diff options
Diffstat (limited to 'string.c')
| -rw-r--r-- | string.c | 38 |
1 files changed, 20 insertions, 18 deletions
@@ -29,14 +29,6 @@ #include <unistd.h> #endif -#define ENC_REPLICATE(name, orig) -#define ENC_ALIAS(name, orig) -#define ENC_DUMMY(name) -#define ENC_DEFINE(name) -#define ENC_SET_BASE(name, orig) -#define ENC_DUMMY_UNICODE(name) -#include <encdb.h> - #define STRING_ENUMERATORS_WANTARRAY 0 /* next major */ #undef rb_str_new_cstr @@ -4479,14 +4471,16 @@ VALUE rb_str_inspect(VALUE str) { rb_encoding *enc = STR_ENC_GET(str); - int encidx = rb_enc_to_index(enc); const char *p, *pend, *prev; char buf[CHAR_ESC_LEN + 1]; VALUE result = rb_str_buf_new(0); rb_encoding *resenc = rb_default_internal_encoding(); int unicode_p = rb_enc_unicode_p(enc); int asciicompat = rb_enc_asciicompat(enc); + static rb_encoding *utf16, *utf32; + if (!utf16) utf16 = rb_enc_find("UTF-16"); + if (!utf32) utf32 = rb_enc_find("UTF-32"); if (resenc == NULL) resenc = rb_default_external_encoding(); if (!rb_enc_asciicompat(resenc)) resenc = rb_usascii_encoding(); rb_enc_associate(result, resenc); @@ -4494,7 +4488,7 @@ rb_str_inspect(VALUE str) p = RSTRING_PTR(str); pend = RSTRING_END(str); prev = p; - if (encidx == ENCIDX_UTF_16) { + if (enc == utf16) { const unsigned char *q = (const unsigned char *)p; if (q[0] == 0xFE && q[1] == 0xFF) enc = rb_enc_find("UTF-16BE"); @@ -4503,7 +4497,7 @@ rb_str_inspect(VALUE str) else unicode_p = 0; } - else if (encidx == ENCIDX_UTF_32) { + else if (enc == utf32) { const unsigned char *q = (const unsigned char *)p; if (q[0] == 0 && q[1] == 0 && q[2] == 0xFE && q[3] == 0xFF) enc = rb_enc_find("UTF-32BE"); @@ -7750,7 +7744,6 @@ str_scrub0(int argc, VALUE *argv, VALUE str) { int cr = ENC_CODERANGE(str); rb_encoding *enc; - int encidx; VALUE repl; if (cr == ENC_CODERANGE_7BIT || cr == ENC_CODERANGE_VALID) @@ -7765,7 +7758,6 @@ str_scrub0(int argc, VALUE *argv, VALUE str) if (rb_enc_dummy_p(enc)) { return Qnil; } - encidx = rb_enc_to_index(enc); #define DEFAULT_REPLACE_CHAR(str) do { \ static const char replace[sizeof(str)-1] = str; \ @@ -7790,7 +7782,7 @@ str_scrub0(int argc, VALUE *argv, VALUE str) replen = RSTRING_LEN(repl); rep7bit_p = (ENC_CODERANGE(repl) == ENC_CODERANGE_7BIT); } - else if (encidx == rb_utf8_encindex()) { + else if (enc == rb_utf8_encoding()) { DEFAULT_REPLACE_CHAR("\xEF\xBF\xBD"); rep7bit_p = FALSE; } @@ -7896,20 +7888,30 @@ str_scrub0(int argc, VALUE *argv, VALUE str) const char *rep; long replen; long mbminlen = rb_enc_mbminlen(enc); + static rb_encoding *utf16be; + static rb_encoding *utf16le; + static rb_encoding *utf32be; + static rb_encoding *utf32le; + if (!utf16be) { + utf16be = rb_enc_find("UTF-16BE"); + utf16le = rb_enc_find("UTF-16LE"); + utf32be = rb_enc_find("UTF-32BE"); + utf32le = rb_enc_find("UTF-32LE"); + } if (!NIL_P(repl)) { rep = RSTRING_PTR(repl); replen = RSTRING_LEN(repl); } - else if (encidx == ENCIDX_UTF_16BE) { + else if (enc == utf16be) { DEFAULT_REPLACE_CHAR("\xFF\xFD"); } - else if (encidx == ENCIDX_UTF_16LE) { + else if (enc == utf16le) { DEFAULT_REPLACE_CHAR("\xFD\xFF"); } - else if (encidx == ENCIDX_UTF_32BE) { + else if (enc == utf32be) { DEFAULT_REPLACE_CHAR("\x00\x00\xFF\xFD"); } - else if (encidx == ENCIDX_UTF_32LE) { + else if (enc == utf32le) { DEFAULT_REPLACE_CHAR("\xFD\xFF\x00\x00"); } else { |