ruby-changes:7843
From: matz <ko1@a...>
Date: Mon, 15 Sep 2008 23:40:21 +0900 (JST)
Subject: [ruby-changes:7843] Ruby:r19364 (trunk): * string.c (rb_str_squeeze_bang): specialized for 7bit characters in
matz 2008-09-15 23:40:00 +0900 (Mon, 15 Sep 2008) New Revision: 19364 http://svn.ruby-lang.org/cgi-bin/viewvc.cgi?view=rev&revision=19364 Log: * string.c (rb_str_squeeze_bang): specialized for 7bit characters in ASCII compatible strings. * string.c (rb_str_count): ditto. * string.c (tr_trans): preserve 7bit/valid coderange flag. * string.c (rb_str_squeeze_bang): preserve previous coderange value. * string.c (rb_str_lstrip_bang): ditto. * string.c (rb_str_rstrip_bang): ditto. * encoding.c (rb_default_external_encoding): preserve default_external_encoding in a static variable. * string.c (single_byte_optimizable): check coderange first, to reduce number of calling rb_enc_from_index(). Modified files: trunk/ChangeLog trunk/encoding.c trunk/string.c Index: encoding.c =================================================================== --- encoding.c (revision 19363) +++ encoding.c (revision 19364) @@ -997,11 +997,15 @@ } static int default_external_index; +static rb_encoding *default_external; rb_encoding * rb_default_external_encoding(void) { - return rb_enc_from_index(default_external_index); + if (!default_external) { + default_external = rb_enc_from_index(default_external_index); + } + return default_external; } VALUE @@ -1028,6 +1032,7 @@ rb_enc_set_default_external(VALUE encoding) { default_external_index = rb_enc_to_index(rb_to_encoding(encoding)); + default_external = 0; } /* Index: ChangeLog =================================================================== --- ChangeLog (revision 19363) +++ ChangeLog (revision 19364) @@ -9,6 +9,27 @@ * io.c (rb_scan_open_args): call FilePathValue before encoding conversion. +Mon Sep 15 22:11:07 2008 Yukihiro Matsumoto <matz@r...> + + * string.c (rb_str_squeeze_bang): specialized for 7bit characters in + ASCII compatible strings. + + * string.c (rb_str_count): ditto. + + * string.c (tr_trans): preserve 7bit/valid coderange flag. + + * string.c (rb_str_squeeze_bang): preserve previous coderange value. + + * string.c (rb_str_lstrip_bang): ditto. + + * string.c (rb_str_rstrip_bang): ditto. + + * encoding.c (rb_default_external_encoding): preserve + default_external_encoding in a static variable. + + * string.c (single_byte_optimizable): check coderange first, to + reduce number of calling rb_enc_from_index(). + Mon Sep 15 20:57:00 2008 Yuki Sonoda (Yugui) <yugui@y...> * lib/matrix.rb (Matrix#eql?): fixed [ruby-dev:36298]. Index: string.c =================================================================== --- string.c (revision 19363) +++ string.c (revision 19364) @@ -115,15 +115,16 @@ static int single_byte_optimizable(VALUE str) { - rb_encoding *enc = STR_ENC_GET(str); + rb_encoding *enc; - if (rb_enc_mbmaxlen(enc) == 1) - return 1; - /* Conservative. It may be ENC_CODERANGE_UNKNOWN. */ if (ENC_CODERANGE(str) == ENC_CODERANGE_7BIT) return 1; + enc = STR_ENC_GET(str); + if (rb_enc_mbmaxlen(enc) == 1) + return 1; + /* Conservative. Possibly single byte. * "\xa1" in Shift_JIS for example. */ return 0; @@ -4325,6 +4326,7 @@ char *s, *send; VALUE hash = 0; int singlebyte = single_byte_optimizable(str); + int cr, cr1, cr2; StringValue(src); StringValue(repl); @@ -4333,6 +4335,12 @@ return rb_str_delete_bang(1, &src, str); } + cr = ENC_CODERANGE(str); + cr1 = ENC_CODERANGE(src); + cr2 = ENC_CODERANGE(repl); + if (cr != cr1 || cr1 != cr2) { + cr = ENC_CODERANGE_UNKNOWN; + } e1 = rb_enc_check(str, src); e2 = rb_enc_check(str, repl); if (e1 == e2) { @@ -4517,6 +4525,7 @@ RSTRING(str)->as.heap.aux.capa = max; } + ENC_CODERANGE_SET(str, cr); if (modify) { rb_enc_associate(str, enc); return str; @@ -4738,6 +4747,8 @@ char *s, *send, *t; int save, modify = 0; int i; + int ascompat, singlebyte = single_byte_optimizable(str); + int cr = ENC_CODERANGE(str); if (argc == 0) { enc = STR_ENC_GET(str); @@ -4757,29 +4768,53 @@ if (!s || RSTRING_LEN(str) == 0) return Qnil; send = RSTRING_END(str); save = -1; - while (s < send) { - unsigned int c = rb_enc_codepoint(s, send, enc); - int clen = rb_enc_codelen(c, enc); + ascompat = rb_enc_asciicompat(enc); - if (c != save || (argc > 0 && !tr_find(c, squeez, del, nodel))) { - if (t != s) rb_enc_mbcput(c, t, enc); - save = c; - t += clen; + if (singlebyte) { + while (s < send) { + unsigned int c = *(unsigned char*)s++; + if (c != save || (argc > 0 && !squeez[c])) { + *t++ = save = c; + } } - s += clen; + } else { + while (s < send) { + unsigned int c; + int clen; + + if (ascompat && (c = *(unsigned char*)s) < 0x80) { + if (c != save || (argc > 0 && !squeez[c])) { + *t++ = save = c; + } + s++; + } + else { + c = rb_enc_codepoint(s, send, enc); + clen = rb_enc_codelen(c, enc); + + if (c != save || (argc > 0 && !tr_find(c, squeez, del, nodel))) { + if (t != s) rb_enc_mbcput(c, t, enc); + save = c; + t += clen; + } + s += clen; + } + } } + *t = '\0'; if (t - RSTRING_PTR(str) != RSTRING_LEN(str)) { STR_SET_LEN(str, t - RSTRING_PTR(str)); modify = 1; } + ENC_CODERANGE_SET(str, cr); if (modify) return str; return Qnil; } -/* +/* * call-seq: * str.squeeze([other_str]*) => new_str * @@ -4864,6 +4899,7 @@ VALUE del = 0, nodel = 0; char *s, *send; int i; + int ascompat; if (argc < 1) { rb_raise(rb_eArgError, "wrong number of arguments"); @@ -4873,22 +4909,36 @@ StringValue(s); enc = rb_enc_check(str, s); - tr_setup_table(s, table,i==0, &del, &nodel, enc); + tr_setup_table(s, table, i==0, &del, &nodel, enc); } s = RSTRING_PTR(str); if (!s || RSTRING_LEN(str) == 0) return INT2FIX(0); send = RSTRING_END(str); i = 0; + ascompat = rb_enc_asciicompat(enc); + while (s < send) { - unsigned int c = rb_enc_codepoint(s, send, enc); - int clen = rb_enc_codelen(c, enc); + unsigned int c; + int clen; - if (tr_find(c, table, del, nodel)) { - i++; + if (ascompat && (c = *(unsigned char*)s) < 0x80) { + clen = 1; + if (table[c]) { + i++; + } + s++; } - s += clen; + else { + c = rb_enc_codepoint(s, send, enc); + clen = rb_enc_codelen(c, enc); + if (tr_find(c, table, del, nodel)) { + i++; + } + s += clen; + } } + return INT2NUM(i); } @@ -5549,8 +5599,10 @@ { rb_encoding *enc; char *s, *t, *e; + int cr = ENC_CODERANGE(str); rb_str_modify(str); + ENC_CODERANGE_SET(str, cr); enc = STR_ENC_GET(str); s = RSTRING_PTR(str); if (!s || RSTRING_LEN(str) == 0) return Qnil; @@ -5612,27 +5664,38 @@ rb_encoding *enc; char *s, *t, *e; int space_seen = Qfalse; + int cr = ENC_CODERANGE(str); rb_str_modify(str); + ENC_CODERANGE_SET(str, cr); enc = STR_ENC_GET(str); s = RSTRING_PTR(str); if (!s || RSTRING_LEN(str) == 0) return Qnil; t = e = RSTRING_END(str); - while (s < e) { - unsigned int cc = rb_enc_codepoint(s, e, enc); - if (!cc || rb_enc_isspace(cc, enc)) { - if (!space_seen) t = s; - space_seen = Qtrue; + if (single_byte_optimizable(str)) { + /* remove trailing '\0's */ + while (s < t && t[-1] == '\0') t--; + + /* remove trailing spaces */ + while (s < t && rb_enc_isspace(*(t-1), enc)) t--; + } else { + while (s < e) { + unsigned int cc = rb_enc_codepoint(s, e, enc); + + if (!cc || rb_enc_isspace(cc, enc)) { + if (!space_seen) t = s; + space_seen = Qtrue; + } + else { + space_seen = Qfalse; + } + s += rb_enc_codelen(cc, enc); } - else { - space_seen = Qfalse; - } - s += rb_enc_codelen(cc, enc); + if (!space_seen) t = s; } - if (!space_seen) t = s; + if (t < e) { - rb_str_modify(str); STR_SET_LEN(str, t-RSTRING_PTR(str)); RSTRING_PTR(str)[RSTRING_LEN(str)] = '\0'; return str; -- ML: ruby-changes@q... Info: http://www.atdot.net/~ko1/quickml/