b: len(a)-len(b))
77 filenames.insert(0, 'root.xml') 97 filenames.insert(0, 'root.xml')
78 98
79 dicts = {}
80
81 for filename in filenames: 99 for filename in filenames:
82 print>>sys.stderr, 'Processing input file %r' % filename 100 print>>sys.stderr, 'Processing input file %r' % filename
83 stem, ext = os.path.splitext(filename) 101 stem, ext = os.path.splitext(filename)
84 if ext != '.xml': 102 if ext != '.xml':
85 continue 103 continue
151 169
152 for elem in supelem.findall('weekendEnd'): 170 for elem in supelem.findall('weekendEnd'):
153 territories = elem.attrib['territories'].split() 171 territories = elem.attrib['territories'].split()
154 if territory in territories or any([r in territories for r in regions]): 172 if territory in territories or any([r in territories for r in regions]):
155 week_data['weekend_end'] = weekdays[elem.attrib['day']] 173 week_data['weekend_end'] = weekdays[elem.attrib['day']]
174
175 zone_formats = data.setdefault('zone_formats', {})
176 for elem in tree.findall('//timeZoneNames/gmtFormat'):
177 if 'draft' not in elem.attrib:
178 zone_formats['gmt'] = unicode(elem.text).replace('{0}', '%s')
179 break
180 for elem in tree.findall('//timeZoneNames/regionFormat'):
181 if 'draft' not in elem.attrib:
182 zone_formats['region'] = unicode(elem.text).replace('{0}', '%s')
183 break
184 for elem in tree.findall('//timeZoneNames/fallbackFormat'):
185 if 'draft' not in elem.attrib:
186 zone_formats['fallback'] = unicode(elem.text) \
187 .replace('{0}', '%(0)s').replace('{1}', '%(1)s')
188 break
156 189
157 time_zones = data.setdefault('time_zones', {}) 190 time_zones = data.setdefault('time_zones', {})
158 for elem in tree.findall('//timeZoneNames/zone'): 191 for elem in tree.findall('//timeZoneNames/zone'):
159 info = {} 192 info = {}
160 city = elem.findtext('exemplarCity') 193 city = elem.findtext('exemplarCity')
162 info['city'] = unicode(city) 195 info['city'] = unicode(city)
163 for child in elem.findall('long/*'): 196 for child in elem.findall('long/*'):
164 info.setdefault('long', {})[child.tag] = unicode(child.text) 197 info.setdefault('long', {})[child.tag] = unicode(child.text)
165 for child in elem.findall('short/*'): 198 for child in elem.findall('short/*'):
166 info.setdefault('short', {})[child.tag] = unicode(child.text) 199 info.setdefault('short', {})[child.tag] = unicode(child.text)
200 for child in elem.findall('usesMetazone'):
201 if 'to' not in child.attrib: # FIXME: support old mappings
202 info['use_metazone'] = child.attrib['mzone']
167 time_zones[elem.attrib['type']] = info 203 time_zones[elem.attrib['type']] = info
168 204
169 zone_aliases = data.setdefault('zone_aliases', {}) 205 meta_zones = data.setdefault('meta_zones', {})
170 if stem == 'root': 206 for elem in tree.findall('//timeZoneNames/metazone'):
171 for elem in sup.findall('//timezoneData/zoneFormatting/zoneItem'): 207 info = {}
172 if 'aliases' in elem.attrib: 208 city = elem.findtext('exemplarCity')
173 canonical_id = elem.attrib['type'] 209 if city:
174 for alias in elem.attrib['aliases'].split(): 210 info['city'] = unicode(city)
175 zone_aliases[alias] = canonical_id 211 for child in elem.findall('long/*'):
212 info.setdefault('long', {})[child.tag] = unicode(child.text)
213 for child in elem.findall('short/*'):
214 info.setdefault('short', {})[child.tag] = unicode(child.text)
215 info['common'] = elem.findtext('commonlyUsed') == 'true'
216 meta_zones[elem.attrib['type']] = info
176 217
177 for calendar in tree.findall('//calendars/calendar'): 218 for calendar in tree.findall('//calendars/calendar'):
178 if calendar.attrib['type'] != 'gregorian': 219 if calendar.attrib['type'] != 'gregorian':
179 # TODO: support other calendar types 220 # TODO: support other calendar types
180 continue 221 continue
210 continue 251 continue
211 widths[int(elem.attrib.get('type'))] = unicode(elem.text) 252 widths[int(elem.attrib.get('type'))] = unicode(elem.text)
212 253
213 eras = data.setdefault('eras', {}) 254 eras = data.setdefault('eras', {})
214 for width in calendar.findall('eras/*'): 255 for width in calendar.findall('eras/*'):
215 ewidth = {'eraNames': 'wide', 'eraAbbr': 'abbreviated'}[width.tag] 256 ewidth = {
257 'eraAbbr': 'abbreviated',
258 'eraNames': 'wide',
259 'eraNarrow': 'narrow',
260 }[width.tag]
216 widths = eras.setdefault(ewidth, {}) 261 widths = eras.setdefault(ewidth, {})
217 for elem in width.findall('era'): 262 for elem in width.findall('era'):
218 if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: 263 if 'draft' in elem.attrib and int(elem.attrib['type']) in widths:
219 continue 264 continue
220 widths[int(elem.attrib.get('type'))] = unicode(elem.text) 265 widths[int(elem.attrib.get('type'))] = unicode(elem.text)
302 currency_names[elem.attrib['type']] = unicode(name) 347 currency_names[elem.attrib['type']] = unicode(name)
303 symbol = elem.findtext('symbol') 348 symbol = elem.findtext('symbol')
304 if symbol: 349 if symbol:
305 currency_symbols[elem.attrib['type']] = unicode(symbol) 350 currency_symbols[elem.attrib['type']] = unicode(symbol)
306 351
307 dicts[stem] = data 352 outfile = open(os.path.join(destdir, 'localedata', stem + '.dat'), 'wb')
308 outfile = open(os.path.join(destdir, stem + '.dat'), 'wb')
309 try: 353 try:
310 pickle.dump(data, outfile, 2) 354 pickle.dump(data, outfile, 2)
311 finally: 355 finally:
312 outfile.close() 356 outfile.close()
313 357
Copyright (C) 2012-2017 Edgewall Software