Mercurial > babel > old > mirror
annotate scripts/import_cldr.py @ 11:11f64b232b04
Add basic support for number format patterns.
author | jonas |
---|---|
date | Wed, 30 May 2007 21:30:14 +0000 |
parents | 0ca5dd65594f |
children | b47c34d42eda |
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 |
10
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
27 weekdays = {'mon': 1, 'tue': 2, 'wed': 3, 'thu': 4, 'fri': 5, 'sat': 6, |
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
28 'sun': 7} |
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'] |
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
107 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
|
108 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
|
109 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
|
110 |
3 | 111 # <localeDisplayNames> |
112 | |
113 territories = data.setdefault('territories', {}) | |
114 for elem in tree.findall('//territories/territory'): | |
115 if 'draft' in elem.attrib and elem.attrib['type'] in territories: | |
116 continue | |
117 territories[elem.attrib['type']] = _text(elem) | |
118 | |
119 languages = data.setdefault('languages', {}) | |
120 for elem in tree.findall('//languages/language'): | |
121 if 'draft' in elem.attrib and elem.attrib['type'] in languages: | |
122 continue | |
123 languages[elem.attrib['type']] = _text(elem) | |
124 | |
125 variants = data.setdefault('variants', {}) | |
126 for elem in tree.findall('//variants/variant'): | |
127 if 'draft' in elem.attrib and elem.attrib['type'] in variants: | |
128 continue | |
129 variants[elem.attrib['type']] = _text(elem) | |
130 | |
131 scripts = data.setdefault('scripts', {}) | |
132 for elem in tree.findall('//scripts/script'): | |
133 if 'draft' in elem.attrib and elem.attrib['type'] in scripts: | |
134 continue | |
135 scripts[elem.attrib['type']] = _text(elem) | |
136 | |
137 # <dates> | |
138 | |
10
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
139 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
|
140 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
|
141 |
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
142 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
|
143 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
|
144 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
|
145 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
|
146 |
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
147 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
|
148 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
|
149 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
|
150 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
|
151 |
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
152 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
|
153 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
|
154 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
|
155 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
|
156 |
0ca5dd65594f
Pull in some supplemental data from the CLDR, for things like the first day of the week.
cmlenz
parents:
3
diff
changeset
|
157 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
|
158 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
|
159 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
|
160 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
|
161 |
3 | 162 time_zones = data.setdefault('time_zones', {}) |
163 for elem in tree.findall('//timeZoneNames/zone'): | |
164 time_zones[elem.tag] = unicode(elem.findtext('displayName')) | |
165 | |
166 for calendar in tree.findall('//calendars/calendar'): | |
167 if calendar.attrib['type'] != 'gregorian': | |
168 # TODO: support other calendar types | |
169 continue | |
170 | |
171 months = data.setdefault('months', {}) | |
172 for ctxt in calendar.findall('months/monthContext'): | |
173 ctxts = months.setdefault(ctxt.attrib['type'], {}) | |
174 for width in ctxt.findall('monthWidth'): | |
175 widths = ctxts.setdefault(width.attrib['type'], {}) | |
176 for elem in width.findall('month'): | |
177 if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: | |
178 continue | |
179 widths[int(elem.attrib.get('type'))] = unicode(elem.text) | |
180 | |
181 days = data.setdefault('days', {}) | |
182 for ctxt in calendar.findall('days/dayContext'): | |
183 ctxts = days.setdefault(ctxt.attrib['type'], {}) | |
184 for width in ctxt.findall('dayWidth'): | |
185 widths = ctxts.setdefault(width.attrib['type'], {}) | |
186 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
|
187 dtype = weekdays[elem.attrib['type']] |
3 | 188 if 'draft' in elem.attrib and dtype in widths: |
189 continue | |
190 widths[dtype] = unicode(elem.text) | |
191 | |
192 quarters = data.setdefault('quarters', {}) | |
193 for ctxt in calendar.findall('quarters/quarterContext'): | |
194 ctxts = quarters.setdefault(ctxt.attrib['type'], {}) | |
195 for width in ctxt.findall('quarterWidth'): | |
196 widths = ctxts.setdefault(width.attrib['type'], {}) | |
197 for elem in width.findall('quarter'): | |
198 if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: | |
199 continue | |
200 widths[int(elem.attrib.get('type'))] = unicode(elem.text) | |
201 | |
202 eras = data.setdefault('eras', {}) | |
203 for width in calendar.findall('eras/*'): | |
204 ewidth = {'eraNames': 'wide', 'eraAbbr': 'abbreviated'}[width.tag] | |
205 widths = eras.setdefault(ewidth, {}) | |
206 for elem in width.findall('era'): | |
207 if 'draft' in elem.attrib and int(elem.attrib['type']) in widths: | |
208 continue | |
209 widths[int(elem.attrib.get('type'))] = unicode(elem.text) | |
210 | |
211 # AM/PM | |
212 periods = data.setdefault('periods', {}) | |
213 for elem in calendar.findall('am'): | |
214 if 'draft' in elem.attrib and elem.tag in periods: | |
215 continue | |
216 periods[elem.tag] = unicode(elem.text) | |
217 for elem in calendar.findall('pm'): | |
218 if 'draft' in elem.attrib and elem.tag in periods: | |
219 continue | |
220 periods[elem.tag] = unicode(elem.text) | |
221 | |
222 date_formats = data.setdefault('date_formats', {}) | |
223 for elem in calendar.findall('dateFormats/dateFormatLength'): | |
224 if 'draft' in elem.attrib and elem.attrib.get('type') in date_formats: | |
225 continue | |
226 try: | |
227 date_formats[elem.attrib.get('type')] = \ | |
11 | 228 dates.parse_pattern(unicode(elem.findtext('dateFormat/pattern'))) |
3 | 229 except ValueError, e: |
230 print e | |
231 | |
232 time_formats = data.setdefault('time_formats', {}) | |
233 for elem in calendar.findall('timeFormats/timeFormatLength'): | |
234 if 'draft' in elem.attrib and elem.attrib.get('type') in time_formats: | |
235 continue | |
236 try: | |
237 time_formats[elem.attrib.get('type')] = \ | |
11 | 238 dates.parse_pattern(unicode(elem.findtext('timeFormat/pattern'))) |
3 | 239 except ValueError, e: |
240 print e | |
241 | |
242 # <numbers> | |
243 | |
244 number_symbols = data.setdefault('number_symbols', {}) | |
245 for elem in tree.findall('//numbers/symbols/*'): | |
246 number_symbols[elem.tag] = unicode(elem.text) | |
247 | |
248 decimal_formats = data.setdefault('decimal_formats', {}) | |
249 for elem in tree.findall('//decimalFormats/decimalFormatLength'): | |
250 if 'draft' in elem.attrib and elem.attrib.get('type') in decimal_formats: | |
251 continue | |
11 | 252 decimal_formats[elem.attrib.get('type')] = numbers.parse_pattern(unicode(elem.findtext('decimalFormat/pattern'))) |
3 | 253 |
254 scientific_formats = data.setdefault('scientific_formats', {}) | |
255 for elem in tree.findall('//scientificFormats/scientificFormatLength'): | |
256 if 'draft' in elem.attrib and elem.attrib.get('type') in scientific_formats: | |
257 continue | |
258 scientific_formats[elem.attrib.get('type')] = unicode(elem.findtext('scientificFormat/pattern')) | |
259 | |
260 currency_formats = data.setdefault('currency_formats', {}) | |
261 for elem in tree.findall('//currencyFormats/currencyFormatLength'): | |
262 if 'draft' in elem.attrib and elem.attrib.get('type') in currency_formats: | |
263 continue | |
264 currency_formats[elem.attrib.get('type')] = unicode(elem.findtext('currencyFormat/pattern')) | |
265 | |
266 percent_formats = data.setdefault('percent_formats', {}) | |
267 for elem in tree.findall('//percentFormats/percentFormatLength'): | |
268 if 'draft' in elem.attrib and elem.attrib.get('type') in percent_formats: | |
269 continue | |
270 percent_formats[elem.attrib.get('type')] = unicode(elem.findtext('percentFormat/pattern')) | |
271 | |
272 currencies = data.setdefault('currencies', {}) | |
273 for elem in tree.findall('//currencies/currency'): | |
274 currencies[elem.attrib['type']] = { | |
275 'display_name': unicode(elem.findtext('displayName')), | |
276 'symbol': unicode(elem.findtext('symbol')) | |
277 } | |
278 | |
279 dicts[stem] = data | |
280 outfile = open(os.path.join(destdir, stem + '.dat'), 'wb') | |
281 try: | |
282 pickle.dump(data, outfile, 2) | |
283 finally: | |
284 outfile.close() | |
285 | |
286 if __name__ == '__main__': | |
287 main() |