ruby-changes:36675
From: nobu <ko1@a...>
Date: Wed, 10 Dec 2014 16:48:35 +0900 (JST)
Subject: [ruby-changes:36675] nobu:r48756 (trunk): string.c: chompped_length chomp_rs
nobu 2014-12-10 16:48:17 +0900 (Wed, 10 Dec 2014) New Revision: 48756 http://svn.ruby-lang.org/cgi-bin/viewvc.cgi?view=revision&revision=48756 Log: string.c: chompped_length chomp_rs * string.c (chompped_length, chomp_rs): extract from rb_str_chomp_bang to share with rb_str_chomp. Modified files: trunk/string.c Index: string.c =================================================================== --- string.c (revision 48755) +++ string.c (revision 48756) @@ -7044,110 +7044,142 @@ rb_str_chop(VALUE str) https://github.com/ruby/ruby/blob/trunk/string.c#L7044 } -/* - * call-seq: - * str.chomp!(separator=$/) -> str or nil - * - * Modifies <i>str</i> in place as described for <code>String#chomp</code>, - * returning <i>str</i>, or <code>nil</code> if no modifications were made. - */ - -static VALUE -rb_str_chomp_bang(int argc, VALUE *argv, VALUE str) +static long +chompped_length(VALUE str, VALUE rs) { rb_encoding *enc; - VALUE rs; int newline; - char *p, *pp, *e; - long len, rslen; + char *pp, *e, *rsptr; + long rslen; + char *const p = RSTRING_PTR(str); + long len = RSTRING_LEN(str); - str_modify_keep_cr(str); - len = RSTRING_LEN(str); - if (len == 0) return Qnil; - p = RSTRING_PTR(str); + if (len == 0) return 0; e = p + len; - if (argc == 0) { - rs = rb_rs; - if (rs == rb_default_rs) { - smart_chomp: - enc = rb_enc_get(str); - if (rb_enc_mbminlen(enc) > 1) { - pp = rb_enc_left_char_head(p, e-rb_enc_mbminlen(enc), e, enc); - if (rb_enc_is_newline(pp, e, enc)) { + enc = rb_enc_get(str); + if (rs == rb_default_rs) { + smart_chomp: + if (rb_enc_mbminlen(enc) > 1) { + pp = rb_enc_left_char_head(p, e-rb_enc_mbminlen(enc), e, enc); + if (rb_enc_is_newline(pp, e, enc)) { + e = pp; + } + pp = e - rb_enc_mbminlen(enc); + if (pp >= p) { + pp = rb_enc_left_char_head(p, pp, e, enc); + if (rb_enc_ascget(pp, e, 0, enc) == '\r') { e = pp; } - pp = e - rb_enc_mbminlen(enc); + } + } + else { + switch (*(e-1)) { /* not e[-1] to get rid of VC bug */ + case '\n': + if (--e > p && *(e-1) == '\r') { + --e; + } + break; + case '\r': + --e; + break; + } + } + return e - p; + } + + RSTRING_GETMEM(rs, rsptr, rslen); + if (rslen == 0) { + if (rb_enc_mbminlen(enc) > 1) { + while (e > p) { + pp = rb_enc_left_char_head(p, e-rb_enc_mbminlen(enc), e, enc); + if (!rb_enc_is_newline(pp, e, enc)) break; + e = pp; + pp -= rb_enc_mbminlen(enc); if (pp >= p) { pp = rb_enc_left_char_head(p, pp, e, enc); if (rb_enc_ascget(pp, e, 0, enc) == '\r') { e = pp; } } - if (e == RSTRING_END(str)) { - return Qnil; - } - len = e - RSTRING_PTR(str); - STR_SET_LEN(str, len); } - else { - if (RSTRING_PTR(str)[len-1] == '\n') { - STR_DEC_LEN(str); - if (RSTRING_LEN(str) > 0 && - RSTRING_PTR(str)[RSTRING_LEN(str)-1] == '\r') { - STR_DEC_LEN(str); - } - } - else if (RSTRING_PTR(str)[len-1] == '\r') { - STR_DEC_LEN(str); - } - else { - return Qnil; - } + } + else { + while (e > p && *(e-1) == '\n') { + --e; + if (e > p && *(e-1) == '\r') + --e; } - RSTRING_PTR(str)[RSTRING_LEN(str)] = '\0'; - return str; } + return e - p; } - else { - rb_scan_args(argc, argv, "01", &rs); + if (rslen > len) return len; + + enc = rb_enc_check(str, rs); + if (is_broken_string(rs)) { + return len; } - if (NIL_P(rs)) return Qnil; - StringValue(rs); - rslen = RSTRING_LEN(rs); - if (rslen == 0) { - while (len>0 && p[len-1] == '\n') { - len--; - if (len>0 && p[len-1] == '\r') - len--; + newline = rsptr[rslen-1]; + if (rslen == rb_enc_mbminlen(enc)) { + if (rslen == 1) { + if (newline == '\n') + goto smart_chomp; } - if (len < RSTRING_LEN(str)) { - STR_SET_LEN(str, len); - RSTRING_PTR(str)[len] = '\0'; - return str; + else { + if (rb_enc_is_newline(rsptr, rsptr+rslen, enc)) + goto smart_chomp; } - return Qnil; } - if (rslen > len) return Qnil; - newline = RSTRING_PTR(rs)[rslen-1]; - if (rslen == 1 && newline == '\n') - goto smart_chomp; - enc = rb_enc_check(str, rs); - if (is_broken_string(rs)) { - return Qnil; - } pp = e - rslen; if (p[len-1] == newline && (rslen <= 1 || - memcmp(RSTRING_PTR(rs), pp, rslen) == 0)) { - if (rb_enc_left_char_head(p, pp, e, enc) != pp) - return Qnil; - if (ENC_CODERANGE(str) != ENC_CODERANGE_7BIT) { - ENC_CODERANGE_CLEAR(str); - } - STR_SET_LEN(str, RSTRING_LEN(str) - rslen); - RSTRING_PTR(str)[RSTRING_LEN(str)] = '\0'; - return str; + memcmp(rsptr, pp, rslen) == 0)) { + if (rb_enc_left_char_head(p, pp, e, enc) == pp) + return len - rslen; + RB_GC_GUARD(rs); + } + return len; +} + +static VALUE +chomp_rs(int argc, const VALUE *argv) +{ + rb_check_arity(argc, 0, 1); + if (argc > 0) { + VALUE rs = argv[0]; + if (!NIL_P(rs)) StringValue(rs); + return rs; + } + else { + return rb_rs; + } +} + +/* + * call-seq: + * str.chomp!(separator=$/) -> str or nil + * + * Modifies <i>str</i> in place as described for <code>String#chomp</code>, + * returning <i>str</i>, or <code>nil</code> if no modifications were made. + */ + +static VALUE +rb_str_chomp_bang(int argc, VALUE *argv, VALUE str) +{ + VALUE rs; + long olen; + str_modify_keep_cr(str); + if ((olen = RSTRING_LEN(str)) > 0 && !NIL_P(rs = chomp_rs(argc, argv))) { + long len; + len = chompped_length(str, rs); + if (len < olen) { + STR_SET_LEN(str, len); + RSTRING_PTR(str)[len] = '\0'; + if (ENC_CODERANGE(str) != ENC_CODERANGE_7BIT) { + ENC_CODERANGE_CLEAR(str); + } + return str; + } } return Qnil; } @@ -7178,9 +7210,9 @@ rb_str_chomp_bang(int argc, VALUE *argv, https://github.com/ruby/ruby/blob/trunk/string.c#L7210 static VALUE rb_str_chomp(int argc, VALUE *argv, VALUE str) { - str = rb_str_dup(str); - rb_str_chomp_bang(argc, argv, str); - return str; + VALUE rs = chomp_rs(argc, argv); + if (NIL_P(rs)) return rb_str_dup(str); + return rb_str_subseq(str, 0, chompped_length(str, rs)); } static long -- ML: ruby-changes@q... Info: http://www.atdot.net/~ko1/quickml/