Mercurial > babel > old > mirror
annotate scripts/import_cldr.py @ 17:aa33ad077d24
Minor date formatting improvements.
| author | cmlenz |
|---|---|
| date | Thu, 31 May 2007 14:20:04 +0000 |
| parents | b47c34d42eda |
| children | 6c2c9fc7d787 |
| rev | line source |
|---|---|
| 3 | 1 #!/usr/bin/env python |
| 2 # -*- coding: utf-8 -*- | |
| 3 # | |
| 4 # Copyright (C) 2007 Edgewall Software | |
| 5 # All rights reserved. | |
| 6 # | |
| 7 # This software is licensed as described in the file COPYING, which | |
| 8 # you should have received as part of this distribution. The terms | |
| 9 # are also available at http://babel.edgewall.org/wiki/License. | |
| 10 # | |
| 11 # This software consists of voluntary contributions made by many | |
| 12 # individuals. For the exact contribution history, see the revision | |
| 13 # history and logs, available at http://babel.edgewall.org/log/. | |
| 14 | |
| 15 import copy | |
| 16 from optparse import OptionParser | |
| 17 import os | |
| 18 import pickle | |
| 19 import sys | |
| 20 try: | |
| 21 from xml.etree.ElementTree import parse | |
| 22 except ImportError: | |
| 23 from elementtree.ElementTree import parse | |
| 24 | |
| 11 | 25 from babel import dates, numbers |
| 3 | 26 |
| 17 | 27 weekdays = {'mon': 0, 'tue': 1, 'wed': 2, 'thu': 3, 'fri': 4, 'sat': 5, |
| 28 'sun': 6} | |
|
10
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
29 |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
30 try: |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
31 any |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
32 except NameError: |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
33 def any(iterable): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
34 return filter(None, list(iterable)) |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
35 |
| 3 | 36 def _parent(locale): |
| 37 parts = locale.split('_') | |
| 38 if len(parts) == 1: | |
| 39 return 'root' | |
| 40 else: | |
| 41 return '_'.join(parts[:-1]) | |
| 42 | |
| 43 def _text(elem): | |
| 44 buf = [elem.text or ''] | |
| 45 for child in elem: | |
| 46 buf.append(_text(child)) | |
| 47 buf.append(elem.tail or '') | |
| 48 return u''.join(filter(None, buf)).strip() | |
| 49 | |
| 50 def main(): | |
| 51 parser = OptionParser(usage='%prog path/to/cldr') | |
| 52 options, args = parser.parse_args() | |
| 53 if len(args) != 1: | |
| 54 parser.error('incorrect number of arguments') | |
| 55 | |
| 56 srcdir = args[0] | |
| 57 destdir = os.path.join(os.path.dirname(os.path.abspath(sys.argv[0])), | |
| 58 '..', 'babel', 'localedata') | |
| 59 | |
|
10
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
60 sup = parse(os.path.join(srcdir, 'supplemental', 'supplementalData.xml')) |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
61 |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
62 # build a territory containment mapping for inheritance |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
63 regions = {} |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
64 for elem in sup.findall('//territoryContainment/group'): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
65 regions[elem.attrib['type']] = elem.attrib['contains'].split() |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
66 from pprint import pprint |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
67 |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
68 # Resolve territory containment |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
69 territory_containment = {} |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
70 region_items = regions.items() |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
71 region_items.sort() |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
72 for group, territory_list in region_items: |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
73 for territory in territory_list: |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
74 containers = territory_containment.setdefault(territory, set([])) |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
75 if group in territory_containment: |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
76 containers |= territory_containment[group] |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
77 containers.add(group) |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
78 |
| 3 | 79 filenames = os.listdir(os.path.join(srcdir, 'main')) |
| 80 filenames.remove('root.xml') | |
| 81 filenames.sort(lambda a,b: len(a)-len(b)) | |
| 82 filenames.insert(0, 'root.xml') | |
| 83 | |
| 84 dicts = {} | |
| 85 | |
| 86 for filename in filenames: | |
| 87 print>>sys.stderr, 'Processing input file %r' % filename | |
| 88 stem, ext = os.path.splitext(filename) | |
| 89 if ext != '.xml': | |
| 90 continue | |
| 91 | |
| 92 data = {} | |
| 93 if stem != 'root': | |
| 94 data.update(copy.deepcopy(dicts[_parent(stem)])) | |
| 95 tree = parse(os.path.join(srcdir, 'main', filename)) | |
| 96 | |
|
10
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
97 language = None |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
98 elem = tree.find('//identity/language') |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
99 if elem is not None: |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
100 language = elem.attrib['type'] |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
101 print>>sys.stderr, ' Language: %r' % language |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
102 |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
103 territory = None |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
104 elem = tree.find('//identity/territory') |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
105 if elem is not None: |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
106 territory = elem.attrib['type'] |
| 15 | 107 else: |
| 108 territory = '001' # world | |
|
10
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
109 print>>sys.stderr, ' Territory: %r' % territory |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
110 regions = territory_containment.get(territory, []) |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
111 print>>sys.stderr, ' Regions: %r' % regions |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
112 |
| 3 | 113 # <localeDisplayNames> |
| 114 | |
| 115 territories = data.setdefault('territories', {}) | |
| 116 for elem in tree.findall('//territories/territory'): | |
| 117 if 'draft' in elem.attrib and elem.attrib['type'] in territories: | |
| 118 continue | |
| 119 territories[elem.attrib['type']] = _text(elem) | |
| 120 | |
| 121 languages = data.setdefault('languages', {}) | |
| 122 for elem in tree.findall('//languages/language'): | |
| 123 if 'draft' in elem.attrib and elem.attrib['type'] in languages: | |
| 124 continue | |
| 125 languages[elem.attrib['type']] = _text(elem) | |
| 126 | |
| 127 variants = data.setdefault('variants', {}) | |
| 128 for elem in tree.findall('//variants/variant'): | |
| 129 if 'draft' in elem.attrib and elem.attrib['type'] in variants: | |
| 130 continue | |
| 131 variants[elem.attrib['type']] = _text(elem) | |
| 132 | |
| 133 scripts = data.setdefault('scripts', {}) | |
| 134 for elem in tree.findall('//scripts/script'): | |
| 135 if 'draft' in elem.attrib and elem.attrib['type'] in scripts: | |
| 136 continue | |
| 137 scripts[elem.attrib['type']] = _text(elem) | |
| 138 | |
| 139 # <dates> | |
| 140 | |
|
10
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
141 week_data = data.setdefault('week_data', {}) |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
142 supelem = sup.find('//weekData') |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
143 |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
144 for elem in supelem.findall('minDays'): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
145 territories = elem.attrib['territories'].split() |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
146 if territory in territories or any([r in territories for r in regions]): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
147 week_data['min_days'] = int(elem.attrib['count']) |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
148 |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
149 for elem in supelem.findall('firstDay'): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
150 territories = elem.attrib['territories'].split() |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
151 if territory in territories or any([r in territories for r in regions]): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
152 week_data['first_day'] = weekdays[elem.attrib['day']] |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
153 |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
154 for elem in supelem.findall('weekendStart'): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
155 territories = elem.attrib['territories'].split() |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
156 if territory in territories or any([r in territories for r in regions]): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
157 week_data['weekend_start'] = weekdays[elem.attrib['day']] |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
158 |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
159 for elem in supelem.findall('weekendEnd'): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
160 territories = elem.attrib['territories'].split() |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
161 if territory in territories or any([r in territories for r in regions]): |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
162 week_data['weekend_end'] = weekdays[elem.attrib['day']] |
|
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
163 |
| 3 | 164 time_zones = data.setdefault('time_zones', {}) |
| 165 for elem in tree.findall('//timeZoneNames/zone'): | |
| 166 time_zones[elem.tag] = unicode(elem.findtext('displayName')) | |
| 167 | |
| 168 for calendar in tree.findall('//calendars/calendar'): | |
| 169 if calendar.attrib['type'] != 'gregorian': | |
| 170 # TODO: support other calendar types | |
| 171 continue | |
| 172 | |
| 173 months = data.setdefault('months', {}) | |
| 174 for ctxt in calendar.findall('months/monthContext'): | |
| 175 ctxts = months.setdefault(ctxt.attrib['type'], {}) | |
| 176 for width in ctxt.findall('monthWidth'): | |
| 177 widths = ctxts.setdefault(width.attrib['type'], {}) | |
| 178 for elem in width.findall('month'): | |
| 179 if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: | |
| 180 continue | |
| 181 widths[int(elem.attrib.get('type'))] = unicode(elem.text) | |
| 182 | |
| 183 days = data.setdefault('days', {}) | |
| 184 for ctxt in calendar.findall('days/dayContext'): | |
| 185 ctxts = days.setdefault(ctxt.attrib['type'], {}) | |
| 186 for width in ctxt.findall('dayWidth'): | |
| 187 widths = ctxts.setdefault(width.attrib['type'], {}) | |
| 188 for elem in width.findall('day'): | |
|
10
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
189 dtype = weekdays[elem.attrib['type']] |
| 3 | 190 if 'draft' in elem.attrib and dtype in widths: |
| 191 continue | |
| 192 widths[dtype] = unicode(elem.text) | |
| 193 | |
| 194 quarters = data.setdefault('quarters', {}) | |
| 195 for ctxt in calendar.findall('quarters/quarterContext'): | |
| 196 ctxts = quarters.setdefault(ctxt.attrib['type'], {}) | |
| 197 for width in ctxt.findall('quarterWidth'): | |
| 198 widths = ctxts.setdefault(width.attrib['type'], {}) | |
| 199 for elem in width.findall('quarter'): | |
| 200 if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: | |
| 201 continue | |
| 202 widths[int(elem.attrib.get('type'))] = unicode(elem.text) | |
| 203 | |
| 204 eras = data.setdefault('eras', {}) | |
| 205 for width in calendar.findall('eras/*'): | |
| 206 ewidth = {'eraNames': 'wide', 'eraAbbr': 'abbreviated'}[width.tag] | |
| 207 widths = eras.setdefault(ewidth, {}) | |
| 208 for elem in width.findall('era'): | |
| 209 if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: | |
| 210 continue | |
| 211 widths[int(elem.attrib.get('type'))] = unicode(elem.text) | |
| 212 | |
| 213 # AM/PM | |
| 214 periods = data.setdefault('periods', {}) | |
| 215 for elem in calendar.findall('am'): | |
| 216 if 'draft' in elem.attrib and elem.tag in periods: | |
| 217 continue | |
| 218 periods[elem.tag] = unicode(elem.text) | |
| 219 for elem in calendar.findall('pm'): | |
| 220 if 'draft' in elem.attrib and elem.tag in periods: | |
| 221 continue | |
| 222 periods[elem.tag] = unicode(elem.text) | |
| 223 | |
| 224 date_formats = data.setdefault('date_formats', {}) | |
| 225 for elem in calendar.findall('dateFormats/dateFormatLength'): | |
| 226 if 'draft' in elem.attrib and elem.attrib.get('type') in date_formats: | |
| 227 continue | |
| 228 try: | |
| 229 date_formats[elem.attrib.get('type')] = \ | |
| 11 | 230 dates.parse_pattern(unicode(elem.findtext('dateFormat/pattern'))) |
| 3 | 231 except ValueError, e: |
| 232 print e | |
| 233 | |
| 234 time_formats = data.setdefault('time_formats', {}) | |
| 235 for elem in calendar.findall('timeFormats/timeFormatLength'): | |
| 236 if 'draft' in elem.attrib and elem.attrib.get('type') in time_formats: | |
| 237 continue | |
| 238 try: | |
| 239 time_formats[elem.attrib.get('type')] = \ | |
| 11 | 240 dates.parse_pattern(unicode(elem.findtext('timeFormat/pattern'))) |
| 3 | 241 except ValueError, e: |
| 242 print e | |
| 243 | |
| 244 # <numbers> | |
| 245 | |
| 246 number_symbols = data.setdefault('number_symbols', {}) | |
| 247 for elem in tree.findall('//numbers/symbols/*'): | |
| 248 number_symbols[elem.tag] = unicode(elem.text) | |
| 249 | |
| 250 decimal_formats = data.setdefault('decimal_formats', {}) | |
| 251 for elem in tree.findall('//decimalFormats/decimalFormatLength'): | |
| 252 if 'draft' in elem.attrib and elem.attrib.get('type') in decimal_formats: | |
| 253 continue | |
| 11 | 254 decimal_formats[elem.attrib.get('type')] = numbers.parse_pattern(unicode(elem.findtext('decimalFormat/pattern'))) |
| 3 | 255 |
| 256 scientific_formats = data.setdefault('scientific_formats', {}) | |
| 257 for elem in tree.findall('//scientificFormats/scientificFormatLength'): | |
| 258 if 'draft' in elem.attrib and elem.attrib.get('type') in scientific_formats: | |
| 259 continue | |
| 260 scientific_formats[elem.attrib.get('type')] = unicode(elem.findtext('scientificFormat/pattern')) | |
| 261 | |
| 262 currency_formats = data.setdefault('currency_formats', {}) | |
| 263 for elem in tree.findall('//currencyFormats/currencyFormatLength'): | |
| 264 if 'draft' in elem.attrib and elem.attrib.get('type') in currency_formats: | |
| 265 continue | |
| 266 currency_formats[elem.attrib.get('type')] = unicode(elem.findtext('currencyFormat/pattern')) | |
| 267 | |
| 268 percent_formats = data.setdefault('percent_formats', {}) | |
| 269 for elem in tree.findall('//percentFormats/percentFormatLength'): | |
| 270 if 'draft' in elem.attrib and elem.attrib.get('type') in percent_formats: | |
| 271 continue | |
| 272 percent_formats[elem.attrib.get('type')] = unicode(elem.findtext('percentFormat/pattern')) | |
| 273 | |
| 274 currencies = data.setdefault('currencies', {}) | |
| 275 for elem in tree.findall('//currencies/currency'): | |
| 276 currencies[elem.attrib['type']] = { | |
| 277 'display_name': unicode(elem.findtext('displayName')), | |
| 278 'symbol': unicode(elem.findtext('symbol')) | |
| 279 } | |
| 280 | |
| 281 dicts[stem] = data | |
| 282 outfile = open(os.path.join(destdir, stem + '.dat'), 'wb') | |
| 283 try: | |
| 284 pickle.dump(data, outfile, 2) | |
| 285 finally: | |
| 286 outfile.close() | |
| 287 | |
| 288 if __name__ == '__main__': | |
| 289 main() |