From cac7af6863a997376bffe00ecf07a62c5618a5d9 Mon Sep 17 00:00:00 2001 From: Eric Smith Date: Mon, 27 Apr 2009 19:04:37 +0000 Subject: Issue #5793: rationalize isdigit / isalpha / tolower, etc. Will port to py3k. Should fix Windows buildbot errors. --- Include/Python.h | 1 + Include/bytes_methods.h | 35 ++---- Include/pyctype.h | 28 +++++ Makefile.pre.in | 2 + Misc/NEWS | 3 + Objects/bytearrayobject.c | 18 +-- Objects/bytes_methods.c | 279 +++++------------------------------------ Objects/stringlib/stringdefs.h | 13 +- PC/VC6/pythoncore.dsp | 4 + PC/VS7.1/pythoncore.vcproj | 3 + PC/VS8.0/pythoncore.vcproj | 8 ++ PC/os2emx/Makefile | 1 + PCbuild/pythoncore.vcproj | 8 ++ Python/pyctype.c | 214 +++++++++++++++++++++++++++++++ Python/pystrtod.c | 24 ++-- 15 files changed, 338 insertions(+), 303 deletions(-) create mode 100644 Include/pyctype.h create mode 100644 Python/pyctype.c diff --git a/Include/Python.h b/Include/Python.h index 7af6b13..9e140a5 100644 --- a/Include/Python.h +++ b/Include/Python.h @@ -134,6 +134,7 @@ #include "compile.h" #include "eval.h" +#include "pyctype.h" #include "pystrtod.h" #include "pystrcmp.h" diff --git a/Include/bytes_methods.h b/Include/bytes_methods.h index 59873f2..4125666 100644 --- a/Include/bytes_methods.h +++ b/Include/bytes_methods.h @@ -34,23 +34,15 @@ extern const char _Py_title__doc__[]; extern const char _Py_capitalize__doc__[]; extern const char _Py_swapcase__doc__[]; -#define FLAG_LOWER 0x01 -#define FLAG_UPPER 0x02 -#define FLAG_ALPHA (FLAG_LOWER|FLAG_UPPER) -#define FLAG_DIGIT 0x04 -#define FLAG_ALNUM (FLAG_ALPHA|FLAG_DIGIT) -#define FLAG_SPACE 0x08 -#define FLAG_XDIGIT 0x10 - -extern const unsigned int _Py_ctype_table[256]; - -#define ISLOWER(c) (_Py_ctype_table[Py_CHARMASK(c)] & FLAG_LOWER) -#define ISUPPER(c) (_Py_ctype_table[Py_CHARMASK(c)] & FLAG_UPPER) -#define ISALPHA(c) (_Py_ctype_table[Py_CHARMASK(c)] & FLAG_ALPHA) -#define ISDIGIT(c) (_Py_ctype_table[Py_CHARMASK(c)] & FLAG_DIGIT) -#define ISXDIGIT(c) (_Py_ctype_table[Py_CHARMASK(c)] & FLAG_XDIGIT) -#define ISALNUM(c) (_Py_ctype_table[Py_CHARMASK(c)] & FLAG_ALNUM) -#define ISSPACE(c) (_Py_ctype_table[Py_CHARMASK(c)] & FLAG_SPACE) +/* These are left in for backward compatibility and will be removed + in 2.8/3.2 */ +#define ISLOWER(c) Py_ISLOWER(c) +#define ISUPPER(c) Py_ISUPPER(c) +#define ISALPHA(c) Py_ISALPHA(c) +#define ISDIGIT(c) Py_ISDIGIT(c) +#define ISXDIGIT(c) Py_ISXDIGIT(c) +#define ISALNUM(c) Py_ISALNUM(c) +#define ISSPACE(c) Py_ISSPACE(c) #undef islower #define islower(c) undefined_islower(c) @@ -67,11 +59,10 @@ extern const unsigned int _Py_ctype_table[256]; #undef isspace #define isspace(c) undefined_isspace(c) -extern const unsigned char _Py_ctype_tolower[256]; -extern const unsigned char _Py_ctype_toupper[256]; - -#define TOLOWER(c) (_Py_ctype_tolower[Py_CHARMASK(c)]) -#define TOUPPER(c) (_Py_ctype_toupper[Py_CHARMASK(c)]) +/* These are left in for backward compatibility and will be removed + in 2.8/3.2 */ +#define TOLOWER(c) Py_TOLOWER(c) +#define TOUPPER(c) Py_TOUPPER(c) #undef tolower #define tolower(c) undefined_tolower(c) diff --git a/Include/pyctype.h b/Include/pyctype.h new file mode 100644 index 0000000..0d211ec --- /dev/null +++ b/Include/pyctype.h @@ -0,0 +1,28 @@ +#ifndef PYCTYPE_H +#define PYCTYPE_H + +#define PY_CTF_LOWER 0x01 +#define PY_CTF_UPPER 0x02 +#define PY_CTF_ALPHA (PY_CTF_LOWER|PY_CTF_UPPER) +#define PY_CTF_DIGIT 0x04 +#define PY_CTF_ALNUM (PY_CTF_ALPHA|PY_CTF_DIGIT) +#define PY_CTF_SPACE 0x08 +#define PY_CTF_XDIGIT 0x10 + +extern const unsigned int _Py_ctype_table[256]; + +#define Py_ISLOWER(c) (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_LOWER) +#define Py_ISUPPER(c) (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_UPPER) +#define Py_ISALPHA(c) (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_ALPHA) +#define Py_ISDIGIT(c) (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_DIGIT) +#define Py_ISXDIGIT(c) (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_XDIGIT) +#define Py_ISALNUM(c) (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_ALNUM) +#define Py_ISSPACE(c) (_Py_ctype_table[Py_CHARMASK(c)] & PY_CTF_SPACE) + +extern const unsigned char _Py_ctype_tolower[256]; +extern const unsigned char _Py_ctype_toupper[256]; + +#define Py_TOLOWER(c) (_Py_ctype_tolower[Py_CHARMASK(c)]) +#define Py_TOUPPER(c) (_Py_ctype_toupper[Py_CHARMASK(c)]) + +#endif /* !PYCTYPE_H */ diff --git a/Makefile.pre.in b/Makefile.pre.in index aa38885..5b77d91 100644 --- a/Makefile.pre.in +++ b/Makefile.pre.in @@ -275,6 +275,7 @@ PYTHON_OBJS= \ Python/mysnprintf.o \ Python/peephole.o \ Python/pyarena.o \ + Python/pyctype.o \ Python/pyfpe.o \ Python/pymath.o \ Python/pystate.o \ @@ -633,6 +634,7 @@ PYTHON_HEADERS= \ Include/pgen.h \ Include/pgenheaders.h \ Include/pyarena.h \ + Include/pyctype.h \ Include/pydebug.h \ Include/pyerrors.h \ Include/pyfpe.h \ diff --git a/Misc/NEWS b/Misc/NEWS index e727d1f..082612e 100644 --- a/Misc/NEWS +++ b/Misc/NEWS @@ -12,6 +12,9 @@ What's New in Python 2.7 alpha 1 Core and Builtins ----------------- +- Issue #5793: Rationalize isdigit / isalpha / tolower, etc. Includes + new Py_ISDIGIT / Py_ISALPHA / Py_TOLOWER, etc. in pctypes.h. + - Issue #4971: Fix titlecase for characters that are their own titlecase, but not their own uppercase. diff --git a/Objects/bytearrayobject.c b/Objects/bytearrayobject.c index ff84495..5591847 100644 --- a/Objects/bytearrayobject.c +++ b/Objects/bytearrayobject.c @@ -2244,16 +2244,16 @@ split_whitespace(const char *s, Py_ssize_t len, Py_ssize_t maxcount) for (i = j = 0; i < len; ) { /* find a token */ - while (i < len && ISSPACE(s[i])) + while (i < len && Py_ISSPACE(s[i])) i++; j = i; - while (i < len && !ISSPACE(s[i])) + while (i < len && !Py_ISSPACE(s[i])) i++; if (j < i) { if (maxcount-- <= 0) break; SPLIT_ADD(s, j, i); - while (i < len && ISSPACE(s[i])) + while (i < len && Py_ISSPACE(s[i])) i++; j = i; } @@ -2478,16 +2478,16 @@ rsplit_whitespace(const char *s, Py_ssize_t len, Py_ssize_t maxcount) for (i = j = len - 1; i >= 0; ) { /* find a token */ - while (i >= 0 && ISSPACE(s[i])) + while (i >= 0 && Py_ISSPACE(s[i])) i--; j = i; - while (i >= 0 && !ISSPACE(s[i])) + while (i >= 0 && !Py_ISSPACE(s[i])) i--; if (j > i) { if (maxcount-- <= 0) break; SPLIT_ADD(s, i + 1, j + 1); - while (i >= 0 && ISSPACE(s[i])) + while (i >= 0 && Py_ISSPACE(s[i])) i--; j = i; } @@ -3054,11 +3054,11 @@ Example: bytearray.fromhex('B9 01EF') -> bytearray(b'\\xb9\\x01\\xef')."); static int hex_digit_to_int(char c) { - if (ISDIGIT(c)) + if (Py_ISDIGIT(c)) return c - '0'; else { - if (ISUPPER(c)) - c = TOLOWER(c); + if (Py_ISUPPER(c)) + c = Py_TOLOWER(c); if (c >= 'a' && c <= 'f') return c - 'a' + 10; } diff --git a/Objects/bytes_methods.c b/Objects/bytes_methods.c index de87905..4d25520 100644 --- a/Objects/bytes_methods.c +++ b/Objects/bytes_methods.c @@ -1,219 +1,6 @@ #include "Python.h" #include "bytes_methods.h" -/* Our own locale-independent ctype.h-like macros */ - -const unsigned int _Py_ctype_table[256] = { - 0, /* 0x0 '\x00' */ - 0, /* 0x1 '\x01' */ - 0, /* 0x2 '\x02' */ - 0, /* 0x3 '\x03' */ - 0, /* 0x4 '\x04' */ - 0, /* 0x5 '\x05' */ - 0, /* 0x6 '\x06' */ - 0, /* 0x7 '\x07' */ - 0, /* 0x8 '\x08' */ - FLAG_SPACE, /* 0x9 '\t' */ - FLAG_SPACE, /* 0xa '\n' */ - FLAG_SPACE, /* 0xb '\v' */ - FLAG_SPACE, /* 0xc '\f' */ - FLAG_SPACE, /* 0xd '\r' */ - 0, /* 0xe '\x0e' */ - 0, /* 0xf '\x0f' */ - 0, /* 0x10 '\x10' */ - 0, /* 0x11 '\x11' */ - 0, /* 0x12 '\x12' */ - 0, /* 0x13 '\x13' */ - 0, /* 0x14 '\x14' */ - 0, /* 0x15 '\x15' */ - 0, /* 0x16 '\x16' */ - 0, /* 0x17 '\x17' */ - 0, /* 0x18 '\x18' */ - 0, /* 0x19 '\x19' */ - 0, /* 0x1a '\x1a' */ - 0, /* 0x1b '\x1b' */ - 0, /* 0x1c '\x1c' */ - 0, /* 0x1d '\x1d' */ - 0, /* 0x1e '\x1e' */ - 0, /* 0x1f '\x1f' */ - FLAG_SPACE, /* 0x20 ' ' */ - 0, /* 0x21 '!' */ - 0, /* 0x22 '"' */ - 0, /* 0x23 '#' */ - 0, /* 0x24 '$' */ - 0, /* 0x25 '%' */ - 0, /* 0x26 '&' */ - 0, /* 0x27 "'" */ - 0, /* 0x28 '(' */ - 0, /* 0x29 ')' */ - 0, /* 0x2a '*' */ - 0, /* 0x2b '+' */ - 0, /* 0x2c ',' */ - 0, /* 0x2d '-' */ - 0, /* 0x2e '.' */ - 0, /* 0x2f '/' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x30 '0' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x31 '1' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x32 '2' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x33 '3' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x34 '4' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x35 '5' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x36 '6' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x37 '7' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x38 '8' */ - FLAG_DIGIT|FLAG_XDIGIT, /* 0x39 '9' */ - 0, /* 0x3a ':' */ - 0, /* 0x3b ';' */ - 0, /* 0x3c '<' */ - 0, /* 0x3d '=' */ - 0, /* 0x3e '>' */ - 0, /* 0x3f '?' */ - 0, /* 0x40 '@' */ - FLAG_UPPER|FLAG_XDIGIT, /* 0x41 'A' */ - FLAG_UPPER|FLAG_XDIGIT, /* 0x42 'B' */ - FLAG_UPPER|FLAG_XDIGIT, /* 0x43 'C' */ - FLAG_UPPER|FLAG_XDIGIT, /* 0x44 'D' */ - FLAG_UPPER|FLAG_XDIGIT, /* 0x45 'E' */ - FLAG_UPPER|FLAG_XDIGIT, /* 0x46 'F' */ - FLAG_UPPER, /* 0x47 'G' */ - FLAG_UPPER, /* 0x48 'H' */ - FLAG_UPPER, /* 0x49 'I' */ - FLAG_UPPER, /* 0x4a 'J' */ - FLAG_UPPER, /* 0x4b 'K' */ - FLAG_UPPER, /* 0x4c 'L' */ - FLAG_UPPER, /* 0x4d 'M' */ - FLAG_UPPER, /* 0x4e 'N' */ - FLAG_UPPER, /* 0x4f 'O' */ - FLAG_UPPER, /* 0x50 'P' */ - FLAG_UPPER, /* 0x51 'Q' */ - FLAG_UPPER, /* 0x52 'R' */ - FLAG_UPPER, /* 0x53 'S' */ - FLAG_UPPER, /* 0x54 'T' */ - FLAG_UPPER, /* 0x55 'U' */ - FLAG_UPPER, /* 0x56 'V' */ - FLAG_UPPER, /* 0x57 'W' */ - FLAG_UPPER, /* 0x58 'X' */ - FLAG_UPPER, /* 0x59 'Y' */ - FLAG_UPPER, /* 0x5a 'Z' */ - 0, /* 0x5b '[' */ - 0, /* 0x5c '\\' */ - 0, /* 0x5d ']' */ - 0, /* 0x5e '^' */ - 0, /* 0x5f '_' */ - 0, /* 0x60 '`' */ - FLAG_LOWER|FLAG_XDIGIT, /* 0x61 'a' */ - FLAG_LOWER|FLAG_XDIGIT, /* 0x62 'b' */ - FLAG_LOWER|FLAG_XDIGIT, /* 0x63 'c' */ - FLAG_LOWER|FLAG_XDIGIT, /* 0x64 'd' */ - FLAG_LOWER|FLAG_XDIGIT, /* 0x65 'e' */ - FLAG_LOWER|FLAG_XDIGIT, /* 0x66 'f' */ - FLAG_LOWER, /* 0x67 'g' */ - FLAG_LOWER, /* 0x68 'h' */ - FLAG_LOWER, /* 0x69 'i' */ - FLAG_LOWER, /* 0x6a 'j' */ - FLAG_LOWER, /* 0x6b 'k' */ - FLAG_LOWER, /* 0x6c 'l' */ - FLAG_LOWER, /* 0x6d 'm' */ - FLAG_LOWER, /* 0x6e 'n' */ - FLAG_LOWER, /* 0x6f 'o' */ - FLAG_LOWER, /* 0x70 'p' */ - FLAG_LOWER, /* 0x71 'q' */ - FLAG_LOWER, /* 0x72 'r' */ - FLAG_LOWER, /* 0x73 's' */ - FLAG_LOWER, /* 0x74 't' */ - FLAG_LOWER, /* 0x75 'u' */ - FLAG_LOWER, /* 0x76 'v' */ - FLAG_LOWER, /* 0x77 'w' */ - FLAG_LOWER, /* 0x78 'x' */ - FLAG_LOWER, /* 0x79 'y' */ - FLAG_LOWER, /* 0x7a 'z' */ - 0, /* 0x7b '{' */ - 0, /* 0x7c '|' */ - 0, /* 0x7d '}' */ - 0, /* 0x7e '~' */ - 0, /* 0x7f '\x7f' */ - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, - 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, -}; - - -const unsigned char _Py_ctype_tolower[256] = { - 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, - 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, - 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, - 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, - 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, - 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f, - 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, - 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, - 0x40, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, - 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, - 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, - 0x78, 0x79, 0x7a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, - 0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, - 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, - 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, - 0x78, 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, - 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, - 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f, - 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, - 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, - 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, - 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf, - 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, - 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, - 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, - 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, - 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, - 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf, - 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, - 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, - 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, - 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff, -}; - -const unsigned char _Py_ctype_toupper[256] = { - 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, - 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, - 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, - 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, - 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, - 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f, - 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, - 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, - 0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, - 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, - 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, - 0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, - 0x60, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, - 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, - 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, - 0x58, 0x59, 0x5a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, - 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, - 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f, - 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, - 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, - 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, - 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf, - 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, - 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, - 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, - 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, - 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, - 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf, - 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, - 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, - 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, - 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff, -}; - - PyDoc_STRVAR_shared(_Py_isspace__doc__, "B.isspace() -> bool\n\ \n\ @@ -228,7 +15,7 @@ _Py_bytes_isspace(const char *cptr, Py_ssize_t len) register const unsigned char *e; /* Shortcut for single character strings */ - if (len == 1 && ISSPACE(*p)) + if (len == 1 && Py_ISSPACE(*p)) Py_RETURN_TRUE; /* Special case for empty strings */ @@ -237,7 +24,7 @@ _Py_bytes_isspace(const char *cptr, Py_ssize_t len) e = p + len; for (; p < e; p++) { - if (!ISSPACE(*p)) + if (!Py_ISSPACE(*p)) Py_RETURN_FALSE; } Py_RETURN_TRUE; @@ -258,7 +45,7 @@ _Py_bytes_isalpha(const char *cptr, Py_ssize_t len) register const unsigned char *e; /* Shortcut for single character strings */ - if (len == 1 && ISALPHA(*p)) + if (len == 1 && Py_ISALPHA(*p)) Py_RETURN_TRUE; /* Special case for empty strings */ @@ -267,7 +54,7 @@ _Py_bytes_isalpha(const char *cptr, Py_ssize_t len) e = p + len; for (; p < e; p++) { - if (!ISALPHA(*p)) + if (!Py_ISALPHA(*p)) Py_RETURN_FALSE; } Py_RETURN_TRUE; @@ -288,7 +75,7 @@ _Py_bytes_isalnum(const char *cptr, Py_ssize_t len) register const unsigned char *e; /* Shortcut for single character strings */ - if (len == 1 && ISALNUM(*p)) + if (len == 1 && Py_ISALNUM(*p)) Py_RETURN_TRUE; /* Special case for empty strings */ @@ -297,7 +84,7 @@ _Py_bytes_isalnum(const char *cptr, Py_ssize_t len) e = p + len; for (; p < e; p++) { - if (!ISALNUM(*p)) + if (!Py_ISALNUM(*p)) Py_RETURN_FALSE; } Py_RETURN_TRUE; @@ -318,7 +105,7 @@ _Py_bytes_isdigit(const char *cptr, Py_ssize_t len) register const unsigned char *e; /* Shortcut for single character strings */ - if (len == 1 && ISDIGIT(*p)) + if (len == 1 && Py_ISDIGIT(*p)) Py_RETURN_TRUE; /* Special case for empty strings */ @@ -327,7 +114,7 @@ _Py_bytes_isdigit(const char *cptr, Py_ssize_t len) e = p + len; for (; p < e; p++) { - if (!ISDIGIT(*p)) + if (!Py_ISDIGIT(*p)) Py_RETURN_FALSE; } Py_RETURN_TRUE; @@ -350,7 +137,7 @@ _Py_bytes_islower(const char *cptr, Py_ssize_t len) /* Shortcut for single character strings */ if (len == 1) - return PyBool_FromLong(ISLOWER(*p)); + return PyBool_FromLong(Py_ISLOWER(*p)); /* Special case for empty strings */ if (len == 0) @@ -359,9 +146,9 @@ _Py_bytes_islower(const char *cptr, Py_ssize_t len) e = p + len; cased = 0; for (; p < e; p++) { - if (ISUPPER(*p)) + if (Py_ISUPPER(*p)) Py_RETURN_FALSE; - else if (!cased && ISLOWER(*p)) + else if (!cased && Py_ISLOWER(*p)) cased = 1; } return PyBool_FromLong(cased); @@ -384,7 +171,7 @@ _Py_bytes_isupper(const char *cptr, Py_ssize_t len) /* Shortcut for single character strings */ if (len == 1) - return PyBool_FromLong(ISUPPER(*p)); + return PyBool_FromLong(Py_ISUPPER(*p)); /* Special case for empty strings */ if (len == 0) @@ -393,9 +180,9 @@ _Py_bytes_isupper(const char *cptr, Py_ssize_t len) e = p + len; cased = 0; for (; p < e; p++) { - if (ISLOWER(*p)) + if (Py_ISLOWER(*p)) Py_RETURN_FALSE; - else if (!cased && ISUPPER(*p)) + else if (!cased && Py_ISUPPER(*p)) cased = 1; } return PyBool_FromLong(cased); @@ -420,7 +207,7 @@ _Py_bytes_istitle(const char *cptr, Py_ssize_t len) /* Shortcut for single character strings */ if (len == 1) - return PyBool_FromLong(ISUPPER(*p)); + return PyBool_FromLong(Py_ISUPPER(*p)); /* Special case for empty strings */ if (len == 0) @@ -432,13 +219,13 @@ _Py_bytes_istitle(const char *cptr, Py_ssize_t len) for (; p < e; p++) { register const unsigned char ch = *p; - if (ISUPPER(ch)) { + if (Py_ISUPPER(ch)) { if (previous_is_cased) Py_RETURN_FALSE; previous_is_cased = 1; cased = 1; } - else if (ISLOWER(ch)) { + else if (Py_ISLOWER(ch)) { if (!previous_is_cased) Py_RETURN_FALSE; previous_is_cased = 1; @@ -473,8 +260,8 @@ _Py_bytes_lower(char *result, const char *cptr, Py_ssize_t len) for (i = 0; i < len; i++) { int c = Py_CHARMASK(result[i]); - if (ISUPPER(c)) - result[i] = TOLOWER(c); + if (Py_ISUPPER(c)) + result[i] = Py_TOLOWER(c); } } @@ -501,8 +288,8 @@ _Py_bytes_upper(char *result, const char *cptr, Py_ssize_t len) for (i = 0; i < len; i++) { int c = Py_CHARMASK(result[i]); - if (ISLOWER(c)) - result[i] = TOUPPER(c); + if (Py_ISLOWER(c)) + result[i] = Py_TOUPPER(c); } } @@ -527,13 +314,13 @@ _Py_bytes_title(char *result, char *s, Py_ssize_t len) */ for (i = 0; i < len; i++) { int c = Py_CHARMASK(*s++); - if (ISLOWER(c)) { + if (Py_ISLOWER(c)) { if (!previous_is_cased) - c = TOUPPER(c); + c = Py_TOUPPER(c); previous_is_cased = 1; - } else if (ISUPPER(c)) { + } else if (Py_ISUPPER(c)) { if (previous_is_cased) - c = TOLOWER(c); + c = Py_TOLOWER(c); previous_is_cased = 1; } else previous_is_cased = 0; @@ -560,16 +347,16 @@ _Py_bytes_capitalize(char *result, char *s, Py_ssize_t len) */ if (0 < len) { int c = Py_CHARMASK(*s++); - if (ISLOWER(c)) - *result = TOUPPER(c); + if (Py_ISLOWER(c)) + *result = Py_TOUPPER(c); else *result = c; result++; } for (i = 1; i < len; i++) { int c = Py_CHARMASK(*s++); - if (ISUPPER(c)) - *result = TOLOWER(c); + if (Py_ISUPPER(c)) + *result = Py_TOLOWER(c); else *result = c; result++; @@ -596,11 +383,11 @@ _Py_bytes_swapcase(char *result, char *s, Py_ssize_t len) */ for (i = 0; i < len; i++) { int c = Py_CHARMASK(*s++); - if (ISLOWER(c)) { - *result = TOUPPER(c); + if (Py_ISLOWER(c)) { + *result = Py_TOUPPER(c); } - else if (ISUPPER(c)) { - *result = TOLOWER(c); + else if (Py_ISUPPER(c)) { + *result = Py_TOLOWER(c); } else *result = c; diff --git a/Objects/stringlib/stringdefs.h b/Objects/stringlib/stringdefs.h index d08971a..f6d0b51 100644 --- a/Objects/stringlib/stringdefs.h +++ b/Objects/stringlib/stringdefs.h @@ -6,15 +6,6 @@ compiled as unicode. */ #define STRINGLIB_IS_UNICODE 0 -/* _tolower and _toupper are defined by SUSv2, but they're not ISO C */ -/* This needs to be cleaned up. See issue 5793. */ -#ifndef _tolower -#define _tolower tolower -#endif -#ifndef _toupper -#define _toupper toupper -#endif - #define STRINGLIB_OBJECT PyStringObject #define STRINGLIB_CHAR char #define STRINGLIB_TYPE_NAME "string" @@ -22,8 +13,8 @@ #define STRINGLIB_EMPTY nullstring #define STRINGLIB_ISDECIMAL(x) ((x >= '0') && (x <= '9')) #define STRINGLIB_TODECIMAL(x) (STRINGLIB_ISDECIMAL(x) ? (x - '0') : -1) -#define STRINGLIB_TOUPPER(x) _toupper(Py_CHARMASK(x)) -#define STRINGLIB_TOLOWER(x) _tolower(Py_CHARMASK(x)) +#define STRINGLIB_TOUPPER Py_TOUPPER +#define STRINGLIB_TOLOWER Py_TOLOWER #define STRINGLIB_FILL memset #define STRINGLIB_STR PyString_AS_STRING #define STRINGLIB_LEN PyString_GET_SIZE diff --git a/PC/VC6/pythoncore.dsp b/PC/VC6/pythoncore.dsp index 38d450b..2cae542 100644 --- a/PC/VC6/pythoncore.dsp +++ b/PC/VC6/pythoncore.dsp @@ -603,6 +603,10 @@ SOURCE=..\..\Python\pyarena.c # End Source File # Begin Source File +SOURCE=..\..\Python\pyctype.c +# End Source File +# Begin Source File + SOURCE=..\..\Python\pyfpe.c # End Source File # Begin Source File diff --git a/PC/VS7.1/pythoncore.vcproj b/PC/VS7.1/pythoncore.vcproj index 672db12..2ee7555 100644 --- a/PC/VS7.1/pythoncore.vcproj +++ b/PC/VS7.1/pythoncore.vcproj @@ -722,6 +722,9 @@ RelativePath="..\..\Python\pyarena.c"> + + + + @@ -1743,6 +1747,10 @@ > + + diff --git a/PC/os2emx/Makefile b/PC/os2emx/Makefile index 82d548e..09ed1a7 100644 --- a/PC/os2emx/Makefile +++ b/PC/os2emx/Makefile @@ -362,6 +362,7 @@ SRC.PYTHON= $(addprefix $(TOP), \ Python/mystrtoul.c \ Python/peephole.c \ Python/pyarena.c \ + Python/pyctype.c \ Python/pyfpe.c \ Python/pymath.c \ Python/pystate.c \ diff --git a/PCbuild/pythoncore.vcproj b/PCbuild/pythoncore.vcproj index 1b9056a..bf4d724 100644 --- a/PCbuild/pythoncore.vcproj +++ b/PCbuild/pythoncore.vcproj @@ -851,6 +851,10 @@ > + + @@ -1743,6 +1747,10 @@ > + + diff --git a/Python/pyctype.c b/Python/pyctype.c new file mode 100644 index 0000000..da117d5 --- /dev/null +++ b/Python/pyctype.c @@ -0,0 +1,214 @@ +#include "Python.h" + +/* Our own locale-independent ctype.h-like macros */ + +const unsigned int _Py_ctype_table[256] = { + 0, /* 0x0 '\x00' */ + 0, /* 0x1 '\x01' */ + 0, /* 0x2 '\x02' */ + 0, /* 0x3 '\x03' */ + 0, /* 0x4 '\x04' */ + 0, /* 0x5 '\x05' */ + 0, /* 0x6 '\x06' */ + 0, /* 0x7 '\x07' */ + 0, /* 0x8 '\x08' */ + PY_CTF_SPACE, /* 0x9 '\t' */ + PY_CTF_SPACE, /* 0xa '\n' */ + PY_CTF_SPACE, /* 0xb '\v' */ + PY_CTF_SPACE, /* 0xc '\f' */ + PY_CTF_SPACE, /* 0xd '\r' */ + 0, /* 0xe '\x0e' */ + 0, /* 0xf '\x0f' */ + 0, /* 0x10 '\x10' */ + 0, /* 0x11 '\x11' */ + 0, /* 0x12 '\x12' */ + 0, /* 0x13 '\x13' */ + 0, /* 0x14 '\x14' */ + 0, /* 0x15 '\x15' */ + 0, /* 0x16 '\x16' */ + 0, /* 0x17 '\x17' */ + 0, /* 0x18 '\x18' */ + 0, /* 0x19 '\x19' */ + 0, /* 0x1a '\x1a' */ + 0, /* 0x1b '\x1b' */ + 0, /* 0x1c '\x1c' */ + 0, /* 0x1d '\x1d' */ + 0, /* 0x1e '\x1e' */ + 0, /* 0x1f '\x1f' */ + PY_CTF_SPACE, /* 0x20 ' ' */ + 0, /* 0x21 '!' */ + 0, /* 0x22 '"' */ + 0, /* 0x23 '#' */ + 0, /* 0x24 '$' */ + 0, /* 0x25 '%' */ + 0, /* 0x26 '&' */ + 0, /* 0x27 "'" */ + 0, /* 0x28 '(' */ + 0, /* 0x29 ')' */ + 0, /* 0x2a '*' */ + 0, /* 0x2b '+' */ + 0, /* 0x2c ',' */ + 0, /* 0x2d '-' */ + 0, /* 0x2e '.' */ + 0, /* 0x2f '/' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x30 '0' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x31 '1' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x32 '2' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x33 '3' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x34 '4' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x35 '5' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x36 '6' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x37 '7' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x38 '8' */ + PY_CTF_DIGIT|PY_CTF_XDIGIT, /* 0x39 '9' */ + 0, /* 0x3a ':' */ + 0, /* 0x3b ';' */ + 0, /* 0x3c '<' */ + 0, /* 0x3d '=' */ + 0, /* 0x3e '>' */ + 0, /* 0x3f '?' */ + 0, /* 0x40 '@' */ + PY_CTF_UPPER|PY_CTF_XDIGIT, /* 0x41 'A' */ + PY_CTF_UPPER|PY_CTF_XDIGIT, /* 0x42 'B' */ + PY_CTF_UPPER|PY_CTF_XDIGIT, /* 0x43 'C' */ + PY_CTF_UPPER|PY_CTF_XDIGIT, /* 0x44 'D' */ + PY_CTF_UPPER|PY_CTF_XDIGIT, /* 0x45 'E' */ + PY_CTF_UPPER|PY_CTF_XDIGIT, /* 0x46 'F' */ + PY_CTF_UPPER, /* 0x47 'G' */ + PY_CTF_UPPER, /* 0x48 'H' */ + PY_CTF_UPPER, /* 0x49 'I' */ + PY_CTF_UPPER, /* 0x4a 'J' */ + PY_CTF_UPPER, /* 0x4b 'K' */ + PY_CTF_UPPER, /* 0x4c 'L' */ + PY_CTF_UPPER, /* 0x4d 'M' */ + PY_CTF_UPPER, /* 0x4e 'N' */ + PY_CTF_UPPER, /* 0x4f 'O' */ + PY_CTF_UPPER, /* 0x50 'P' */ + PY_CTF_UPPER, /* 0x51 'Q' */ + PY_CTF_UPPER, /* 0x52 'R' */ + PY_CTF_UPPER, /* 0x53 'S' */ + PY_CTF_UPPER, /* 0x54 'T' */ + PY_CTF_UPPER, /* 0x55 'U' */ + PY_CTF_UPPER, /* 0x56 'V' */ + PY_CTF_UPPER, /* 0x57 'W' */ + PY_CTF_UPPER, /* 0x58 'X' */ + PY_CTF_UPPER, /* 0x59 'Y' */ + PY_CTF_UPPER, /* 0x5a 'Z' */ + 0, /* 0x5b '[' */ + 0, /* 0x5c '\\' */ + 0, /* 0x5d ']' */ + 0, /* 0x5e '^' */ + 0, /* 0x5f '_' */ + 0, /* 0x60 '`' */ + PY_CTF_LOWER|PY_CTF_XDIGIT, /* 0x61 'a' */ + PY_CTF_LOWER|PY_CTF_XDIGIT, /* 0x62 'b' */ + PY_CTF_LOWER|PY_CTF_XDIGIT, /* 0x63 'c' */ + PY_CTF_LOWER|PY_CTF_XDIGIT, /* 0x64 'd' */ + PY_CTF_LOWER|PY_CTF_XDIGIT, /* 0x65 'e' */ + PY_CTF_LOWER|PY_CTF_XDIGIT, /* 0x66 'f' */ + PY_CTF_LOWER, /* 0x67 'g' */ + PY_CTF_LOWER, /* 0x68 'h' */ + PY_CTF_LOWER, /* 0x69 'i' */ + PY_CTF_LOWER, /* 0x6a 'j' */ + PY_CTF_LOWER, /* 0x6b 'k' */ + PY_CTF_LOWER, /* 0x6c 'l' */ + PY_CTF_LOWER, /* 0x6d 'm' */ + PY_CTF_LOWER, /* 0x6e 'n' */ + PY_CTF_LOWER, /* 0x6f 'o' */ + PY_CTF_LOWER, /* 0x70 'p' */ + PY_CTF_LOWER, /* 0x71 'q' */ + PY_CTF_LOWER, /* 0x72 'r' */ + PY_CTF_LOWER, /* 0x73 's' */ + PY_CTF_LOWER, /* 0x74 't' */ + PY_CTF_LOWER, /* 0x75 'u' */ + PY_CTF_LOWER, /* 0x76 'v' */ + PY_CTF_LOWER, /* 0x77 'w' */ + PY_CTF_LOWER, /* 0x78 'x' */ + PY_CTF_LOWER, /* 0x79 'y' */ + PY_CTF_LOWER, /* 0x7a 'z' */ + 0, /* 0x7b '{' */ + 0, /* 0x7c '|' */ + 0, /* 0x7d '}' */ + 0, /* 0x7e '~' */ + 0, /* 0x7f '\x7f' */ + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, +}; + + +const unsigned char _Py_ctype_tolower[256] = { + 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, + 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, + 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, + 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, + 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, + 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f, + 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, + 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, + 0x40, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, + 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, + 0x78, 0x79, 0x7a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, + 0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, + 0x68, 0x69, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, + 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, + 0x78, 0x79, 0x7a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, + 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, + 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f, + 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, + 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, + 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, + 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf, + 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, + 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, + 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, + 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, + 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, + 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf, + 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, + 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, + 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, + 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff, +}; + +const unsigned char _Py_ctype_toupper[256] = { + 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, + 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, + 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, + 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, + 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, + 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f, + 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, + 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, + 0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, + 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, + 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, + 0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, + 0x60, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, + 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, + 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, + 0x58, 0x59, 0x5a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, + 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, + 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f, + 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, + 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, + 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, + 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf, + 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, + 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, + 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, + 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, + 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, + 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf, + 0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, + 0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, + 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, + 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff, +}; + diff --git a/Python/pystrtod.c b/Python/pystrtod.c index ea7f1f5..e677305 100644 --- a/Python/pystrtod.c +++ b/Python/pystrtod.c @@ -3,12 +3,6 @@ #include #include -/* ascii character tests (as opposed to locale tests) */ -#define ISSPACE(c) ((c) == ' ' || (c) == '\f' || (c) == '\n' || \ - (c) == '\r' || (c) == '\t' || (c) == '\v') -#define ISDIGIT(c) ((c) >= '0' && (c) <= '9') - - /** * PyOS_ascii_strtod: * @nptr: the string to convert to a numeric value. @@ -81,7 +75,7 @@ PyOS_ascii_strtod(const char *nptr, char **endptr) p = nptr; /* Skip leading space */ - while (ISSPACE(*p)) + while (Py_ISSPACE(*p)) p++; /* Process leading sign, if present */ @@ -124,7 +118,7 @@ PyOS_ascii_strtod(const char *nptr, char **endptr) goto invalid_string; /* Check that what's left begins with a digit or decimal point */ - if (!ISDIGIT(*p) && *p != '.') + if (!Py_ISDIGIT(*p) && *p != '.') goto invalid_string; digits_pos = p; @@ -135,7 +129,7 @@ PyOS_ascii_strtod(const char *nptr, char **endptr) swapped for the current locale's decimal point before we call strtod. On the other hand, if we find the current locale's decimal point then the input is invalid. */ - while (ISDIGIT(*p)) + while (Py_ISDIGIT(*p)) p++; if (*p == '.') @@ -143,14 +137,14 @@ PyOS_ascii_strtod(const char *nptr, char **endptr) decimal_point_pos = p++; /* locate end of number */ - while (ISDIGIT(*p)) + while (Py_ISDIGIT(*p)) p++; if (*p == 'e' || *p == 'E') p++; if (*p == '+' || *p == '-') p++; - while (ISDIGIT(*p)) + while (Py_ISDIGIT(*p)) p++; end = p; } @@ -244,7 +238,7 @@ change_decimal_from_locale_to_dot(char* buffer) if (*buffer == '+' || *buffer == '-') buffer++; - while (isdigit(Py_CHARMASK(*buffer))) + while (Py_ISDIGIT(*buffer)) buffer++; if (strncmp(buffer, decimal_point, decimal_point_len) == 0) { *buffer = '.'; @@ -306,7 +300,7 @@ ensure_minimum_exponent_length(char* buffer, size_t buf_size) /* Find the end of the exponent, keeping track of leading zeros. */ - while (*p && isdigit(Py_CHARMASK(*p))) { + while (*p && Py_ISDIGIT(*p)) { if (in_leading_zeros && *p == '0') ++leading_zero_cnt; if (*p != '0') @@ -369,11 +363,11 @@ ensure_decimal_point(char* buffer, size_t buf_size) /* Skip leading sign, if present. I think this could only ever be '-', but it can't hurt to check for both. */ ++p; - while (*p && isdigit(Py_CHARMASK(*p))) + while (*p && Py_ISDIGIT(*p)) ++p; if (*p == '.') { - if (isdigit(Py_CHARMASK(*(p+1)))) { + if (Py_ISDIGIT(*(p+1))) { /* Nothing to do, we already have a decimal point and a digit after it */ } -- cgit v0.12