1 | n/a | # |
---|
2 | n/a | # Copyright (c) 2008-2012 Stefan Krah. All rights reserved. |
---|
3 | n/a | # |
---|
4 | n/a | # Redistribution and use in source and binary forms, with or without |
---|
5 | n/a | # modification, are permitted provided that the following conditions |
---|
6 | n/a | # are met: |
---|
7 | n/a | # |
---|
8 | n/a | # 1. Redistributions of source code must retain the above copyright |
---|
9 | n/a | # notice, this list of conditions and the following disclaimer. |
---|
10 | n/a | # |
---|
11 | n/a | # 2. Redistributions in binary form must reproduce the above copyright |
---|
12 | n/a | # notice, this list of conditions and the following disclaimer in the |
---|
13 | n/a | # documentation and/or other materials provided with the distribution. |
---|
14 | n/a | # |
---|
15 | n/a | # THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS "AS IS" AND |
---|
16 | n/a | # ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
---|
17 | n/a | # IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
---|
18 | n/a | # ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE |
---|
19 | n/a | # FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
---|
20 | n/a | # DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS |
---|
21 | n/a | # OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) |
---|
22 | n/a | # HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT |
---|
23 | n/a | # LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY |
---|
24 | n/a | # OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF |
---|
25 | n/a | # SUCH DAMAGE. |
---|
26 | n/a | # |
---|
27 | n/a | |
---|
28 | n/a | |
---|
29 | n/a | # Generate PEP-3101 format strings. |
---|
30 | n/a | |
---|
31 | n/a | |
---|
32 | n/a | import os, sys, locale, random |
---|
33 | n/a | import platform, subprocess |
---|
34 | n/a | from test.support import import_fresh_module |
---|
35 | n/a | from distutils.spawn import find_executable |
---|
36 | n/a | |
---|
37 | n/a | C = import_fresh_module('decimal', fresh=['_decimal']) |
---|
38 | n/a | P = import_fresh_module('decimal', blocked=['_decimal']) |
---|
39 | n/a | |
---|
40 | n/a | |
---|
41 | n/a | windows_lang_strings = [ |
---|
42 | n/a | "chinese", "chinese-simplified", "chinese-traditional", "czech", "danish", |
---|
43 | n/a | "dutch", "belgian", "english", "australian", "canadian", "english-nz", |
---|
44 | n/a | "english-uk", "english-us", "finnish", "french", "french-belgian", |
---|
45 | n/a | "french-canadian", "french-swiss", "german", "german-austrian", |
---|
46 | n/a | "german-swiss", "greek", "hungarian", "icelandic", "italian", "italian-swiss", |
---|
47 | n/a | "japanese", "korean", "norwegian", "norwegian-bokmal", "norwegian-nynorsk", |
---|
48 | n/a | "polish", "portuguese", "portuguese-brazil", "russian", "slovak", "spanish", |
---|
49 | n/a | "spanish-mexican", "spanish-modern", "swedish", "turkish", |
---|
50 | n/a | ] |
---|
51 | n/a | |
---|
52 | n/a | preferred_encoding = { |
---|
53 | n/a | 'cs_CZ': 'ISO8859-2', |
---|
54 | n/a | 'cs_CZ.iso88592': 'ISO8859-2', |
---|
55 | n/a | 'czech': 'ISO8859-2', |
---|
56 | n/a | 'eesti': 'ISO8859-1', |
---|
57 | n/a | 'estonian': 'ISO8859-1', |
---|
58 | n/a | 'et_EE': 'ISO8859-15', |
---|
59 | n/a | 'et_EE.ISO-8859-15': 'ISO8859-15', |
---|
60 | n/a | 'et_EE.iso885915': 'ISO8859-15', |
---|
61 | n/a | 'et_EE.iso88591': 'ISO8859-1', |
---|
62 | n/a | 'fi_FI.iso88591': 'ISO8859-1', |
---|
63 | n/a | 'fi_FI': 'ISO8859-15', |
---|
64 | n/a | 'fi_FI@euro': 'ISO8859-15', |
---|
65 | n/a | 'fi_FI.iso885915@euro': 'ISO8859-15', |
---|
66 | n/a | 'finnish': 'ISO8859-1', |
---|
67 | n/a | 'lv_LV': 'ISO8859-13', |
---|
68 | n/a | 'lv_LV.iso885913': 'ISO8859-13', |
---|
69 | n/a | 'nb_NO': 'ISO8859-1', |
---|
70 | n/a | 'nb_NO.iso88591': 'ISO8859-1', |
---|
71 | n/a | 'bokmal': 'ISO8859-1', |
---|
72 | n/a | 'nn_NO': 'ISO8859-1', |
---|
73 | n/a | 'nn_NO.iso88591': 'ISO8859-1', |
---|
74 | n/a | 'no_NO': 'ISO8859-1', |
---|
75 | n/a | 'norwegian': 'ISO8859-1', |
---|
76 | n/a | 'nynorsk': 'ISO8859-1', |
---|
77 | n/a | 'ru_RU': 'ISO8859-5', |
---|
78 | n/a | 'ru_RU.iso88595': 'ISO8859-5', |
---|
79 | n/a | 'russian': 'ISO8859-5', |
---|
80 | n/a | 'ru_RU.KOI8-R': 'KOI8-R', |
---|
81 | n/a | 'ru_RU.koi8r': 'KOI8-R', |
---|
82 | n/a | 'ru_RU.CP1251': 'CP1251', |
---|
83 | n/a | 'ru_RU.cp1251': 'CP1251', |
---|
84 | n/a | 'sk_SK': 'ISO8859-2', |
---|
85 | n/a | 'sk_SK.iso88592': 'ISO8859-2', |
---|
86 | n/a | 'slovak': 'ISO8859-2', |
---|
87 | n/a | 'sv_FI': 'ISO8859-1', |
---|
88 | n/a | 'sv_FI.iso88591': 'ISO8859-1', |
---|
89 | n/a | 'sv_FI@euro': 'ISO8859-15', |
---|
90 | n/a | 'sv_FI.iso885915@euro': 'ISO8859-15', |
---|
91 | n/a | 'uk_UA': 'KOI8-U', |
---|
92 | n/a | 'uk_UA.koi8u': 'KOI8-U' |
---|
93 | n/a | } |
---|
94 | n/a | |
---|
95 | n/a | integers = [ |
---|
96 | n/a | "", |
---|
97 | n/a | "1", |
---|
98 | n/a | "12", |
---|
99 | n/a | "123", |
---|
100 | n/a | "1234", |
---|
101 | n/a | "12345", |
---|
102 | n/a | "123456", |
---|
103 | n/a | "1234567", |
---|
104 | n/a | "12345678", |
---|
105 | n/a | "123456789", |
---|
106 | n/a | "1234567890", |
---|
107 | n/a | "12345678901", |
---|
108 | n/a | "123456789012", |
---|
109 | n/a | "1234567890123", |
---|
110 | n/a | "12345678901234", |
---|
111 | n/a | "123456789012345", |
---|
112 | n/a | "1234567890123456", |
---|
113 | n/a | "12345678901234567", |
---|
114 | n/a | "123456789012345678", |
---|
115 | n/a | "1234567890123456789", |
---|
116 | n/a | "12345678901234567890", |
---|
117 | n/a | "123456789012345678901", |
---|
118 | n/a | "1234567890123456789012", |
---|
119 | n/a | ] |
---|
120 | n/a | |
---|
121 | n/a | numbers = [ |
---|
122 | n/a | "0", "-0", "+0", |
---|
123 | n/a | "0.0", "-0.0", "+0.0", |
---|
124 | n/a | "0e0", "-0e0", "+0e0", |
---|
125 | n/a | ".0", "-.0", |
---|
126 | n/a | ".1", "-.1", |
---|
127 | n/a | "1.1", "-1.1", |
---|
128 | n/a | "1e1", "-1e1" |
---|
129 | n/a | ] |
---|
130 | n/a | |
---|
131 | n/a | # Get the list of available locales. |
---|
132 | n/a | if platform.system() == 'Windows': |
---|
133 | n/a | locale_list = windows_lang_strings |
---|
134 | n/a | else: |
---|
135 | n/a | locale_list = ['C'] |
---|
136 | n/a | if os.path.isfile("/var/lib/locales/supported.d/local"): |
---|
137 | n/a | # On Ubuntu, `locale -a` gives the wrong case for some locales, |
---|
138 | n/a | # so we get the correct names directly: |
---|
139 | n/a | with open("/var/lib/locales/supported.d/local") as f: |
---|
140 | n/a | locale_list = [loc.split()[0] for loc in f.readlines() \ |
---|
141 | n/a | if not loc.startswith('#')] |
---|
142 | n/a | elif find_executable('locale'): |
---|
143 | n/a | locale_list = subprocess.Popen(["locale", "-a"], |
---|
144 | n/a | stdout=subprocess.PIPE).communicate()[0] |
---|
145 | n/a | try: |
---|
146 | n/a | locale_list = locale_list.decode() |
---|
147 | n/a | except UnicodeDecodeError: |
---|
148 | n/a | # Some distributions insist on using latin-1 characters |
---|
149 | n/a | # in their locale names. |
---|
150 | n/a | locale_list = locale_list.decode('latin-1') |
---|
151 | n/a | locale_list = locale_list.split('\n') |
---|
152 | n/a | try: |
---|
153 | n/a | locale_list.remove('') |
---|
154 | n/a | except ValueError: |
---|
155 | n/a | pass |
---|
156 | n/a | |
---|
157 | n/a | # Debian |
---|
158 | n/a | if os.path.isfile("/etc/locale.alias"): |
---|
159 | n/a | with open("/etc/locale.alias") as f: |
---|
160 | n/a | while 1: |
---|
161 | n/a | try: |
---|
162 | n/a | line = f.readline() |
---|
163 | n/a | except UnicodeDecodeError: |
---|
164 | n/a | continue |
---|
165 | n/a | if line == "": |
---|
166 | n/a | break |
---|
167 | n/a | if line.startswith('#'): |
---|
168 | n/a | continue |
---|
169 | n/a | x = line.split() |
---|
170 | n/a | if len(x) == 2: |
---|
171 | n/a | if x[0] in locale_list: |
---|
172 | n/a | locale_list.remove(x[0]) |
---|
173 | n/a | |
---|
174 | n/a | # FreeBSD |
---|
175 | n/a | if platform.system() == 'FreeBSD': |
---|
176 | n/a | # http://www.freebsd.org/cgi/query-pr.cgi?pr=142173 |
---|
177 | n/a | # en_GB.US-ASCII has 163 as the currency symbol. |
---|
178 | n/a | for loc in ['it_CH.ISO8859-1', 'it_CH.ISO8859-15', 'it_CH.UTF-8', |
---|
179 | n/a | 'it_IT.ISO8859-1', 'it_IT.ISO8859-15', 'it_IT.UTF-8', |
---|
180 | n/a | 'sl_SI.ISO8859-2', 'sl_SI.UTF-8', |
---|
181 | n/a | 'en_GB.US-ASCII']: |
---|
182 | n/a | try: |
---|
183 | n/a | locale_list.remove(loc) |
---|
184 | n/a | except ValueError: |
---|
185 | n/a | pass |
---|
186 | n/a | |
---|
187 | n/a | # Print a testcase in the format of the IBM tests (for runtest.c): |
---|
188 | n/a | def get_preferred_encoding(): |
---|
189 | n/a | loc = locale.setlocale(locale.LC_CTYPE) |
---|
190 | n/a | if loc in preferred_encoding: |
---|
191 | n/a | return preferred_encoding[loc] |
---|
192 | n/a | else: |
---|
193 | n/a | return locale.getpreferredencoding() |
---|
194 | n/a | |
---|
195 | n/a | def printit(testno, s, fmt, encoding=None): |
---|
196 | n/a | if not encoding: |
---|
197 | n/a | encoding = get_preferred_encoding() |
---|
198 | n/a | try: |
---|
199 | n/a | result = format(P.Decimal(s), fmt) |
---|
200 | n/a | fmt = str(fmt.encode(encoding))[2:-1] |
---|
201 | n/a | result = str(result.encode(encoding))[2:-1] |
---|
202 | n/a | if "'" in result: |
---|
203 | n/a | sys.stdout.write("xfmt%d format %s '%s' -> \"%s\"\n" |
---|
204 | n/a | % (testno, s, fmt, result)) |
---|
205 | n/a | else: |
---|
206 | n/a | sys.stdout.write("xfmt%d format %s '%s' -> '%s'\n" |
---|
207 | n/a | % (testno, s, fmt, result)) |
---|
208 | n/a | except Exception as err: |
---|
209 | n/a | sys.stderr.write("%s %s %s\n" % (err, s, fmt)) |
---|
210 | n/a | |
---|
211 | n/a | |
---|
212 | n/a | # Check if an integer can be converted to a valid fill character. |
---|
213 | n/a | def check_fillchar(i): |
---|
214 | n/a | try: |
---|
215 | n/a | c = chr(i) |
---|
216 | n/a | c.encode('utf-8').decode() |
---|
217 | n/a | format(P.Decimal(0), c + '<19g') |
---|
218 | n/a | return c |
---|
219 | n/a | except: |
---|
220 | n/a | return None |
---|
221 | n/a | |
---|
222 | n/a | # Generate all unicode characters that are accepted as |
---|
223 | n/a | # fill characters by decimal.py. |
---|
224 | n/a | def all_fillchars(): |
---|
225 | n/a | for i in range(0, 0x110002): |
---|
226 | n/a | c = check_fillchar(i) |
---|
227 | n/a | if c: yield c |
---|
228 | n/a | |
---|
229 | n/a | # Return random fill character. |
---|
230 | n/a | def rand_fillchar(): |
---|
231 | n/a | while 1: |
---|
232 | n/a | i = random.randrange(0, 0x110002) |
---|
233 | n/a | c = check_fillchar(i) |
---|
234 | n/a | if c: return c |
---|
235 | n/a | |
---|
236 | n/a | # Generate random format strings |
---|
237 | n/a | # [[fill]align][sign][#][0][width][.precision][type] |
---|
238 | n/a | def rand_format(fill, typespec='EeGgFfn%'): |
---|
239 | n/a | active = sorted(random.sample(range(7), random.randrange(8))) |
---|
240 | n/a | have_align = 0 |
---|
241 | n/a | s = '' |
---|
242 | n/a | for elem in active: |
---|
243 | n/a | if elem == 0: # fill+align |
---|
244 | n/a | s += fill |
---|
245 | n/a | s += random.choice('<>=^') |
---|
246 | n/a | have_align = 1 |
---|
247 | n/a | elif elem == 1: # sign |
---|
248 | n/a | s += random.choice('+- ') |
---|
249 | n/a | elif elem == 2 and not have_align: # zeropad |
---|
250 | n/a | s += '0' |
---|
251 | n/a | elif elem == 3: # width |
---|
252 | n/a | s += str(random.randrange(1, 100)) |
---|
253 | n/a | elif elem == 4: # thousands separator |
---|
254 | n/a | s += ',' |
---|
255 | n/a | elif elem == 5: # prec |
---|
256 | n/a | s += '.' |
---|
257 | n/a | s += str(random.randrange(100)) |
---|
258 | n/a | elif elem == 6: |
---|
259 | n/a | if 4 in active: c = typespec.replace('n', '') |
---|
260 | n/a | else: c = typespec |
---|
261 | n/a | s += random.choice(c) |
---|
262 | n/a | return s |
---|
263 | n/a | |
---|
264 | n/a | # Partially brute force all possible format strings containing a thousands |
---|
265 | n/a | # separator. Fall back to random where the runtime would become excessive. |
---|
266 | n/a | # [[fill]align][sign][#][0][width][,][.precision][type] |
---|
267 | n/a | def all_format_sep(): |
---|
268 | n/a | for align in ('', '<', '>', '=', '^'): |
---|
269 | n/a | for fill in ('', 'x'): |
---|
270 | n/a | if align == '': fill = '' |
---|
271 | n/a | for sign in ('', '+', '-', ' '): |
---|
272 | n/a | for zeropad in ('', '0'): |
---|
273 | n/a | if align != '': zeropad = '' |
---|
274 | n/a | for width in ['']+[str(y) for y in range(1, 15)]+['101']: |
---|
275 | n/a | for prec in ['']+['.'+str(y) for y in range(15)]: |
---|
276 | n/a | # for type in ('', 'E', 'e', 'G', 'g', 'F', 'f', '%'): |
---|
277 | n/a | type = random.choice(('', 'E', 'e', 'G', 'g', 'F', 'f', '%')) |
---|
278 | n/a | yield ''.join((fill, align, sign, zeropad, width, ',', prec, type)) |
---|
279 | n/a | |
---|
280 | n/a | # Partially brute force all possible format strings with an 'n' specifier. |
---|
281 | n/a | # [[fill]align][sign][#][0][width][,][.precision][type] |
---|
282 | n/a | def all_format_loc(): |
---|
283 | n/a | for align in ('', '<', '>', '=', '^'): |
---|
284 | n/a | for fill in ('', 'x'): |
---|
285 | n/a | if align == '': fill = '' |
---|
286 | n/a | for sign in ('', '+', '-', ' '): |
---|
287 | n/a | for zeropad in ('', '0'): |
---|
288 | n/a | if align != '': zeropad = '' |
---|
289 | n/a | for width in ['']+[str(y) for y in range(1, 20)]+['101']: |
---|
290 | n/a | for prec in ['']+['.'+str(y) for y in range(1, 20)]: |
---|
291 | n/a | yield ''.join((fill, align, sign, zeropad, width, prec, 'n')) |
---|
292 | n/a | |
---|
293 | n/a | # Generate random format strings with a unicode fill character |
---|
294 | n/a | # [[fill]align][sign][#][0][width][,][.precision][type] |
---|
295 | n/a | def randfill(fill): |
---|
296 | n/a | active = sorted(random.sample(range(5), random.randrange(6))) |
---|
297 | n/a | s = '' |
---|
298 | n/a | s += str(fill) |
---|
299 | n/a | s += random.choice('<>=^') |
---|
300 | n/a | for elem in active: |
---|
301 | n/a | if elem == 0: # sign |
---|
302 | n/a | s += random.choice('+- ') |
---|
303 | n/a | elif elem == 1: # width |
---|
304 | n/a | s += str(random.randrange(1, 100)) |
---|
305 | n/a | elif elem == 2: # thousands separator |
---|
306 | n/a | s += ',' |
---|
307 | n/a | elif elem == 3: # prec |
---|
308 | n/a | s += '.' |
---|
309 | n/a | s += str(random.randrange(100)) |
---|
310 | n/a | elif elem == 4: |
---|
311 | n/a | if 2 in active: c = 'EeGgFf%' |
---|
312 | n/a | else: c = 'EeGgFfn%' |
---|
313 | n/a | s += random.choice(c) |
---|
314 | n/a | return s |
---|
315 | n/a | |
---|
316 | n/a | # Generate random format strings with random locale setting |
---|
317 | n/a | # [[fill]align][sign][#][0][width][,][.precision][type] |
---|
318 | n/a | def rand_locale(): |
---|
319 | n/a | try: |
---|
320 | n/a | loc = random.choice(locale_list) |
---|
321 | n/a | locale.setlocale(locale.LC_ALL, loc) |
---|
322 | n/a | except locale.Error as err: |
---|
323 | n/a | pass |
---|
324 | n/a | active = sorted(random.sample(range(5), random.randrange(6))) |
---|
325 | n/a | s = '' |
---|
326 | n/a | have_align = 0 |
---|
327 | n/a | for elem in active: |
---|
328 | n/a | if elem == 0: # fill+align |
---|
329 | n/a | s += chr(random.randrange(32, 128)) |
---|
330 | n/a | s += random.choice('<>=^') |
---|
331 | n/a | have_align = 1 |
---|
332 | n/a | elif elem == 1: # sign |
---|
333 | n/a | s += random.choice('+- ') |
---|
334 | n/a | elif elem == 2 and not have_align: # zeropad |
---|
335 | n/a | s += '0' |
---|
336 | n/a | elif elem == 3: # width |
---|
337 | n/a | s += str(random.randrange(1, 100)) |
---|
338 | n/a | elif elem == 4: # prec |
---|
339 | n/a | s += '.' |
---|
340 | n/a | s += str(random.randrange(100)) |
---|
341 | n/a | s += 'n' |
---|
342 | n/a | return s |
---|