1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502
|
import sys
from pypy.interpreter.baseobjspace import W_Root, BufferInterfaceNotFound
from pypy.interpreter.error import OperationError, oefmt
from pypy.interpreter.gateway import WrappedDefault, interp2app, unwrap_spec
from pypy.interpreter.typedef import (
GetSetProperty, TypeDef, generic_new_descr, interp_attrproperty,
interp_attrproperty_w)
from pypy.module._codecs import interp_codecs
from pypy.module._io.interp_iobase import W_IOBase, convert_size, trap_eintr
from pypy.module.posix.interp_posix import device_encoding
from rpython.rlib.rarithmetic import intmask, r_uint, r_ulonglong
from rpython.rlib.rbigint import rbigint
from rpython.rlib.rstring import StringBuilder
from rpython.rlib.rutf8 import (check_utf8, next_codepoint_pos,
codepoints_in_utf8, codepoints_in_utf8,
Utf8StringBuilder)
from rpython.rlib import rlocale, jit
from rpython.rlib.objectmodel import always_inline
STATE_ZERO, STATE_OK, STATE_DETACHED = range(3)
SEEN_CR = 1
SEEN_LF = 2
SEEN_CRLF = 4
SEEN_ALL = SEEN_CR | SEEN_LF | SEEN_CRLF
_WINDOWS = sys.platform == 'win32'
def make_newlines_dict(space):
return {
SEEN_CR: space.newtext("\r", 1),
SEEN_LF: space.newtext("\n", 1),
SEEN_CRLF: space.newtext("\r\n", 2),
SEEN_CR | SEEN_LF: space.newtuple(
[space.newtext("\r", 1),
space.newtext("\n", 1)]),
SEEN_CR | SEEN_CRLF: space.newtuple(
[space.newtext("\r", 1),
space.newtext("\r\n", 2)]),
SEEN_LF | SEEN_CRLF: space.newtuple(
[space.newtext("\n", 1),
space.newtext("\r\n", 2)]),
SEEN_CR | SEEN_LF | SEEN_CRLF: space.newtuple(
[space.newtext("\r", 1),
space.newtext("\n", 1),
space.newtext("\r\n", 2)]),
}
def io_check_errors(space, w_errors):
# Make sure w_errors is valid:
# - no \0
# - can be encoded into utf8, no surrogates
# - if dev_mode, is a valid error handler
errors = space.text0_w(w_errors)
space.encode_unicode_object(w_errors, 'utf8', 'strict')
if not space.sys.get_flag('dev_mode'):
return
interp_codecs.lookup_error(space, errors)
class W_IncrementalNewlineDecoder(W_Root):
seennl = 0
pendingcr = False
w_decoder = None
def __init__(self, space):
pass
@unwrap_spec(translate=int)
def descr_init(self, space, w_decoder, translate, w_errors=None):
self.w_decoder = w_decoder
self.translate = translate
if space.is_none(w_errors) or w_errors is None:
w_errors = space.newtext("strict")
if not space.isinstance_w(w_errors, space.w_text):
raise oefmt(space.w_TypeError, "TextIOWrapper() argument 'errors' "
"must be str or None, not %N", w_errors)
io_check_errors(space, w_errors)
self.w_errors = w_errors
self.seennl = 0
def newlines_get_w(self, space):
return space.fromcache(make_newlines_dict).get(self.seennl, space.w_None)
@unwrap_spec(final=int)
def decode_w(self, space, w_input, final=False):
if self.w_decoder is None:
raise oefmt(space.w_ValueError,
"IncrementalNewlineDecoder.__init__ not called")
# decode input (with the eventual \r from a previous pass)
if not space.is_w(self.w_decoder, space.w_None):
w_output = space.call_method(self.w_decoder, "decode",
w_input, space.newbool(bool(final)))
else:
w_output = w_input
if not space.isinstance_w(w_output, space.w_unicode):
raise oefmt(space.w_TypeError,
"decoder should return a string result")
output, output_len = space.utf8_len_w(w_output)
w_result = self._decoder_w_unboxed(space, output, output_len, final)
if w_result is None:
return w_output
return w_result
def _decoder_w_unboxed(self, space, output, output_len, final):
changed = False
if self.pendingcr and (final or output_len):
output = '\r' + output
self.pendingcr = False
output_len += 1
changed = True
# retain last \r even when not translating data:
# then readline() is sure to get \r\n in one pass
if not final and output_len > 0:
last = len(output) - 1
assert last >= 0
if output[last] == '\r':
output = output[:last]
self.pendingcr = True
output_len -= 1
changed = True
if output_len == 0:
return space.newutf8("", 0)
# Record which newlines are read and do newline translation if
# desired, all in one pass.
seennl = self.seennl
rpos = output.find('\r')
if rpos < 0:
# If no \r, quick scan for a possible "\n" character.
# (there's nothing else to be done, even when in translation mode)
if not (seennl & SEEN_LF) and output.find('\n') >= 0:
self.seennl |= SEEN_LF
# Finished: we have scanned for newlines, and none of them
# need translating.
if not changed:
return None # means we can reuse the w_output in the caller
elif not self.translate:
i = 0
while i < len(output):
if seennl == SEEN_ALL:
break
c = output[i]
i += 1
if c == '\n':
seennl |= SEEN_LF
elif c == '\r':
if i < len(output) and output[i] == '\n':
seennl |= SEEN_CRLF
i += 1
else:
seennl |= SEEN_CR
if not changed:
self.seennl |= seennl
return None
else:
changed = True
assert rpos >= 0
# Translate!
builder = StringBuilder(len(output))
if output.find("\n", 0, rpos) >= 0:
seennl |= SEEN_LF
builder.append_slice(output, 0, rpos)
i = rpos
while i < len(output):
c = output[i]
i += 1
if c == '\n':
seennl |= SEEN_LF
elif c == '\r':
if i < len(output) and output[i] == '\n':
seennl |= SEEN_CRLF
i += 1
output_len -= 1
else:
seennl |= SEEN_CR
builder.append('\n')
continue
builder.append(c)
output = builder.build()
self.seennl |= seennl
return space.newutf8(output, output_len)
def reset_w(self, space):
self.seennl = 0
self.pendingcr = False
if self.w_decoder and not space.is_w(self.w_decoder, space.w_None):
space.call_method(self.w_decoder, "reset")
def getstate_u(self, space):
if self.w_decoder and not space.is_w(self.w_decoder, space.w_None):
w_state = space.call_method(self.w_decoder, "getstate")
w_buffer, w_flag = space.unpackiterable(w_state, 2)
flag = space.r_longlong_w(w_flag)
else:
w_buffer = space.newbytes("")
flag = 0
flag <<= 1
if self.pendingcr:
flag |= 1
return w_buffer, flag
def getstate_w(self, space):
w_buffer, flag = self.getstate_u(space)
return space.newtuple2(w_buffer, space.newint(flag))
def setstate_w(self, space, w_state):
w_buffer, w_flag = space.unpackiterable(w_state, 2)
flag = space.r_longlong_w(w_flag)
self.pendingcr = bool(flag & 1)
flag >>= 1
if self.w_decoder and not space.is_w(self.w_decoder, space.w_None):
w_state = space.newtuple2(w_buffer, space.newint(flag))
space.call_method(self.w_decoder, "setstate", w_state)
W_IncrementalNewlineDecoder.typedef = TypeDef(
'_io.IncrementalNewlineDecoder',
__new__ = generic_new_descr(W_IncrementalNewlineDecoder),
__init__ = interp2app(W_IncrementalNewlineDecoder.descr_init),
decode = interp2app(W_IncrementalNewlineDecoder.decode_w),
reset = interp2app(W_IncrementalNewlineDecoder.reset_w),
getstate = interp2app(W_IncrementalNewlineDecoder.getstate_w),
setstate = interp2app(W_IncrementalNewlineDecoder.setstate_w),
newlines = GetSetProperty(W_IncrementalNewlineDecoder.newlines_get_w),
)
class W_TextIOBase(W_IOBase):
w_encoding = None
def __init__(self, space):
W_IOBase.__init__(self, space)
def read_w(self, space, w_size=None):
self._unsupportedoperation(space, "read")
def readline_w(self, space, w_limit=None):
self._unsupportedoperation(space, "readline")
def write_w(self, space, w_data):
self._unsupportedoperation(space, "write")
def detach_w(self, space):
self._unsupportedoperation(space, "detach")
def errors_get_w(self, space):
return space.w_None
def newlines_get_w(self, space):
return space.w_None
W_TextIOBase.typedef = TypeDef(
'_io._TextIOBase', W_IOBase.typedef,
__new__ = generic_new_descr(W_TextIOBase),
read = interp2app(W_TextIOBase.read_w),
readline = interp2app(W_TextIOBase.readline_w),
write = interp2app(W_TextIOBase.write_w),
detach = interp2app(W_TextIOBase.detach_w),
encoding = interp_attrproperty_w("w_encoding", W_TextIOBase),
newlines = GetSetProperty(W_TextIOBase.newlines_get_w),
errors = GetSetProperty(W_TextIOBase.errors_get_w),
)
def _determine_encoding(space, encoding, w_buffer):
if encoding == "locale":
pass
elif encoding is None:
_maybe_warn_encoding(space)
else:
return space.newtext(encoding)
# Try os.device_encoding(fileno) which is interp_posix.device_encoding
try:
w_fileno = space.call_method(w_buffer, 'fileno')
except OperationError as e:
from pypy.module._io.interp_io import Cache
if not (e.match(space, space.w_AttributeError) or
e.match(space, space.fromcache(Cache).w_unsupportedoperation)):
raise
else:
try:
w_encoding = device_encoding(space, space.int_w(w_fileno))
except OverflowError:
raise oefmt(space.w_OverflowError,
"Python int too large to convert to C int")
else:
if space.isinstance_w(w_encoding, space.w_unicode):
return w_encoding
if space.sys.get_flag('utf8_mode'):
# if device_encoding returns None, CPython does
# _Py_GetLocaleEncoding, which has this at the top
return space.newtext("utf-8")
# On legacy systems or darwin, try app-level
# _bootlocale.getprefferedencoding(False)
try:
w_locale = space.call_method(space.builtin, '__import__',
space.newtext('_bootlocale'))
w_encoding = space.call_method(w_locale, 'getpreferredencoding',
space.w_False)
except OperationError as e:
# getpreferredencoding() may also raise ImportError
if not e.match(space, space.w_ImportError):
raise
return space.newtext('ascii')
else:
if space.isinstance_w(w_encoding, space.w_text):
return w_encoding
raise oefmt(space.w_IOError, "could not determine default encoding")
@unwrap_spec(stacklevel=int)
def text_encoding(space, w_encoding, stacklevel=2):
"""
A helper function to choose the text encoding.
When encoding is not None, just return it.
Otherwise, return the default text encoding (i.e. "locale").
This function emits an EncodingWarning if *encoding* is None and
sys.flags.warn_default_encoding is non-zero.
This can be used in APIs with an encoding=None parameter
that pass it to TextIOWrapper or open.
However, please consider using encoding="utf-8" for new APIs.
"""
if space.is_none(w_encoding):
_maybe_warn_encoding(space, stacklevel + 1)
if space.sys.get_flag('utf8_mode') == 0:
return space.newtext("locale")
else:
return space.newtext("utf-8")
return w_encoding
def _maybe_warn_encoding(space, stacklevel=1):
if space.sys.get_flag('warn_default_encoding') != 0:
space.warn(space.newtext("'encoding' argument not specified."),
space.w_EncodingWarning, stacklevel)
class PositionCookie(object):
def __init__(self, bigint):
self.start_pos = bigint.ulonglongmask()
bigint = bigint.rshift(r_ulonglong.BITS)
x = intmask(bigint.uintmask())
assert x >= 0
self.dec_flags = x
bigint = bigint.rshift(r_uint.BITS)
x = intmask(bigint.uintmask())
assert x >= 0
self.bytes_to_feed = x
bigint = bigint.rshift(r_uint.BITS)
x = intmask(bigint.uintmask())
assert x >= 0
self.chars_to_skip = x
bigint = bigint.rshift(r_uint.BITS)
self.need_eof = bigint.tobool()
def pack(self):
# The meaning of a tell() cookie is: seek to position, set the
# decoder flags to dec_flags, read bytes_to_feed bytes, feed them
# into the decoder with need_eof as the EOF flag, then skip
# chars_to_skip characters of the decoded result. For most simple
# decoders, tell() will often just give a byte offset in the file.
rb = rbigint.fromrarith_int
res = rb(self.start_pos)
bits = r_ulonglong.BITS
res = res.or_(rb(r_uint(self.dec_flags)).lshift(bits))
bits += r_uint.BITS
res = res.or_(rb(r_uint(self.bytes_to_feed)).lshift(bits))
bits += r_uint.BITS
res = res.or_(rb(r_uint(self.chars_to_skip)).lshift(bits))
bits += r_uint.BITS
return res.or_(rb(r_uint(self.need_eof)).lshift(bits))
def __repr__(self):
return "coookie w/start_pos %d dec_flags %d bytes_to_feed %d chars_to_skip %d need_eof %d" %(self.start_pos, self.dec_flags, self.bytes_to_feed, self.chars_to_skip, self.need_eof)
class PositionSnapshot:
_immutable_fields_ = ["flags", "input"]
def __init__(self, flags, input):
self.flags = flags
self.input = input
class DecodeBuffer(object):
def __init__(self, text=None, ulen=-1):
# self.text is a valid utf-8 string
if text is not None:
assert ulen >= 0
self.text = text
self.pos = 0 # in utf8
self.upos = 0 # decoded_chars_used in CPython, in codepoints
self.ulen = ulen
def set(self, space, w_decoded):
# set_chars_decoded in CPython
check_decoded(space, w_decoded)
self.text, self.ulen = space.utf8_len_w(w_decoded)
self.pos = 0
self.upos = 0
def reset(self):
self.text = None
self.pos = 0
self.upos = 0
self.ulen = -1
def get_chars(self, size):
""" returns a tuple (utf8, lgt) """
# get_decoded_chars in CPython
if self.text is None or size == 0:
return "", 0
lgt = self.ulen
available = lgt - self.upos
if size < 0 or size > available:
size = available
assert size >= 0
if self.pos > 0 or size < available:
start = self.pos
pos = start
for i in range(size):
pos = next_codepoint_pos(self.text, pos)
self.upos += 1
assert start >= 0
assert pos >= 0
chars = self.text[start:pos]
self.pos = pos
else:
chars = self.text
self.pos = len(self.text)
self.upos = lgt
size = lgt
return chars, size
def has_data(self):
return (self.text is not None and not self.exhausted())
def exhausted(self):
return self.pos >= len(self.text)
def find_newline_universal(self, limit):
# Universal newline search. Find any of \r, \r\n, \n
# The decoder ensures that \r\n are not split in two pieces
if limit < 0:
limit = sys.maxint
scanned = 0
while scanned < limit:
if self.exhausted():
return False
ch = self.text[self.pos]
self._advance_codepoint()
scanned += 1
if ch == '\n':
return True
if ch == '\r':
if scanned >= limit:
return False
if self.exhausted():
return True
ch = self.text[self.pos]
if ch == '\n':
self.pos += 1
self.upos += 1
return True
else:
return True
return False
def find_crlf(self, limit):
if limit < 0:
limit = sys.maxint
scanned = 0
while scanned < limit:
if self.exhausted():
return False
ch = self.text[self.pos]
scanned += 1
if ch == '\r':
self.pos += 1
self.upos += 1
if scanned >= limit:
return False
if self.exhausted():
# This is the tricky case: we found a \r right at the end,
# un-consume it
self.pos -= 1
self.upos -= 1
return False
if self.text[self.pos] == '\n':
self.pos += 1
self.upos += 1
return True
else:
self._advance_codepoint()
return False
def find_char(self, marker, limit):
# only works for ascii markers!
assert 0 <= ord(marker) < 128
# ascii fast path
if self.ulen == len(self.text):
end = len(self.text)
if limit >= 0:
end = min(end, self.pos + limit)
pos = self.pos
assert pos >= 0
assert end >= 0
pos = self.text.find(marker, pos, end)
if pos >= 0:
self.pos = self.upos = pos + 1
return True
else:
self.pos = self.upos = end
return False
if limit < 0:
limit = sys.maxint
# XXX it might be better to search for the marker quickly, then compute
# the new upos afterwards.
scanned = 0
while scanned < limit:
# don't use next_char here, since that computes a slice etc
if self.exhausted():
return False
# this is never true if self.text[pos] is part of a larger char
found = self.text[self.pos] == marker
if found:
self.pos += 1
self.upos += 1
return True
self._advance_codepoint()
scanned += 1
return False
@always_inline
def _advance_codepoint(self):
# must only be called after checking self.exhausted()!
self.pos = next_codepoint_pos(self.text, self.pos)
self.upos += 1
def check_decoded(space, w_decoded):
if not space.isinstance_w(w_decoded, space.w_unicode):
msg = "decoder should return a string result, not '%T'"
raise oefmt(space.w_TypeError, msg, w_decoded)
return w_decoded
def unwrap_newline(space, w_newline):
if space.is_none(w_newline):
newline = None
else:
newline = space.utf8_w(w_newline)
if newline and newline not in ('\n', '\r\n', '\r'):
raise oefmt(space.w_ValueError,
"illegal newline value: %R", w_newline)
return newline
class W_TextIOWrapper(W_TextIOBase):
def __init__(self, space):
W_TextIOBase.__init__(self, space)
self.state = STATE_ZERO
self.w_encoder = None
self.w_decoder = None
self.decoded = DecodeBuffer()
self.pending_bytes = None # list of bytes objects waiting to be
# written, or NULL
self.pending_bytes_count = 0
self.chunk_size = 8192
self.b2cratio = 0.0
self.readuniversal = False
self.readtranslate = False
self.readnl = None
self.encodefunc = None # Specialized encoding func (see below)
self.encoding_start_of_stream = False # Whether or not it's the start
# of the stream
self.snapshot = None
@unwrap_spec(encoding="text0_or_none", line_buffering=int, write_through=int)
def descr_init(self, space, w_buffer, encoding=None,
w_errors=None, w_newline=None, line_buffering=0,
write_through=0):
self.state = STATE_ZERO
self.w_buffer = w_buffer
self.w_encoding = w_encoding = _determine_encoding(space, encoding, w_buffer)
if space.is_none(w_errors) or w_errors is None:
w_errors = space.newtext("strict")
if not space.isinstance_w(w_errors, space.w_text):
raise oefmt(space.w_TypeError, "TextIOWrapper() argument 'errors' "
"must be str or None, not %N", w_errors)
io_check_errors(space, w_errors)
self.w_errors = w_errors
newline = unwrap_newline(space, w_newline)
self.line_buffering = bool(line_buffering)
self.write_through = bool(write_through)
self._set_newline(newline)
self._set_encoder_decoder(w_encoding, w_errors)
self.seekable = space.is_true(space.call_method(w_buffer, "seekable"))
self.telling = self.seekable
self.has_read1 = space.findattr(w_buffer, space.newtext("read1"))
self._fix_encoder_state()
self.state = STATE_OK
def _set_newline(self, newline):
self.readuniversal = not newline # null or empty
self.readtranslate = newline is None
self.readnl = newline
self.writetranslate = (newline != '')
if not self.readuniversal:
self.writenl = self.readnl
if self.writenl == '\n':
self.writenl = None
elif _WINDOWS:
self.writenl = "\r\n"
else:
self.writenl = None
def _set_encoder_decoder(self, w_encoding, w_errors):
space = self.space
w_codec = interp_codecs.lookup_codec(space,
space.text_w(w_encoding))
if not space.is_true(space.getattr(w_codec,
space.newtext('_is_text_encoding'))):
msg = ("%R is not a text encoding; "
"use codecs.open() to handle arbitrary codecs")
raise oefmt(space.w_LookupError, msg, w_encoding)
# build the decoder object
if space.is_true(space.call_method(self.w_buffer, "readable")):
self.w_decoder = space.call_method(w_codec,
"incrementaldecoder", w_errors)
if self.readuniversal:
self.w_decoder = space.call_function(
space.gettypeobject(W_IncrementalNewlineDecoder.typedef),
self.w_decoder, space.newbool(self.readtranslate))
# build the encoder object
if space.is_true(space.call_method(self.w_buffer, "writable")):
self.w_encoder = space.call_method(w_codec,
"incrementalencoder", w_errors)
def _fix_encoder_state(self):
space = self.space
self.encoding_start_of_stream = False
if self.seekable and self.w_encoder:
self.encoding_start_of_stream = True
w_cookie = space.call_method(self.w_buffer, "tell")
if not space.eq_w(w_cookie, space.newint(0)):
self.encoding_start_of_stream = False
space.call_method(self.w_encoder, "setstate", space.newint(0))
def reconfigure(self, space, __args__):
"""
Reconfigure the text stream with new parameters.
This also does an implicit stream flush.
"""
# XXX quite annoying, kwonly args can't easily support unwrapped None
# as the default, do our own argument parsing
args_w, kwargs_w = __args__.unpack()
if args_w:
raise oefmt(space.w_TypeError, "reconfigure() takes no positional arguments")
# for all arguments passing w_None means "keep value", with two exceptions:
# 1) if encoding is given but not errors, set errors to strict
# 2) newline=None means universal newline support
w_encoding = kwargs_w.pop("encoding", space.w_None)
w_errors = kwargs_w.pop("errors", space.w_None)
w_newline = kwargs_w.pop("newline", None)
if (not space.is_none(w_encoding) and
not space.isinstance_w(w_encoding, space.w_text)):
raise oefmt(space.w_TypeError,
"reconfigure argument 'encoding' must be str or None, not %T",
w_encoding)
if (not space.is_none(w_errors) and
not space.isinstance_w(w_errors, space.w_text)):
raise oefmt(space.w_TypeError,
"reconfigure argument 'errors' must be str or None, not %T",
w_errors)
if (not space.is_none(w_newline) and
not space.isinstance_w(w_newline, space.w_text)):
raise oefmt(space.w_TypeError,
"reconfigure argument 'newline' must be str or None, not %T",
w_newline)
w_line_buffering = kwargs_w.pop("line_buffering", space.w_None)
w_write_through = kwargs_w.pop("write_through", space.w_None)
if kwargs_w:
key, w_value = kwargs_w.popitem()
raise oefmt(space.w_TypeError, "%8 is an invalid keyword argument for reconfigure()", key)
if self.decoded.text is not None:
if (not space.is_none(w_encoding) or
not space.is_none(w_errors) or
w_newline is not None):
self._unsupportedoperation(
space, "It is not possible to set the encoding "
"or newline of stream after the first read")
newline = None
if w_newline is not None:
newline = unwrap_newline(space, w_newline)
line_buffering = self.line_buffering
if not space.is_none(w_line_buffering):
line_buffering = bool(space.int_w(w_line_buffering))
write_through = self.write_through
if not space.is_none(w_write_through):
write_through = bool(space.int_w(w_write_through))
space.call_method(self, "flush")
if w_newline is not None:
self._set_newline(newline)
if not space.is_none(w_errors):
io_check_errors(space, w_errors)
# if encoding is specified but not errors, set errors to strict
if not space.is_none(w_encoding):
if space.is_none(w_errors):
w_errors = space.newtext("strict")
if not space.is_none(w_encoding) or not space.is_none(w_errors) or (w_newline is not None and self.readuniversal):
if space.is_none(w_encoding):
w_encoding = self.w_encoding
elif space.eq_w(w_encoding, space.newtext("locale")):
w_encoding = space.newtext("utf-8")
if space.is_none(w_errors):
w_errors = self.w_errors
# NB we also need to call _set_encoder_decoder if the newline
# changed to readuniversal, to get newline translation
self._set_encoder_decoder(w_encoding, w_errors)
self.w_encoding = w_encoding
self.w_errors = w_errors
self.line_buffering = line_buffering
self.write_through = write_through
self._fix_encoder_state()
self.b2cratio = 0.0
def _check_init(self, space):
if self.state == STATE_ZERO:
raise oefmt(space.w_ValueError,
"I/O operation on uninitialized object")
def _check_attached(self, space):
if jit.promote(self.state) == STATE_DETACHED:
raise oefmt(space.w_ValueError,
"underlying buffer has been detached")
self._check_init(space)
def _check_closed(self, space, message=None):
self._check_init(space)
W_TextIOBase._check_closed(self, space, message)
def __w_attr_repr(self, space, name):
w_attr = space.findattr(self, space.newtext(name))
if w_attr is None:
return space.newtext("")
return space.mod(space.newtext("%s=%%r " % name), w_attr)
def descr_repr(self, space):
self._check_init(space)
w_args = space.newtuple([self.__w_attr_repr(space, 'name'),
self.__w_attr_repr(space, 'mode'),
self.w_encoding])
return space.mod(
space.newtext("<_io.TextIOWrapper %s%sencoding=%r>"), w_args
)
def readable_w(self, space):
self._check_attached(space)
return space.call_method(self.w_buffer, "readable")
def writable_w(self, space):
self._check_attached(space)
return space.call_method(self.w_buffer, "writable")
def seekable_w(self, space):
self._check_attached(space)
return space.call_method(self.w_buffer, "seekable")
def isatty_w(self, space):
self._check_attached(space)
return space.call_method(self.w_buffer, "isatty")
def fileno_w(self, space):
self._check_attached(space)
return space.call_method(self.w_buffer, "fileno")
def closed_get_w(self, space):
self._check_attached(space)
return space.getattr(self.w_buffer, space.newtext("closed"))
def newlines_get_w(self, space):
self._check_attached(space)
if self.w_decoder is None:
return space.w_None
return space.findattr(self.w_decoder, space.newtext("newlines"))
def name_get_w(self, space):
self._check_attached(space)
return space.getattr(self.w_buffer, space.newtext("name"))
def flush_w(self, space):
self._check_attached(space)
self._check_closed(space)
self.telling = self.seekable
self._writeflush(space)
space.call_method(self.w_buffer, "flush")
@unwrap_spec(w_pos = WrappedDefault(None))
def truncate_w(self, space, w_pos=None):
self._check_attached(space)
space.call_method(self, "flush")
return space.call_method(self.w_buffer, "truncate", w_pos)
def close_w(self, space):
self._check_attached(space)
if space.is_true(space.getattr(self.w_buffer,
space.newtext("closed"))):
return
try:
space.call_method(self, "flush")
except OperationError as e:
try:
ret = space.call_method(self.w_buffer, "close")
except OperationError as e2:
e2.chain_exceptions(space, e)
raise
else:
ret = space.call_method(self.w_buffer, "close")
self.maybe_unregister_rpython_finalizer_io(space)
return ret
def _dealloc_warn_w(self, space, w_source):
space.call_method(self.w_buffer, "_dealloc_warn", w_source)
# _____________________________________________________________
# read methods
def _read_chunk(self, space, size_hint):
"""Read and decode the next chunk of data from the BufferedReader.
The return value is True unless EOF was reached. The decoded string
is placed in self.decoded (replacing its previous value).
The entire input chunk is sent to the decoder, though some of it may
remain buffered in the decoder, yet to be converted."""
if not self.w_decoder:
self._unsupportedoperation(space, "not readable")
if self.telling:
# To prepare for tell(), we need to snapshot a point in the file
# where the decoder's input buffer is empty.
w_decoder = self.w_decoder
# fast path for the common case of decoder being
# W_IncrementalNewlineDecoder. avoids (un)wrapping the tuple too
if type(w_decoder) is W_IncrementalNewlineDecoder:
w_dec_buffer, dec_flags = w_decoder.getstate_u(space)
else:
w_state = space.call_method(self.w_decoder, "getstate")
if (not space.isinstance_w(w_state, space.w_tuple)
or space.len_w(w_state) != 2):
raise oefmt(space.w_TypeError, "illegal decoder state")
# Given this, we know there was a valid snapshot point
# len(dec_buffer) bytes ago with decoder state (b'', dec_flags).
w_dec_buffer, w_dec_flags = space.unpackiterable(w_state, 2)
if not space.isinstance_w(w_dec_buffer, space.w_bytes):
msg = ("illegal decoder state: the first value should be a "
"bytes object not '%T'")
raise oefmt(space.w_TypeError, msg, w_dec_buffer)
dec_flags = space.int_w(w_dec_flags)
dec_buffer = space.bytes_w(w_dec_buffer)
else:
dec_buffer = None
dec_flags = 0
# Read a chunk, decode it, and put the result in self.decoded
if size_hint > 0:
size_hint = int(max(self.b2cratio, 1.0) * float(size_hint))
chunk_size = max(self.chunk_size, size_hint)
func_name = "read1" if self.has_read1 else "read"
w_input = space.call_method(self.w_buffer, func_name,
space.newint(chunk_size))
try:
input_buf = w_input.buffer_w(space, space.BUF_SIMPLE)
except BufferInterfaceNotFound:
msg = ("underlying %s() should have returned a bytes-like "
"object, not '%T'")
raise oefmt(space.w_TypeError, msg, func_name, w_input)
nbytes = input_buf.getlength()
eof = nbytes == 0
w_decoder = self.w_decoder
if type(w_decoder) is W_IncrementalNewlineDecoder:
w_decoded = w_decoder.decode_w(space, w_input, eof)
else:
w_decoded = space.call_method(w_decoder, "decode",
w_input, space.newbool(eof))
self.decoded.set(space, w_decoded)
nchars = space.len_w(w_decoded)
if nchars > 0:
eof = False
self.b2cratio = float(nbytes) / float(nchars)
else:
self.b2cratio = 0.0
if self.telling:
# At the snapshot point, len(dec_buffer) bytes before the read,
# the next input to be decoded is dec_buffer + input_chunk.
next_input = dec_buffer + input_buf.as_str()
self.snapshot = PositionSnapshot(dec_flags, next_input)
return not eof
def _ensure_data(self, space, size_hint):
while not self.decoded.has_data():
try:
if not self._read_chunk(space, size_hint):
self.decoded.reset()
self.snapshot = None
return False
except OperationError as e:
if trap_eintr(space, e):
continue
raise
return True
def next_w(self, space):
self._check_attached(space)
self.telling = False
try:
return W_TextIOBase.next_w(self, space)
except OperationError as e:
if e.match(space, space.w_StopIteration):
self.telling = self.seekable
raise
def read_w(self, space, w_size=None):
self._check_attached(space)
self._check_closed(space)
if not self.w_decoder:
self._unsupportedoperation(space, "not readable")
size = convert_size(space, w_size)
self._writeflush(space)
if size < 0:
return self._read_all(space)
else:
return self._read(space, size)
def _read_all(self, space):
w_bytes = space.call_method(self.w_buffer, "read")
w_decoded = space.call_method(self.w_decoder, "decode", w_bytes, space.w_True)
check_decoded(space, w_decoded)
w_result = space.newutf8(*self.decoded.get_chars(-1))
w_final = space.add(w_result, w_decoded)
if self.snapshot:
# CPython GH-35928
self.decoded.reset()
self.snapshot = None
return w_final
def _read(self, space, size):
remaining = size
builder = Utf8StringBuilder(size)
# Keep reading chunks until we have n characters to return
while remaining > 0:
if not self._ensure_data(space, remaining):
break
data, size = self.decoded.get_chars(remaining)
builder.append_utf8(data, size)
remaining -= size
return space.newutf8(builder.build(), builder.getlength())
def _scan_line_ending(self, limit):
if self.readtranslate:
# Newlines are already translated, only search for \n
return self.decoded.find_char('\n', limit)
if self.readuniversal:
return self.decoded.find_newline_universal(limit)
else:
# Non-universal mode.
newline = self.readnl
if newline == '\r\n':
return self.decoded.find_crlf(limit)
else:
return self.decoded.find_char(newline[0], limit)
def readline_w(self, space, w_limit=None):
self._check_attached(space)
self._check_closed(space)
self._writeflush(space)
limit = convert_size(space, w_limit)
text, lgt = self._readline(space, limit)
return space.newutf8(text, lgt)
def _readline(self, space, limit):
# This is a separate function so that readline_w() can be jitted.
remnant = None
remnant_ulen = -1
builder = Utf8StringBuilder()
while True:
# First, get some data if necessary
has_data = self._ensure_data(space, 0)
if not has_data:
# end of file
if remnant:
builder.append_utf8(remnant, remnant_ulen)
break
if remnant:
assert not self.readtranslate and self.readnl == '\r\n'
assert self.decoded.pos == 0
if remnant == '\r' and self.decoded.text[0] == '\n':
builder.append_utf8('\r\n', 2)
self.decoded.pos = 1
self.decoded.upos = 1
remnant = None
remnant_ulen = -1
break
else:
builder.append_utf8(remnant, remnant_ulen)
remnant = None
remnant_ulen = -1
continue
if limit >= 0:
remaining = limit - builder.getlength()
assert remaining >= 0
else:
remaining = -1
start = self.decoded.pos
ustart = self.decoded.upos
assert start >= 0
found = self._scan_line_ending(remaining)
end_scan = self.decoded.pos
uend_scan = self.decoded.upos
if end_scan > start:
builder.append_utf8_slice(self.decoded.text, start, end_scan, uend_scan - ustart)
if found or (limit >= 0 and builder.getlength() >= limit):
break
# There may be some remaining chars we'll have to prepend to the
# next chunk of data
if not self.decoded.exhausted():
remnant, remnant_ulen = self.decoded.get_chars(-1)
# We have consumed the buffer
self.decoded.reset()
result = builder.build()
lgt = builder.getlength()
return (result, lgt)
# _____________________________________________________________
# write methods
def write_w(self, space, w_text):
self._check_attached(space)
self._check_closed(space)
if not self.w_encoder:
self._unsupportedoperation(space, "not writable")
if not space.isinstance_w(w_text, space.w_unicode):
raise oefmt(space.w_TypeError,
"unicode argument expected, got '%T'", w_text)
text, textlen = space.utf8_len_w(w_text)
haslf = False
if (self.writetranslate and self.writenl) or self.line_buffering:
if text.find('\n') >= 0:
haslf = True
if haslf and self.writetranslate and self.writenl:
w_text = space.call_method(w_text, "replace", space.newutf8('\n', 1),
space.newutf8(self.writenl, codepoints_in_utf8(self.writenl)))
text = space.utf8_w(w_text)
needflush = False
text_needflush = False
if self.write_through:
text_needflush = True
if self.line_buffering and (haslf or text.find('\r') >= 0):
needflush = True
# XXX What if we were just reading?
if self.encodefunc:
# XXX CPython has an optimization if
# - isascii(w_text) and
# - len(w_text) <= chunk_size and
# - is_asciicompat_encoding(self.encodefunc)
# then w_bytes = w_text
w_bytes = self.encodefunc(space, w_text, self.errors)
self.encoding_start_of_stream = False
else:
w_bytes = space.call_method(self.w_encoder, "encode", w_text)
b = space.bytes_w(w_bytes)
if len(b) >= self.chunk_size:
# _textiowrapper_writeflush() calls buffer.write().
# self->pending_bytes can be appended during buffer->write()
# or other thread.
# We need to loop until buffer becomes empty.
# https://github.com/python/cpython/issues/118138
# https://github.com/python/cpython/issues/119506
while self.pending_bytes:
self._writeflush(space)
if not self.pending_bytes:
self.pending_bytes = [b]
else:
self.pending_bytes.append(b)
self.pending_bytes_count += len(b)
if (self.pending_bytes_count >= self.chunk_size or
needflush or text_needflush):
self._writeflush(space)
if needflush:
space.call_method(self.w_buffer, "flush")
if self.snapshot:
# CPython GH-35928
self.decoded.reset()
self.snapshot = None
if self.w_decoder:
space.call_method(self.w_decoder, "reset")
return space.newint(textlen)
def _writeflush(self, space):
# jit inlinable fast path
if not self.pending_bytes:
return
self._really_flush(space)
def _really_flush(self, space):
pending_bytes = ''.join(self.pending_bytes)
self.pending_bytes = None
self.pending_bytes_count = 0
while True:
try:
space.call_method(self.w_buffer, "write",
space.newbytes(pending_bytes))
except OperationError as e:
if trap_eintr(space, e):
continue
raise
else:
break
def detach_w(self, space):
self._check_attached(space)
space.call_method(self, "flush")
w_buffer = self.w_buffer
self.w_buffer = None
self.state = STATE_DETACHED
return w_buffer
# _____________________________________________________________
# seek/tell
def _decoder_setstate(self, space, cookie):
# When seeking to the start of the stream, we call decoder.reset()
# rather than decoder.getstate().
# This is for a few decoders such as utf-16 for which the state value
# at start is not (b"", 0) but e.g. (b"", 2) (meaning, in the case of
# utf-16, that we are expecting a BOM).
if cookie.start_pos == 0 and cookie.dec_flags == 0:
space.call_method(self.w_decoder, "reset")
else:
space.call_method(self.w_decoder, "setstate",
space.newtuple2(space.newbytes(""),
space.newint(cookie.dec_flags)))
def _encoder_reset(self, space, start_of_stream):
if start_of_stream:
space.call_method(self.w_encoder, "reset")
self.encoding_start_of_stream = True
else:
space.call_method(self.w_encoder, "setstate", space.newint(0))
self.encoding_start_of_stream = False
def _encoder_setstate(self, space, cookie):
self._encoder_reset(space,
cookie.start_pos == 0 and cookie.dec_flags == 0)
@unwrap_spec(whence=int)
def seek_w(self, space, w_pos, whence=0):
self._check_attached(space)
if not self.seekable:
self._unsupportedoperation(space,
"underlying stream is not seekable")
if whence == 1:
# seek relative to current position
if not space.eq_w(w_pos, space.newint(0)):
self._unsupportedoperation(
space, "can't do nonzero cur-relative seeks")
# Seeking to the current position should attempt to sync the
# underlying buffer with the current position.
w_pos = space.call_method(self, "tell")
elif whence == 2:
# seek relative to end of file
if not space.eq_w(w_pos, space.newint(0)):
self._unsupportedoperation(
space, "can't do nonzero end-relative seeks")
space.call_method(self, "flush")
self.decoded.reset()
self.snapshot = None
if self.w_decoder:
space.call_method(self.w_decoder, "reset")
w_res = space.call_method(self.w_buffer, "seek",
w_pos, space.newint(whence))
if self.w_encoder:
# If seek() == 0, we are at the start of stream
start_of_stream = space.eq_w(w_res, space.newint(0))
self._encoder_reset(space, start_of_stream)
return w_res
elif whence != 0:
raise oefmt(space.w_ValueError,
"invalid whence (%d, should be 0, 1 or 2)",
whence)
if space.is_true(space.lt(w_pos, space.newint(0))):
raise oefmt(space.w_ValueError,
"negative seek position %R", w_pos)
space.call_method(self, "flush")
# The strategy of seek() is to go back to the safe start point and
# replay the effect of read(chars_to_skip) from there.
cookie = PositionCookie(space.bigint_w(w_pos))
# Seek back to the safe start point
space.call_method(self.w_buffer, "seek", space.newint(cookie.start_pos))
self.decoded.reset()
self.snapshot = None
# Restore the decoder to its state from the safe start point.
if self.w_decoder:
self._decoder_setstate(space, cookie)
if cookie.chars_to_skip:
# Just like _read_chunk, feed the decoder and save a snapshot.
w_chunk = space.call_method(self.w_buffer, "read",
space.newint(cookie.bytes_to_feed))
if not space.isinstance_w(w_chunk, space.w_bytes):
msg = "underlying read() should have returned " \
"a bytes object, not '%T'"
raise oefmt(space.w_TypeError, msg, w_chunk)
self.snapshot = PositionSnapshot(cookie.dec_flags,
space.bytes_w(w_chunk))
w_decoded = space.call_method(self.w_decoder, "decode",
w_chunk, space.newbool(bool(cookie.need_eof)))
w_decoded = check_decoded(space, w_decoded)
# Skip chars_to_skip of the decoded characters
if space.len_w(w_decoded) < cookie.chars_to_skip:
raise oefmt(space.w_IOError,
"can't restore logical file position")
self.decoded.set(space, w_decoded)
self.decoded.pos = w_decoded._index_to_byte(cookie.chars_to_skip)
self.decoded.upos = cookie.chars_to_skip
else:
self.snapshot = PositionSnapshot(cookie.dec_flags, "")
# Finally, reset the encoder (merely useful for proper BOM handling)
if self.w_encoder:
self._encoder_setstate(space, cookie)
return w_pos
def tell_w(self, space):
self._check_closed(space)
if not self.seekable:
self._unsupportedoperation(space,
"underlying stream is not seekable")
if not self.telling:
raise oefmt(space.w_IOError,
"telling position disabled by next() call")
self._writeflush(space)
space.call_method(self, "flush")
w_pos = space.call_method(self.w_buffer, "tell")
if self.w_decoder is None or self.snapshot is None:
assert not self.decoded.text
return w_pos
cookie = PositionCookie(space.bigint_w(w_pos))
# Skip backward to the snapshot point (see _read_chunk)
cookie.dec_flags = self.snapshot.flags
input = self.snapshot.input
if len(input) > cookie.start_pos:
# Must be a device or something that reports 0 for tell()
# CPython GH-95782
return space.newint(0)
cookie.start_pos -= len(input)
# How many decoded characters have been used up since the snapshot?
if self.decoded.pos == 0:
# We haven't moved from the snapshot point.
return space.newlong_from_rbigint(cookie.pack())
# chars_to_skip = codepoints_in_utf8(
# self.decoded.text, end=self.decoded.pos)
chars_to_skip = self.decoded.upos
# Starting from the snapshot position, we will walk the decoder
# forward until it gives us enough decoded characters.
w_saved_state = space.call_method(self.w_decoder, "getstate")
skip_bytes = 0
# Fast search for an acceptable start point, close to our
# current pos
skip_bytes = int(self.b2cratio * chars_to_skip)
skip_back = 1;
# ??? this assert is in CPython but is not triggered??
# assert skip_back <= len(input)
while skip_bytes > 0:
# Decode up to temptative start point
self._decoder_setstate(space, cookie)
w_decoded = space.call_method(self.w_decoder, "decode",
space.newbytes(input[0:skip_bytes])
)
check_decoded(space, w_decoded)
chars_decoded = space.len_w(w_decoded)
if chars_decoded <= chars_to_skip:
w_state = space.call_method(self.w_decoder, "getstate")
w_dec_buffer, w_flags = space.unpackiterable(w_state, 2)
dec_buffer_len = space.len_w(w_dec_buffer)
if (dec_buffer_len == 0):
# Before pos and no bytes buffered in decoder => OK
cookie.dec_flags = space.int_w(w_flags)
chars_to_skip -= chars_decoded
break
# Skip back by buffered amount and reset heuristic
skip_bytes -= dec_buffer_len;
skip_back = 1;
else:
# We're too far ahead, skip back a bit
skip_bytes -= skip_back;
skip_back *= 2;
if skip_bytes <= 0:
skip_bytes = 0
self._decoder_setstate(space, cookie)
cookie.start_pos += skip_bytes
cookie.chars_to_skip = chars_to_skip
if chars_to_skip == 0:
space.call_method(self.w_decoder, "setstate", w_saved_state)
return space.newlong_from_rbigint(cookie.pack())
try:
# Note our initial start point
self._decoder_setstate(space, cookie)
# Feed the decoder one byte at a time. As we go, note the nearest
# "safe start point" before the current location (a point where
# the decoder has nothing buffered, so seek() can safely start
# from there and advance to this location).
chars_decoded = 0
i = skip_bytes
while i < len(input):
w_decoded = space.call_method(self.w_decoder, "decode",
space.newbytes(input[i]))
check_decoded(space, w_decoded)
chars_decoded += space.len_w(w_decoded)
cookie.bytes_to_feed += 1
w_state = space.call_method(self.w_decoder, "getstate")
w_dec_buffer, w_flags = space.unpackiterable(w_state, 2)
dec_buffer_len = space.len_w(w_dec_buffer)
if dec_buffer_len == 0 and chars_decoded <= chars_to_skip:
# Decoder buffer is empty, so this is a safe start point.
cookie.start_pos += cookie.bytes_to_feed
chars_to_skip -= chars_decoded
assert chars_to_skip >= 0
cookie.dec_flags = space.int_w(w_flags)
cookie.bytes_to_feed = 0
chars_decoded = 0
if chars_decoded >= chars_to_skip:
break
i += 1
else:
# We didn't get enough decoded data; signal EOF to get more.
w_decoded = space.call_method(self.w_decoder, "decode",
space.newbytes(""),
space.newint(1)) # final=1
check_decoded(space, w_decoded)
chars_decoded += space.len_w(w_decoded)
cookie.need_eof = 1
if chars_decoded < chars_to_skip:
raise oefmt(space.w_IOError,
"can't reconstruct logical file position")
finally:
space.call_method(self.w_decoder, "setstate", w_saved_state)
# The returned cookie corresponds to the last safe start point.
cookie.chars_to_skip = chars_to_skip
return space.newlong_from_rbigint(cookie.pack())
def chunk_size_get_w(self, space):
self._check_attached(space)
return space.newint(self.chunk_size)
def chunk_size_set_w(self, space, w_size):
self._check_attached(space)
size = space.int_w(w_size)
if size <= 0:
raise oefmt(space.w_ValueError,
"a strictly positive integer is required")
self.chunk_size = size
W_TextIOWrapper.typedef = TypeDef(
'_io.TextIOWrapper', W_TextIOBase.typedef,
__new__ = generic_new_descr(W_TextIOWrapper),
__init__ = interp2app(W_TextIOWrapper.descr_init),
__repr__ = interp2app(W_TextIOWrapper.descr_repr),
__next__ = interp2app(W_TextIOWrapper.next_w),
__getstate__ = interp2app(W_TextIOWrapper.getstate_w),
read = interp2app(W_TextIOWrapper.read_w),
readline = interp2app(W_TextIOWrapper.readline_w),
write = interp2app(W_TextIOWrapper.write_w),
seek = interp2app(W_TextIOWrapper.seek_w),
tell = interp2app(W_TextIOWrapper.tell_w),
detach = interp2app(W_TextIOWrapper.detach_w),
flush = interp2app(W_TextIOWrapper.flush_w),
truncate = interp2app(W_TextIOWrapper.truncate_w),
close = interp2app(W_TextIOWrapper.close_w),
reconfigure = interp2app(W_TextIOWrapper.reconfigure),
line_buffering = interp_attrproperty("line_buffering", W_TextIOWrapper,
wrapfn="newbool"),
write_through = interp_attrproperty("write_through", W_TextIOWrapper,
wrapfn="newbool"),
readable = interp2app(W_TextIOWrapper.readable_w),
writable = interp2app(W_TextIOWrapper.writable_w),
seekable = interp2app(W_TextIOWrapper.seekable_w),
isatty = interp2app(W_TextIOWrapper.isatty_w),
fileno = interp2app(W_TextIOWrapper.fileno_w),
_dealloc_warn = interp2app(W_TextIOWrapper._dealloc_warn_w),
name = GetSetProperty(W_TextIOWrapper.name_get_w),
buffer = interp_attrproperty_w("w_buffer", cls=W_TextIOWrapper),
closed = GetSetProperty(W_TextIOWrapper.closed_get_w),
errors = interp_attrproperty_w("w_errors", cls=W_TextIOWrapper),
newlines = GetSetProperty(W_TextIOWrapper.newlines_get_w),
_CHUNK_SIZE = GetSetProperty(
W_TextIOWrapper.chunk_size_get_w, W_TextIOWrapper.chunk_size_set_w
),
)
|