diff options
| author | Yukihiro Matsumoto <matz@ruby-lang.org> | 1995-03-15 14:59:18 +0900 |
|---|---|---|
| committer | Takashi Kokubun <takashikkbn@gmail.com> | 2019-08-17 22:09:31 +0900 |
| commit | d349889e770a2078c247d9d28070e86a54b856f4 (patch) | |
| tree | 8bdbf87f3ff78ef237c4ff8efa061fbb8dc5a226 | |
| parent | 881c5a9c320c637ee0f6526b40cf70c1379ab656 (diff) | |
| download | ruby-d349889e770a2078c247d9d28070e86a54b856f4.tar.gz | |
version 0.69v0_69
https://cache.ruby-lang.org/pub/ruby/1.0/ruby-0.69.tar.gz
Wed Mar 15 14:59:18 1995 Yukihiro Matsumoto (matz@ix-02)
* version 0.69
* eval.c(method_missing): unknownから名称変更．
* eval.c(single_method_added): 特異メソッドが定義された時に呼ばれ
るメソッド．hookとして使える．実際に定義される直前に呼ばれる．
Tue Mar 14 14:46:44 1995 Yukihiro Matsumoto (matz@ix-02)
* ruby.c(proc_options): 引数の解析を自分でやることにより引数指定の
方法がperlに近付いた．getopt_longはもう使わない．
* dir.c(glob): `{}'のネストを許すようにした．
Mon Mar 13 17:56:25 1995 Yukihiro Matsumoto (matz@ix-02)
* glob.c: Glob(ワイルドカードオブジェクト)はなくなった．ワイルドカー
ドの展開はDir.glob(文字列)を使う．ワイルドカードのマッチは正規表
現で代用．
Fri Mar 10 18:35:46 1995 Yukihiro Matsumoto (matz@ix-02)
* eval.c: Mathのようなモジュールは自分自身でextendする．
* eval.c: クラスやモジュールを定義した既に同名のものがあれば追加定
義となるように．ただし．superクラスの違いなどはチェックする．
* regex.c: debug.
* math.c: 定数PIとEを定義．
Thu Mar 9 21:35:12 1995 Yukihiro Matsumoto (matz@ix-02)
* regex.c: EUC,SJISモードでは0x80以上の8進，16進リテラルを禁止．
* regex.c: クラス内でも数値リテラル・文字クラスが使えるようした．
Wed Mar 8 17:39:05 1995 Yukihiro Matsumoto (matz@ix-02)
* regex.c: \200など括弧の数以上の表現は8進リテラルと解釈する．ただ
し，\1から\9までは例外．
* regex.c: \9以上のリファレンスも有効にした．
Tue Mar 7 14:26:01 1995 Yukihiro Matsumoto (matz@ix-02)
* eval.c(public/private): スコープ制御メソッドの名称変更．静的なア
クセスも出来るようにしてみたが，不採用．
Mon Mar 6 19:34:32 1995 Yukihiro Matsumoto (matz@ix-02)
* eval.c(inlcude): メソッド化．動的にモジュールをインクルードでき
るように．さらに任意のオブジェクトにもモジュールをインクルードで
きるメソッド `extend'も用意した．
* parse.y: 文法からincludeを削除．メソッド化．
Tue Feb 28 15:35:10 1995 Yukihiro Matsumoto (matz@ix-02)
* parse.y: 配列，連想配列の最後に`,'をおけるように．
| -rw-r--r-- | ChangeLog | 63 | ||||
| -rw-r--r-- | MANIFEST | 6 | ||||
| -rw-r--r-- | Makefile.in | 24 | ||||
| -rw-r--r-- | class.c | 6 | ||||
| -rw-r--r-- | dict.c | 3 | ||||
| -rw-r--r-- | dir.c | 105 | ||||
| -rw-r--r-- | env.h | 3 | ||||
| -rw-r--r-- | error.c | 9 | ||||
| -rw-r--r-- | etc.c | 15 | ||||
| -rw-r--r-- | eval.c | 167 | ||||
| -rw-r--r-- | file.c | 56 | ||||
| -rw-r--r-- | getopt.c | 660 | ||||
| -rw-r--r-- | getopt.h | 128 | ||||
| -rw-r--r-- | getopt1.c | 162 | ||||
| -rw-r--r-- | glob.c | 678 | ||||
| -rw-r--r-- | gnuglob.c | 580 | ||||
| -rw-r--r-- | ident.h | 11 | ||||
| -rw-r--r-- | inits.c | 1 | ||||
| -rw-r--r-- | io.c | 3 | ||||
| -rw-r--r-- | math.c | 24 | ||||
| -rw-r--r-- | node.h | 9 | ||||
| -rw-r--r-- | object.c | 26 | ||||
| -rw-r--r-- | parse.y | 290 | ||||
| -rw-r--r-- | process.c | 29 | ||||
| -rw-r--r-- | re.c | 163 | ||||
| -rw-r--r-- | re.h | 17 | ||||
| -rw-r--r-- | regex.c | 1329 | ||||
| -rw-r--r-- | regex.h | 52 | ||||
| -rw-r--r-- | ruby.1 | 49 | ||||
| -rw-r--r-- | ruby.c | 195 | ||||
| -rw-r--r-- | ruby.h | 2 | ||||
| -rw-r--r-- | sample/Artistic | 0 | ||||
| -rw-r--r-- | sample/evaldef.rb | 4 | ||||
| -rw-r--r-- | sample/export.rb | 32 | ||||
| -rw-r--r-- | sample/exyacc.rb | 2 | ||||
| -rw-r--r-- | sample/from.rb | 4 | ||||
| -rw-r--r-- | sample/opt_x.rb | 0 | ||||
| -rw-r--r-- | sample/ruby-mode.el | 120 | ||||
| -rwxr-xr-x | sample/uumerge.rb | 20 | ||||
| -rw-r--r-- | spec | 118 | ||||
| -rw-r--r-- | string.c | 25 | ||||
| -rw-r--r-- | util.c | 53 | ||||
| -rw-r--r-- | util.h | 16 | ||||
| -rw-r--r-- | variable.c | 11 | ||||
| -rw-r--r-- | version.h | 4 |
45 files changed, 2081 insertions, 3193 deletions
@@ -1,3 +1,66 @@ +Wed Mar 15 14:59:18 1995 Yukihiro Matsumoto (matz@ix-02) + + * version 0.69 + + * eval.c(method_missing): unknown̾ѹ + + * eval.c(single_method_added): ðۥåɤ줿˸ƤФ + åɡhookȤƻȤ롥ºݤľ˸ƤФ롥 + +Tue Mar 14 14:46:44 1995 Yukihiro Matsumoto (matz@ix-02) + + * ruby.c(proc_options): βϤʬǤ뤳Ȥˤ + ˡperl˶դgetopt_longϤ⤦Ȥʤ + + * dir.c(glob): `{}'ΥͥȤ褦ˤ + +Mon Mar 13 17:56:25 1995 Yukihiro Matsumoto (matz@ix-02) + + * glob.c: Glob(磻ɥɥ֥)Ϥʤʤä磻ɥ + ɤŸDir.glob(ʸ)Ȥ磻ɥɤΥޥåɽ + ѡ + +Fri Mar 10 18:35:46 1995 Yukihiro Matsumoto (matz@ix-02) + + * eval.c: MathΤ褦ʥ⥸塼ϼʬȤextend롥 + + * eval.c: 饹⥸塼Ʊ̾ΤΤɲ + Ȥʤ褦ˡsuper饹ΰ㤤ʤɤϥå롥 + + * regex.c: debug. + + * math.c: PIE + +Thu Mar 9 21:35:12 1995 Yukihiro Matsumoto (matz@ix-02) + + * regex.c: EUC,SJIS⡼ɤǤ0x80ʾ8ʡ16ʥƥػߡ + + * regex.c: 饹Ǥͥƥ롦ʸ饹Ȥ褦 + +Wed Mar 8 17:39:05 1995 Yukihiro Matsumoto (matz@ix-02) + + * regex.c: \200ʤɳ̤οʾɽ8ʥƥȲ᤹롥 + \1\9ޤǤ㳰 + + * regex.c: \9ʾΥեͭˤ + +Tue Mar 7 14:26:01 1995 Yukihiro Matsumoto (matz@ix-02) + + * eval.c(public/private): åɤ̾ѹŪʥ + 褦ˤƤߤԺѡ + +Mon Mar 6 19:34:32 1995 Yukihiro Matsumoto (matz@ix-02) + + * eval.c(inlcude): åɲưŪ˥⥸塼롼ɤǤ + 褦ˡǤդΥ֥Ȥˤ⥸塼롼ɤ + å `extend'Ѱդ + + * parse.y: ʸˡincludeåɲ + +Tue Feb 28 15:35:10 1995 Yukihiro Matsumoto (matz@ix-02) + + * parse.y: ϢκǸ`,'褦ˡ + Fri Feb 24 13:15:43 1995 Yukihiro Matsumoto (matz@ix-02) * version 0.68 @@ -27,11 +27,7 @@ file.c fnmatch.c fnmatch.h gc.c -getopt.c -getopt.h -getopt1.c glob.c -gnuglob.c ident.h inits.c io.c @@ -62,6 +58,8 @@ st.h string.c struct.c time.c +util.h +util.c variable.c version.c version.h diff --git a/Makefile.in b/Makefile.in index b0c6eb5ea9..8f818473ad 100644 --- a/Makefile.in +++ b/Makefile.in @@ -28,7 +28,6 @@ bindir = $(exec_prefix)/bin HDRS = defines.h \ dln.h \ - getopt.h \ ident.h \ io.h \ node.h \ @@ -54,10 +53,7 @@ OBJS = array.o \ file.o \ fnmatch.o \ gc.o \ - getopt.o \ - getopt1.o \ glob.o \ - gnuglob.o \ inits.o \ io.o \ main.o \ @@ -79,6 +75,7 @@ OBJS = array.o \ string.o \ struct.o \ time.o \ + util.o \ variable.o \ version.o \ $(MISSING) @@ -142,19 +139,19 @@ array.o: array.c ruby.h config.h defines.h bignum.o: bignum.c ruby.h config.h defines.h class.o: class.c ruby.h config.h defines.h env.h node.h st.h compar.o: compar.c ruby.h config.h defines.h +cons.o: cons.c ruby.h config.h defines.h dbm.o: dbm.c ruby.h config.h defines.h dict.o: dict.c ruby.h config.h defines.h st.h dir.o: dir.c ruby.h config.h defines.h -dln.o: dln.c defines.h dln.h +dln.o: dln.c config.h defines.h dln.h st.h enum.o: enum.c ruby.h config.h defines.h error.o: error.c ruby.h config.h defines.h env.h etc.o: etc.c ruby.h config.h defines.h eval.o: eval.c ruby.h config.h defines.h ident.h env.h node.h st.h file.o: file.c ruby.h config.h defines.h io.h -fnmatch.o: fnmatch.c fnmatch.h -gc.o: gc.c ruby.h config.h defines.h env.h st.h -glob.o: glob.c ruby.h config.h defines.h fnmatch.h -gnuglob.o: gnuglob.c fnmatch.h +fnmatch.o: fnmatch.c config.h fnmatch.h +gc.o: gc.c ruby.h config.h defines.h env.h st.h node.h +glob.o: glob.c config.h fnmatch.h inits.o: inits.c ruby.h config.h defines.h io.o: io.c ruby.h config.h defines.h io.h main.o: main.c @@ -166,13 +163,16 @@ process.o: process.c ruby.h config.h defines.h st.h random.o: random.c ruby.h config.h defines.h range.o: range.c ruby.h config.h defines.h re.o: re.c ruby.h config.h defines.h re.h regex.h -regex.o: regex.c defines.h regex.h -ruby.o: ruby.c ruby.h config.h defines.h re.h regex.h getopt.h +regex.o: regex.c config.h defines.h regex.h util.h +ruby.o: ruby.c ruby.h config.h defines.h re.h regex.h +signal.o: signal.c ruby.h config.h defines.h socket.o: socket.c ruby.h config.h defines.h io.h sprintf.o: sprintf.c ruby.h config.h defines.h -st.o: st.c st.h +st.o: st.c config.h st.h string.o: string.c ruby.h config.h defines.h re.h regex.h struct.o: struct.c ruby.h config.h defines.h env.h +tclglob.o: tclglob.c time.o: time.c ruby.h config.h defines.h +util.o: util.c defines.h config.h util.h variable.o: variable.c ruby.h config.h defines.h env.h ident.h st.h version.o: version.c ruby.h config.h defines.h version.h @@ -230,7 +230,7 @@ rb_define_method(class, name, func, argc) VALUE (*func)(); int argc; { - rb_add_method(class, rb_intern(name), NEW_CFUNC(func, argc), 0); + rb_add_method(class, rb_intern(name), NEW_CFUNC(func, argc), NOEX_PUBLIC); } void @@ -238,7 +238,7 @@ rb_undef_method(class, name) struct RClass *class; char *name; { - rb_add_method(class, rb_intern(name), Qnil, 0); + rb_add_method(class, rb_intern(name), Qnil, NOEX_PUBLIC); } void @@ -248,7 +248,7 @@ rb_define_private_method(class, name, func, argc) VALUE (*func)(); int argc; { - rb_add_method(class, rb_intern(name), NEW_CFUNC(func, argc), 1); + rb_add_method(class, rb_intern(name), NEW_CFUNC(func, argc), NOEX_PRIVATE); } VALUE @@ -615,12 +615,13 @@ Init_Dict() rb_define_method(C_Dict,"has_value", Fdic_has_value, 1); envtbl = obj_alloc(C_Object); + rb_extend_object(envtbl, M_Enumerable); + rb_define_single_method(envtbl,"[]", Fgetenv, 1); rb_define_single_method(envtbl,"[]=", Fsetenv, 2); rb_define_single_method(envtbl,"each", Fenv_each, 0); rb_define_single_method(envtbl,"delete", Fenv_delete, 1); rb_define_single_method(envtbl,"to_s", Fenv_to_s, 0); - rb_include_module(CLASS_OF(envtbl), M_Enumerable); rb_define_variable("$ENV", &envtbl, Qnil, rb_readonly_hook, 0); rb_define_const(C_Kernel, "ENV", envtbl); @@ -237,6 +237,108 @@ Sdir_rmdir(obj, dir) return TRUE; } +#define isdelim(c) ((c)==' '||(c)=='\t'||(c)=='\n'||(c)=='\0') + +char **glob_filename(); + +static void +push_globs(ary, s) + VALUE ary; + char *s; +{ + char **fnames, **ff; + + fnames = glob_filename(s); + if (fnames == (char**)-1) rb_sys_fail(s); + ff = fnames; + while (*ff) { + ary_push(ary, str_new2(*ff)); + free(*ff); + ff++; + } + free(fnames); +} + +static int +push_braces(ary, s) + VALUE ary; + char *s; +{ + char buf[MAXPATHLEN]; + char *p, *t, *b; + char *lbrace, *rbrace; + + p = s; + lbrace = rbrace = Qnil; + while (*p) { + if (*p == '{' && !lbrace) lbrace = p; + if (*p == '}' && lbrace) rbrace = p; + *p++; + } + + if (lbrace) { + memcpy(buf, s, lbrace-s); + b = buf + (lbrace-s); + p = lbrace; + while (*p != '}') { + t = p + 1; + for (p = t; *p!='}' && *p!=','; p++) { + /* skip inner braces */ + if (*p == '{') while (*p!='}') p++; + } + memcpy(b, t, p-t); + strcpy(b+(p-t), rbrace+1); + push_braces(ary, buf); + } + } + else { + push_globs(ary, s); + } +} + +static VALUE +Sdir_glob(dir, str) + VALUE dir; + struct RString *str; +{ + char *p, *pend; + char buf[MAXPATHLEN]; + char *t, *t0; + int nest; + VALUE ary; + + Check_Type(str, T_STRING); + + ary = ary_new(); + + p = str->ptr; + pend = p + str->len; + + while (p < pend) { + t = buf; + while (p < pend && isdelim(*p)) p++; + while (p < pend && !isdelim(*p)) { + *t++ = *p++; + } + *t = '\0'; + t0 = buf; + nest = 0; + while (t0 < t) { + if (*t0 == '{') nest+=2; + if (*t0 == '}') nest+=3; + t0++; + } + if (nest == 0) { + push_globs(ary, buf); + } + else if (nest % 5 == 0) { + push_braces(ary, buf); + } + /* else unmatched braces */ + } + return ary; +} + Init_Dir() { extern VALUE M_Enumerable; @@ -261,4 +363,7 @@ Init_Dir() rb_define_single_method(C_Dir,"rmdir", Sdir_rmdir, 1); rb_define_single_method(C_Dir,"delete", Sdir_rmdir, 1); rb_define_single_method(C_Dir,"unlink", Sdir_rmdir, 1); + + rb_define_single_method(C_Dir,"glob", Sdir_glob, 1); + rb_define_single_method(C_Dir,"[]", Sdir_glob, 1); } @@ -32,4 +32,7 @@ extern int rb_in_eval; extern struct RClass *the_class; +#define NOEX_PUBLIC 0 +#define NOEX_PRIVATE 1 + #endif /* ENV_H */ @@ -25,8 +25,13 @@ err_sprintf(buf, fmt, args) char *buf, *fmt; va_list args; { - sprintf(buf, "%s:%d: ", sourcefile, sourceline); - vsprintf((char*)buf+strlen(buf), fmt, args); + if (!sourcefile) { + vsprintf(buf, fmt, args); + } + else { + sprintf(buf, "%s:%d: ", sourcefile, sourceline); + vsprintf((char*)buf+strlen(buf), fmt, args); + } if (buf[strlen(buf)-1] != '\n') strcat(buf, "\n"); } @@ -179,14 +179,15 @@ VALUE M_Etc; Init_Etc() { M_Etc = rb_define_module("Etc"); + rb_extend_object(M_Etc, M_Etc); - rb_define_module_function(M_Etc, "getlogin", Fetc_getlogin, 0); + rb_define_method(M_Etc, "getlogin", Fetc_getlogin, 0); - rb_define_module_function(M_Etc, "getpwuid", Fetc_getpwuid, -1); - rb_define_module_function(M_Etc, "getpwnam", Fetc_getpwnam, 1); - rb_define_module_function(M_Etc, "passwd", Fetc_passwd, 0); + rb_define_method(M_Etc, "getpwuid", Fetc_getpwuid, -1); + rb_define_method(M_Etc, "getpwnam", Fetc_getpwnam, 1); + rb_define_method(M_Etc, "passwd", Fetc_passwd, 0); - rb_define_module_function(M_Etc, "getgrgid", Fetc_getgrgid, 1); - rb_define_module_function(M_Etc, "getgrnam", Fetc_getgrnam, 1); - rb_define_module_function(M_Etc, "group", Fetc_group, 0); + rb_define_method(M_Etc, "getgrgid", Fetc_getgrgid, 1); + rb_define_method(M_Etc, "getgrnam", Fetc_getgrnam, 1); + rb_define_method(M_Etc, "group", Fetc_group, 0); } @@ -448,6 +448,10 @@ ruby_init(argc, argv, envp) POP_SCOPE(); the_scope = top_scope; + if (state == TAG_EXIT) { + rb_trap_exit(); + exit(FIX2UINT(last_val)); + } if (state) { PUSH_TAG(); error_print(); @@ -926,11 +930,21 @@ rb_eval(node) int argc; VALUE *argv; /* used in SETUP_ARGS */ PUSH_ITER(ITER_NOT); - recv = node->nd_recv?rb_eval(node->nd_recv):Qself; + recv = rb_eval(node->nd_recv); SETUP_ARGS; POP_ITER(); - return rb_call(CLASS_OF(recv),recv,node->nd_mid,argc,argv, - node->nd_recv?0:1); + return rb_call(CLASS_OF(recv),recv,node->nd_mid,argc,argv,0); + } + break; + + case NODE_FCALL: + { + int argc; VALUE *argv; /* used in SETUP_ARGS */ + + PUSH_ITER(ITER_NOT); + SETUP_ARGS; + POP_ITER(); + return rb_call(CLASS_OF(Qself),Qself,node->nd_mid,argc,argv,1); } break; @@ -1093,6 +1107,9 @@ rb_eval(node) case NODE_CONST: return node->nd_cval; + case NODE_NTH_REF: + return re_nth_match(node->nd_nth); + case NODE_HASH: { NODE *list; @@ -1139,7 +1156,6 @@ rb_eval(node) case NODE_STR2: case NODE_XSTR2: case NODE_DREGX: - case NODE_DGLOB: { VALUE str, str2; NODE *list = node->nd_next; @@ -1166,9 +1182,6 @@ rb_eval(node) if (nd_type(node) == NODE_XSTR2) { return rb_xstring(str); } - if (nd_type(node) == NODE_DGLOB) { - return glob_new(str); - } return str; } @@ -1210,12 +1223,15 @@ rb_eval(node) Fail("Can't define method \"%s\" for nil", rb_id2name(node->nd_mid)); } - rb_add_method(rb_single_class(recv),node->nd_mid,node->nd_defn,0); + rb_funcall(recv, rb_intern("single_method_added"), + 1, INT2FIX(node->nd_mid)); + rb_add_method(rb_single_class(recv),node->nd_mid,node->nd_defn, + NOEX_PUBLIC); } return Qnil; case NODE_UNDEF: - rb_add_method(the_class, node->nd_mid, Qnil, 0); + rb_add_method(the_class, node->nd_mid, Qnil, NOEX_PUBLIC); return Qnil; case NODE_ALIAS: @@ -1225,23 +1241,42 @@ rb_eval(node) case NODE_CLASS: { VALUE super, class; + struct RClass *tmp; if (node->nd_super) { - super = rb_id2class(node->nd_super); + super = rb_const_get(the_class, node->nd_super); if (super == Qnil) { Fail("undefined superclass %s", rb_id2name(node->nd_super)); } } else { - super = C_Object; - } - if (verbose && rb_id2class(node->nd_cname)) { - Warning("redefine class %s", rb_id2name(node->nd_cname)); + super = Qnil; } - class = rb_define_class_id(node->nd_cname, super); - rb_const_set(the_class, node->nd_cname, class); + if (rb_const_bound(the_class, node->nd_cname)) { + class = rb_const_get(the_class, node->nd_cname); + if (super) { + if (TYPE(class) != T_CLASS) + Fail("%s is not a class", rb_id2name(node->nd_cname)); + tmp = RCLASS(class)->super; + while (FL_TEST(tmp, FL_SINGLE)) { + tmp = RCLASS(tmp)->super; + } + while (TYPE(tmp) == T_ICLASS) { + tmp = RCLASS(tmp)->super; + } + if (tmp != super) + Fail("%s's superclass differs", + rb_id2name(node->nd_cname)); + } + Warning("extending class %s", rb_id2name(node->nd_cname)); + } + else { + if (super == Qnil) super = C_Object; + class = rb_define_class_id(node->nd_cname, super); + rb_const_set(the_class, node->nd_cname, class); + } module_setup(class, node->nd_body); return class; @@ -1251,29 +1286,21 @@ rb_eval(node) { VALUE module; - if (verbose && rb_id2class(node->nd_cname)) { - Warning("redefine module %s", rb_id2name(node->nd_cname)); + if (rb_const_bound(the_class, node->nd_cname)) { + module = rb_const_get(the_class, node->nd_cname); + if (TYPE(module) != T_MODULE) + Fail("%s is not a module", rb_id2name(node->nd_cname)); + Warning("extending module %s", rb_id2name(node->nd_cname)); + } + else { + module = rb_define_module_id(node->nd_cname); + rb_const_set(the_class, node->nd_cname, module); } - - module = rb_define_module_id(node->nd_cname); - rb_const_set(the_class, node->nd_cname, module); module_setup(module, node->nd_body); return module; } - case NODE_INC: - { - struct RClass *module; - - module = (struct RClass*)rb_const_get(CLASS_OF(Qself), node->nd_modl); - if (module == Qnil) { - Fail("undefined module %s", rb_id2name(node->nd_modl)); - } - rb_include_module(the_class, module); - } - return Qnil; - default: Bug("unknown node type %d", nd_type(node)); } @@ -1706,7 +1733,7 @@ rb_ensure(b_proc, data1, e_proc, data2) static int last_noex; static VALUE -Funknown(argc, argv, obj) +Fmissing(argc, argv, obj) int argc; VALUE *argv; VALUE obj; @@ -1731,7 +1758,7 @@ Funknown(argc, argv, obj) /* fake environment */ PUSH_ENV(); env = the_env->prev; - MEMCPY(the_env, the_env->prev->prev, struct ENVIRON, 1); + MEMCPY(the_env, env->prev, struct ENVIRON, 1); the_env->prev = env; Fail(format, @@ -1758,23 +1785,24 @@ rb_undefined(obj, id, argc, argv, noex) last_noex = noex; - return rb_funcall2(obj, rb_intern("unknown"), argc+1, nargv); + return rb_funcall2(obj, rb_intern("method_missing"), argc+1, nargv); } static VALUE -rb_call(class, recv, mid, argc, argv, func) +rb_call(class, recv, mid, argc, argv, scope) struct RClass *class; VALUE recv; ID mid; int argc; VALUE *argv; - int func; + int scope; { NODE *body; int noex; VALUE result; struct cache_entry *ent; int itr; + enum node_type type; /* is it in the method cache? */ ent = cache + EXPR1(class, mid); @@ -1793,8 +1821,13 @@ rb_call(class, recv, mid, argc, argv, func) mid = id; } - if (!func && noex) { - return rb_undefined(recv, mid, argc, argv, 1); + switch (noex) { + case NOEX_PUBLIC: + break; + case NOEX_PRIVATE: + if (scope == 0) /* receiver specified */ + return rb_undefined(recv, mid, argc, argv, 1); + break; } switch (iter->iter) { @@ -1806,6 +1839,13 @@ rb_call(class, recv, mid, argc, argv, func) itr = ITER_NOT; break; } + + type = nd_type(body); + if (type == NODE_ZSUPER) { + /* for re-scoped method */ + return rb_call(class->super, recv, mid, argc, argv, scope?scope:1); + } + PUSH_ITER(itr); PUSH_SELF(recv); PUSH_ENV(); @@ -1814,7 +1854,7 @@ rb_call(class, recv, mid, argc, argv, func) the_env->argc = argc; the_env->argv = argv; - switch (nd_type(body)) { + switch (type) { case NODE_CFUNC: { int len = body->nd_argc; @@ -1831,7 +1871,7 @@ rb_call(class, recv, mid, argc, argv, func) result = (*body->nd_cfnc)(argc, argv, recv); break; case 0: - result = (*body->nd_cfnc)(recv); + result = (*body->nd_cfnc)(recv); break; case 1: result = (*body->nd_cfnc)(recv, argv[0]); @@ -1930,8 +1970,6 @@ rb_call(class, recv, mid, argc, argv, func) /* for attr get/set */ case NODE_ATTRSET: case NODE_IVAR: - /* for exported method */ - case NODE_ZSUPER: return rb_eval(body); default: @@ -2281,8 +2319,41 @@ addpath(path) } } -extern VALUE C_Kernel; +static VALUE +Fmod_include(argc, argv, module) + int argc; + VALUE *argv; + struct RClass *module; +{ + int i; + for (i=0; i<argc; i++) { + rb_include_module(module, argv[i]); + } + return (VALUE)module; +} + +static VALUE +Ftop_include(argc, argv) +{ + return Fmod_include(argc, argv, C_Object); +} + +static VALUE +Fobj_extend(argc, argv, obj) +{ + return Fmod_include(argc, argv, rb_single_class(obj)); +} + +void +rb_extend_object(obj, module) + VALUE obj, module; +{ + rb_include_module(rb_single_class(obj), module); +} + +extern VALUE C_Kernel; +extern VALUE C_Module; Init_eval() { @@ -2298,7 +2369,11 @@ Init_eval() rb_define_private_method(C_Kernel, "eval", Feval, 1); rb_define_private_method(C_Kernel, "iterator_p", Fiterator_p, 0); rb_define_method(C_Kernel, "apply", Fapply, -1); - rb_define_method(C_Kernel, "unknown", Funknown, -1); + rb_define_method(C_Kernel, "method_missing", Fmissing, -1); + + rb_define_method(C_Module, "include", Fmod_include, -1); + rb_define_method(CLASS_OF(TopSelf), "include", Ftop_include, -1); + rb_define_method(C_Object, "extend", Fobj_extend, -1); } Init_load() @@ -1,4 +1,3 @@ - /************************************************ file.c - @@ -56,60 +55,24 @@ file_open(fname, mode) } static int -apply2files0(func, args, arg, gl) - void (*func)(); +apply2files(func, args, arg) + int (*func)(); struct RArray *args; void *arg; - int gl; { - int i, n; + int i; VALUE path; - for (i=n=0; i<args->len; i++) { + for (i=0; i<args->len; i++) { path = args->ptr[i]; - if (TYPE(path) == T_STRING) { - if (gl) { - char buf[MAXPATHLEN]; - char *p, *s; - - s = buf; - p = RSTRING(path)->ptr; - while (*p) { - switch (*s = *p++) { - case '*': case '?': - case '[': case '{': - path = glob_new(path); - goto glob; - case '\\': - if (*p == '\0') break; - *s = *p++; - } - s++; - } - *s = '\0'; - (*func)(buf, arg); - } - else { - (*func)(RSTRING(path)->ptr, arg); - } - n++; - } - else { - extern VALUE C_Glob; - - if (!obj_is_kind_of(path, C_Glob)) { - WrongType(path, T_STRING); - } - glob: - n += apply2files0(func, rb_to_a(path), arg, 0); - } + Check_Type(path, T_STRING); + if ((*func)(RSTRING(path)->ptr, arg) < 0) + rb_sys_fail(RSTRING(path)->ptr); } - return n; + return args->len; } -#define apply2files(func,args,arg) apply2files0(func,args,arg,1) - static VALUE Ffile_tell(obj) VALUE obj; @@ -1131,6 +1094,7 @@ Init_File() rb_define_method(M_FileTest, "k", Ftest_sticky, 1); C_File = rb_define_class("File", C_IO); + rb_extend_object(C_File, M_FileTest); rb_define_single_method(C_File, "stat", Sfile_stat, 1); rb_define_single_method(C_File, "lstat", Sfile_lstat, 1); @@ -1154,8 +1118,6 @@ Init_File() rb_define_single_method(C_File, "umask", Sfile_umask, -1); rb_define_single_method(C_File, "truncate", Sfile_truncate, 2); - rb_include_module(CLASS_OF(C_File), M_FileTest); - rb_define_method(C_File, "stat", Ffile_stat, 0); rb_define_method(C_File, "lstat", Ffile_lstat, 0); diff --git a/getopt.c b/getopt.c deleted file mode 100644 index bfc4013b29..0000000000 --- a/getopt.c +++ /dev/null @@ -1,660 +0,0 @@ -/* Getopt for GNU. - NOTE: getopt is now part of the C library, so if you don't know what - "Keep this file name-space clean" means, talk to roland@gnu.ai.mit.edu - before changing it! - - Copyright (C) 1987, 88, 89, 90, 91, 1992 Free Software Foundation, Inc. - - This program is free software; you can redistribute it and/or modify it - under the terms of the GNU Library General Public License as published - by the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU Library General Public - License along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ - -#include "config.h" - -#include <stdio.h> - -/* This needs to come after some library #include - to get __GNU_LIBRARY__ defined. */ -#ifdef __GNU_LIBRARY__ -#include <stdlib.h> -#include <string.h> -#endif /* GNU C library. */ - - -#ifndef __STDC__ -#define const -#endif - -/* If GETOPT_COMPAT is defined, `+' as well as `--' can introduce a - long-named option. Because this is not POSIX.2 compliant, it is - being phased out. */ -#define GETOPT_COMPAT - -/* This version of `getopt' appears to the caller like standard Unix `getopt' - but it behaves differently for the user, since it allows the user - to intersperse the options with the other arguments. - - As `getopt' works, it permutes the elements of ARGV so that, - when it is done, all the options precede everything else. Thus - all application programs are extended to handle flexible argument order. - - Setting the environment variable POSIXLY_CORRECT disables permutation. - Then the behavior is completely standard. - - GNU application programs can use a third alternative mode in which - they can distinguish the relative order of options and other arguments. */ - -#include "getopt.h" - -/* For communication from `getopt' to the caller. - When `getopt' finds an option that takes an argument, - the argument value is returned here. - Also, when `ordering' is RETURN_IN_ORDER, - each non-option ARGV-element is returned here. */ - -char *optarg = 0; - -/* Index in ARGV of the next element to be scanned. - This is used for communication to and from the caller - and for communication between successive calls to `getopt'. - - On entry to `getopt', zero means this is the first call; initialize. - - When `getopt' returns EOF, this is the index of the first of the - non-option elements that the caller should itself scan. - - Otherwise, `optind' communicates from one call to the next - how much of ARGV has been scanned so far. */ - -int optind = 0; - -/* The next char to be scanned in the option-element - in which the last option character we returned was found. - This allows us to pick up the scan where we left off. - - If this is zero, or a null string, it means resume the scan - by advancing to the next ARGV-element. */ - -static char *nextchar; - -/* Callers store zero here to inhibit the error message - for unrecognized options. */ - -int opterr = 1; - -/* Describe how to deal with options that follow non-option ARGV-elements. - - If the caller did not specify anything, - the default is REQUIRE_ORDER if the environment variable - POSIXLY_CORRECT is defined, PERMUTE otherwise. - - REQUIRE_ORDER means don't recognize them as options; - stop option processing when the first non-option is seen. - This is what Unix does. - This mode of operation is selected by either setting the environment - variable POSIXLY_CORRECT, or using `+' as the first character - of the list of option characters. - - PERMUTE is the default. We permute the contents of ARGV as we scan, - so that eventually all the non-options are at the end. This allows options - to be given in any order, even with programs that were not written to - expect this. - - RETURN_IN_ORDER is an option available to programs that were written - to expect options and other ARGV-elements in any order and that care about - the ordering of the two. We describe each non-option ARGV-element - as if it were the argument of an option with character code 1. - Using `-' as the first character of the list of option characters - selects this mode of operation. - - The special argument `--' forces an end of option-scanning regardless - of the value of `ordering'. In the case of RETURN_IN_ORDER, only - `--' can cause `getopt' to return EOF with `optind' != ARGC. */ - -static enum -{ - REQUIRE_ORDER, PERMUTE, RETURN_IN_ORDER -} ordering; - -#ifdef __GNU_LIBRARY__ -#include <string.h> -#define my_index strchr -#define my_bcopy(src, dst, n) memcpy ((dst), (src), (n)) -#else - -/* Avoid depending on library functions or files - whose names are inconsistent. */ - -char *getenv (); - -static char * -my_index (string, chr) - char *string; - int chr; -{ - while (*string) - { - if (*string == chr) - return string; - string++; - } - return 0; -} - -static void -my_bcopy (from, to, size) - char *from, *to; - int size; -{ - int i; - for (i = 0; i < size; i++) - to[i] = from[i]; -} -#endif /* GNU C library. */ - -/* Handle permutation of arguments. */ - -/* Describe the part of ARGV that contains non-options that have - been skipped. `first_nonopt' is the index in ARGV of the first of them; - `last_nonopt' is the index after the last of them. */ - -static int first_nonopt; -static int last_nonopt; - -/* Exchange two adjacent subsequences of ARGV. - One subsequence is elements [first_nonopt,last_nonopt) - which contains all the non-options that have been skipped so far. - The other is elements [last_nonopt,optind), which contains all - the options processed since those non-options were skipped. - - `first_nonopt' and `last_nonopt' are relocated so that they describe - the new indices of the non-options in ARGV after they are moved. */ - -static void -exchange (argv) - char **argv; -{ - int nonopts_size = (last_nonopt - first_nonopt) * sizeof (char *); - char **temp = (char **) malloc (nonopts_size); - - /* Interchange the two blocks of data in ARGV. */ - - my_bcopy (&argv[first_nonopt], temp, nonopts_size); - my_bcopy (&argv[last_nonopt], &argv[first_nonopt], - (optind - last_nonopt) * sizeof (char *)); - my_bcopy (temp, &argv[first_nonopt + optind - last_nonopt], nonopts_size); - - free(temp); - - /* Update records for the slots the non-options now occupy. */ - - first_nonopt += (optind - last_nonopt); - last_nonopt = optind; -} - -/* Scan elements of ARGV (whose length is ARGC) for option characters - given in OPTSTRING. - - If an element of ARGV starts with '-', and is not exactly "-" or "--", - then it is an option element. The characters of this element - (aside from the initial '-') are option characters. If `getopt' - is called repeatedly, it returns successively each of the option characters - from each of the option elements. - - If `getopt' finds another option character, it returns that character, - updating `optind' and `nextchar' so that the next call to `getopt' can - resume the scan with the following option character or ARGV-element. - - If there are no more option characters, `getopt' returns `EOF'. - Then `optind' is the index in ARGV of the first ARGV-element - that is not an option. (The ARGV-elements have been permuted - so that those that are not options now come last.) - - OPTSTRING is a string containing the legitimate option characters. - If an option character is seen that is not listed in OPTSTRING, - return '?' after printing an error message. If you set `opterr' to - zero, the error message is suppressed but we still return '?'. - - If a char in OPTSTRING is followed by a colon, that means it wants an arg, - so the following text in the same ARGV-element, or the text of the following - ARGV-element, is returned in `optarg'. Two colons mean an option that - wants an optional arg; if there is text in the current ARGV-element, - it is returned in `optarg', otherwise `optarg' is set to zero. - - If OPTSTRING starts with `-' or `+', it requests different methods of - handling the non-option ARGV-elements. - See the comments about RETURN_IN_ORDER and REQUIRE_ORDER, above. - - Long-named options begin with `--' instead of `-'. - Their names may be abbreviated as long as the abbreviation is unique - or is an exact match for some defined option. If they have an - argument, it follows the option name in the same ARGV-element, separated - from the option name by a `=', or else the in next ARGV-element. - When `getopt' finds a long-named option, it returns 0 if that option's - `flag' field is nonzero, the value of the option's `val' field - if the `flag' field is zero. - - The elements of ARGV aren't really const, because we permute them. - But we pretend they're const in the prototype to be compatible - with other systems. - - LONGOPTS is a vector of `struct option' terminated by an - element containing a name which is zero. - - LONGIND returns the index in LONGOPT of the long-named option found. - It is only valid when a long-named option has been found by the most - recent call. - - If LONG_ONLY is nonzero, '-' as well as '--' can introduce - long-named options. */ - -int -_getopt_internal (argc, argv, optstring, longopts, longind, long_only) - int argc; - char *const *argv; - const char *optstring; - const struct option *longopts; - int *longind; - int long_only; -{ - int option_index; - - optarg = 0; - - /* Initialize the internal data when the first call is made. - Start processing options with ARGV-element 1 (since ARGV-element 0 - is the program name); the sequence of previously skipped - non-option ARGV-elements is empty. */ - - if (optind == 0) - { - first_nonopt = last_nonopt = optind = 1; - - nextchar = NULL; - - /* Determine how to handle the ordering of options and nonoptions. */ - - if (optstring[0] == '-') - { - ordering = RETURN_IN_ORDER; - ++optstring; - } - else if (optstring[0] == '+') - { - ordering = REQUIRE_ORDER; - ++optstring; - } - else if (getenv ("POSIXLY_CORRECT") != NULL) - ordering = REQUIRE_ORDER; - else - ordering = PERMUTE; - } - - if (nextchar == NULL || *nextchar == '\0') - { - if (ordering == PERMUTE) - { - /* If we have just processed some options following some non-options, - exchange them so that the options come first. */ - - if (first_nonopt != last_nonopt && last_nonopt != optind) - exchange ((char **) argv); - else if (last_nonopt != optind) - first_nonopt = optind; - - /* Now skip any additional non-options - and extend the range of non-options previously skipped. */ - - while (optind < argc - && (argv[optind][0] != '-' || argv[optind][1] == '\0') -#ifdef GETOPT_COMPAT - && (longopts == NULL - || argv[optind][0] != '+' || argv[optind][1] == '\0') -#endif /* GETOPT_COMPAT */ - ) - optind++; - last_nonopt = optind; - } - - /* Special ARGV-element `--' means premature end of options. - Skip it like a null option, - then exchange with previous non-options as if it were an option, - then skip everything else like a non-option. */ - - if (optind != argc && !strcmp (argv[optind], "--")) - { - optind++; - - if (first_nonopt != last_nonopt && last_nonopt != optind) - exchange ((char **) argv); - else if (first_nonopt == last_nonopt) - first_nonopt = optind; - last_nonopt = argc; - - optind = argc; - } - - /* If we have done all the ARGV-elements, stop the scan - and back over any non-options that we skipped and permuted. */ - - if (optind == argc) - { - /* Set the next-arg-index to point at the non-options - that we previously skipped, so the caller will digest them. */ - if (first_nonopt != last_nonopt) - optind = first_nonopt; - return EOF; - } - - /* If we have come to a non-option and did not permute it, - either stop the scan or describe it to the caller and pass it by. */ - - if ((argv[optind][0] != '-' || argv[optind][1] == '\0') -#ifdef GETOPT_COMPAT - && (longopts == NULL - || argv[optind][0] != '+' || argv[optind][1] == '\0') -#endif /* GETOPT_COMPAT */ - ) - { - if (ordering == REQUIRE_ORDER) - return EOF; - optarg = argv[optind++]; - return 1; - } - - /* We have found another option-ARGV-element. - Start decoding its characters. */ - - nextchar = (argv[optind] + 1 - + (longopts != NULL && argv[optind][1] == '-')); - } - - if (longopts != NULL - && ((argv[optind][0] == '-' - && (argv[optind][1] == '-' || long_only)) -#ifdef GETOPT_COMPAT - || argv[optind][0] == '+' -#endif /* GETOPT_COMPAT */ - )) - { - const struct option *p; - char *s = nextchar; - int exact = 0; - int ambig = 0; - const struct option *pfound = NULL; - int indfound = 0; - extern int strncmp(); - - while (*s && *s != '=') - s++; - - /* Test all options for either exact match or abbreviated matches. */ - for (p = longopts, option_index = 0; p->name; - p++, option_index++) - if (!strncmp (p->name, nextchar, s - nextchar)) - { - if (s - nextchar == strlen (p->name)) - { - /* Exact match found. */ - pfound = p; - indfound = option_index; - exact = 1; - break; - } - else if (pfound == NULL) - { - /* First nonexact match found. */ - pfound = p; - indfound = option_index; - } - else - /* Second nonexact match found. */ - ambig = 1; - } - - if (ambig && !exact) - { - if (opterr) - fprintf (stderr, "%s: option `%s' is ambiguous\n", - argv[0], argv[optind]); - nextchar += strlen (nextchar); - optind++; - return '?'; - } - - if (pfound != NULL) - { - option_index = indfound; - optind++; - if (*s) - { - /* Don't test has_arg with >, because some C compilers don't - allow it to be used on enums. */ - if (pfound->has_arg) - optarg = s + 1; - else - { - if (opterr) - { - if (argv[optind - 1][1] == '-') - /* --option */ - fprintf (stderr, - "%s: option `--%s' doesn't allow an argument\n", - argv[0], pfound->name); - else - /* +option or -option */ - fprintf (stderr, - "%s: option `%c%s' doesn't allow an argument\n", - argv[0], argv[optind - 1][0], pfound->name); - } - nextchar += strlen (nextchar); - return '?'; - } - } - else if (pfound->has_arg == 1) - { - if (optind < argc) - optarg = argv[optind++]; - else - { - if (opterr) - fprintf (stderr, "%s: option `%s' requires an argument\n", - argv[0], argv[optind - 1]); - nextchar += strlen (nextchar); - return '?'; - } - } - nextchar += strlen (nextchar); - if (longind != NULL) - *longind = option_index; - if (pfound->flag) - { - *(pfound->flag) = pfound->val; - return 0; - } - return pfound->val; - } - /* Can't find it as a long option. If this is not getopt_long_only, - or the option starts with '--' or is not a valid short - option, then it's an error. - Otherwise interpret it as a short option. */ - if (!long_only || argv[optind][1] == '-' -#ifdef GETOPT_COMPAT - || argv[optind][0] == '+' -#endif /* GETOPT_COMPAT */ - || my_index (optstring, *nextchar) == NULL) - { - if (opterr) - { - if (argv[optind][1] == '-') - /* --option */ - fprintf (stderr, "%s: unrecognized option `--%s'\n", - argv[0], nextchar); - else - /* +option or -option */ - fprintf (stderr, "%s: unrecognized option `%c%s'\n", - argv[0], argv[optind][0], nextchar); - } - nextchar = (char *) ""; - optind++; - return '?'; - } - } - - /* Look at and handle the next option-character. */ - - { - char c = *nextchar++; - char *temp = my_index (optstring, c); - - /* Increment `optind' when we start to process its last character. */ - if (*nextchar == '\0') - ++optind; - - if (temp == NULL || c == ':') - { - if (opterr) - { - if (c < 040 || c >= 0177) - fprintf (stderr, "%s: unrecognized option, character code 0%o\n", - argv[0], c); - else - fprintf (stderr, "%s: unrecognized option `-%c'\n", argv[0], c); - } - return '?'; - } - if (temp[1] == ':') - { - if (temp[2] == ':') - { - /* This is an option that accepts an argument optionally. */ - if (*nextchar != '\0') - { - optarg = nextchar; - optind++; - } - else - optarg = 0; - nextchar = NULL; - } - else - { - /* This is an option that requires an argument. */ - if (*nextchar != '\0') - { - optarg = nextchar; - /* If we end this ARGV-element by taking the rest as an arg, - we must advance to the next element now. */ - optind++; - } - else if (optind == argc) - { - if (opterr) - fprintf (stderr, "%s: option `-%c' requires an argument\n", - argv[0], c); - c = '?'; - } - else - /* We already incremented `optind' once; - increment it again when taking next ARGV-elt as argument. */ - optarg = argv[optind++]; - nextchar = NULL; - } - } - return c; - } -} - -int -getopt (argc, argv, optstring) - int argc; - char *const *argv; - const char *optstring; -{ - return _getopt_internal (argc, argv, optstring, - (const struct option *) 0, - (int *) 0, - 0); -} - -#ifdef TEST - -/* Compile with -DTEST to make an executable for use in testing - the above definition of `getopt'. */ - -int -main (argc, argv) - int argc; - char **argv; -{ - int c; - int digit_optind = 0; - - while (1) - { - int this_option_optind = optind ? optind : 1; - - c = getopt (argc, argv, "abc:d:0123456789"); - if (c == EOF) - break; - - switch (c) - { - case '0': - case '1': - case '2': - case '3': - case '4': - case '5': - case '6': - case '7': - case '8': - case '9': - if (digit_optind != 0 && digit_optind != this_option_optind) - printf ("digits occur in two different argv-elements.\n"); - digit_optind = this_option_optind; - printf ("option %c\n", c); - break; - - case 'a': - printf ("option a\n"); - break; - - case 'b': - printf ("option b\n"); - break; - - case 'c': - printf ("option c with value `%s'\n", optarg); - break; - - case '?': - break; - - default: - printf ("?? getopt returned character code 0%o ??\n", c); - } - } - - if (optind < argc) - { - printf ("non-option ARGV-elements: "); - while (optind < argc) - printf ("%s ", argv[optind++]); - printf ("\n"); - } - - exit (0); -} - -#endif /* TEST */ diff --git a/getopt.h b/getopt.h deleted file mode 100644 index de027434f7..0000000000 --- a/getopt.h +++ /dev/null @@ -1,128 +0,0 @@ -/* Declarations for getopt. - Copyright (C) 1989, 1990, 1991, 1992 Free Software Foundation, Inc. - - This program is free software; you can redistribute it and/or modify it - under the terms of the GNU Library General Public License as published - by the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU Library General Public - License along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ - -#ifndef _GETOPT_H -#define _GETOPT_H 1 - -#ifdef __cplusplus -extern "C" { -#endif - -/* For communication from `getopt' to the caller. - When `getopt' finds an option that takes an argument, - the argument value is returned here. - Also, when `ordering' is RETURN_IN_ORDER, - each non-option ARGV-element is returned here. */ - -extern char *optarg; - -/* Index in ARGV of the next element to be scanned. - This is used for communication to and from the caller - and for communication between successive calls to `getopt'. - - On entry to `getopt', zero means this is the first call; initialize. - - When `getopt' returns EOF, this is the index of the first of the - non-option elements that the caller should itself scan. - - Otherwise, `optind' communicates from one call to the next - how much of ARGV has been scanned so far. */ - -extern int optind; - -/* Callers store zero here to inhibit the error message `getopt' prints - for unrecognized options. */ - -extern int opterr; - -/* Describe the long-named options requested by the application. - The LONG_OPTIONS argument to getopt_long or getopt_long_only is a vector - of `struct option' terminated by an element containing a name which is - zero. - - The field `has_arg' is: - no_argument (or 0) if the option does not take an argument, - required_argument (or 1) if the option requires an argument, - optional_argument (or 2) if the option takes an optional argument. - - If the field `flag' is not NULL, it points to a variable that is set - to the value given in the field `val' when the option is found, but - left unchanged if the option is not found. - - To have a long-named option do something other than set an `int' to - a compiled-in constant, such as set a value from `optarg', set the - option's `flag' field to zero and its `val' field to a nonzero - value (the equivalent single-letter option character, if there is - one). For long options that have a zero `flag' field, `getopt' - returns the contents of the `val' field. */ - -struct option -{ -#if __STDC__ - const char *name; -#else - char *name; -#endif - /* has_arg can't be an enum because some compilers complain about - type mismatches in all the code that assumes it is an int. */ - int has_arg; - int *flag; - int val; -}; - -/* Names for the values of the `has_arg' field of `struct option'. */ - -enum _argtype -{ - no_argument, - required_argument, - optional_argument -}; - -#if __STDC__ -#if defined(__GNU_LIBRARY__) -/* Many other libraries have conflicting prototypes for getopt, with - differences in the consts, in stdlib.h. To avoid compilation - errors, only prototype getopt for the GNU C library. */ -extern int getopt (int argc, char *const *argv, const char *shortopts); -#else /* not __GNU_LIBRARY__ */ -extern int getopt (); -#endif /* not __GNU_LIBRARY__ */ -extern int getopt_long (int argc, char *const *argv, const char *shortopts, - const struct option *longopts, int *longind); -extern int getopt_long_only (int argc, char *const *argv, - const char *shortopts, - const struct option *longopts, int *longind); - -/* Internal only. Users should not call this directly. */ -extern int _getopt_internal (int argc, char *const *argv, - const char *shortopts, - const struct option *longopts, int *longind, - int long_only); -#else /* not __STDC__ */ -extern int getopt (); -extern int getopt_long (); -extern int getopt_long_only (); - -extern int _getopt_internal (); -#endif /* not __STDC__ */ - -#ifdef __cplusplus -} -#endif - -#endif /* _GETOPT_H */ diff --git a/getopt1.c b/getopt1.c deleted file mode 100644 index 7a2bbae344..0000000000 --- a/getopt1.c +++ /dev/null @@ -1,162 +0,0 @@ -/* Getopt for GNU. - Copyright (C) 1987, 88, 89, 90, 91, 1992 Free Software Foundation, Inc. - -This file is part of the libiberty library. -Libiberty is free software; you can redistribute it and/or -modify it under the terms of the GNU Library General Public -License as published by the Free Software Foundation; either -version 2 of the License, or (at your option) any later version. - -Libiberty is distributed in the hope that it will be useful, -but WITHOUT ANY WARRANTY; without even the implied warranty of -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU -Library General Public License for more details. - -You should have received a copy of the GNU Library General Public -License along with libiberty; see the file COPYING.LIB. If -not, write to the Free Software Foundation, Inc., 675 Mass Ave, -Cambridge, MA 02139, USA. */ - -#include "config.h" - -#ifdef LIBC -/* For when compiled as part of the GNU C library. */ -#include <ansidecl.h> -#endif - -#include "getopt.h" - -#ifndef __STDC__ -#define const -#endif - -#if defined(STDC_HEADERS) || defined(__GNU_LIBRARY__) || defined (LIBC) -#include <stdlib.h> -#else /* STDC_HEADERS or __GNU_LIBRARY__ */ -char *getenv (); -#endif /* STDC_HEADERS or __GNU_LIBRARY__ */ - -#if !defined (NULL) -#define NULL 0 -#endif - -int -getopt_long (argc, argv, options, long_options, opt_index) - int argc; - char *const *argv; - const char *options; - const struct option *long_options; - int *opt_index; -{ - return _getopt_internal (argc, argv, options, long_options, opt_index, 0); -} - -/* Like getopt_long, but '-' as well as '--' can indicate a long option. - If an option that starts with '-' (not '--') doesn't match a long option, - but does match a short option, it is parsed as a short option - instead. */ - -int -getopt_long_only (argc, argv, options, long_options, opt_index) - int argc; - char *const *argv; - const char *options; - const struct option *long_options; - int *opt_index; -{ - return _getopt_internal (argc, argv, options, long_options, opt_index, 1); -} - -#ifdef TEST - -#include <stdio.h> - -int -main (argc, argv) - int argc; - char **argv; -{ - int c; - int digit_optind = 0; - - while (1) - { - int this_option_optind = optind ? optind : 1; - int option_index = 0; - static struct option long_options[] = - { - {"add", 1, 0, 0}, - {"append", 0, 0, 0}, - {"delete", 1, 0, 0}, - {"verbose", 0, 0, 0}, - {"create", 0, 0, 0}, - {"file", 1, 0, 0}, - {0, 0, 0, 0} - }; - - c = getopt_long (argc, argv, "abc:d:0123456789", - long_options, &option_index); - if (c == EOF) - break; - - switch (c) - { - case 0: - printf ("option %s", long_options[option_index].name); - if (optarg) - printf (" with arg %s", optarg); - printf ("\n"); - break; - - case '0': - case '1': - case '2': - case '3': - case '4': - case '5': - case '6': - case '7': - case '8': - case '9': - if (digit_optind != 0 && digit_optind != this_option_optind) - printf ("digits occur in two different argv-elements.\n"); - digit_optind = this_option_optind; - printf ("option %c\n", c); - break; - - case 'a': - printf ("option a\n"); - break; - - case 'b': - printf ("option b\n"); - break; - - case 'c': - printf ("option c with value `%s'\n", optarg); - break; - - case 'd': - printf ("option d with value `%s'\n", optarg); - break; - - case '?': - break; - - default: - printf ("?? getopt returned character code 0%o ??\n", c); - } - } - - if (optind < argc) - { - printf ("non-option ARGV-elements: "); - while (optind < argc) - printf ("%s ", argv[optind++]); - printf ("\n"); - } - - exit (0); -} - -#endif /* TEST */ @@ -1,184 +1,580 @@ -/************************************************ +/* File-name wildcard pattern matching for GNU. + Copyright (C) 1985, 1988, 1989 Free Software Foundation, Inc. - glob.c - + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 1, or (at your option) + any later version. - $Author: matz $ - $Date: 1994/12/09 09:47:52 $ - created at: Mon Sep 12 18:56:43 JST 1994 + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. -************************************************/ + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ + +/* To whomever it may concern: I have never seen the code which most + Unix programs use to perform this function. I wrote this from scratch + based on specifications for the pattern matching. --RMS. */ + +#include "config.h" + +#if defined (SHELL) +# if defined (HAVE_STDLIB_H) +# include <stdlib.h> +# else +# include "ansi_stdlib.h" +# endif /* HAVE_STDLIB_H */ +# include <config.h> +#endif + +#include <sys/types.h> + +#if !defined (SHELL) && (defined (_POSIX_VERSION) || defined (USGr3)) +# if !defined (HAVE_DIRENT_H) +# define HAVE_DIRENT_H +# endif /* !HAVE_DIRENT_H */ +#endif /* !SHELL && (_POSIX_VERSION || USGr3) */ + +#if defined (HAVE_DIRENT_H) +# include <dirent.h> +# if !defined (direct) +# define direct dirent +# endif /* !direct */ +# define D_NAMLEN(d) strlen ((d)->d_name) +#else /* !HAVE_DIRENT_H */ +# define D_NAMLEN(d) ((d)->d_namlen) +# if defined (USG) +# if defined (Xenix) +# include <sys/ndir.h> +# else /* !Xenix (but USG...) */ +# include "ndir.h" +# endif /* !Xenix */ +# else /* !USG */ +# include <sys/dir.h> +# endif /* !USG */ +#endif /* !HAVE_DIRENT_H */ + +#if defined (_POSIX_SOURCE) +/* Posix does not require that the d_ino field be present, and some + systems do not provide it. */ +# define REAL_DIR_ENTRY(dp) 1 +#else +# define REAL_DIR_ENTRY(dp) (dp->d_ino != 0) +#endif /* _POSIX_SOURCE */ + +#if defined (USG) || defined (NeXT) +# if !defined (HAVE_STRING_H) +# define HAVE_STRING_H +# endif /* !HAVE_STRING_H */ +#endif /* USG || NeXT */ + +#if defined (HAVE_STRING_H) +# include <string.h> +#else /* !HAVE_STRING_H */ +# include <strings.h> +#endif /* !HAVE_STRING_H */ + +#if defined (USG) +# if !defined (isc386) +# include <memory.h> +# endif /* !isc386 */ +# if defined (RISC6000) +extern void bcopy (); +# else /* !RISC6000 */ +# define bcopy(s, d, n) ((void) memcpy ((d), (s), (n))) +# endif /* !RISC6000 */ +#endif /* USG */ + +#if defined(HAVE_ALLOCA_H) && !defined(__GNUC__) +#include <alloca.h> +#else +char *alloca (); +#endif -#include <sys/param.h> -#include "ruby.h" #include "fnmatch.h" -char *strdup(); +/* If the opendir () on your system lets you open non-directory files, + then we consider that not robust. Define OPENDIR_NOT_ROBUST in the + SYSDEP_CFLAGS for your machines entry in machines.h. */ +#if defined (OPENDIR_NOT_ROBUST) +# if defined (SHELL) +# include "posixstat.h" +# else /* !SHELL */ +# include <sys/stat.h> +# endif /* !SHELL */ +#endif /* OPENDIR_NOT_ROBUST */ + +extern void *xmalloc (), *xrealloc (); +#if !defined (HAVE_STDLIB_H) +extern void free (); +#endif /* !HAVE_STDLIB_H */ + +#if !defined (NULL) +# if defined (__STDC__) +# define NULL ((void *) 0) +# else +# define NULL 0x0 +# endif /* __STDC__ */ +#endif /* !NULL */ -VALUE C_Glob; +#if defined (SHELL) +extern int interrupt_state; +#endif /* SHELL */ -struct glob_data { - char **globs; -}; +/* Global variable which controls whether or not * matches .*. + Non-zero means don't match .*. */ +int noglob_dot_filenames = 1; -static ID id_data; +/* Global variable to return to signify an error in globbing. */ +char *glob_error_return; + +/* Return nonzero if PATTERN has any special globbing chars in it. */ +int +glob_pattern_p (pattern) + char *pattern; +{ + register char *p = pattern; + register char c; + int open = 0; + + while ((c = *p++) != '\0') + switch (c) + { + case '?': + case '*': + return (1); + + case '[': /* Only accept an open brace if there is a close */ + open++; /* brace to match it. Bracket expressions must be */ + continue; /* complete, according to Posix.2 */ + case ']': + if (open) + return (1); + continue; + + case '\\': + if (*p++ == '\0') + return (0); + } + + return (0); +} + +/* Remove backslashes quoting characters in PATHNAME by modifying PATHNAME. */ static void -glob_free(data) - struct glob_data *data; +dequote_pathname (pathname) + char *pathname; { - char **globs = data->globs; - while (*globs) { - free(*globs); - globs++; + register int i, j; + + for (i = j = 0; pathname && pathname[i]; ) + { + if (pathname[i] == '\\') + i++; + + pathname[j++] = pathname[i++]; + + if (!pathname[i - 1]) + break; } - free(data->globs); + pathname[j] = '\0'; } -#define isdelim(c) ((c)==' '||(c)=='\t'||(c)=='\n'||(c)=='\0') + +/* Return a vector of names of files in directory DIR + whose names match glob pattern PAT. + The names are not in any particular order. + Wildcards at the beginning of PAT do not match an initial period. + + The vector is terminated by an element that is a null pointer. -char *strchr(); -char *strdup(); + To free the space allocated, first free the vector's elements, + then free the vector. -static int -expand_brace(s, data, len) - char *s; - struct glob_data *data; - int len; + Return 0 if cannot get enough memory to hold the pointer + and the names. + + Return -1 if cannot access directory DIR. + Look in errno for more information. */ + +char ** +glob_vector (pat, dir) + char *pat; + char *dir; { - char org[MAXPATHLEN], path[MAXPATHLEN]; - char *pre, *post, *head, *p, *t; - - strcpy(org, s); - pre = strchr(org, '{'); - if (pre) post = strchr(pre, '}'); - if (!pre || !post) { - data->globs[len++] = strdup(s); - REALLOC_N(data->globs, char*, len+1); - return len; + struct globval + { + struct globval *next; + char *name; + }; + + DIR *d; + register struct direct *dp; + struct globval *lastlink; + register struct globval *nextlink; + register char *nextname; + unsigned int count; + int lose, skip; + register char **name_vector; + register unsigned int i; +#if defined (OPENDIR_NOT_ROBUST) + struct stat finfo; + + if (stat (dir, &finfo) < 0) + return ((char **) &glob_error_return); + + if (!S_ISDIR (finfo.st_mode)) + return ((char **) &glob_error_return); +#endif /* OPENDIR_NOT_ROBUST */ + + d = opendir (dir); + if (d == NULL) + return ((char **) &glob_error_return); + + lastlink = 0; + count = 0; + lose = 0; + skip = 0; + + /* If PAT is empty, skip the loop, but return one (empty) filename. */ + if (!pat || !*pat) + { + nextlink = (struct globval *)alloca (sizeof (struct globval)); + nextlink->next = lastlink; + nextname = (char *) xmalloc (1); + if (!nextname) + lose = 1; + else + { + lastlink = nextlink; + nextlink->name = nextname; + nextname[0] = '\0'; + count++; + } + skip = 1; } - memcpy(path, org, pre - org); - p = org + (pre - org) + 1; - head = path + (pre - org); + /* Scan the directory, finding all names that match. + For each name that matches, allocate a struct globval + on the stack and store the name in it. + Chain those structs together; lastlink is the front of the chain. */ + while (!skip) + { + int flags; /* Flags passed to fnmatch (). */ +#if defined (SHELL) + /* Make globbing interruptible in the bash shell. */ + if (interrupt_state) + { + closedir (d); + lose = 1; + goto lost; + } +#endif /* SHELL */ + + dp = readdir (d); + if (dp == NULL) + break; + + /* If this directory entry is not to be used, try again. */ + if (!REAL_DIR_ENTRY (dp)) + continue; - while (p < post) { - t = p; - while (t < post) { - if (*t == ',') break; - t++; + /* If a dot must be explicity matched, check to see if they do. */ + if (noglob_dot_filenames && dp->d_name[0] == '.' && pat[0] != '.') + continue; + + flags = (noglob_dot_filenames ? FNM_PERIOD : 0) | FNM_PATHNAME; + + if (fnmatch (pat, dp->d_name, flags) != FNM_NOMATCH) + { + nextlink = (struct globval *) alloca (sizeof (struct globval)); + nextlink->next = lastlink; + nextname = (char *) xmalloc (D_NAMLEN (dp) + 1); + if (nextname == NULL) + { + lose = 1; + break; + } + lastlink = nextlink; + nextlink->name = nextname; + bcopy (dp->d_name, nextname, D_NAMLEN (dp) + 1); + ++count; } - memcpy(head, p, t-p); - strcpy(head+(t-p), post+1); - len = expand_brace(path, data, len); - p = t + 1; } - return len; -} + (void) closedir (d); -static VALUE -glob_new0(class, str) - VALUE class; - struct RString *str; -{ - VALUE new; - struct glob_data *data; - char *p1, *p2, *pend, *s; - int len = 0; - - new = obj_alloc(class); - Make_Data_Struct(new, id_data, struct glob_data, Qnil, glob_free, data); - data->globs = ALLOC_N(char*, 1); - - p1 = p2 = str->ptr; - pend = p1 + str->len; - while (p1 < pend) { - char s[MAXPATHLEN]; - int d; - - while (isdelim(*p1)) p1++; - p2 = p1; - while (!isdelim(*p2)) p2++; - d = p2 - p1; - memcpy(s, p1, d); - s[d] = '\0'; - len = expand_brace(s, data, len); - p1 = p2; + if (!lose) + { + name_vector = (char **) xmalloc ((count + 1) * sizeof (char *)); + lose |= name_vector == NULL; } - data->globs[len] = Qnil; - return new; -} + /* Have we run out of memory? */ + lost: + if (lose) + { + /* Here free the strings we have got. */ + while (lastlink) + { + free (lastlink->name); + lastlink = lastlink->next; + } +#if defined (SHELL) + if (interrupt_state) + throw_to_top_level (); +#endif /* SHELL */ + return (NULL); + } -VALUE -glob_new(str) - struct RString *str; -{ - return glob_new0(C_Glob, str); + /* Copy the name pointers from the linked list into the vector. */ + for (i = 0; i < count; ++i) + { + name_vector[i] = lastlink->name; + lastlink = lastlink->next; + } + + name_vector[count] = NULL; + return (name_vector); } + +/* Return a new array which is the concatenation of each string in ARRAY + to DIR. This function expects you to pass in an allocated ARRAY, and + it takes care of free()ing that array. Thus, you might think of this + function as side-effecting ARRAY. */ +static char ** +glob_dir_to_array (dir, array) + char *dir, **array; +{ + register unsigned int i, l; + int add_slash; + char **result; + + l = strlen (dir); + if (l == 0) + return (array); + + add_slash = dir[l - 1] != '/'; -char **glob_filename(); + i = 0; + while (array[i] != NULL) + ++i; + + result = (char **) xmalloc ((i + 1) * sizeof (char *)); + if (result == NULL) + return (NULL); + + for (i = 0; array[i] != NULL; i++) + { + result[i] = (char *) xmalloc (l + (add_slash ? 1 : 0) + + strlen (array[i]) + 1); + if (result[i] == NULL) + return (NULL); + sprintf (result[i], "%s%s%s", dir, add_slash ? "/" : "", array[i]); + } + result[i] = NULL; -static VALUE -Fglob_each(glob) - VALUE glob; + /* Free the input array. */ + for (i = 0; array[i] != NULL; i++) + free (array[i]); + free ((char *) array); + + return (result); +} + +/* Do globbing on PATHNAME. Return an array of pathnames that match, + marking the end of the array with a null-pointer as an element. + If no pathnames match, then the array is empty (first element is null). + If there isn't enough memory, then return NULL. + If a file system error occurs, return -1; `errno' has the error code. */ +char ** +glob_filename (pathname) + char *pathname; { - struct glob_data *data; - char **patv, **fnames, **ff; - - Get_Data_Struct(glob, id_data, struct glob_data, data); - for (patv = data->globs; *patv; patv++) { - if (!glob_pattern_p(*patv)) { - rb_yield(str_new2(*patv)); - continue; - } - fnames = glob_filename(*patv); - if (fnames == (char**)-1) rb_sys_fail(*patv); - if (fnames[0] == Qnil) { - rb_yield(str_new2(*patv)); + char *strrchr(); + + char **result; + unsigned int result_size; + char *directory_name, *filename; + unsigned int directory_len; + + result = (char **) xmalloc (sizeof (char *)); + result_size = 1; + if (result == NULL) + return (NULL); + + result[0] = NULL; + + /* Find the filename. */ + filename = strrchr (pathname, '/'); + if (filename == NULL) + { + filename = pathname; + directory_name = ""; + directory_len = 0; + } + else + { + directory_len = (filename - pathname) + 1; + directory_name = (char *) alloca (directory_len + 1); + + bcopy (pathname, directory_name, directory_len); + directory_name[directory_len] = '\0'; + ++filename; + } + + /* If directory_name contains globbing characters, then we + have to expand the previous levels. Just recurse. */ + if (glob_pattern_p (directory_name)) + { + char **directories; + register unsigned int i; + + if (directory_name[directory_len - 1] == '/') + directory_name[directory_len - 1] = '\0'; + + directories = glob_filename (directory_name); + + if (directories == NULL) + goto memory_error; + else if (directories == (char **)&glob_error_return) + return ((char **) &glob_error_return); + else if (*directories == NULL) + { + free ((char *) directories); + return ((char **) &glob_error_return); } - else { - ff = fnames; - while (*ff) { - rb_yield(str_new2(*ff)); - free(*ff); - ff++; + + /* We have successfully globbed the preceding directory name. + For each name in DIRECTORIES, call glob_vector on it and + FILENAME. Concatenate the results together. */ + for (i = 0; directories[i] != NULL; ++i) + { + char **temp_results; + + /* Scan directory even on a NULL pathname. That way, `*h/' + returns only directories ending in `h', instead of all + files ending in `h' with a `/' appended. */ + temp_results = glob_vector (filename, directories[i]); + + /* Handle error cases. */ + if (temp_results == NULL) + goto memory_error; + else if (temp_results == (char **)&glob_error_return) + /* This filename is probably not a directory. Ignore it. */ + ; + else + { + char **array; + register unsigned int l; + + array = glob_dir_to_array (directories[i], temp_results); + l = 0; + while (array[l] != NULL) + ++l; + + result = + (char **)xrealloc(result, (result_size + l) * sizeof (char *)); + + if (result == NULL) + goto memory_error; + + for (l = 0; array[l] != NULL; ++l) + result[result_size++ - 1] = array[l]; + + result[result_size - 1] = NULL; + + /* Note that the elements of ARRAY are not freed. */ + free ((char *) array); } } - free(fnames); + /* Free the directories. */ + for (i = 0; directories[i]; i++) + free (directories[i]); + + free ((char *) directories); + + return (result); } - return Qnil; -} -VALUE -Fglob_match(glob, str) - VALUE glob; - struct RString *str; -{ - struct glob_data *data; - char **patv; - - Check_Type(str, T_STRING); - Get_Data_Struct(glob, id_data, struct glob_data, data); - patv = data->globs; - while (*patv) { - if (fnmatch(*patv, str->ptr, 0) != FNM_NOMATCH) - return TRUE; - patv++; + /* If there is only a directory name, return it. */ + if (*filename == '\0') + { + result = (char **) xrealloc ((char *) result, 2 * sizeof (char *)); + if (result == NULL) + return (NULL); + result[0] = (char *) xmalloc (directory_len + 1); + if (result[0] == NULL) + goto memory_error; + bcopy (directory_name, result[0], directory_len + 1); + result[1] = NULL; + return (result); } - return FALSE; -} + else + { + char **temp_results; -extern VALUE M_Enumerable; + /* There are no unquoted globbing characters in DIRECTORY_NAME. + Dequote it before we try to open the directory since there may + be quoted globbing characters which should be treated verbatim. */ + if (directory_len > 0) + dequote_pathname (directory_name); -Init_Glob() -{ - C_Glob = rb_define_class("Glob", C_Object); - rb_include_module(C_Glob, M_Enumerable); + /* We allocated a small array called RESULT, which we won't be using. + Free that memory now. */ + free (result); + + /* Just return what glob_vector () returns appended to the + directory name. */ + temp_results = + glob_vector (filename, (directory_len == 0 ? "." : directory_name)); - rb_define_single_method(C_Glob, "new", glob_new0, 1); + if (temp_results == NULL || temp_results == (char **)&glob_error_return) + return (temp_results); - rb_define_method(C_Glob, "each", Fglob_each, 0); - rb_define_method(C_Glob, "=~", Fglob_match, 1); + return (glob_dir_to_array (directory_name, temp_results)); + } + + /* We get to memory_error if the program has run out of memory, or + if this is the shell, and we have been interrupted. */ + memory_error: + if (result != NULL) + { + register unsigned int i; + for (i = 0; result[i] != NULL; ++i) + free (result[i]); + free ((char *) result); + } +#if defined (SHELL) + if (interrupt_state) + throw_to_top_level (); +#endif /* SHELL */ + return (NULL); +} + +#if defined (TEST) + +main (argc, argv) + int argc; + char **argv; +{ + unsigned int i; + + for (i = 1; i < argc; ++i) + { + char **value = glob_filename (argv[i]); + if (value == NULL) + puts ("Out of memory."); + else if (value == &glob_error_return) + perror (argv[i]); + else + for (i = 0; value[i] != NULL; i++) + puts (value[i]); + } - id_data = rb_intern("data"); + exit (0); } +#endif /* TEST. */ diff --git a/gnuglob.c b/gnuglob.c deleted file mode 100644 index 69af7b1618..0000000000 --- a/gnuglob.c +++ /dev/null @@ -1,580 +0,0 @@ -/* File-name wildcard pattern matching for GNU. - Copyright (C) 1985, 1988, 1989 Free Software Foundation, Inc. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 1, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; if not, write to the Free Software - Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ - -/* To whomever it may concern: I have never seen the code which most - Unix programs use to perform this function. I wrote this from scratch - based on specifications for the pattern matching. --RMS. */ - -#include "config.h" - -#if defined (SHELL) -# if defined (HAVE_STDLIB_H) -# include <stdlib.h> -# else -# include "ansi_stdlib.h" -# endif /* HAVE_STDLIB_H */ -# include <config.h> -#endif - -#include <sys/types.h> - -#if !defined (SHELL) && (defined (_POSIX_VERSION) || defined (USGr3)) -# if !defined (HAVE_DIRENT_H) -# define HAVE_DIRENT_H -# endif /* !HAVE_DIRENT_H */ -#endif /* !SHELL && (_POSIX_VERSION || USGr3) */ - -#if defined (HAVE_DIRENT_H) -# include <dirent.h> -# if !defined (direct) -# define direct dirent -# endif /* !direct */ -# define D_NAMLEN(d) strlen ((d)->d_name) -#else /* !HAVE_DIRENT_H */ -# define D_NAMLEN(d) ((d)->d_namlen) -# if defined (USG) -# if defined (Xenix) -# include <sys/ndir.h> -# else /* !Xenix (but USG...) */ -# include "ndir.h" -# endif /* !Xenix */ -# else /* !USG */ -# include <sys/dir.h> -# endif /* !USG */ -#endif /* !HAVE_DIRENT_H */ - -#if defined (_POSIX_SOURCE) -/* Posix does not require that the d_ino field be present, and some - systems do not provide it. */ -# define REAL_DIR_ENTRY(dp) 1 -#else -# define REAL_DIR_ENTRY(dp) (dp->d_ino != 0) -#endif /* _POSIX_SOURCE */ - -#if defined (USG) || defined (NeXT) -# if !defined (HAVE_STRING_H) -# define HAVE_STRING_H -# endif /* !HAVE_STRING_H */ -#endif /* USG || NeXT */ - -#if defined (HAVE_STRING_H) -# include <string.h> -#else /* !HAVE_STRING_H */ -# include <strings.h> -#endif /* !HAVE_STRING_H */ - -#if defined (USG) -# if !defined (isc386) -# include <memory.h> -# endif /* !isc386 */ -# if defined (RISC6000) -extern void bcopy (); -# else /* !RISC6000 */ -# define bcopy(s, d, n) ((void) memcpy ((d), (s), (n))) -# endif /* !RISC6000 */ -#endif /* USG */ - -#if defined(HAVE_ALLOCA_H) && !defined(__GNUC__) -#include <alloca.h> -#else -char *alloca (); -#endif - -#include "fnmatch.h" - -/* If the opendir () on your system lets you open non-directory files, - then we consider that not robust. Define OPENDIR_NOT_ROBUST in the - SYSDEP_CFLAGS for your machines entry in machines.h. */ -#if defined (OPENDIR_NOT_ROBUST) -# if defined (SHELL) -# include "posixstat.h" -# else /* !SHELL */ -# include <sys/stat.h> -# endif /* !SHELL */ -#endif /* OPENDIR_NOT_ROBUST */ - -extern void *xmalloc (), *xrealloc (); -#if !defined (HAVE_STDLIB_H) -extern void free (); -#endif /* !HAVE_STDLIB_H */ - -#if !defined (NULL) -# if defined (__STDC__) -# define NULL ((void *) 0) -# else -# define NULL 0x0 -# endif /* __STDC__ */ -#endif /* !NULL */ - -#if defined (SHELL) -extern int interrupt_state; -#endif /* SHELL */ - -/* Global variable which controls whether or not * matches .*. - Non-zero means don't match .*. */ -int noglob_dot_filenames = 1; - -/* Global variable to return to signify an error in globbing. */ -char *glob_error_return; - - -/* Return nonzero if PATTERN has any special globbing chars in it. */ -int -glob_pattern_p (pattern) - char *pattern; -{ - register char *p = pattern; - register char c; - int open = 0; - - while ((c = *p++) != '\0') - switch (c) - { - case '?': - case '*': - return (1); - - case '[': /* Only accept an open brace if there is a close */ - open++; /* brace to match it. Bracket expressions must be */ - continue; /* complete, according to Posix.2 */ - case ']': - if (open) - return (1); - continue; - - case '\\': - if (*p++ == '\0') - return (0); - } - - return (0); -} - -/* Remove backslashes quoting characters in PATHNAME by modifying PATHNAME. */ -static void -dequote_pathname (pathname) - char *pathname; -{ - register int i, j; - - for (i = j = 0; pathname && pathname[i]; ) - { - if (pathname[i] == '\\') - i++; - - pathname[j++] = pathname[i++]; - - if (!pathname[i - 1]) - break; - } - pathname[j] = '\0'; -} - - -/* Return a vector of names of files in directory DIR - whose names match glob pattern PAT. - The names are not in any particular order. - Wildcards at the beginning of PAT do not match an initial period. - - The vector is terminated by an element that is a null pointer. - - To free the space allocated, first free the vector's elements, - then free the vector. - - Return 0 if cannot get enough memory to hold the pointer - and the names. - - Return -1 if cannot access directory DIR. - Look in errno for more information. */ - -char ** -glob_vector (pat, dir) - char *pat; - char *dir; -{ - struct globval - { - struct globval *next; - char *name; - }; - - DIR *d; - register struct direct *dp; - struct globval *lastlink; - register struct globval *nextlink; - register char *nextname; - unsigned int count; - int lose, skip; - register char **name_vector; - register unsigned int i; -#if defined (OPENDIR_NOT_ROBUST) - struct stat finfo; - - if (stat (dir, &finfo) < 0) - return ((char **) &glob_error_return); - - if (!S_ISDIR (finfo.st_mode)) - return ((char **) &glob_error_return); -#endif /* OPENDIR_NOT_ROBUST */ - - d = opendir (dir); - if (d == NULL) - return ((char **) &glob_error_return); - - lastlink = 0; - count = 0; - lose = 0; - skip = 0; - - /* If PAT is empty, skip the loop, but return one (empty) filename. */ - if (!pat || !*pat) - { - nextlink = (struct globval *)alloca (sizeof (struct globval)); - nextlink->next = lastlink; - nextname = (char *) xmalloc (1); - if (!nextname) - lose = 1; - else - { - lastlink = nextlink; - nextlink->name = nextname; - nextname[0] = '\0'; - count++; - } - skip = 1; - } - - /* Scan the directory, finding all names that match. - For each name that matches, allocate a struct globval - on the stack and store the name in it. - Chain those structs together; lastlink is the front of the chain. */ - while (!skip) - { - int flags; /* Flags passed to fnmatch (). */ -#if defined (SHELL) - /* Make globbing interruptible in the bash shell. */ - if (interrupt_state) - { - closedir (d); - lose = 1; - goto lost; - } -#endif /* SHELL */ - - dp = readdir (d); - if (dp == NULL) - break; - - /* If this directory entry is not to be used, try again. */ - if (!REAL_DIR_ENTRY (dp)) - continue; - - /* If a dot must be explicity matched, check to see if they do. */ - if (noglob_dot_filenames && dp->d_name[0] == '.' && pat[0] != '.') - continue; - - flags = (noglob_dot_filenames ? FNM_PERIOD : 0) | FNM_PATHNAME; - - if (fnmatch (pat, dp->d_name, flags) != FNM_NOMATCH) - { - nextlink = (struct globval *) alloca (sizeof (struct globval)); - nextlink->next = lastlink; - nextname = (char *) xmalloc (D_NAMLEN (dp) + 1); - if (nextname == NULL) - { - lose = 1; - break; - } - lastlink = nextlink; - nextlink->name = nextname; - bcopy (dp->d_name, nextname, D_NAMLEN (dp) + 1); - ++count; - } - } - (void) closedir (d); - - if (!lose) - { - name_vector = (char **) xmalloc ((count + 1) * sizeof (char *)); - lose |= name_vector == NULL; - } - - /* Have we run out of memory? */ - lost: - if (lose) - { - /* Here free the strings we have got. */ - while (lastlink) - { - free (lastlink->name); - lastlink = lastlink->next; - } -#if defined (SHELL) - if (interrupt_state) - throw_to_top_level (); -#endif /* SHELL */ - return (NULL); - } - - /* Copy the name pointers from the linked list into the vector. */ - for (i = 0; i < count; ++i) - { - name_vector[i] = lastlink->name; - lastlink = lastlink->next; - } - - name_vector[count] = NULL; - return (name_vector); -} - -/* Return a new array which is the concatenation of each string in ARRAY - to DIR. This function expects you to pass in an allocated ARRAY, and - it takes care of free()ing that array. Thus, you might think of this - function as side-effecting ARRAY. */ -static char ** -glob_dir_to_array (dir, array) - char *dir, **array; -{ - register unsigned int i, l; - int add_slash; - char **result; - - l = strlen (dir); - if (l == 0) - return (array); - - add_slash = dir[l - 1] != '/'; - - i = 0; - while (array[i] != NULL) - ++i; - - result = (char **) xmalloc ((i + 1) * sizeof (char *)); - if (result == NULL) - return (NULL); - - for (i = 0; array[i] != NULL; i++) - { - result[i] = (char *) xmalloc (l + (add_slash ? 1 : 0) - + strlen (array[i]) + 1); - if (result[i] == NULL) - return (NULL); - sprintf (result[i], "%s%s%s", dir, add_slash ? "/" : "", array[i]); - } - result[i] = NULL; - - /* Free the input array. */ - for (i = 0; array[i] != NULL; i++) - free (array[i]); - free ((char *) array); - - return (result); -} - -/* Do globbing on PATHNAME. Return an array of pathnames that match, - marking the end of the array with a null-pointer as an element. - If no pathnames match, then the array is empty (first element is null). - If there isn't enough memory, then return NULL. - If a file system error occurs, return -1; `errno' has the error code. */ -char ** -glob_filename (pathname) - char *pathname; -{ - char *strrchr(); - - char **result; - unsigned int result_size; - char *directory_name, *filename; - unsigned int directory_len; - - result = (char **) xmalloc (sizeof (char *)); - result_size = 1; - if (result == NULL) - return (NULL); - - result[0] = NULL; - - /* Find the filename. */ - filename = strrchr (pathname, '/'); - if (filename == NULL) - { - filename = pathname; - directory_name = ""; - directory_len = 0; - } - else - { - directory_len = (filename - pathname) + 1; - directory_name = (char *) alloca (directory_len + 1); - - bcopy (pathname, directory_name, directory_len); - directory_name[directory_len] = '\0'; - ++filename; - } - - /* If directory_name contains globbing characters, then we - have to expand the previous levels. Just recurse. */ - if (glob_pattern_p (directory_name)) - { - char **directories; - register unsigned int i; - - if (directory_name[directory_len - 1] == '/') - directory_name[directory_len - 1] = '\0'; - - directories = glob_filename (directory_name); - - if (directories == NULL) - goto memory_error; - else if (directories == (char **)&glob_error_return) - return ((char **) &glob_error_return); - else if (*directories == NULL) - { - free ((char *) directories); - return ((char **) &glob_error_return); - } - - /* We have successfully globbed the preceding directory name. - For each name in DIRECTORIES, call glob_vector on it and - FILENAME. Concatenate the results together. */ - for (i = 0; directories[i] != NULL; ++i) - { - char **temp_results; - - /* Scan directory even on a NULL pathname. That way, `*h/' - returns only directories ending in `h', instead of all - files ending in `h' with a `/' appended. */ - temp_results = glob_vector (filename, directories[i]); - - /* Handle error cases. */ - if (temp_results == NULL) - goto memory_error; - else if (temp_results == (char **)&glob_error_return) - /* This filename is probably not a directory. Ignore it. */ - ; - else - { - char **array; - register unsigned int l; - - array = glob_dir_to_array (directories[i], temp_results); - l = 0; - while (array[l] != NULL) - ++l; - - result = - (char **)xrealloc(result, (result_size + l) * sizeof (char *)); - - if (result == NULL) - goto memory_error; - - for (l = 0; array[l] != NULL; ++l) - result[result_size++ - 1] = array[l]; - - result[result_size - 1] = NULL; - - /* Note that the elements of ARRAY are not freed. */ - free ((char *) array); - } - } - /* Free the directories. */ - for (i = 0; directories[i]; i++) - free (directories[i]); - - free ((char *) directories); - - return (result); - } - - /* If there is only a directory name, return it. */ - if (*filename == '\0') - { - result = (char **) xrealloc ((char *) result, 2 * sizeof (char *)); - if (result == NULL) - return (NULL); - result[0] = (char *) xmalloc (directory_len + 1); - if (result[0] == NULL) - goto memory_error; - bcopy (directory_name, result[0], directory_len + 1); - result[1] = NULL; - return (result); - } - else - { - char **temp_results; - - /* There are no unquoted globbing characters in DIRECTORY_NAME. - Dequote it before we try to open the directory since there may - be quoted globbing characters which should be treated verbatim. */ - if (directory_len > 0) - dequote_pathname (directory_name); - - /* We allocated a small array called RESULT, which we won't be using. - Free that memory now. */ - free (result); - - /* Just return what glob_vector () returns appended to the - directory name. */ - temp_results = - glob_vector (filename, (directory_len == 0 ? "." : directory_name)); - - if (temp_results == NULL || temp_results == (char **)&glob_error_return) - return (temp_results); - - return (glob_dir_to_array (directory_name, temp_results)); - } - - /* We get to memory_error if the program has run out of memory, or - if this is the shell, and we have been interrupted. */ - memory_error: - if (result != NULL) - { - register unsigned int i; - for (i = 0; result[i] != NULL; ++i) - free (result[i]); - free ((char *) result); - } -#if defined (SHELL) - if (interrupt_state) - throw_to_top_level (); -#endif /* SHELL */ - return (NULL); -} - -#if defined (TEST) - -main (argc, argv) - int argc; - char **argv; -{ - unsigned int i; - - for (i = 1; i < argc; ++i) - { - char **value = glob_filename (argv[i]); - if (value == NULL) - puts ("Out of memory."); - else if (value == &glob_error_return) - perror (argv[i]); - else - for (i = 0; value[i] != NULL; i++) - puts (value[i]); - } - - exit (0); -} -#endif /* TEST. */ @@ -14,12 +14,13 @@ #ifndef IDENT_H #define IDENT_H +#define ID_SCOPE_SHIFT 3 #define ID_SCOPE_MASK 0x07 #define ID_LOCAL 0x00 -#define ID_ATTRSET 0x04 -#define ID_INSTANCE 0x02 -#define ID_GLOBAL 0x03 -#define ID_CONST 0x06 -#define ID_VARMASK 0x02 +#define ID_INSTANCE 0x01 +#define ID_GLOBAL 0x02 +#define ID_ATTRSET 0x03 +#define ID_CONST 0x04 +#define ID_NTHREF 0x05 #endif @@ -28,7 +28,6 @@ rb_call_inits() Init_Struct(); Init_String(); Init_Regexp(); - Init_Glob(); Init_pack(); Init_Cons(); Init_Range(); @@ -1395,6 +1395,8 @@ Init_IO() rb_define_const(C_Object, "STDERR", rb_stderr); argf = obj_alloc(C_Object); + rb_extend_object(argf, M_Enumerable); + rb_define_variable("$<", &argf, Qnil, rb_readonly_hook, 0); rb_define_variable("$ARGF", &argf, Qnil, rb_readonly_hook, 0); @@ -1411,7 +1413,6 @@ Init_IO() rb_define_single_method(argf, "to_s", Farg_filename, 0); rb_define_single_method(argf, "filename", Farg_filename, 0); rb_define_single_method(argf, "file", Farg_file, 0); - rb_include_module(CLASS_OF(argf), M_Enumerable); filename = str_new2("-"); rb_define_variable("$FILENAME", &filename, Qnil, rb_readonly_hook, 0); @@ -108,14 +108,18 @@ Fmath_sqrt(obj, x) Init_Math() { M_Math = rb_define_module("Math"); - - rb_define_module_function(M_Math, "atan2", Fmath_atan2, 2); - rb_define_module_function(M_Math, "cos", Fmath_cos, 1); - rb_define_module_function(M_Math, "sin", Fmath_sin, 1); - rb_define_module_function(M_Math, "tan", Fmath_tan, 1); - - rb_define_module_function(M_Math, "exp", Fmath_exp, 1); - rb_define_module_function(M_Math, "log", Fmath_log, 1); - rb_define_module_function(M_Math, "log10", Fmath_log10, 1); - rb_define_module_function(M_Math, "sqrt", Fmath_sqrt, 1); + rb_extend_object(M_Math, M_Math); + + rb_define_const(M_Math, "PI", float_new(M_PI)); + rb_define_const(M_Math, "E", float_new(M_E)); + + rb_define_method(M_Math, "atan2", Fmath_atan2, 2); + rb_define_method(M_Math, "cos", Fmath_cos, 1); + rb_define_method(M_Math, "sin", Fmath_sin, 1); + rb_define_method(M_Math, "tan", Fmath_tan, 1); + + rb_define_method(M_Math, "exp", Fmath_exp, 1); + rb_define_method(M_Math, "log", Fmath_log, 1); + rb_define_method(M_Math, "log10", Fmath_log10, 1); + rb_define_method(M_Math, "sqrt", Fmath_sqrt, 1); } @@ -38,6 +38,7 @@ enum node_type { NODE_OP_ASGN1, NODE_OP_ASGN2, NODE_CALL, + NODE_FCALL, NODE_SUPER, NODE_ZSUPER, NODE_ARRAY, @@ -56,13 +57,13 @@ enum node_type { NODE_IVAR, NODE_CVAR, NODE_CONST, + NODE_NTH_REF, NODE_LIT, NODE_STR, NODE_STR2, NODE_XSTR, NODE_XSTR2, NODE_DREGX, - NODE_DGLOB, NODE_ARGS, NODE_DEFN, NODE_DEFS, @@ -71,7 +72,6 @@ enum node_type { NODE_CLASS, NODE_MODULE, NODE_CREF, - NODE_INC, NODE_DOT3, NODE_ATTRSET, NODE_SELF, @@ -173,6 +173,8 @@ typedef struct RNode { #define nd_state u3.state #define nd_rval u3.node +#define nd_nth u2.argc + #define NEW_METHOD(n,x) newnode(NODE_METHOD,x,n,0) #define NEW_FBODY(n,i,o) newnode(NODE_FBODY,n,i,o) #define NEW_DEFN(i,a,d,p) newnode(NODE_DEFN,p,i,NEW_RFUNC(a,d)) @@ -216,12 +218,14 @@ typedef struct RNode { #define NEW_LVAR2(v) newnode(NODE_LVAR2,v,0,0) #define NEW_IVAR(v) newnode(NODE_IVAR,v,0,0) #define NEW_CVAR(v) newnode(NODE_CVAR,v,0,cref_list) +#define NEW_NTH_REF(n) newnode(NODE_NTH_REF,0,n,0) #define NEW_LIT(l) newnode(NODE_LIT,l,0,0) #define NEW_STR(s) newnode(NODE_STR,s,0,0) #define NEW_STR2(s) newnode(NODE_STR2,s,0,0) #define NEW_XSTR(s) newnode(NODE_XSTR,s,0,0) #define NEW_XSTR2(s) newnode(NODE_XSTR2,s,0,0) #define NEW_CALL(r,m,a) newnode(NODE_CALL,r,m,a) +#define NEW_FCALL(m,a) newnode(NODE_FCALL,0,m,a) #define NEW_SUPER(a) newnode(NODE_SUPER,0,0,a) #define NEW_ZSUPER() newnode(NODE_ZSUPER,0,0,0) #define NEW_ARGS(f,r) newnode(NODE_ARGS,0,r,f) @@ -232,7 +236,6 @@ typedef struct RNode { #define NEW_CREF0() (cref_list=newnode(NODE_CREF,the_class,0,0)) #define NEW_CREF(b) (cref_list=newnode(NODE_CREF,0,0,cref_list)) #define NEW_CBODY(b) (cref_list->nd_body=NEW_SCOPE(b),cref_list) -#define NEW_INC(m) newnode(NODE_INC,m,0,0) #define NEW_DOT3(b,e) newnode(NODE_DOT3,b,e,0) #define NEW_ATTRSET(a) newnode(NODE_ATTRSET,a,0,0) #define NEW_SELF() newnode(NODE_SELF,0,0,0) @@ -208,6 +208,13 @@ Fobj_init_object(obj) } static VALUE +Fobj_s_added(obj, id) + VALUE obj, id; +{ + return Qnil; +} + +static VALUE Fnil_to_s(obj) VALUE obj; { @@ -296,8 +303,8 @@ Fcls_attr(argc, argv, class) return Qnil; } -static VALUE -Fcls_export_internal(argc, argv, ex) +void +method_visibility(argc, argv, ex) int argc; VALUE *argv; int ex; @@ -316,24 +323,23 @@ Fcls_export_internal(argc, argv, ex) } rb_export_method(self, id, ex); } - return Qnil; } static VALUE -Fcls_export(argc, argv) +Fcls_public(argc, argv) int argc; VALUE *argv; { - Fcls_export_internal(argc, argv, 0); + method_visibility(argc, argv, NOEX_PUBLIC); return Qnil; } static VALUE -Fcls_unexport(argc, argv) +Fcls_private(argc, argv) int argc; VALUE *argv; { - Fcls_export_internal(argc, argv, 1); + method_visibility(argc, argv, NOEX_PRIVATE); return Qnil; } @@ -429,6 +435,7 @@ Init_Object() rb_define_alias(C_Kernel, "format", "sprintf"); rb_define_private_method(C_Object, "init_object", Fobj_init_object, -1); + rb_define_private_method(C_Object, "single_method_added", Fobj_s_added, 1); rb_define_method(C_Object, "clone", Fobj_clone, 0); @@ -440,8 +447,9 @@ Init_Object() rb_define_method(C_Module, "to_s", Fcls_to_s, 0); rb_define_method(C_Module, "clone", Fcant_clone, 0); rb_define_private_method(C_Module, "attr", Fcls_attr, -1); - rb_define_method(C_Module, "export", Fcls_export, -1); - rb_define_method(C_Module, "unexport", Fcls_unexport, -1); + + rb_define_method(C_Module, "public", Fcls_public, -1); + rb_define_method(C_Module, "private", Fcls_private, -1); rb_define_method(C_Class, "new", Fcls_new, -1); @@ -24,9 +24,9 @@ #define is_local_id(id) (is_id_nonop(id)&&((id)&ID_SCOPE_MASK)==ID_LOCAL) #define is_global_id(id) (is_id_nonop(id)&&((id)&ID_SCOPE_MASK)==ID_GLOBAL) #define is_instance_id(id) (is_id_nonop(id)&&((id)&ID_SCOPE_MASK)==ID_INSTANCE) -#define is_variable_id(id) (is_id_nonop(id)&&((id)&ID_VARMASK)) #define is_attrset_id(id) (is_id_nonop(id)&&((id)&ID_SCOPE_MASK)==ID_ATTRSET) #define is_const_id(id) (is_id_nonop(id)&&((id)&ID_SCOPE_MASK)==ID_CONST) +#define is_nthref_id(id) (((id)&ID_SCOPE_MASK)==ID_NTHREF) struct op_tbl { ID token; @@ -94,7 +94,6 @@ static void top_local_setup(); MODULE DEF UNDEF - INCLUDE BEGIN RESQUE ENSURE @@ -126,16 +125,16 @@ static void top_local_setup(); WHILE_MOD ALIAS -%token <id> IDENTIFIER GVAR IVAR CONSTANT -%token <val> INTEGER FLOAT STRING XSTRING REGEXP GLOB -%token <node> STRING2 XSTRING2 DREGEXP DGLOB +%token <id> IDENTIFIER GVAR IVAR CONSTANT NTH_REF +%token <val> INTEGER FLOAT STRING XSTRING REGEXP +%token <node> STRING2 XSTRING2 DREGEXP -%type <node> singleton inc_list +%type <node> singleton %type <val> literal numeric %type <node> compexpr exprs expr arg primary var_ref %type <node> if_tail opt_else case_body cases resque ensure %type <node> call_args call_args0 args args2 opt_args -%type <node> f_arglist f_args assoc_list assocs assoc +%type <node> f_arglist f_args array assoc_list assocs assoc %type <node> mlhs mlhs_head mlhs_tail lhs iter_var opt_iter_var %type <id> superclass variable symbol %type <id> fname op rest_arg @@ -245,7 +244,7 @@ expr : mlhs '=' args2 } | IDENTIFIER call_args0 { - $$ = NEW_CALL(Qnil, $1, $2); + $$ = NEW_FCALL($1, $2); } | primary '.' IDENTIFIER call_args0 { @@ -266,12 +265,6 @@ expr : mlhs '=' args2 { $$ = NEW_ALIAS($2, $4); } - | INCLUDE inc_list - { - if (cur_mid || in_single) - Error("include appeared in method definition"); - $$ = $2; - } | expr IF_MOD expr { $$ = NEW_IF(cond($3), $1, Qnil); @@ -322,7 +315,7 @@ lhs : variable { $$ = asignable($1, Qnil); } - | primary '[' opt_args rbracket + | primary '[' opt_args opt_nl rbracket { $$ = aryset($1, $3, Qnil); } @@ -331,25 +324,6 @@ lhs : variable $$ = attrset($1, $3, Qnil); } -inc_list : CONSTANT - { - $$ = NEW_INC($1); - } - | inc_list comma CONSTANT - { - $$ = block_append($1, NEW_INC($3)); - } - | error - { - lex_state = EXPR_BEG; - $$ = Qnil; - } - | inc_list comma error - { - lex_state = EXPR_BEG; - $$ = $1; - } - fname : IDENTIFIER | CONSTANT | op @@ -389,9 +363,9 @@ arg : variable '=' arg value_expr($3); $$ = asignable($1, $3); } - | primary '[' opt_args rbracket '=' arg + | primary '[' opt_args opt_nl rbracket '=' arg { - $$ = aryset($1, $3, $6); + $$ = aryset($1, $3, $7); } | primary '.' IDENTIFIER '=' arg { @@ -416,12 +390,12 @@ arg : variable '=' arg } $$ = asignable($1, call_op(val, $2, 1, $3)); } - | primary '[' opt_args rbracket OP_ASGN arg + | primary '[' opt_args opt_nl rbracket OP_ASGN arg { - NODE *args = NEW_LIST($6); + NODE *args = NEW_LIST($7); if ($3) list_concat(args, $3); - $$ = NEW_OP_ASGN1($1, $5, args); + $$ = NEW_OP_ASGN1($1, $6, args); } | primary '.' IDENTIFIER OP_ASGN arg { @@ -577,8 +551,8 @@ call_args : /* none */ { $$ = Qnil; } - | call_args0 - | '*' arg + | call_args0 opt_nl + | '*' arg opt_nl { $$ = $2; } @@ -624,6 +598,12 @@ args2 : args } } +array : /* none */ + { + $$ = Qnil; + } + | args trailer + primary : literal { $$ = NEW_LIT($1); @@ -639,7 +619,6 @@ primary : literal } | XSTRING2 | DREGEXP - | DGLOB | var_ref | SUPER '(' call_args rparen { @@ -653,12 +632,12 @@ primary : literal Error("super called outside of method"); $$ = NEW_ZSUPER(); } - | primary '[' opt_args rbracket + | primary '[' opt_args opt_nl rbracket { value_expr($1); $$ = NEW_CALL($1, AREF, $3); } - | LBRACK opt_args rbracket + | LBRACK array rbracket { if ($2 == Qnil) $$ = NEW_ZARRAY(); /* zero length array*/ @@ -726,13 +705,13 @@ primary : literal if (nd_type($1) == NODE_LVAR || nd_type($1) == NODE_LVAR2 || nd_type($1) == NODE_CVAR) { - $1 = NEW_CALL(Qnil, $1->nd_vid, Qnil); + $1 = NEW_FCALL($1->nd_vid, Qnil); } $$ = NEW_ITER($3, $1, $5); } | IDENTIFIER '(' call_args rparen { - $$ = NEW_CALL(Qnil, $1, $3); + $$ = NEW_FCALL($1, $3); } | primary '.' IDENTIFIER '(' call_args rparen { @@ -832,7 +811,7 @@ primary : literal compexpr END { - $$ = NEW_DEFN($2, $4, $5, class_nest?1:0); + $$ = NEW_DEFN($2, $4, $5, class_nest?0:1); local_pop(); cur_mid = Qnil; } @@ -918,7 +897,6 @@ literal : numeric $$ = INT2FIX($2); } | REGEXP - | GLOB symbol : fname | IVAR @@ -931,6 +909,7 @@ variable : IDENTIFIER | IVAR | GVAR | CONSTANT + | NTH_REF | NIL { $$ = NIL; @@ -1037,7 +1016,6 @@ singleton : var_ref case NODE_XSTR: case NODE_XSTR2: case NODE_DREGX: - case NODE_DGLOB: case NODE_LIT: case NODE_ARRAY: case NODE_ZARRAY: @@ -1052,8 +1030,8 @@ assoc_list : /* none */ { $$ = Qnil; } - | assocs - | args + | assocs trailer + | args trailer { if ($1->nd_alen%2 != 0) { Error("odd number list for Dict"); @@ -1078,6 +1056,10 @@ opt_term : /* none */ opt_nl : /* none */ | nl +trailer : /* none */ + | nl + | comma + term : sc | nl @@ -1091,10 +1073,12 @@ rparen : ')' { yyerrok; } rbracket : ']' { yyerrok; } rbrace : '}' { yyerrok; } comma : ',' { yyerrok; } + %% #include <ctype.h> #include <sys/types.h> #include "regex.h" +#include "util.h" #define is_identchar(c) ((c)!=-1&&(isalnum(c) || (c) == '_' || ismbchar(c))) @@ -1107,12 +1091,11 @@ VALUE newfloat(); VALUE newinteger(); char *strdup(); -#define EXPAND_B 1 -#define LEAVE_BS 2 - static NODE *var_extend(); static void read_escape(); +#define LEAVE_BS 1 + static char *lex_p; static int lex_len; @@ -1132,6 +1115,22 @@ lex_setsrc(src, ptr, len) #define nextc() ((--lex_len>=0)?(*lex_p++):-1) #define pushback() (lex_len++, lex_p--) +#define SCAN_HEX(i) \ +do { \ + int numlen; \ + i=scan_hex(lex_p, 2, &numlen); \ + lex_p += numlen; \ + lex_len -= numlen; \ +} while (0) + +#define SCAN_OCT(i) \ +do { \ + int numlen; \ + i=scan_oct(lex_p, 3, &numlen); \ + lex_p += numlen; \ + lex_len -= numlen; \ +} while (0) + #define tokfix() (tokenbuf[tokidx]='\0') #define tok() tokenbuf #define toklen() tokidx @@ -1187,18 +1186,37 @@ parse_regx() continue; case '\\': - if ((c = nextc()) == -1) { + switch (c = nextc()) { + case -1: sourceline = re_start; Error("unterminated regexp meets end of file"); return 0; - } - else if (c == '\n') { + + case '\n': sourceline++; - } - else if (in_brack && c == 'b') { - tokadd('\b'); - } - else { + break; + + case '\\': + tokadd('\\'); + tokadd('\\'); + break; + + case '1': case '2': case '3': + case '4': case '5': case '6': + case '7': case '8': case '9': + case '0': case 'x': + tokadd('\\'); + tokadd(c); + break; + + case 'b': + if (!in_brack) { + tokadd('\\'); + tokadd('b'); + break; + } + /* fall through */ + default: pushback(); read_escape(LEAVE_BS); } @@ -1285,11 +1303,11 @@ parse_string(term) tokadd(c); } else { - int flags = EXPAND_B; - if (term != '"') flags |= LEAVE_BS; - pushback(); - read_escape(flags); - } + int flags = 0; + if (term != '"') flags = LEAVE_BS; + pushback(); + read_escape(flags); + } continue; } tokadd(c); @@ -1342,7 +1360,6 @@ static struct kwtable { "for", FOR, EXPR_BEG, "if", IF, EXPR_BEG, "in", IN, EXPR_BEG, - "include", INCLUDE, EXPR_BEG, "module", MODULE, EXPR_BEG, "nil", NIL, EXPR_END, "or", OR, EXPR_BEG, @@ -1438,14 +1455,6 @@ retry: return '='; case '<': - if (lex_state == EXPR_BEG) { - if (parse_string('>') == STRING) { - yylval.val = glob_new(yylval.val); - return GLOB; - } - nd_set_type(yylval.node, NODE_DGLOB); - return DGLOB; - } lex_state = EXPR_BEG; if ((c = nextc()) == '=') { if ((c = nextc()) == '>') { @@ -1533,7 +1542,7 @@ retry: case '?': if ((c = nextc()) == '\\') { newtok(); - read_escape(EXPAND_B); + read_escape(0); c = tok()[0]; } c &= 0xff; @@ -1830,8 +1839,8 @@ retry: return c; case '$': + lex_state = EXPR_END; newtok(); - tokadd(c); c = nextc(); switch (c) { case '*': /* $*: argv */ @@ -1854,17 +1863,42 @@ retry: case '<': /* $<: reading filename */ case '>': /* $>: default output handle */ case '"': /* $": already loaded files */ + tokadd('$'); tokadd(c); tokfix(); yylval.id = rb_intern(tok()); - lex_state = EXPR_END; return GVAR; + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + while (isdigit(c)) { + tokadd(c); + c = nextc(); + } + pushback(); + tokfix(); + { + ID id = atoi(tok()); + id <<= ID_SCOPE_SHIFT; + id |= ID_NTHREF; + yylval.id = id; + return NTH_REF; + } + + case '0': default: if (!is_identchar(c)) { pushback(); return '$'; } + tokadd('$'); } break; @@ -2082,87 +2116,36 @@ read_escape(flag) tokadd(033); break; - case 'M': - if ((c = nextc()) != '-') { - Error("Invalid escape character syntax"); - tokadd('\0'); - return; - } - if ((c = nextc()) == '\\') { - read_escape(flag); - tokenbuf[tokidx-1] |= 0200; /* kludge */ - } + case 'c': + if (c == '?') + tokadd(0177); else { - tokadd((c & 0xff) | 0200); + if (islower(c)) + c = toupper(c); + tokadd(c ^ 64); } break; - case 'C': - if ((c = nextc()) != '-') { - Error("Invalid escape character syntax"); - tokadd('\0'); - return; - } - case '^': - if ((c = nextc())== '\\') { - read_escape (flag); - tokenbuf[tokidx-1] &= 0237; /* kludge */ - } - else if (c == '?') - tokadd(0177); - else - tokadd(c & 0237); - break; - case '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7': { /* octal constant */ - register int i = c - '0'; - register int count = 0; - - while (++count < 3) { - if ((c = nextc()) >= '0' && c <= '7') { - i *= 8; - i += c - '0'; - } - else { - pushback(); - break; - } - } - tokadd(i&0xff); + pushback(); + SCAN_OCT(c); + tokadd(c); } break; case 'x': /* hex constant */ { - register int i = 0; - register int count = 0; - - while (++count < 3) { - if ((c = nextc()) >= '0' && c <= '9') { - i *= 16; - i += c - '0'; - } - else if ((int)strchr("abcdefABCDEF", (c = nextc()))) { - i *= 16; - i += toupper(c) - 'A' + 10; - } - else { - pushback(); - break; - } - } - tokadd(i&0xff); + SCAN_HEX(c); + tokadd(c); } break; case 'b': /* backspace */ - if (flag & EXPAND_B) { - tokadd('\b'); - return; - } - /* go turough */ + tokadd('\b'); + return; + default: if (flag & LEAVE_BS) { tokadd('\\'); @@ -2398,6 +2381,9 @@ gettable(id) else if (is_const_id(id)) { return NEW_CVAR(id); } + else if (is_nthref_id(id)) { + return NEW_NTH_REF(id>>ID_SCOPE_SHIFT); + } } static NODE* @@ -2405,14 +2391,12 @@ asignable(id, val) ID id; NODE *val; { - NODE *lhs; + NODE *lhs = Qnil; if (id == SELF) { - lhs = Qnil; Error("Can't change the value of self"); } else if (id == NIL) { - lhs = Qnil; Error("Can't asign to nil"); } else if (id == _LINE_ || id == _FILE_) { @@ -2432,6 +2416,9 @@ asignable(id, val) Error("class constant asigned in method body"); lhs = NEW_CASGN(id, val); } + else if (is_nthref_id(id)) { + Error("Can't set variable $%d", id>>ID_SCOPE_SHIFT); + } else { Bug("bad id for variable"); } @@ -2476,7 +2463,6 @@ value_expr(node) case NODE_FAIL: case NODE_WHILE: case NODE_WHILE2: - case NODE_INC: case NODE_CLASS: case NODE_MODULE: case NODE_DEFN: @@ -2699,8 +2685,8 @@ void yyappend_print() { eval_tree = - block_append(eval_tree, NEW_CALL(Qnil, rb_intern("print"), - NEW_ARRAY(NEW_GVAR(rb_intern("$_"))))); + block_append(eval_tree, NEW_FCALL(rb_intern("print"), + NEW_ARRAY(NEW_GVAR(rb_intern("$_"))))); } void @@ -2719,7 +2705,7 @@ yywhole_loop(chop, split) block_append(NEW_CALL(NEW_GVAR(rb_intern("$_")), rb_intern("chop"), Qnil), eval_tree); } - eval_tree = NEW_WHILE(NEW_CALL(0,rb_intern("gets"),0),eval_tree); + eval_tree = NEW_WHILE(NEW_FCALL(rb_intern("gets"),0),eval_tree); } static struct op_tbl rb_op_tbl[] = { @@ -2790,7 +2776,7 @@ rb_intern(name) return id; id = ++last_id; - id <<= 3; + id <<= ID_SCOPE_SHIFT; switch (name[0]) { case '$': id |= ID_GLOBAL; @@ -492,6 +492,7 @@ Init_process() rb_define_private_method(C_Kernel, "sleep", Fsleep, -1); M_Process = rb_define_module("Process"); + rb_extend_object(M_Process, M_Process); rb_define_single_method(M_Process, "fork", Ffork, 0); rb_define_single_method(M_Process, "_exit", Ffork, 1); @@ -499,25 +500,25 @@ Init_process() rb_define_single_method(M_Process, "waitpid", Fwaitpid, 2); rb_define_single_method(M_Process, "kill", Fkill, -1); - rb_define_module_function(M_Process, "pid", get_pid, 0); - rb_define_module_function(M_Process, "ppid", get_ppid, 0); + rb_define_method(M_Process, "pid", get_pid, 0); + rb_define_method(M_Process, "ppid", get_ppid, 0); - rb_define_module_function(M_Process, "getpgrp", Fproc_getpgrp, -1); - rb_define_module_function(M_Process, "setpgrp", Fproc_setpgrp, 2); + rb_define_method(M_Process, "getpgrp", Fproc_getpgrp, -1); + rb_define_method(M_Process, "setpgrp", Fproc_setpgrp, 2); - rb_define_module_function(M_Process, "getpriority", Fproc_getpriority, 2); - rb_define_module_function(M_Process, "setpriority", Fproc_setpriority, 3); + rb_define_method(M_Process, "getpriority", Fproc_getpriority, 2); + rb_define_method(M_Process, "setpriority", Fproc_setpriority, 3); rb_define_const(M_Process, "PRIO_PROCESS", INT2FIX(PRIO_PROCESS)); rb_define_const(M_Process, "PRIO_PGRP", INT2FIX(PRIO_PGRP)); rb_define_const(M_Process, "PRIO_USER", INT2FIX(PRIO_USER)); - rb_define_module_function(M_Process, "uid", Fproc_getuid, 0); - rb_define_module_function(M_Process, "uid=", Fproc_setuid, 1); - rb_define_module_function(M_Process, "gid", Fproc_getgid, 0); - rb_define_module_function(M_Process, "gid=", Fproc_setgid, 1); - rb_define_module_function(M_Process, "euid", Fproc_geteuid, 0); - rb_define_module_function(M_Process, "euid=", Fproc_seteuid, 1); - rb_define_module_function(M_Process, "egid", Fproc_getegid, 0); - rb_define_module_function(M_Process, "egid=", Fproc_setegid, 1); + rb_define_method(M_Process, "uid", Fproc_getuid, 0); + rb_define_method(M_Process, "uid=", Fproc_setuid, 1); + rb_define_method(M_Process, "gid", Fproc_getgid, 0); + rb_define_method(M_Process, "gid=", Fproc_setgid, 1); + rb_define_method(M_Process, "euid", Fproc_geteuid, 0); + rb_define_method(M_Process, "euid=", Fproc_seteuid, 1); + rb_define_method(M_Process, "egid", Fproc_getegid, 0); + rb_define_method(M_Process, "egid=", Fproc_setegid, 1); } @@ -13,7 +13,6 @@ #include "ruby.h" #include "re.h" -/* Generate compiled regular expressions */ #if 'a' == 97 /* it's ascii */ static char casetable[] = { '\000', '\001', '\002', '\003', '\004', '\005', '\006', '\007', @@ -102,22 +101,16 @@ int len; rp = ALLOC(Regexp); MEMZERO((char *)rp, Regexp, 1); - rp->pat.buffer = ALLOC_N(char, 16); - rp->pat.allocated = 16; - rp->pat.fastmap = ALLOC_N(char, 256); + rp->buffer = ALLOC_N(char, 16); + rp->allocated = 16; + rp->fastmap = ALLOC_N(char, 256); - if ((err = re_compile_pattern(s, (size_t)len, &(rp->pat))) != NULL) + if ((err = re_compile_pattern(s, (size_t)len, rp)) != NULL) Fail("%s: /%s/", err, s); return rp; } -struct match { - UINT len; - char *ptr; - struct re_registers regs; -}; - struct match last_match; VALUE ignorecase; @@ -134,14 +127,20 @@ research(reg, str, start) if (FL_TEST(reg, FL_USER1)) { casefold = TRUE; } - if (casefold) - reg->ptr->pat.translate = casetable; - else - reg->ptr->pat.translate = NULL; + if (casefold) { + if (reg->ptr->translate != casetable) { + reg->ptr->translate = casetable; + reg->ptr->fastmap_accurate = 0; + } + } + else if (reg->ptr->translate) { + reg->ptr->translate = NULL; + reg->ptr->fastmap_accurate = 0; + } if (start > str->len) return -1; - result = re_search(&(reg->ptr->pat), str->ptr, str->len, - start, str->len - start, &(reg->ptr->regs)); + result = re_search(reg->ptr, str->ptr, str->len, + start, str->len - start, &last_match.regs); if (result >= 0) { last_match.len = str->len; @@ -153,24 +152,23 @@ research(reg, str, start) } memcpy(last_match.ptr, str->ptr, last_match.len); last_match.ptr[last_match.len] = '\0'; - last_match.regs = reg->ptr->regs; } return result; } -static VALUE -nth_match(nth) +VALUE +re_nth_match(nth) int nth; { int start, end, len; - if (nth >= RE_NREGS) { - Fail("match out of range %d, %d", nth, RE_NREGS); + if (nth >= last_match.regs.num_regs) { + return Qnil; } - start = last_match.regs.start[nth]; + start = BEG(nth); if (start == -1) return Qnil; - end = last_match.regs.end[nth]; + end = END(nth); len = end - start; return str_new(last_match.ptr + start, len); } @@ -179,7 +177,7 @@ VALUE re_last_match(id) ID id; { - return nth_match(0); + return re_nth_match(0); } static VALUE @@ -187,8 +185,8 @@ re_match_pre() { struct match *match; - if (last_match.regs.start[0] == -1) return Qnil; - return str_new(last_match.ptr, last_match.regs.start[0]); + if (BEG(0) == -1) return Qnil; + return str_new(last_match.ptr, BEG(0)); } static VALUE @@ -196,24 +194,22 @@ re_match_post() { struct match *match; - if (last_match.regs.start[0] == -1) return Qnil; - return str_new(last_match.ptr+last_match.regs.end[0], - last_match.len-last_match.regs.end[0]); + if (BEG(0) == -1) return Qnil; + return str_new(last_match.ptr+END(0), + last_match.len-END(0)); } static VALUE re_match_last() { - struct match *match; - int i; + int i, len; - if (last_match.regs.start[0] == -1) return Qnil; + if (BEG(0) == -1) return Qnil; - for (i=0; i<RE_NREGS; i++) { - if (last_match.regs.start[i] == -1) break; + for (i=last_match.regs.num_regs-1; BEG(i) == -1 && i > 0; i--) { } - if (i == RE_NREGS) return Qnil; - return nth_match(i-1); + if (i == 0) return Qnil; + return re_nth_match(i); } static VALUE @@ -221,7 +217,7 @@ get_match_data(id, nth) ID id; int nth; { - return nth_match(nth); + return re_nth_match(nth); } static void @@ -229,6 +225,10 @@ free_match(data) struct match *data; { free(data->ptr); + if (data->regs.allocated > 0) { + free(data->regs.beg); + free(data->regs.end); + } } static VALUE @@ -242,7 +242,8 @@ get_match() data->len = last_match.len; data->ptr = ALLOC_N(char, last_match.len+1); memcpy(data->ptr, last_match.ptr, data->len+1); - data->regs = last_match.regs; + data->regs.allocated = 0; + re_copy_registers(&data->regs, &last_match.regs); return data_new(data, free_match, Qnil); } @@ -280,8 +281,8 @@ void reg_free(rp) Regexp *rp; { - free(rp->pat.buffer); - free(rp->pat.fastmap); + free(rp->buffer); + free(rp->fastmap); free(rp); } @@ -472,10 +473,6 @@ re_regsub(str) if (c == '\\' && (*s == '\\' || *s == '&')) p = ++s; } else { - -#define BEG(no) last_match.regs.start[no] -#define END(no) last_match.regs.end[no] - if (BEG(no) == -1) continue; str_cat(val, last_match.ptr+BEG(no), END(no)-BEG(no)); } @@ -491,6 +488,19 @@ re_regsub(str) return val; } +static VALUE +kcode() +{ + switch (re_syntax_options & RE_MBCTYPE_MASK) { + case RE_MBCTYPE_SJIS: + return str_new2("SJIS"); + case RE_MBCTYPE_EUC: + return str_new2("EUC"); + default: + return str_new2("NONE"); + } +} + void rb_set_kanjicode(code) char *code; @@ -500,59 +510,46 @@ rb_set_kanjicode(code) switch (code[0]) { case 'E': case 'e': - obscure_syntax &= ~RE_MBCTYPE_MASK; - obscure_syntax |= RE_MBCTYPE_EUC; + re_syntax_options &= ~RE_MBCTYPE_MASK; + re_syntax_options |= RE_MBCTYPE_EUC; break; case 'S': case 's': - obscure_syntax &= ~RE_MBCTYPE_MASK; - obscure_syntax |= RE_MBCTYPE_SJIS; + re_syntax_options &= ~RE_MBCTYPE_MASK; + re_syntax_options |= RE_MBCTYPE_SJIS; break; default: case 'N': case 'n': set_no_conversion: - obscure_syntax &= ~RE_MBCTYPE_MASK; + re_syntax_options &= ~RE_MBCTYPE_MASK; break; } - re_set_syntax(obscure_syntax); + re_set_syntax(re_syntax_options); } -static VALUE -kanji_var_get() +void +rb_setup_kcode() { - switch (obscure_syntax & RE_MBCTYPE_MASK) { - case RE_MBCTYPE_SJIS: - return str_new2("SJIS"); - case RE_MBCTYPE_EUC: - return str_new2("EUC"); - default: - return Qnil; - } + rb_define_const(C_Object, "KCODE", kcode()); } -static void -kanji_var_set(val) - struct RString *val; -{ - if (val == Qnil) rb_set_kanjicode(Qnil); - Check_Type(val, T_STRING); - rb_set_kanjicode(val->ptr); -} +VALUE rb_readonly_hook(); void Init_Regexp() { int i; - obscure_syntax = RE_NO_BK_PARENS | RE_NO_BK_VBAR - | RE_CONTEXT_INDEP_OPS | RE_INTERVALS - | RE_NO_BK_CURLY_BRACES - | RE_MBCTYPE_EUC; - - for (i=0; i<RE_NREGS; i++) { - last_match.regs.start[i] = last_match.regs.end[i] = -1; - } + re_set_syntax(RE_NO_BK_PARENS | RE_NO_BK_VBAR + | RE_AWK_CLASS_HACK + | RE_INTERVALS + | RE_NO_BK_BRACES + | RE_BACKSLASH_ESCAPE_IN_LISTS +#ifdef DEFAULT_MBCTYPE + | DEFAULT_MBCTYPE +#endif +); rb_define_variable("$~", Qnil, get_match, set_match, 0); @@ -561,18 +558,6 @@ Init_Regexp() rb_define_variable("$'", Qnil, re_match_post, Qnil, 0); rb_define_variable("$+", Qnil, re_match_last, Qnil, 0); - rb_define_variable("$1", Qnil, get_match_data, Qnil, 1); - rb_define_variable("$2", Qnil, get_match_data, Qnil, 2); - rb_define_variable("$3", Qnil, get_match_data, Qnil, 3); - rb_define_variable("$4", Qnil, get_match_data, Qnil, 4); - rb_define_variable("$5", Qnil, get_match_data, Qnil, 5); - rb_define_variable("$6", Qnil, get_match_data, Qnil, 6); - rb_define_variable("$7", Qnil, get_match_data, Qnil, 7); - rb_define_variable("$8", Qnil, get_match_data, Qnil, 8); - rb_define_variable("$9", Qnil, get_match_data, Qnil, 9); - - rb_define_variable("$KANJI", Qnil, kanji_var_get, kanji_var_set, 0); - rb_define_variable("$=", &ignorecase, Qnil, Qnil, 0); C_Regexp = rb_define_class("Regexp", C_Object); @@ -18,10 +18,19 @@ #include <stdio.h> #include "regex.h" -typedef struct Regexp { - struct re_pattern_buffer pat; - struct re_registers regs; -} Regexp; + +typedef struct re_pattern_buffer Regexp; + +struct match { + UINT len; + char *ptr; + struct re_registers regs; +}; + +extern struct match last_match; + +#define BEG(no) last_match.regs.beg[no] +#define END(no) last_match.regs.end[no] VALUE re_regcomp(); VALUE re_regsub(); @@ -25,27 +25,15 @@ On the other hand, if you compile with both -Dtest and -Dcanned you can run some tests we've already thought of. */ -#include "config.h" - /* We write fatal error messages on standard error. */ #include <stdio.h> /* isalpha(3) etc. are used for the character classes. */ #include <ctype.h> +#include <sys/types.h> -#ifdef emacs - -/* The `emacs' switch turns on certain special matching commands - that make sense only in emacs. */ - -#include "lisp.h" -#include "buffer.h" -#include "syntax.h" - -#else /* not emacs */ - +#include "config.h" #include "defines.h" -#include <sys/types.h> #ifdef __STDC__ #define P(s) s @@ -74,6 +62,9 @@ char *alloca (); #endif #endif /* __GNUC__ */ +#define RE_ALLOCATE alloca +#define FREE_VARIABLES() alloca (0) + #define FREE_AND_RETURN_VOID(stackb) return #define FREE_AND_RETURN(stackb,val) return(val) #define DOUBLE_STACK(stackx,stackb,len) \ @@ -82,12 +73,38 @@ char *alloca (); /* Only copy what is in use. */ \ (unsigned char **) memcpy (stackx, stackb, len * sizeof (char *))) #else /* NO_ALLOCA defined */ + +#define RE_ALLOCATE malloc + +#define FREE_VAR(var) if (var) free (var); var = NULL +#define FREE_VARIABLES() \ + do { \ + FREE_VAR (regstart); \ + FREE_VAR (regend); \ + FREE_VAR (best_regstart); \ + FREE_VAR (best_regend); \ + FREE_VAR (reg_info); \ + } while (0) + #define FREE_AND_RETURN_VOID(stackb) free(stackb);return #define FREE_AND_RETURN(stackb,val) free(stackb);return(val) #define DOUBLE_STACK(stackx,stackb,len) \ (unsigned char **)xrealloc (stackb, 2 * len * sizeof (unsigned char *)) #endif /* NO_ALLOCA */ +#define RE_TALLOC(n,t) ((t*)RE_ALLOCATE((n)*sizeof(t))) +#define TMALLOC(n,t) ((t*)xmalloc((n)*sizeof(t))) +#define TREALLOC(s,n,t) (s=((t*)xrealloc(s,(n)*sizeof(t)))) + +/* Get the interface, including the syntax bits. */ +#include "regex.h" + +static int re_search_2 P((struct re_pattern_buffer *, char *, int, + char *, int, int, int, + struct re_registers *)); +static int re_match_2 P((struct re_pattern_buffer *, char *, int, + char *, int, int, struct re_registers *)); + static void store_jump P((char *, int, char *)); static void insert_jump P((int, char *, char *, char *)); static void store_jump_n P((char *, int, char *, unsigned)); @@ -95,7 +112,6 @@ static void insert_jump_n P((int, char *, char *, char *, unsigned)); static void insert_op_2 P((int, char *, char *, int, int )); static int memcmp_translate P((unsigned char *, unsigned char *, int, unsigned char *)); -long re_set_syntax P((long)); /* Define the syntax stuff, so we can do the \<, \>, etc. */ @@ -107,16 +123,12 @@ long re_set_syntax P((long)); #define SYNTAX(c) re_syntax_table[c] - -#ifdef SYNTAX_TABLE - -char *re_syntax_table; - -#else /* not SYNTAX_TABLE */ - static char re_syntax_table[256]; static void init_syntax_once P((void)); +#undef P + +#include "util.h" static void init_syntax_once () @@ -147,20 +159,11 @@ init_syntax_once () done = 1; } -#endif /* SYNTAX_TABLE */ -#undef P -#endif /* emacs */ - - /* Sequents are missing isgraph. */ #ifndef isgraph #define isgraph(c) (isprint((c)) && !isspace((c))) #endif -/* Get the interface, including the syntax bits. */ -#include "regex.h" - - /* These are the command codes that appear in compiled regular expressions, one per byte. Some command codes are followed by argument bytes. A command code can specify any interpretation @@ -226,22 +229,10 @@ enum regexpcode duplicate, /* Match a duplicate of something remembered. Followed by one byte containing the index of the memory register. */ - before_dot, /* Succeeds if before point. */ - at_dot, /* Succeeds if at point. */ - after_dot, /* Succeeds if after point. */ - begbuf, /* Succeeds if at beginning of buffer. */ - endbuf, /* Succeeds if at end of buffer. */ wordchar, /* Matches any word-constituent character. */ notwordchar, /* Matches any char that is not a word-constituent. */ - wordbeg, /* Succeeds if at word beginning. */ - wordend, /* Succeeds if at word end. */ wordbound, /* Succeeds if at a word boundary. */ notwordbound,/* Succeeds if not at a word boundary. */ - syntaxspec, /* Matches any character whose syntax is specified. - followed by a byte which contains a syntax code, - e.g., Sword. */ - notsyntaxspec /* Matches any character whose syntax differs from - that specified. */ }; @@ -301,39 +292,20 @@ re_set_syntax (syntax) { long ret; - ret = obscure_syntax; - obscure_syntax = syntax; + ret = re_syntax_options; + re_syntax_options = syntax; return ret; } /* Set by re_set_syntax to the current regexp syntax to recognize. */ -#ifdef EUC -#define DEFAULT_MBCTYPE RE_MBCTYPE_EUC -#else -#ifdef SJIS -#define DEFAULT_MBCTYPE RE_MBCTYPE_SJIS -#else -#define DEFAULT_MBCTYPE 0 -#endif -#endif -long obscure_syntax = DEFAULT_MBCTYPE; +long re_syntax_options = DEFAULT_MBCTYPE; /* Macros for re_compile_pattern, which is found below these definitions. */ #define CHAR_CLASS_MAX_LENGTH 6 - -/* Fetch the next character in the uncompiled pattern, translating it if - necessary. */ +/* Fetch the next character in the uncompiled pattern. */ #define PATFETCH(c) \ - {if (p == pend) goto end_of_pattern; \ - c = * (unsigned char *) p++; \ - if (translate && !ismbchar (c)) \ - c = (unsigned char) translate[(unsigned char) c]; } - -/* Fetch the next character in the uncompiled pattern, with no - translation. */ -#define PATFETCH_RAW(c) \ {if (p == pend) goto end_of_pattern; \ c = * (unsigned char *) p++; } @@ -436,20 +408,13 @@ long obscure_syntax = DEFAULT_MBCTYPE; }; */ static void -#ifdef __STDC__ -set_list_bits (unsigned short c1, unsigned short c2, - unsigned char *b, const unsigned char *translate) -#else -set_list_bits (c1, c2, b, translate) - unsigned short c1, c2; - unsigned char *b; - const unsigned char *translate; -#endif +set_list_bits (c1, c2, b) + unsigned short c1, c2; + unsigned char *b; { - enum regexpcode op = (enum regexpcode) b[-2]; unsigned char sbc_size = b[-1]; unsigned short mbc_size = EXTRACT_UNSIGNED (&b[sbc_size]); - unsigned short c, beg, end, upb; + unsigned short beg, end, upb; if (c1 > c2) return; @@ -464,15 +429,9 @@ set_list_bits (c1, c2, b, translate) memset (&b[sbc_size], 0, (1 << BYTEWIDTH) / BYTEWIDTH - sbc_size); b[-1] = sbc_size = (1 << BYTEWIDTH) / BYTEWIDTH; } - if (!translate) { - for (; c1 <= upb; c1++) - if (!ismbchar (c1)) - SET_LIST_BIT (c1); - } - else - for (; c1 <= upb; c1++) + for (; c1 <= upb; c1++) if (!ismbchar (c1)) - SET_LIST_BIT (translate[c1]); + SET_LIST_BIT (c1); if (c2 < 1 << BYTEWIDTH) return; c1 = 0x8000; /* The first wide char */ @@ -513,41 +472,37 @@ set_list_bits (c1, c2, b, translate) } static int -#ifdef __STDC__ -is_in_list (unsigned short c, const unsigned char *b) -#else is_in_list (c, b) - unsigned short c; - const unsigned char *b; -#endif + unsigned short c; + const unsigned char *b; { - unsigned short size; - int in = (enum regexpcode) b[-1] == charset_not; + unsigned short size; - size = *b++; - if (c < 1 << BYTEWIDTH) { - if (c / BYTEWIDTH < size && b[c / BYTEWIDTH] & 1 << c % BYTEWIDTH) - in = !in; - } - else { - unsigned short i, j; + size = *b++; + if (c < 1 << BYTEWIDTH) { + if (c / BYTEWIDTH < size && b[c / BYTEWIDTH] & 1 << c % BYTEWIDTH) { + return 1; + } + } + else { + unsigned short i, j; - b += size + 2; - size = EXTRACT_UNSIGNED (&b[-2]); + b += size + 2; + size = EXTRACT_UNSIGNED (&b[-2]); - for (i = 0, j = size; i < j; ) { - unsigned short k = (i + j) >> 1; + for (i = 0, j = size; i < j; ) { + unsigned short k = (i + j) >> 1; - if (c > EXTRACT_MBC (&b[k*4 + 2])) - i = k + 1; - else - j = k; + if (c > EXTRACT_MBC (&b[k*4 + 2])) + i = k + 1; + else + j = k; + } + if (i < size && EXTRACT_MBC (&b[i*4]) <= c + && ((unsigned char) c != '\n' && (unsigned char) c != '\0')) + return 1; } - if (i < size && EXTRACT_MBC (&b[i*4]) <= c - && ((unsigned char) c != '\n' && (unsigned char) c != '\0')) - in = !in; - } - return in; + return 0; } /* re_compile_pattern takes a regular-expression string @@ -571,88 +526,85 @@ re_compile_pattern (pattern, size, bufp) size_t size; struct re_pattern_buffer *bufp; { - register char *b = bufp->buffer; - register char *p = pattern; - char *pend = pattern + size; - register unsigned c, c1; - char *p0; - unsigned char *translate = (unsigned char *) bufp->translate; - - /* Address of the count-byte of the most recently inserted `exactn' - command. This makes it possible to tell whether a new exact-match - character can be added to that command or requires a new `exactn' - command. */ - - char *pending_exact = 0; - - /* Address of the place where a forward-jump should go to the end of - the containing expression. Each alternative of an `or', except the - last, ends with a forward-jump of this sort. */ - - char *fixup_jump = 0; - - /* Address of start of the most recently finished expression. - This tells postfix * where to find the start of its operand. */ - - char *laststart = 0; - - /* In processing a repeat, 1 means zero matches is allowed. */ - - char zero_times_ok; - - /* In processing a repeat, 1 means many matches is allowed. */ - - char many_times_ok; - - /* Address of beginning of regexp, or inside of last \(. */ - - char *begalt = b; - - /* In processing an interval, at least this many matches must be made. */ - int lower_bound; - - /* In processing an interval, at most this many matches can be made. */ - int upper_bound; - - /* Place in pattern (i.e., the {) to which to go back if the interval - is invalid. */ - char *beg_interval = 0; - - /* Stack of information saved by \( and restored by \). - Four stack elements are pushed by each \(: + register char *b = bufp->buffer; + register char *p = pattern; + char *pend = pattern + size; + register unsigned c, c1; + char *p0; + int numlen; + + /* Address of the count-byte of the most recently inserted `exactn' + command. This makes it possible to tell whether a new exact-match + character can be added to that command or requires a new `exactn' + command. */ + + char *pending_exact = 0; + + /* Address of the place where a forward-jump should go to the end of + the containing expression. Each alternative of an `or', except the + last, ends with a forward-jump of this sort. */ + + char *fixup_jump = 0; + + /* Address of start of the most recently finished expression. + This tells postfix * where to find the start of its operand. */ + + char *laststart = 0; + + /* In processing a repeat, 1 means zero matches is allowed. */ + + char zero_times_ok; + + /* In processing a repeat, 1 means many matches is allowed. */ + + char many_times_ok; + + /* Address of beginning of regexp, or inside of last \(. */ + + char *begalt = b; + + /* In processing an interval, at least this many matches must be made. */ + int lower_bound; + + /* In processing an interval, at most this many matches can be made. */ + int upper_bound; + + /* Place in pattern (i.e., the {) to which to go back if the interval + is invalid. */ + char *beg_interval = 0; + + /* Stack of information saved by \( and restored by \). + Four stack elements are pushed by each \(: First, the value of b. Second, the value of fixup_jump. Third, the value of regnum. Fourth, the value of begalt. */ + + int stackb[40]; + int *stackp = stackb; + int *stacke = stackb + 40; + int *stackt; + + /* Counts \('s as they are encountered. Remembered for the matching \), + where it becomes the register number to put in the stop_memory + command. */ + + int regnum = 1; + int range = 0; - int stackb[40]; - int *stackp = stackb; - int *stacke = stackb + 40; - int *stackt; - - /* Counts \('s as they are encountered. Remembered for the matching \), - where it becomes the register number to put in the stop_memory - command. */ - - int regnum = 1; - - bufp->fastmap_accurate = 0; - -#ifndef emacs -#ifndef SYNTAX_TABLE - /* Initialize the syntax table. */ - init_syntax_once(); -#endif -#endif - - if (bufp->allocated == 0) - { - bufp->allocated = INIT_BUF_SIZE; - if (bufp->buffer) - /* EXTEND_BUFFER loses when bufp->allocated is 0. */ - bufp->buffer = (char *) xrealloc (bufp->buffer, INIT_BUF_SIZE); - else - /* Caller did not allocate a buffer. Do it for them. */ + bufp->fastmap_accurate = 0; + + /* Initialize the syntax table. */ + init_syntax_once(); + + if (bufp->allocated == 0) + { + bufp->allocated = INIT_BUF_SIZE; + if (bufp->buffer) + /* EXTEND_BUFFER loses when bufp->allocated is 0. */ + bufp->buffer = (char *) xrealloc (bufp->buffer, INIT_BUF_SIZE); + else + /* Caller did not allocate a buffer. Do it for them. */ bufp->buffer = (char *) xmalloc (INIT_BUF_SIZE); if (!bufp->buffer) goto memory_exhausted; begalt = b = bufp->buffer; @@ -669,23 +621,18 @@ re_compile_pattern (pattern, size, bufp) char *p1 = p; /* When testing what follows the $, look past the \-constructs that don't consume anything. */ - if (! (obscure_syntax & RE_CONTEXT_INDEP_OPS)) + if (! (re_syntax_options & RE_CONTEXT_INDEP_OPS)) while (p1 != pend) { if (*p1 == '\\' && p1 + 1 != pend - && (p1[1] == '<' || p1[1] == '>' - || p1[1] == '`' || p1[1] == '\'' -#ifdef emacs - || p1[1] == '=' -#endif - || p1[1] == 'b' || p1[1] == 'B')) + && (p1[1] == 'b' || p1[1] == 'B')) p1 += 2; else break; } - if (obscure_syntax & RE_TIGHT_VBAR) + if (re_syntax_options & RE_TIGHT_VBAR) { - if (! (obscure_syntax & RE_CONTEXT_INDEP_OPS) && p1 != pend) + if (! (re_syntax_options & RE_CONTEXT_INDEP_OPS) && p1 != pend) goto normal_char; /* Make operand of last vbar end before this `$'. */ if (fixup_jump) @@ -697,14 +644,14 @@ re_compile_pattern (pattern, size, bufp) /* $ means succeed if at end of line, but only in special contexts. If validly in the middle of a pattern, it is a normal character. */ - if ((obscure_syntax & RE_CONTEXTUAL_INVALID_OPS) && p1 != pend) + if ((re_syntax_options & RE_CONTEXTUAL_INVALID_OPS) && p1 != pend) goto invalid_pattern; if (p1 == pend || *p1 == '\n' - || (obscure_syntax & RE_CONTEXT_INDEP_OPS) - || (obscure_syntax & RE_NO_BK_PARENS + || (re_syntax_options & RE_CONTEXT_INDEP_OPS) + || (re_syntax_options & RE_NO_BK_PARENS ? *p1 == ')' : *p1 == '\\' && p1[1] == ')') - || (obscure_syntax & RE_NO_BK_VBAR + || (re_syntax_options & RE_NO_BK_VBAR ? *p1 == '|' : *p1 == '\\' && p1[1] == '|')) { @@ -717,15 +664,15 @@ re_compile_pattern (pattern, size, bufp) /* ^ means succeed if at beg of line, but only if no preceding pattern. */ - if ((obscure_syntax & RE_CONTEXTUAL_INVALID_OPS) && laststart) + if ((re_syntax_options & RE_CONTEXTUAL_INVALID_OPS) && laststart) goto invalid_pattern; if (laststart && p - 2 >= pattern && p[-2] != '\n' - && !(obscure_syntax & RE_CONTEXT_INDEP_OPS)) + && !(re_syntax_options & RE_CONTEXT_INDEP_OPS)) goto normal_char; - if (obscure_syntax & RE_TIGHT_VBAR) + if (re_syntax_options & RE_TIGHT_VBAR) { if (p != pattern + 1 - && ! (obscure_syntax & RE_CONTEXT_INDEP_OPS)) + && ! (re_syntax_options & RE_CONTEXT_INDEP_OPS)) goto normal_char; BUFPUSH (begline); begalt = b; @@ -736,17 +683,17 @@ re_compile_pattern (pattern, size, bufp) case '+': case '?': - if ((obscure_syntax & RE_BK_PLUS_QM) - || (obscure_syntax & RE_LIMITED_OPS)) + if ((re_syntax_options & RE_BK_PLUS_QM) + || (re_syntax_options & RE_LIMITED_OPS)) goto normal_char; handle_plus: case '*': /* If there is no previous pattern, char not special. */ if (!laststart) { - if (obscure_syntax & RE_CONTEXTUAL_INVALID_OPS) + if (re_syntax_options & RE_CONTEXTUAL_INVALID_OPS) goto invalid_pattern; - else if (! (obscure_syntax & RE_CONTEXT_INDEP_OPS)) + else if (! (re_syntax_options & RE_CONTEXT_INDEP_OPS)) goto normal_char; } /* If there is a sequence of repetition chars, @@ -762,10 +709,10 @@ re_compile_pattern (pattern, size, bufp) PATFETCH (c); if (c == '*') ; - else if (!(obscure_syntax & RE_BK_PLUS_QM) + else if (!(re_syntax_options & RE_BK_PLUS_QM) && (c == '+' || c == '?')) ; - else if ((obscure_syntax & RE_BK_PLUS_QM) + else if ((re_syntax_options & RE_BK_PLUS_QM) && c == '\\') { /* int c1; */ @@ -846,7 +793,7 @@ re_compile_pattern (pattern, size, bufp) /* Clear the whole map */ memset (b, 0, (1 << BYTEWIDTH) / BYTEWIDTH + 2); - if ((obscure_syntax & RE_HAT_NOT_NEWLINE) && b[-2] == charset_not) + if ((re_syntax_options & RE_HAT_NOT_NEWLINE) && b[-2] == charset_not) SET_LIST_BIT ('\n'); @@ -854,6 +801,7 @@ re_compile_pattern (pattern, size, bufp) while (1) { int size; + unsigned last = -1; if ((size = EXTRACT_UNSIGNED (&b[(1 << BYTEWIDTH) / BYTEWIDTH]))) { /* Ensure the space is enough to hold another interval @@ -862,150 +810,131 @@ re_compile_pattern (pattern, size, bufp) while (b + size + 1 > bufp->buffer + bufp->allocated) EXTEND_BUFFER; } - /* Don't translate while fetching, in case it's a range bound. - When we set the bit for the character, we translate it. */ - PATFETCH_RAW (c); - - /* If set, \ escapes characters when inside [...]. */ - if ((obscure_syntax & RE_AWK_CLASS_HACK) && c == '\\') - { - PATFETCH(c1); - if (ismbchar (c1)) { - unsigned char c2; + range_retry: + PATFETCH (c); - PATFETCH_RAW (c2); - c1 = c1 << 8 | c2; - set_list_bits (c1, c1, (unsigned char *) b, translate); - continue; - } - SET_LIST_BIT (c1); - continue; - } - if (c == ']') - { - if (p == p0 + 1) - { + if (c == ']') { + if (p == p0 + 1) { /* If this is an empty bracket expression. */ - if ((obscure_syntax & RE_NO_EMPTY_BRACKETS) + if ((re_syntax_options & RE_NO_EMPTY_BRACKETS) && p == pend) - goto invalid_pattern; - } + goto invalid_pattern; + } else /* Stop if this isn't merely a ] inside a bracket expression, but rather the end of a bracket expression. */ - break; - } + break; + } if (ismbchar (c)) { - unsigned char c2; - - PATFETCH_RAW (c2); - c = c << 8 | c2; + PATFETCH (c1); + c = c << 8 | c1; } - /* Get a range. */ - if (p[0] == '-' && p[1] != ']') - { - PATFETCH_RAW (c1); - /* Don't translate the range bounds while fetching them. */ - PATFETCH_RAW (c1); - if (ismbchar (c)) { - unsigned char c2; - PATFETCH_RAW (c2); - c1 = c1 << 8 | c2; + /* \ escapes characters when inside [...]. */ + if (c == '\\') { + PATFETCH(c); + switch (c) { + case 'w': + for (c = 0; c < 256; c++) + if (isalnum(c)) + SET_LIST_BIT (c); + last = -1; + continue; + + case 'W': + if (re_syntax_options & RE_MBCTYPE_MASK) + goto invalid_char; + for (c = 0; c < 256; c++) + if (!isalnum(c)) + SET_LIST_BIT (c); + last = -1; + continue; + + case 's': + for (c = 0; c < 256; c++) + if (isspace(c)) + SET_LIST_BIT (c); + last = -1; + continue; + + case 'S': + if (re_syntax_options & RE_MBCTYPE_MASK) + goto invalid_char; + for (c = 0; c < 256; c++) + if (!isspace(c)) + SET_LIST_BIT (c); + last = -1; + continue; + + case 'd': + for (c = '0'; c <= '9'; c++) + SET_LIST_BIT (c); + last = -1; + continue; + + case 'D': + if (re_syntax_options & RE_MBCTYPE_MASK) + goto invalid_char; + for (c = 0; c < '0'; c++) + SET_LIST_BIT (c); + for (c = '9' + 1; c < 256; c++) + SET_LIST_BIT (c); + last = -1; + continue; + + case 'x': + c = scan_hex(p, 2, &numlen); + if ((re_syntax_options & RE_MBCTYPE_MASK) && (c > 0x7f)) + goto invalid_char; + p += numlen; + break; + + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + PATUNFETCH; + c = scan_oct(p, 3, &numlen); + if ((re_syntax_options & RE_MBCTYPE_MASK) && (c > 0x7f)) + goto invalid_char; + p += numlen; + break; + + default: + if (ismbchar (c)) { + PATFETCH(c1); + c = c << 8 | c1; + } + break; } - - if ((obscure_syntax & RE_NO_EMPTY_RANGES) && c > c1) - goto invalid_pattern; - - if ((obscure_syntax & RE_NO_HYPHEN_RANGE_END) - && c1 == '-' && *p != ']') + } + + /* Get a range. */ + if (range) { + if (last > c) goto invalid_pattern; - - set_list_bits (c, c1, (unsigned char *) b, translate); - } - else if ((obscure_syntax & RE_CHAR_CLASSES) - && c == '[' && p[0] == ':') - { - /* Longest valid character class word has six characters. */ - char str[CHAR_CLASS_MAX_LENGTH]; - PATFETCH_RAW (c); - c1 = 0; - /* If no ] at end. */ - if (p == pend) + + if ((re_syntax_options & RE_NO_HYPHEN_RANGE_END) + && c == '-' && *p != ']') goto invalid_pattern; - while (1) - { - /* Don't translate the ``character class'' characters. */ - PATFETCH_RAW (c); - if (c == ':' || c == ']' || p == pend - || c1 == CHAR_CLASS_MAX_LENGTH) - break; - str[c1++] = c; - } - str[c1] = '\0'; - if (p == pend - || c == ']' /* End of the bracket expression. */ - || p[0] != ']' - || p + 1 == pend - || (strcmp (str, "alpha") != 0 - && strcmp (str, "upper") != 0 - && strcmp (str, "lower") != 0 - && strcmp (str, "digit") != 0 - && strcmp (str, "alnum") != 0 - && strcmp (str, "xdigit") != 0 - && strcmp (str, "space") != 0 - && strcmp (str, "print") != 0 - && strcmp (str, "punct") != 0 - && strcmp (str, "graph") != 0 - && strcmp (str, "cntrl") != 0)) - { - /* Undo the ending character, the letters, and leave - the leading : and [ (but set bits for them). */ - c1++; - while (c1--) - PATUNFETCH; -#if 1 /* The original was: */ - SET_LIST_BIT ('['); - SET_LIST_BIT (':'); -#else /* I think this is the right way. */ - if (translate) { - SET_LIST_BIT (translate['[']); - SET_LIST_BIT (trasnlate[':']); - } - else { - SET_LIST_BIT ('['); - SET_LIST_BIT (':'); - } -#endif - } - else - { - /* The ] at the end of the character class. */ - PATFETCH (c); - if (c != ']') - goto invalid_pattern; - for (c = 0; c < (1 << BYTEWIDTH); c++) - { - if ((strcmp (str, "alpha") == 0 && isalpha (c)) - || (strcmp (str, "upper") == 0 && isupper (c)) - || (strcmp (str, "lower") == 0 && islower (c)) - || (strcmp (str, "digit") == 0 && isdigit (c)) - || (strcmp (str, "alnum") == 0 && isalnum (c)) - || (strcmp (str, "xdigit") == 0 && isxdigit (c)) - || (strcmp (str, "space") == 0 && isspace (c)) - || (strcmp (str, "print") == 0 && isprint (c)) - || (strcmp (str, "punct") == 0 && ispunct (c)) - || (strcmp (str, "graph") == 0 && isgraph (c)) - || (strcmp (str, "cntrl") == 0 && iscntrl (c))) - SET_LIST_BIT (c); - } - } - } - else if (translate && c < 1 << BYTEWIDTH) - SET_LIST_BIT (translate[c]); + + range = 0; + if (last < 1 << BYTEWIDTH && c < 1 << BYTEWIDTH) { + for (;last<=c;last++) + SET_LIST_BIT (last); + } + else + set_list_bits (last, c, (unsigned char*)b); + } + else if (p[0] == '-' && p[1] != ']') { + last = c; + PATFETCH(c1); + range = 1; + goto range_retry; + } + else if (c < 1 << BYTEWIDTH) + SET_LIST_BIT (c); else - set_list_bits (c, c, (unsigned char *) b, translate); + set_list_bits (c, c, (unsigned char*)b); } /* Discard any character set/class bitmap bytes that are all @@ -1019,46 +948,46 @@ re_compile_pattern (pattern, size, bufp) break; case '(': - if (! (obscure_syntax & RE_NO_BK_PARENS)) + if (! (re_syntax_options & RE_NO_BK_PARENS)) goto normal_char; else goto handle_open; case ')': - if (! (obscure_syntax & RE_NO_BK_PARENS)) + if (! (re_syntax_options & RE_NO_BK_PARENS)) goto normal_char; else goto handle_close; case '\n': - if (! (obscure_syntax & RE_NEWLINE_OR)) + if (! (re_syntax_options & RE_NEWLINE_OR)) goto normal_char; else goto handle_bar; case '|': - if ((obscure_syntax & RE_CONTEXTUAL_INVALID_OPS) + if ((re_syntax_options & RE_CONTEXTUAL_INVALID_OPS) && (! laststart || p == pend)) goto invalid_pattern; - else if (! (obscure_syntax & RE_NO_BK_VBAR)) + else if (! (re_syntax_options & RE_NO_BK_VBAR)) goto normal_char; else goto handle_bar; case '{': - if (! ((obscure_syntax & RE_NO_BK_CURLY_BRACES) - && (obscure_syntax & RE_INTERVALS))) + if (! ((re_syntax_options & RE_NO_BK_CURLY_BRACES) + && (re_syntax_options & RE_INTERVALS))) goto normal_char; else goto handle_interval; case '\\': if (p == pend) goto invalid_pattern; - PATFETCH_RAW (c); + PATFETCH(c); switch (c) { case '(': - if (obscure_syntax & RE_NO_BK_PARENS) + if (re_syntax_options & RE_NO_BK_PARENS) goto normal_backsl; handle_open: if (stackp == stacke) goto nesting_too_deep; @@ -1066,11 +995,8 @@ re_compile_pattern (pattern, size, bufp) /* Laststart should point to the start_memory that we are about to push (unless the pattern has RE_NREGS or more ('s). */ *stackp++ = b - bufp->buffer; - if (regnum < RE_NREGS) - { - BUFPUSH (start_memory); - BUFPUSH (regnum); - } + BUFPUSH (start_memory); + BUFPUSH (regnum); *stackp++ = fixup_jump ? fixup_jump - bufp->buffer + 1 : 0; *stackp++ = regnum++; *stackp++ = begalt - bufp->buffer; @@ -1080,29 +1006,26 @@ re_compile_pattern (pattern, size, bufp) break; case ')': - if (obscure_syntax & RE_NO_BK_PARENS) + if (re_syntax_options & RE_NO_BK_PARENS) goto normal_backsl; handle_close: if (stackp == stackb) goto unmatched_close; begalt = *--stackp + bufp->buffer; if (fixup_jump) store_jump (fixup_jump, jump, b); - if (stackp[-1] < RE_NREGS) - { - BUFPUSH (stop_memory); - BUFPUSH (stackp[-1]); - } + BUFPUSH (stop_memory); + BUFPUSH (stackp[-1]); stackp -= 2; fixup_jump = *stackp ? *stackp + bufp->buffer - 1 : 0; laststart = *--stackp + bufp->buffer; break; case '|': - if ((obscure_syntax & RE_LIMITED_OPS) - || (obscure_syntax & RE_NO_BK_VBAR)) + if ((re_syntax_options & RE_LIMITED_OPS) + || (re_syntax_options & RE_NO_BK_VBAR)) goto normal_backsl; handle_bar: - if (obscure_syntax & RE_LIMITED_OPS) + if (re_syntax_options & RE_LIMITED_OPS) goto normal_char; /* Insert before the previous alternative a jump which jumps to this alternative if the former fails. */ @@ -1130,10 +1053,10 @@ re_compile_pattern (pattern, size, bufp) break; case '{': - if (! (obscure_syntax & RE_INTERVALS) + if (! (re_syntax_options & RE_INTERVALS) /* Let \{ be a literal. */ - || ((obscure_syntax & RE_INTERVALS) - && (obscure_syntax & RE_NO_BK_CURLY_BRACES)) + || ((re_syntax_options & RE_INTERVALS) + && (re_syntax_options & RE_NO_BK_CURLY_BRACES)) /* If it's the string "\{". */ || (p - 2 == pattern && p == pend)) goto normal_backsl; @@ -1142,7 +1065,7 @@ re_compile_pattern (pattern, size, bufp) /* If there is no previous pattern, this isn't an interval. */ if (!laststart) { - if (obscure_syntax & RE_CONTEXTUAL_INVALID_OPS) + if (re_syntax_options & RE_CONTEXTUAL_INVALID_OPS) goto invalid_pattern; else goto normal_backsl; @@ -1156,15 +1079,15 @@ re_compile_pattern (pattern, size, bufp) || *laststart == charset_not || *laststart == start_memory || (*laststart == exactn && laststart[1] == 1) - || (! (obscure_syntax & RE_NO_BK_REFS) + || (! (re_syntax_options & RE_NO_BK_REFS) && *laststart == duplicate))) { - if (obscure_syntax & RE_NO_BK_CURLY_BRACES) + if (re_syntax_options & RE_NO_BK_CURLY_BRACES) goto normal_char; /* Posix extended syntax is handled in previous statement; this is for Posix basic syntax. */ - if (obscure_syntax & RE_INTERVALS) + if (re_syntax_options & RE_INTERVALS) goto invalid_pattern; goto normal_backsl; @@ -1180,7 +1103,7 @@ re_compile_pattern (pattern, size, bufp) } if (upper_bound < 0) upper_bound = lower_bound; - if (! (obscure_syntax & RE_NO_BK_CURLY_BRACES)) + if (! (re_syntax_options & RE_NO_BK_CURLY_BRACES)) { if (c != '\\') goto invalid_pattern; @@ -1188,10 +1111,10 @@ re_compile_pattern (pattern, size, bufp) } if (c != '}' || lower_bound < 0 || upper_bound > RE_DUP_MAX || lower_bound > upper_bound - || ((obscure_syntax & RE_NO_BK_CURLY_BRACES) + || ((re_syntax_options & RE_NO_BK_CURLY_BRACES) && p != pend && *p == '{')) { - if (obscure_syntax & RE_NO_BK_CURLY_BRACES) + if (re_syntax_options & RE_NO_BK_CURLY_BRACES) goto unfetch_interval; else goto invalid_pattern; @@ -1273,27 +1196,6 @@ re_compile_pattern (pattern, size, bufp) goto normal_char; break; -#ifdef emacs - case '=': - BUFPUSH (at_dot); - break; - - case 's': - laststart = b; - BUFPUSH (syntaxspec); - PATFETCH (c); - BUFPUSH (syntax_spec_code[c]); - break; - - case 'S': - laststart = b; - BUFPUSH (notsyntaxspec); - PATFETCH (c); - BUFPUSH (syntax_spec_code[c]); - break; -#endif /* emacs */ - -#ifdef RUBY case 's': case 'S': case 'd': @@ -1334,7 +1236,6 @@ re_compile_pattern (pattern, size, bufp) 2 + EXTRACT_UNSIGNED (&b[(1 << BYTEWIDTH) / BYTEWIDTH])*4); b += b[-1] + 2 + EXTRACT_UNSIGNED (&b[b[-1]])*4; break; -#endif /* RUBY */ case 'w': laststart = b; @@ -1345,15 +1246,7 @@ re_compile_pattern (pattern, size, bufp) laststart = b; BUFPUSH (notwordchar); break; -#ifndef RUBY - case '<': - BUFPUSH (wordbeg); - break; - case '>': - BUFPUSH (wordend); - break; -#endif /* RUBY */ case 'b': BUFPUSH (wordbound); break; @@ -1362,33 +1255,50 @@ re_compile_pattern (pattern, size, bufp) BUFPUSH (notwordbound); break; - case '`': - BUFPUSH (begbuf); - break; + /* hex */ + case 'x': + c1 = 0; + c = scan_hex(p, 2, &numlen); + if ((re_syntax_options & RE_MBCTYPE_MASK) && (c > 0x7f)) + goto invalid_char; + p += numlen; + goto numeric_char; + + /* octal */ + case '0': + c1 = 0; + c = scan_oct(p, 3, &numlen); + p += numlen; + goto numeric_char; + + /* back-ref or octal */ + case '1': case '2': case '3': + case '4': case '5': case '6': + case '7': case '8': case '9': + { + char *p_save; - case '\'': - BUFPUSH (endbuf); - break; + PATUNFETCH; + p_save = p; + + c1 = 0; + GET_UNSIGNED_NUMBER(c1); + + if (c1 >= regnum) { + if (c1 < RE_NREGS) + goto invalid_pattern; + + /* need to get octal */ + p = p_save; + c = scan_oct(p_save, 3, &numlen); + if ((re_syntax_options & RE_MBCTYPE_MASK) && (c > 0x7f)) + goto invalid_char; + p = p_save + numlen; + c1 = 0; + goto numeric_char; + } + } - case '1': - case '2': - case '3': - case '4': - case '5': - case '6': - case '7': - case '8': - case '9': - if (obscure_syntax & RE_NO_BK_REFS) - goto normal_char; - c1 = c - '0'; - if (c1 >= regnum) - { - if (obscure_syntax & RE_NO_EMPTY_BK_REF) - goto invalid_pattern; - else - goto normal_char; - } /* Can't back reference to a subexpression if inside of it. */ for (stackt = stackp - 2; stackt > stackb; stackt -= 4) if (*stackt == c1) @@ -1400,7 +1310,7 @@ re_compile_pattern (pattern, size, bufp) case '+': case '?': - if (obscure_syntax & RE_BK_PLUS_QM) + if (re_syntax_options & RE_BK_PLUS_QM) goto handle_plus; else goto normal_backsl; @@ -1408,10 +1318,6 @@ re_compile_pattern (pattern, size, bufp) default: normal_backsl: - /* You might think it would be useful for \ to mean - not to translate; but if we don't translate it - it will never match anything. */ - if (translate && !ismbchar (c)) c = (unsigned char) translate[c]; goto normal_char; } break; @@ -1421,16 +1327,17 @@ re_compile_pattern (pattern, size, bufp) c1 = 0; if (ismbchar (c)) { c1 = c; - PATFETCH_RAW (c); + PATFETCH(c); } + numeric_char: if (!pending_exact || pending_exact + *pending_exact + 1 != b || *pending_exact >= (c1 ? 0176 : 0177) || *p == '*' || *p == '^' - || ((obscure_syntax & RE_BK_PLUS_QM) + || ((re_syntax_options & RE_BK_PLUS_QM) ? *p == '\\' && (p[1] == '+' || p[1] == '?') : (*p == '+' || *p == '?')) - || ((obscure_syntax & RE_INTERVALS) - && ((obscure_syntax & RE_NO_BK_CURLY_BRACES) + || ((re_syntax_options & RE_INTERVALS) + && ((re_syntax_options & RE_NO_BK_CURLY_BRACES) ? *p == '{' : (p[0] == '\\' && p[1] == '{')))) { @@ -1454,16 +1361,20 @@ re_compile_pattern (pattern, size, bufp) if (stackp != stackb) goto unmatched_open; bufp->used = b - bufp->buffer; + bufp->re_nsub = regnum; return 0; + invalid_char: + return "Invalid character in regular expression"; + invalid_pattern: return "Invalid regular expression"; unmatched_open: - return "Unmatched \\("; + return "Unmatched ("; unmatched_close: - return "Unmatched \\)"; + return "Unmatched )"; end_of_pattern: return "Premature end of regular expression"; @@ -1600,20 +1511,14 @@ re_compile_fastmap (bufp) register unsigned char *p = pattern; register unsigned char *pend = pattern + size; register int j, k; - unsigned char *translate = (unsigned char *) bufp->translate; + unsigned char *translate = (unsigned char *)bufp->translate; unsigned is_a_succeed_n; -#ifndef NO_ALLOCA - unsigned char *stackb[NFAILURES]; - unsigned char **stackp = stackb; - -#else unsigned char **stackb; unsigned char **stackp; - stackb = (unsigned char **) xmalloc (NFAILURES * sizeof (unsigned char *)); + stackb = RE_TALLOC(NFAILURES, unsigned char*); stackp = stackb; -#endif /* NO_ALLOCA */ memset (fastmap, 0, (1 << BYTEWIDTH)); bufp->fastmap_accurate = 1; bufp->can_be_null = 0; @@ -1633,26 +1538,15 @@ re_compile_fastmap (bufp) #endif { case exactn: -#if 1 /* The original was: */ if (translate) fastmap[translate[p[1]]] = 1; else fastmap[p[1]] = 1; -#else /* The compiled pattern has already been translated. (not true for ruby) */ - fastmap[p[1]] = 1; -#endif break; case begline: - case before_dot: - case at_dot: - case after_dot: - case begbuf: - case endbuf: case wordbound: case notwordbound: - case wordbeg: - case wordend: continue; case endline: @@ -1748,41 +1642,16 @@ re_compile_fastmap (bufp) fastmap[j] = 1; break; -#ifdef emacs - case syntaxspec: - k = *p++; - for (j = 0; j < (1 << BYTEWIDTH); j++) - if (SYNTAX (j) == (enum syntaxcode) k) - fastmap[j] = 1; - break; - - case notsyntaxspec: - k = *p++; - for (j = 0; j < (1 << BYTEWIDTH); j++) - if (SYNTAX (j) != (enum syntaxcode) k) - fastmap[j] = 1; - break; - -#else /* not emacs */ - case syntaxspec: - case notsyntaxspec: - break; -#endif /* not emacs */ - case charset: /* NOTE: Charset for single-byte chars never contain multi-byte char. See set_list_bits(). */ for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--) if (p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH))) { -#if 1 /* The original was: */ if (translate) fastmap[translate[j]] = 1; else fastmap[j] = 1; -#else /* The compiled pattern has already been translated. */ - fastmap[j] = 1; -#endif } { unsigned short size; @@ -1830,7 +1699,7 @@ re_compile_fastmap (bufp) p += p[-1] + 2; size = EXTRACT_UNSIGNED (&p[-2]); - c = 0x8000; + c = 0x80; for (j = 0; j < size; j++) { for (beg = (unsigned char) p[j*4 + 0]; c < beg; c++) if (ismbchar (c)) @@ -1868,8 +1737,7 @@ re_search (pbufp, string, size, startpos, range, regs) int size, startpos, range; struct re_registers *regs; { - return re_search_2 (pbufp, (char *) 0, 0, string, size, startpos, range, - regs, size); + return re_search_2 (pbufp, (char*)0, 0, string, size, startpos, range, regs); } @@ -1881,23 +1749,20 @@ re_search (pbufp, string, size, startpos, range, regs) - 1, etc. STRING1 and STRING2 are of SIZE1 and SIZE2, respectively. In REGS, return the indices of the virtual concatenation of STRING1 and STRING2 that matched the entire PBUFP->buffer and its contained - subexpressions. Do not consider matching one past the index MSTOP in - the virtual concatenation of STRING1 and STRING2. + subexpressions. The value returned is the position in the strings at which the match was found, or -1 if no match was found, or -2 if error (such as failure stack overflow). */ -int -re_search_2 (pbufp, string1, size1, string2, size2, startpos, range, - regs, mstop) +static int +re_search_2 (pbufp, string1, size1, string2, size2, startpos, range, regs) struct re_pattern_buffer *pbufp; char *string1, *string2; int size1, size2; int startpos; register int range; struct re_registers *regs; - int mstop; { register char *fastmap = pbufp->fastmap; register unsigned char *translate = (unsigned char *) pbufp->translate; @@ -1917,19 +1782,9 @@ re_search_2 (pbufp, string1, size1, string2, size2, startpos, range, range = total_size - startpos; /* Update the fastmap now if not correct already. */ - if (fastmap && !pbufp->fastmap_accurate) - re_compile_fastmap (pbufp); - - /* If the search isn't to be a backwards one, don't waste time in a - long search for a pattern that says it is anchored. */ - if (pbufp->used > 0 && (enum regexpcode) pbufp->buffer[0] == begbuf - && range > 0) - { - if (startpos > 0) - return -1; - else - range = 1; - } + if (fastmap && !pbufp->fastmap_accurate) { + re_compile_fastmap (pbufp); + } while (1) { @@ -1987,7 +1842,7 @@ re_search_2 (pbufp, string1, size1, string2, size2, startpos, range, return -1; val = re_match_2 (pbufp, string1, size1, string2, size2, startpos, - regs, mstop); + regs); if (val >= 0) return startpos; if (val == -2) @@ -2039,7 +1894,6 @@ re_search_2 (pbufp, string1, size1, string2, size2, startpos, range, -#ifndef emacs /* emacs never uses this. */ int re_match (pbufp, string, size, pos, regs) struct re_pattern_buffer *pbufp; @@ -2049,7 +1903,6 @@ re_match (pbufp, string, size, pos, regs) { return re_match_2 (pbufp, (char *) 0, 0, string, size, pos, regs, size); } -#endif /* not emacs */ /* The following are used for re_match_2, defined below: */ @@ -2086,7 +1939,7 @@ struct register_info fail. The `+ 2' refers to PATTERN_PLACE and STRING_PLACE, which are arguments to the PUSH_FAILURE_POINT macro. */ -#define MAX_NUM_FAILURE_ITEMS (RE_NREGS * NUM_REG_ITEMS + 2) +#define MAX_NUM_FAILURE_ITEMS (num_regs * NUM_REG_ITEMS + 2) /* We push this many things on the stack whenever we fail. */ @@ -2102,8 +1955,8 @@ struct register_info long last_used_reg, this_reg; \ \ /* Find out how many registers are active or have been matched. \ - (Aside from register zero, which is only set at the end.) */ \ - for (last_used_reg = RE_NREGS - 1; last_used_reg > 0; last_used_reg--)\ + (Aside from register zero, which is only set at the end.) */ \ + for (last_used_reg = num_regs - 1; last_used_reg > 0; last_used_reg--)\ if (regstart[last_used_reg] != (unsigned char *)(-1L)) \ break; \ \ @@ -2113,6 +1966,7 @@ struct register_info unsigned int len = stacke - stackb; \ if (len > re_max_failures * MAX_NUM_FAILURE_ITEMS) \ { \ + FREE_VARIABLES(); \ FREE_AND_RETURN(stackb,(-2)); \ } \ \ @@ -2179,8 +2033,8 @@ struct register_info registers corresponding to the subexpressions of which we currently are inside. */ #define SET_REGS_MATCHED \ - { unsigned this_reg; \ - for (this_reg = 0; this_reg < RE_NREGS; this_reg++) \ + { unsigned this_reg; \ + for (this_reg = 0; this_reg < num_regs; this_reg++) \ { \ if (IS_ACTIVE(reg_info[this_reg])) \ MATCHED_SOMETHING(reg_info[this_reg]) = 1; \ @@ -2209,15 +2063,37 @@ struct register_info (SYNTAX ((d) == end1 ? *string2 : (d) == string2 - 1 ? *(end1 - 1) : *(d))\ == Sword) +static void +init_regs(regs, num_regs) + struct re_registers *regs; + unsigned num_regs; +{ + int i; + + regs->num_regs = num_regs; + if (num_regs < RE_NREGS) + num_regs = RE_NREGS; + + if (regs->allocated == 0) { + regs->beg = TMALLOC(num_regs, int); + regs->end = TMALLOC(num_regs, int); + regs->allocated = num_regs; + } + else if (regs->allocated < num_regs) { + TREALLOC(regs->beg, num_regs, int); + TREALLOC(regs->end, num_regs, int); + } + for (i=0; i<num_regs; i++) { + regs->beg[i] = regs->end[i] = -1; + } +} /* Match the pattern described by PBUFP against the virtual concatenation of STRING1 and STRING2, which are of SIZE1 and SIZE2, respectively. Start the match at index POS in the virtual concatenation of STRING1 and STRING2. In REGS, return the indices of the virtual concatenation of STRING1 and STRING2 that matched the - entire PBUFP->buffer and its contained subexpressions. Do not - consider matching one past the index MSTOP in the virtual - concatenation of STRING1 and STRING2. + entire PBUFP->buffer and its contained subexpressions. If pbufp->fastmap is nonzero, then it had better be up to date. @@ -2229,20 +2105,21 @@ struct register_info error (such as match stack overflow). Otherwise the value is the length of the substring which was matched. */ -int -re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) +static int +re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs) struct re_pattern_buffer *pbufp; char *string1_arg, *string2_arg; int size1, size2; int pos; struct re_registers *regs; - int mstop; { register unsigned char *p = (unsigned char *) pbufp->buffer; /* Pointer to beyond end of buffer. */ register unsigned char *pend = p + pbufp->used; + unsigned num_regs = pbufp->re_nsub + 1; + unsigned char *string1 = (unsigned char *) string1_arg; unsigned char *string2 = (unsigned char *) string2_arg; unsigned char *end1; /* Just past end of first string. */ @@ -2267,9 +2144,6 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) ``dummy''; if a failure happens and the failure point is a dummy, it gets discarded and the next next one is tried. */ -#ifndef NO_ALLOCA - unsigned char *initial_stack[MAX_NUM_FAILURE_ITEMS * NFAILURES]; -#endif unsigned char **stackb; unsigned char **stackp; unsigned char **stacke; @@ -2283,8 +2157,8 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) stopped matching the regnum-th subexpression. (The zeroth register keeps track of what the whole pattern matches.) */ - unsigned char *regstart[RE_NREGS]; - unsigned char *regend[RE_NREGS]; + unsigned char **regstart = RE_TALLOC(num_regs, unsigned char*); + unsigned char **regend = RE_TALLOC(num_regs, unsigned char*); /* The is_active field of reg_info helps us keep track of which (possibly nested) subexpressions we are currently in. The matched_something @@ -2293,8 +2167,7 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) subexpression. These two fields get reset each time through any loop their register is in. */ - struct register_info reg_info[RE_NREGS]; - + struct register_info *reg_info = RE_TALLOC(num_regs, struct register_info); /* The following record the register info as found in the above variables when we find a match better than any we've seen before. @@ -2302,15 +2175,16 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) turn happens only if we have not yet matched the entire string. */ unsigned best_regs_set = 0; - unsigned char *best_regstart[RE_NREGS]; - unsigned char *best_regend[RE_NREGS]; + unsigned char **best_regstart = RE_TALLOC(num_regs, unsigned char*); + unsigned char **best_regend = RE_TALLOC(num_regs, unsigned char*); + + if (regs) { + init_regs(regs, num_regs); + } + /* Initialize the stack. */ -#ifdef NO_ALLOCA - stackb = (unsigned char **) xmalloc (MAX_NUM_FAILURE_ITEMS * NFAILURES * sizeof (char *)); -#else - stackb = initial_stack; -#endif + stackb = RE_TALLOC(MAX_NUM_FAILURE_ITEMS * NFAILURES, unsigned char*); stackp = stackb; stacke = &stackb[MAX_NUM_FAILURE_ITEMS * NFAILURES]; @@ -2321,41 +2195,27 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) /* Initialize subexpression text positions to -1 to mark ones that no \( or ( and \) or ) has been seen for. Also set all registers to inactive and mark them as not having matched anything or ever - failed. */ - for (mcnt = 0; mcnt < RE_NREGS; mcnt++) - { + failed. */ + for (mcnt = 0; mcnt < num_regs; mcnt++) { regstart[mcnt] = regend[mcnt] = (unsigned char *) (-1L); IS_ACTIVE (reg_info[mcnt]) = 0; MATCHED_SOMETHING (reg_info[mcnt]) = 0; - } + } - if (regs) - for (mcnt = 0; mcnt < RE_NREGS; mcnt++) - regs->start[mcnt] = regs->end[mcnt] = -1; - /* Set up pointers to ends of strings. Don't allow the second string to be empty unless both are empty. */ - if (size2 == 0) - { + if (size2 == 0) { string2 = string1; size2 = size1; string1 = 0; size1 = 0; - } + } end1 = string1 + size1; end2 = string2 + size2; /* Compute where to stop matching, within the two strings. */ - if (mstop <= size1) - { - end_match_1 = string1 + mstop; - end_match_2 = string2; - } - else - { - end_match_1 = end1; - end_match_2 = string2 + mstop - size1; - } + end_match_1 = end1; + end_match_2 = string2 + size2 - size1; /* `p' scans through the pattern as `d' scans through the data. `dend' is the end of the input string that `d' points within. `d' is @@ -2405,7 +2265,7 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) best_regs_set = 1; best_regend[0] = d; /* Never use regstart[0]. */ - for (mcnt = 1; mcnt < RE_NREGS; mcnt++) + for (mcnt = 1; mcnt < num_regs; mcnt++) { best_regstart[mcnt] = regstart[mcnt]; best_regend[mcnt] = regend[mcnt]; @@ -2420,7 +2280,7 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) /* Restore best match. */ d = best_regend[0]; - for (mcnt = 0; mcnt < RE_NREGS; mcnt++) + for (mcnt = 0; mcnt < num_regs; mcnt++) { regstart[mcnt] = best_regstart[mcnt]; regend[mcnt] = best_regend[mcnt]; @@ -2432,23 +2292,23 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) to indices. */ if (regs) { - regs->start[0] = pos; + regs->beg[0] = pos; if (MATCHING_IN_FIRST_STRING) regs->end[0] = d - string1; else regs->end[0] = d - string2 + size1; - for (mcnt = 1; mcnt < RE_NREGS; mcnt++) + for (mcnt = 1; mcnt < num_regs; mcnt++) { if (regend[mcnt] == (unsigned char *)(-1L)) { - regs->start[mcnt] = -1; + regs->beg[mcnt] = -1; regs->end[mcnt] = -1; continue; } if (IS_IN_FIRST_STRING (regstart[mcnt])) - regs->start[mcnt] = regstart[mcnt] - string1; + regs->beg[mcnt] = regstart[mcnt] - string1; else - regs->start[mcnt] = regstart[mcnt] - string2 + size1; + regs->beg[mcnt] = regstart[mcnt] - string2 + size1; if (IS_IN_FIRST_STRING (regend[mcnt])) regs->end[mcnt] = regend[mcnt] - string1; @@ -2456,6 +2316,7 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) regs->end[mcnt] = regend[mcnt] - string2 + size1; } } + FREE_VARIABLES(); FREE_AND_RETURN(stackb, (d - pos - (MATCHING_IN_FIRST_STRING ? string1 : @@ -2586,7 +2447,7 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) break; } if ((translate ? translate[*d] : *d) == '\n' - || ((obscure_syntax & RE_DOT_NOT_NULL) + || ((re_syntax_options & RE_DOT_NOT_NULL) && (translate ? translate[*d] : *d) == '\000')) goto fail; SET_REGS_MATCHED; @@ -2597,9 +2458,7 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) case charset_not: { int not; /* Nonzero for charset_not. */ - register int c; - if (*(p - 1) == (unsigned char) charset_not) - not = 1; + int c; PREFETCH; /* Fetch a data character. */ @@ -2613,14 +2472,17 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) c = (unsigned char) translate[c]; not = is_in_list (c, p); + if (*(p - 1) == (unsigned char) charset_not) + not = !not; p += 1 + *p + 2 + EXTRACT_UNSIGNED (&p[1 + *p])*4; if (!not) goto fail; SET_REGS_MATCHED; + d++; if (d != dend && c >= 1 << BYTEWIDTH) - d++; + d++; break; } @@ -2686,14 +2548,17 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) if (p1[3] == (unsigned char) exactn && p1[5] != c) p[-3] = (unsigned char) finalize_jump; else if (p1[3] == (unsigned char) charset - || p1[3] == (unsigned char) charset_not) - { + || p1[3] == (unsigned char) charset_not) { + int not; if (ismbchar (c)) c = c << 8 | p2[3]; /* `is_in_list()' is TRUE if c would match */ /* That means it is not safe to finalize. */ - if (!is_in_list (c, p1 + 4)) - p[-3] = (unsigned char) finalize_jump; + not = is_in_list (c, p1 + 4); + if (p1[3] == (unsigned char) charset_not) + not = !not; + if (!not) + p[-3] = (unsigned char) finalize_jump; } } } @@ -2798,60 +2663,6 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) goto fail; break; - case wordbeg: - if (IS_A_LETTER (d) && (AT_STRINGS_BEG || !IS_A_LETTER (d - 1))) - break; - goto fail; - - case wordend: - /* Have to check if AT_STRINGS_BEG before looking at d - 1. */ - if (!AT_STRINGS_BEG && IS_A_LETTER (d - 1) - && (!IS_A_LETTER (d) || AT_STRINGS_END)) - break; - goto fail; - -#ifdef emacs - case before_dot: - if (PTR_CHAR_POS (d) >= point) - goto fail; - break; - - case at_dot: - if (PTR_CHAR_POS (d) != point) - goto fail; - break; - - case after_dot: - if (PTR_CHAR_POS (d) <= point) - goto fail; - break; - - case wordchar: - mcnt = (int) Sword; - goto matchsyntax; - - case syntaxspec: - mcnt = *p++; - matchsyntax: - PREFETCH; - if (SYNTAX (*d++) != (enum syntaxcode) mcnt) goto fail; - SET_REGS_MATCHED; - break; - - case notwordchar: - mcnt = (int) Sword; - goto matchnotsyntax; - - case notsyntaxspec: - mcnt = *p++; - matchnotsyntax: - PREFETCH; - if (SYNTAX (*d++) == (enum syntaxcode) mcnt) goto fail; - SET_REGS_MATCHED; - break; - -#else /* not emacs */ - case wordchar: PREFETCH; if (!IS_A_LETTER (d)) @@ -2868,25 +2679,6 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) SET_REGS_MATCHED; break; - case before_dot: - case at_dot: - case after_dot: - case syntaxspec: - case notsyntaxspec: - break; - -#endif /* not emacs */ - - case begbuf: - if (AT_STRINGS_BEG) - break; - goto fail; - - case endbuf: - if (AT_STRINGS_END) - break; - goto fail; - case exactn: /* Match the next few pattern characters exactly. mcnt is how many characters to match. */ @@ -2922,27 +2714,8 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) { do { -#if 0 - /* this code suppose that multi-byte chars are not splited - in string1 and string2. If you want to check this with - speed cost, change `#if 0' here and next to `#if 1'. */ - unsigned char c; - -#endif PREFETCH; -#if 0 - c = *d++; - if (ismbchar (c)) { - if (c != (unsigned char) *p++ - || !--mcnt - || d == dend) - goto fail; - c = *d++; - } - if (c != (unsigned char) *p++) goto fail; -#else if (*d++ != *p++) goto fail; -#endif } while (--mcnt); } @@ -2973,8 +2746,8 @@ re_match_2 (pbufp, string1_arg, size1, string2_arg, size2, pos, regs, mstop) /* Restore register info. */ last_used_reg = (long) *--stackp; - /* Make the ones that weren't saved -1 or 0 again. */ - for (this_reg = RE_NREGS - 1; this_reg > last_used_reg; this_reg--) + /* Make the ones that weren't saved -1 or 0 again. */ + for (this_reg = num_regs - 1; this_reg > last_used_reg; this_reg--) { regend[this_reg] = (unsigned char *) (-1L); regstart[this_reg] = (unsigned char *) (-1L); @@ -3023,218 +2796,22 @@ memcmp_translate (s1, s2, len, translate) return 0; } - - -/* Entry points compatible with 4.2 BSD regex library. */ - -#if !defined(emacs) && !defined(GAWK) && !defined(RUBY) - -static struct re_pattern_buffer re_comp_buf; - -char * -re_comp (s) - char *s; -{ - if (!s) - { - if (!re_comp_buf.buffer) - return "No previous regular expression"; - return 0; - } - - if (!re_comp_buf.buffer) - { - if (!(re_comp_buf.buffer = (char *) xmalloc (200))) - return "Memory exhausted"; - re_comp_buf.allocated = 200; - if (!(re_comp_buf.fastmap = (char *) xmalloc (1 << BYTEWIDTH))) - return "Memory exhausted"; - } - return re_compile_pattern (s, strlen (s), &re_comp_buf); -} - -int -re_exec (s) - char *s; -{ - int len = strlen (s); - return 0 <= re_search (&re_comp_buf, s, len, 0, len, - (struct re_registers *) 0); -} -#endif /* not emacs && not GAWK && not RUBY */ - - - -#ifdef test - -#ifdef atarist -long _stksize = 2L; /* reserve memory for stack */ -#endif -#include <stdio.h> - -/* Indexed by a character, gives the upper case equivalent of the - character. */ - -char upcase[0400] = - { 000, 001, 002, 003, 004, 005, 006, 007, - 010, 011, 012, 013, 014, 015, 016, 017, - 020, 021, 022, 023, 024, 025, 026, 027, - 030, 031, 032, 033, 034, 035, 036, 037, - 040, 041, 042, 043, 044, 045, 046, 047, - 050, 051, 052, 053, 054, 055, 056, 057, - 060, 061, 062, 063, 064, 065, 066, 067, - 070, 071, 072, 073, 074, 075, 076, 077, - 0100, 0101, 0102, 0103, 0104, 0105, 0106, 0107, - 0110, 0111, 0112, 0113, 0114, 0115, 0116, 0117, - 0120, 0121, 0122, 0123, 0124, 0125, 0126, 0127, - 0130, 0131, 0132, 0133, 0134, 0135, 0136, 0137, - 0140, 0101, 0102, 0103, 0104, 0105, 0106, 0107, - 0110, 0111, 0112, 0113, 0114, 0115, 0116, 0117, - 0120, 0121, 0122, 0123, 0124, 0125, 0126, 0127, - 0130, 0131, 0132, 0173, 0174, 0175, 0176, 0177, - 0200, 0201, 0202, 0203, 0204, 0205, 0206, 0207, - 0210, 0211, 0212, 0213, 0214, 0215, 0216, 0217, - 0220, 0221, 0222, 0223, 0224, 0225, 0226, 0227, - 0230, 0231, 0232, 0233, 0234, 0235, 0236, 0237, - 0240, 0241, 0242, 0243, 0244, 0245, 0246, 0247, - 0250, 0251, 0252, 0253, 0254, 0255, 0256, 0257, - 0260, 0261, 0262, 0263, 0264, 0265, 0266, 0267, - 0270, 0271, 0272, 0273, 0274, 0275, 0276, 0277, - 0300, 0301, 0302, 0303, 0304, 0305, 0306, 0307, - 0310, 0311, 0312, 0313, 0314, 0315, 0316, 0317, - 0320, 0321, 0322, 0323, 0324, 0325, 0326, 0327, - 0330, 0331, 0332, 0333, 0334, 0335, 0336, 0337, - 0340, 0341, 0342, 0343, 0344, 0345, 0346, 0347, - 0350, 0351, 0352, 0353, 0354, 0355, 0356, 0357, - 0360, 0361, 0362, 0363, 0364, 0365, 0366, 0367, - 0370, 0371, 0372, 0373, 0374, 0375, 0376, 0377 - }; - -#ifdef canned - -#include "tests.h" - -typedef enum { extended_test, basic_test } test_type; - -/* Use this to run the tests we've thought of. */ - void -main () +re_copy_registers(regs1, regs2) + struct re_registers *regs1, *regs2; { - test_type t = extended_test; + int i; - if (t == basic_test) - { - printf ("Running basic tests:\n\n"); - test_posix_basic (); - } - else if (t == extended_test) - { - printf ("Running extended tests:\n\n"); - test_posix_extended (); + if (regs1->allocated == 0) { + regs1->beg = TMALLOC(regs2->num_regs, int); + regs1->end = TMALLOC(regs2->num_regs, int); } -} - -#else /* not canned */ - -/* Use this to run interactive tests. */ - -int -main (argc, argv) - int argc; - char **argv; -{ - char pat[80]; - struct re_pattern_buffer buf; - int i; - char c; - char fastmap[(1 << BYTEWIDTH)]; - - /* Allow a command argument to specify the style of syntax. */ - if (argc > 1) - obscure_syntax = atol (argv[1]); - - buf.allocated = 40; - buf.buffer = (char *) xmalloc (buf.allocated); - buf.fastmap = fastmap; - buf.translate = upcase; - - while (1) - { - gets (pat); - - if (*pat) - { - re_compile_pattern (pat, strlen(pat), &buf); - - for (i = 0; i < buf.used; i++) - printchar (buf.buffer[i]); - - putchar ('\n'); - - printf ("%d allocated, %d used.\n", buf.allocated, buf.used); - - re_compile_fastmap (&buf); - printf ("Allowed by fastmap: "); - for (i = 0; i < (1 << BYTEWIDTH); i++) - if (fastmap[i]) printchar (i); - putchar ('\n'); - } - - gets (pat); /* Now read the string to match against */ - - i = re_match (&buf, pat, strlen (pat), 0, 0); - printf ("Match value %d.\n", i); + else if (regs1->allocated < regs2->num_regs) { + TREALLOC(regs1->beg, regs2->num_regs, int); + TREALLOC(regs1->end, regs2->num_regs, int); } -} - -#endif - - -#ifdef NOTDEF -print_buf (bufp) - struct re_pattern_buffer *bufp; -{ - int i; - - printf ("buf is :\n----------------\n"); - for (i = 0; i < bufp->used; i++) - printchar (bufp->buffer[i]); - - printf ("\n%d allocated, %d used.\n", bufp->allocated, bufp->used); - - printf ("Allowed by fastmap: "); - for (i = 0; i < (1 << BYTEWIDTH); i++) - if (bufp->fastmap[i]) - printchar (i); - printf ("\nAllowed by translate: "); - if (bufp->translate) - for (i = 0; i < (1 << BYTEWIDTH); i++) - if (bufp->translate[i]) - printchar (i); - printf ("\nfastmap is%s accurate\n", bufp->fastmap_accurate ? "" : "n't"); - printf ("can %s be null\n----------", bufp->can_be_null ? "" : "not"); -} -#endif /* NOTDEF */ - -printchar (c) - char c; -{ - if (c < 040 || c >= 0177) - { - putchar ('\\'); - putchar (((c >> 6) & 3) + '0'); - putchar (((c >> 3) & 7) + '0'); - putchar ((c & 7) + '0'); + for (i=0; i<regs2->num_regs; i++) { + regs1->beg[i] = regs2->beg[i]; + regs1->end[i] = regs2->end[i]; } - else - putchar (c); -} - -error (string) - char *string; -{ - puts (string); - exit (1); } -#endif /* test */ @@ -38,10 +38,10 @@ /* This defines the various regexp syntaxes. */ -extern long obscure_syntax; +extern long re_syntax_options; -/* The following bits are used in the obscure_syntax variable to choose among +/* The following bits are used in the re_syntax_options variable to choose among alternative regexp syntaxes. */ /* If this bit is set, plain parentheses serve as grouping, and backslash @@ -93,6 +93,7 @@ extern long obscure_syntax; If set, then { and } serve as interval operators and \{ and \} are literals. */ #define RE_NO_BK_CURLY_BRACES (1L << 8) +#define RE_NO_BK_BRACES RE_NO_BK_CURLY_BRACES /* If this bit is set, then character classes are supported; they are: [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:], @@ -142,6 +143,9 @@ extern long obscure_syntax; If it isn't, then it can. */ #define RE_NO_HYPHEN_RANGE_END (1L << 18) +/* If this bit is not set, then \ inside a bracket expression is literal. + If set, then such a \ quotes the following character. */ +#define RE_BACKSLASH_ESCAPE_IN_LISTS (1L << 19) /* Define combinations of bits for the standard possibilities. */ #define RE_SYNTAX_POSIX_AWK (RE_NO_BK_PARENS | RE_NO_BK_VBAR \ @@ -165,15 +169,25 @@ extern long obscure_syntax; | RE_NO_HYPHEN_RANGE_END) /* For multi-byte char support */ -#define RE_MBCTYPE_EUC (1L << 19) -#define RE_MBCTYPE_SJIS (1L << 20) +#define RE_MBCTYPE_EUC (1L << 20) +#define RE_MBCTYPE_SJIS (1L << 21) #define RE_MBCTYPE_MASK (RE_MBCTYPE_EUC | RE_MBCTYPE_SJIS) +#ifdef EUC +#define DEFAULT_MBCTYPE RE_MBCTYPE_EUC +#else +#ifdef SJIS +#define DEFAULT_MBCTYPE RE_MBCTYPE_SJIS +#else +#define DEFAULT_MBCTYPE 0 +#endif +#endif + #undef ismbchar #define ismbchar(c) \ - (obscure_syntax & RE_MBCTYPE_EUC \ + (re_syntax_options & RE_MBCTYPE_EUC \ ? ((unsigned char) (c) >= 0x80) \ - : (obscure_syntax & RE_MBCTYPE_SJIS \ + : (re_syntax_options & RE_MBCTYPE_SJIS \ ? (( 0x80 <= (unsigned char) (c) \ && (unsigned char) (c) <= 0x9f) \ || (0xe0 <= (unsigned char) (c))) \ @@ -193,6 +207,9 @@ struct re_pattern_buffer comparing, or zero for no translation. The translation is applied to a pattern when it is compiled and to data when it is matched. */ + + + long re_nsub; /* Number of subexpressions found by the compiler. */ char fastmap_accurate; /* Set to zero when a new pattern is stored, set to one when the fastmap is updated from it. */ @@ -223,8 +240,10 @@ struct re_pattern_buffer struct re_registers { - int start[RE_NREGS]; - int end[RE_NREGS]; + unsigned allocated; + unsigned num_regs; + int *beg; + int *end; }; @@ -236,14 +255,10 @@ extern char *re_compile_pattern (char *, size_t, struct re_pattern_buffer *); extern void re_compile_fastmap (struct re_pattern_buffer *); extern int re_search (struct re_pattern_buffer *, char*, int, int, int, struct re_registers *); -extern int re_search_2 (struct re_pattern_buffer *, char *, int, - char *, int, int, int, - struct re_registers *, int); extern int re_match (struct re_pattern_buffer *, char *, int, int, struct re_registers *); -extern int re_match_2 (struct re_pattern_buffer *, char *, int, - char *, int, int, struct re_registers *, int); extern long re_set_syntax (long syntax); +extern void re_copy_registers (struct re_registers*, struct re_registers*); #ifndef RUBY /* 4.2 bsd compatibility. */ @@ -256,15 +271,10 @@ extern int re_exec (char *); extern char *re_compile_pattern (); /* Is this really advertised? */ extern void re_compile_fastmap (); -extern int re_search (), re_search_2 (); -extern int re_match (), re_match_2 (); +extern int re_search (); +extern int re_match (); extern long re_set_syntax(); - -#ifndef RUBY -/* 4.2 bsd compatibility. */ -extern char *re_comp (); -extern int re_exec (); -#endif +extern void re_copy_registers (); #endif /* __STDC__ */ @@ -39,6 +39,15 @@ ruby \- ֥Ȼظץȸ .B ruby ץϰʲΰդ. .TP 5 +.B \-0 +ϥ쥳ɥѥ졼(`$/')8ʿǻꤹ. + +ꤷʤϥ̥륭饯ѥ졼ˤʤ롣 +¾ΥåäƤ褤 + +\-00, ѥ饰ե⡼, \-0777(Υɤʸ¸ +ߤʤΤ)ե٤ɤ߹⡼ɤǤ. +.TP 5 .B \-a `\-n'`\-p'ȤȤѤ, ȥץåȥ⡼ɤONˤ. ȥץåȥ⡼ɤǤϳƥ롼פƬ, @@ -55,21 +64,20 @@ ruby \- ֥Ȼظץȸ ץȤؤΥѥΤߤԤ, ¹Ԥʤ. ѥ뽪λ, ʸˡ顼̵, "Syntax OK"ȽϤ. .TP 5 -.B \-C " code" +.B \-K " c" .B ruby νɤꤹ. .B ruby -ϻꤷʸ `E'ޤ`e'Ϥޤʸ䥢 -ե뤬EUCǵҤƤȲꤹ. Ʊͤ`S' -`s'ξSJISȤƽ. `N'ϴʤ. -եȤEUC. +ϻꤷʸ `E'ޤ`e'ξʸ䥢ե +뤬EUCǵҤƤȲꤹ. Ʊͤ`S'ޤ`s'ξ +SJISȤƽ. `N'ϴʤ. ǥեȤ +EUC. .nf .ne 2 - ruby -C EUC -e 'print "ƥ"' - ruby -Ceuc -e 'print "ƥ"' - ruby -C 'Shift JIS' -e 'print "ƥ"' - ruby -CN 'print "ƥ"' + ruby -CE -e 'print "ƥ"' + ruby -Cs -e 'print "ƥ"' + ruby -Cn -e 'print "ƥ"' .fi ΥץϾʸɤμưȽ̵ǽɲä줿 @@ -83,11 +91,11 @@ ruby \- ֥Ȼظץȸ ޥɥ饤饹ץȤꤹ. \-eץդ ˤϰ饹ץȥե̾ʤ. .TP 5 -.B \-F " ʸ" +.B \-F" ʸ" ϥեɥѥ졼(`$;')ͤʸ˥åȤ. awk Ʊ̾ΥץƱƯ. .TP 5 -.B \-i " extension" +.B \-i" extension" ǻꤵ줿եƤ֤(in-place edit) Ȥꤹ. ΥեϳĥҤĤ¸. .nf @@ -105,7 +113,7 @@ ruby \- ֥Ȼظץȸ .fi .TP 5 -.B \-I " directory" +.B \-I" directory" եɤѥ(ɲ). ꤵ줿ǥ쥯 ȥ .B ruby @@ -139,15 +147,6 @@ ruby \- ֥Ȼظץȸ .fi .TP 5 -.B \-R ʸ(ޤ8ʿ) -ϥ쥳ɥѥ졼(`$/')ͤʸ˥åȤ. ʸ -٤ƿ(0-7ޤ)鹽Ƥˤ, 8ʿǻ -ꤵ줿ɤʸѥ졼Ȥ. - -ξ, \-R0, ѥ饰ե⡼, \-R777(Υɤ -ʸ¸ߤʤΤ)ե٤ɤ߹⡼ɤ -Ǥ. -.TP 5 .B \-s ץ̾³, \-ǻϤޤᤷ, Ʊ̾ѿ ͤꤹ. `\-\-'ʤʹߤϲԤʤʤ. @@ -207,14 +206,14 @@ $VERBOSEåȤ. ѿåȤƤ, Ĥ .fi .TP 5 -.B \-x +.B \-x"directory" åΥץȤФƼ¹Ԥ. #!ǻϤޤ, "ruby"ȤʸޤԤޤǤɤФ. ץȤν EOF(եν), ^D(ȥD), ^Z(ȥZ) ޤͽ``__END__''ǻꤹ. -.TP 5 -.B \-X " directory" -ץȼ¹˻ꤵ줿ǥ쥯ȥ˰ܤ. + +ǥ쥯ȥ̾ꤹȡץȼ¹˻ꤵ줿ǥ +쥯ȥ˰ܤ. .TP 5 .B \-y, \-\-yydebug ѥǥХå⡼. ѥιʸϤβɽ @@ -18,20 +18,8 @@ #include <sys/stat.h> #include <signal.h> -#include "getopt.h" - static int version, copyright; -static struct option long_options[] = -{ - {"debug", 0, 0, 'd'}, - {"yydebug", 0, 0, 'y'}, - {"verbose", 0, 0, 'v'}, - {"version", 0, &version, 1}, - {"copyright", 0, ©right, 1}, - {0, 0, 0, 0} -}; - int debug = 0; int verbose = 0; static int sflag = FALSE; @@ -67,12 +55,11 @@ proc_options(argcp, argvp) { int argc = *argcp; char **argv = *argvp; + int script_given, do_search; + char *s; + extern VALUE rb_load_path; - extern char *optarg; - extern int optind; - int c, i, j, script_given, do_search, opt_index; extern VALUE RS, ORS, FS; - char *src; if (argc == 0) return; @@ -80,114 +67,155 @@ proc_options(argcp, argvp) script_given = FALSE; do_search = FALSE; - optind = 0; - while ((c = getopt_long(argc, argv, "+acC:de:F:i:I:lnpR:svxX:yS", - long_options, &opt_index)) != EOF) { - switch (c) { + for (argc--,argv++; argc > 0; argc--,argv++) { + if (argv[0][0] != '-' || !argv[0][1]) break; + + s = argv[0]+1; + reswitch: + switch (*s) { + case 'a': + do_split = TRUE; + s++; + goto reswitch; + case 'p': do_print = TRUE; /* through */ case 'n': do_loop = TRUE; - break; + s++; + goto reswitch; case 'd': debug = TRUE; - break; + s++; + goto reswitch; case 'y': yydebug = 1; - break; + s++; + goto reswitch; case 'v': verbose = TRUE; show_version(); - break; + s++; + goto reswitch; + + case 'c': + do_check = TRUE; + s++; + goto reswitch; + + case 's': + sflag = TRUE; + s++; + goto reswitch; + + case 'l': + do_line = TRUE; + ORS = RS; + s++; + goto reswitch; + + case 'S': + do_search = TRUE; + s++; + goto reswitch; case 'e': script_given++; if (script == 0) script = "-e"; - lex_setsrc("-e", optarg, strlen(optarg)); + if (argv[1]) { + lex_setsrc("-e", argv[1], strlen(argv[1])); + argc--,argv++; + } + else { + lex_setsrc("-e", "", 0); + } yyparse(); break; case 'i': - inplace = strdup(optarg); - break; - - case 'c': - do_check = TRUE; + inplace = strdup(s+1); break; case 'x': xflag = TRUE; + s++; + if (*s && chdir(s) < 0) { + Fatal("Can't chdir to %s", s); + } break; - case 'X': - if (chdir(optarg) < 0) - Fatal("Can't chdir to %s", optarg); + case 'F': + FS = str_new2(s+1); break; - case 's': - sflag = TRUE; - break; + case 'K': + s++; + rb_set_kanjicode(s); + s++; + goto reswitch; - case 'l': - do_line = TRUE; - ORS = RS; + case 'I': + ary_unshift(rb_load_path, str_new2(s+1)); break; - case 'R': + case '0': { - char *p = optarg; - - while (*p) { - if (*p < '0' || '7' < *p) { - break; - } - p++; - } - if (*p) { - RS = str_new2(optarg); + int numlen; + int v; + char c; + + v = scan_oct(s, 4, &numlen); + s += numlen; + if (v > 0377) RS = Qnil; + else if (v == 0 && numlen >= 2) { + RS = str_new2("\n\n"); } else { - int i = strtoul(optarg, Qnil, 8); - - if (i == 0) RS = str_new(0, 0); - else if (i > 0xff) RS = Qnil; - else { - char c = i; - RS = str_new(&c, 1); - } + c = v & 0xff; + RS = str_new(&c, 1); } } - break; + goto reswitch; - case 'F': - FS = str_new2(optarg); - break; + case 'u': + case 'U': - case 'a': - do_split = TRUE; - break; - - case 'C': - rb_set_kanjicode(optarg); - break; - - case 'S': - do_search = TRUE; - break; - - case 'I': - ary_unshift(rb_load_path, str_new2(optarg)); + case '-': + if (!s[1]) { + argc--,argv++; + goto switch_end; + } + s++; + if (strcmp("copyright", s) == 0) + copyright = 1; + else if (strcmp("debug", s) == 0) + debug = 1; + else if (strcmp("version", s) == 0) + version = 1; + else if (strcmp("verbose", s) == 0) + verbose = 1; + else if (strcmp("yydebug", s) == 0) + yydebug = 1; + else { + Fatal("Unrecognized long option: --%s",s); + } break; default: + Fatal("Unrecognized switch: -%s",s); + + case 0: break; } } + switch_end: + if (*argvp[0] == Qnil) return; + if (version) { show_version(); exit(0); @@ -196,28 +224,28 @@ proc_options(argcp, argvp) show_copyright(); } - if (argv[0] == Qnil) return; + rb_setup_kcode(); if (script_given == 0) { - if (argc == optind) { /* no more args */ + if (argc == 0) { /* no more args */ if (verbose) exit(0); script = "-"; load_stdin(); } else { - script = argv[optind]; + script = argv[0]; if (do_search) { script = dln_find_file(script, getenv("PATH")); - if (!script) script = argv[optind]; + if (!script) script = argv[0]; } load_file(script, 1); - optind++; + argc--,argv++; } } xflag = FALSE; - *argvp += optind; - *argcp -= optind; + *argvp = argv; + *argcp = argc; if (sflag) { char *s; @@ -239,6 +267,7 @@ proc_options(argcp, argvp) } *argcp = argc; *argvp = argv; } + } static void @@ -161,7 +161,7 @@ struct RArray { struct RRegexp { struct RBasic basic; - struct Regexp *ptr; + struct re_pattern_buffer *ptr; UINT len; char *str; }; diff --git a/sample/Artistic b/sample/Artistic deleted file mode 100644 index e69de29bb2..0000000000 --- a/sample/Artistic +++ /dev/null diff --git a/sample/evaldef.rb b/sample/evaldef.rb index a705300a94..2cedd54999 100644 --- a/sample/evaldef.rb +++ b/sample/evaldef.rb @@ -18,8 +18,8 @@ class Bar : Foo end end -f = foo.new -b = bar.new +f = Foo.new +b = Bar.new b.foo b.bar diff --git a/sample/export.rb b/sample/export.rb index a6dfa8e40e..2d05d8afd7 100644 --- a/sample/export.rb +++ b/sample/export.rb @@ -1,18 +1,38 @@ # method access permission # output: # foobar -# foo +# Foo class Foo - export :printf + public :printf + def baz + print "baz\n" + end + private :baz + + def quux + print "in QUUX " + baz() + end end def foobar print "foobar\n" end -f = foo.new -#foo.unexport :printf -foo.export :foobar +f = Foo.new +#Foo.private :printf +Foo.public :foobar f.foobar -f.printf "%s\n", foo +f.printf "%s\n", Foo + +f.quux + +class Bar : Foo + def quux + super + baz() + end +end + +Bar.new.quux diff --git a/sample/exyacc.rb b/sample/exyacc.rb index ad6191c03d..cd1170feb4 100644 --- a/sample/exyacc.rb +++ b/sample/exyacc.rb @@ -1,4 +1,4 @@ -#! /usr/local/bin/ruby -Cn +#! /usr/local/bin/ruby -Kn # usage: exyacc.rb [yaccfiles] # this is coverted from exyacc.pl in the camel book diff --git a/sample/from.rb b/sample/from.rb index f6602943af..a8baf4f375 100644 --- a/sample/from.rb +++ b/sample/from.rb @@ -17,8 +17,8 @@ module ParseDate if date.sub(/\s+(\d+:\d+(:\d+)?)/, ' ') time = $1 end - if date =~ /19\d\d/ - year = $& + if date =~ /19(\d\d)/ + year = $1 end if date.sub(/\s*(\d+)\s+(#{MONTHPAT})\S*\s+/, ' ') dayofmonth = $1 diff --git a/sample/opt_x.rb b/sample/opt_x.rb deleted file mode 100644 index e69de29bb2..0000000000 --- a/sample/opt_x.rb +++ /dev/null diff --git a/sample/ruby-mode.el b/sample/ruby-mode.el index aa91be330d..fd44dc661b 100644 --- a/sample/ruby-mode.el +++ b/sample/ruby-mode.el @@ -18,7 +18,7 @@ (defconst ruby-block-end-re "end") (defconst ruby-delimiter - (concat "[$/<(){}#\"'`]\\|\\[\\|\\]\\|\\b\\(" + (concat "[$/(){}#\"'`]\\|\\[\\|\\]\\|\\b\\(" ruby-block-beg-re "\\|" ruby-block-end-re "\\)\\b") ) @@ -178,9 +178,7 @@ The variable ruby-indent-level controls the amount of indentation. (t (goto-char indent-point) (setq in-string t)))) - ((or (string= "/" w) - (string= "<" w)) - (if (string= "<" w) (setq w ">")) + ((string= "/" w) (let (c) (save-excursion (goto-char pnt) @@ -263,6 +261,120 @@ The variable ruby-indent-level controls the amount of indentation. (error (format "bad string %s" w))))))) (list in-string in-paren (car nest) depth))) +(defun ruby-parse-region (start end) + (let ((indent-point end) + (indent 0) + (in-string nil) + (in-paren nil) + (depth 0) + (nest nil)) + (save-excursion + (if start + (goto-char start) + (ruby-beginning-of-defun)) + (while (and (> indent-point (point)) + (re-search-forward ruby-delimiter indent-point t)) + (let ((w (buffer-substring (match-beginning 0) (match-end 0))) + (pnt (match-beginning 0))) + (cond + ((or (string= "\"" w) ;skip string + (string= "'" w) + (string= "`" w)) + (cond + ((string= w (char-to-string (char-after (point)))) + (forward-char 1)) + ((re-search-forward (format "[^\\]%s" w) indent-point t) + nil) + (t + (goto-char indent-point) + (setq in-string t)))) + ((or (string= "/" w) + (string= "<" w)) + (if (string= "<" w) (setq w ">")) + (let (c) + (save-excursion + (goto-char pnt) + (skip-chars-backward " \t") + (setq c (char-after (1- (point)))) + (if c (setq c (char-syntax c)))) + (if (or (eq c ?.) + (and (eq c ?w) + (save-excursion + (forward-word -1) + (or + (looking-at ruby-block-beg-re) + (looking-at ruby-block-mid-re))))) + (cond + ((string= w (char-to-string (char-after (point)))) + (forward-char 1)) + ((re-search-forward (format "[^\\]%s" w) indent-point t) + nil) + (t + (goto-char indent-point) + (setq in-string t)))))) + ((string= "$" w) ;skip $char + (forward-char 1)) + ((string= "#" w) ;skip comment + (forward-line 1)) + ((string= "(" w) ;skip to matching paren + (let ((orig depth)) + (setq nest (cons (point) nest)) + (setq depth (1+ depth)) + (while (and (/= depth orig) + (re-search-forward "[()]" indent-point t)) + (cond + ((= (char-after (match-beginning 0)) ?\( ) + (setq nest (cons (point) nest)) + (setq depth (1+ depth))) + (t + (setq nest (cdr nest)) + (setq depth (1- depth))))) + (if (> depth orig) (setq in-paren ?\()))) + ((string= "[" w) ;skip to matching paren + (let ((orig depth)) + (setq nest (cons (point) nest)) + (setq depth (1+ depth)) + (while (and (/= depth orig) + (re-search-forward "\\[\\|\\]" indent-point t)) + (cond + ((= (char-after (match-beginning 0)) ?\[ ) + (setq nest (cons (point) nest)) + (setq depth (1+ depth))) + (t + (setq nest (cdr nest)) + (setq depth (1- depth))))) + (if (> depth orig) (setq in-paren ?\[)))) + ((string= "{" w) ;skip to matching paren + (let ((orig depth)) + (setq nest (cons (point) nest)) + (setq depth (1+ depth)) + (while (and (/= depth orig) + (re-search-forward "[{}]" indent-point t)) + (cond + ((= (char-after (match-beginning 0)) ?{ ) + (setq nest (cons (point) nest)) + (setq depth (1+ depth))) + (t + (setq nest (cdr nest)) + (setq depth (1- depth))))) + (if (> depth orig) (setq in-paren ?{)))) + ((string-match ruby-block-end-re w) + (setq nest (cdr nest)) + (setq depth (1- depth))) + ((string-match ruby-block-beg-re w) + (let (c) + (save-excursion + (goto-char pnt) + (skip-chars-backward " \t") + (setq c (char-after (1- (point))))) + (if (or (null c) (= c ?\n) (= c ?\;)) + (progn + (setq nest (cons (point) nest)) + (setq depth (1+ depth)))))) + (t + (error (format "bad string %s" w))))))) + (list in-string in-paren (car nest) depth))) + (defun ruby-calculate-indent (&optional parse-start) (save-excursion (beginning-of-line) diff --git a/sample/uumerge.rb b/sample/uumerge.rb index d0bd016036..ac6e1c6849 100755 --- a/sample/uumerge.rb +++ b/sample/uumerge.rb @@ -1,5 +1,10 @@ #!/usr/local/bin/ruby +if $ARGV[0] == "-c" + out_stdout = 1; + $ARGV.shift +end + while gets() if /^begin\s*(\d*)\s*(\S*)/ $mode, $file = $1, $2 @@ -8,8 +13,13 @@ while gets() end end -fail "missing begin" if !$sawbegin; -OUT = open($file, "w") if $file != ""; +fail "missing begin" if ! $sawbegin; + +if out_stdout + out = STDOUT +else + out = open($file, "w") if $file != ""; +end while gets() if /^end/ @@ -19,9 +29,9 @@ while gets() sub(/[a-z]+$/, ""); # handle stupid trailing lowercase letters continue if /[a-z]/ continue if !(((($_[0] - 32) & 077) + 2) / 3 == $_.length / 4) - OUT << $_.unpack("u"); + out << $_.unpack("u"); end -fail "missing end" if ! $sawend; -File.chmod $mode.oct, $file; +fail "missing end" if !$sawend; +File.chmod $mode.oct, $file if ! out_stdout exit 0; @@ -68,13 +68,13 @@ tab) CR(carriage return)(form feed)Ǥ롥(newline) ͽϰʲ̤Ǥ - alias def if resque when - and else in retry while - begin elsif include return yield - break end module self __END__ - case ensure nil super __FILE__ - class fail or then __LINE__ - continue for redo undef + alias def if retry while + and else in return yield + begin elsif module self __END__ + break end nil super __FILE__ + case ensure or then __LINE__ + class fail redo undef + continue for resque when ͽϥ饹̾å̾ѿ̾ʤɤѤ뤳ȤϤǤʤ @@ -120,9 +120,7 @@ RubyǤnilʳɾ롥CPerlʤɤȤϰۤʤꡤ0 \# ʸ`#'Τ \nnn 8ʿɽ(n0-7) \xnn 16ʿɽ(n0-9,a-f) - \^c ȥʸ(cASCIIʸ) - \C-c ȥʸ(Ʊ) - \M-c ʸ(c|0x80) + \cx ȥʸ(xASCIIʸ) \x ʸxΤ ޤ`#'ˤѿŸԤ롥륯(`'')dz @@ -166,17 +164,6 @@ RubyǤshΤ褦˥ޥɤμ¹Է̤ʸƥΤ褦˻ȤȤ ¾ʸƱХåå嵭ˡѿŸͭǤ롥 -** 磻ɥɼ - - <...> - - * Ǥդʸ(ʸޤ)Ȱ - ? Ǥդ1ʸȰ - [ ] []Τ줫1ʸȰ - {..} {}(ޤǶڤ줿)줫ʸȰ - -¾ʸƱХåå嵭ˡѿŸͭǤ롥 - ** ѿŸ ֥륯(`"')ǰϤޤ줿ʸޥʸɽ @@ -363,9 +350,10 @@ Rubyѿϥ(ͭϰ)ȼ̿(ͭ)ˤä4ʬव졤 Ǥ⾮ʸǤʤѿ̾Ȥϼ̻Ҥ֤̾㤦ΤǽʣƤ ʤ -饹ʸγǻꤵ줿åɤȥ饹Moduleunexportåɤ -ꤵ줿åɤϴؿŪåɤȸƤФ졤ؿǤƤӽФ -Ǥʤ(DEF) +饹ModuleƤå(public,protected,private)ǥ +ɤθƤӽФ椹뤳Ȥ롥protectedޤ privateǻꤵ +줿¤줿åɤϴؿǤƤӽФȤ褺 +privateåɤƱ饹줿åɤ餷ƤӽФʤ ** SUPER @@ -805,17 +793,18 @@ beginresqueǻȤbeginϤᤫ¹Ԥ롥㳰Ԥʤä ǻꤹ롥¤ӤκǸ`*'硤¿Ϳ줿 °ϡǸΰȤͿ(ʤˤϥ顼) -åɤˤϴؿŪåɤ̾åɤ롥ؿŪåɤϴؿ -ǤƤӽФȤǤäƳ륯饹ȤΥ֥饹Υ -åɤ餷ƤӽФȤǤʤ -˥åɤ硤饹γˤdefϴؿŪ -ɤ饹ˤdef̾Υåɤ롥 -ѡ饹ΥåɤˤåɤβĻϥ -ѡ饹ΥåɤΤΤѤ +åɤˤϸƤӽФ¤ä뤳Ȥ衤¤ä줿åɤϡ +ؿǤƤӽФʤ(protected)뤤ϴؿǤƱ +饹Υåɤ餷Ƥ٤ʤ(private) + +˥åɤ硤饹γˤdefϥǥե +Ǥprotectedåɤ饹ˤdefpublic +ɤ롥ѡ饹Υåɤˤ +åɤβĻϥѡ饹ΥåɤΤΤѤ -åɤβĻѹˤModule饹export/unexport -ɤѤ롥 +åɤβĻѹˤModule饹Ƥpublic, +protected, privateγƥåɤѤ롥 ** ðۥå @@ -835,19 +824,6 @@ beginresqueǻȤbeginϤᤫ¹Ԥ롥㳰Ԥʤä ֥饹ˤѾ롥ؤХ饹ðۥåɤ¾Υ֥ Ȼظƥˤ륯饹åɤƯ롥 -** INCLUDE - -⥸塼롼ɤ뤳Ȥˤäơ饹ޤϥ⥸塼˵ǽ -ɲäǤ롥ߤΥ饹ޤϥ⥸塼(ȥåץ٥Ǥ -Object饹)˻ꤷ⥸塼롼ɤ빽ϰʲ̤ -롥 - - include ⥸塼̾ [`,' ⥸塼̾..] - -⥸塼롼ɤȡΥ⥸塼(ӤΥ⥸塼뤬 -˥롼ɤƤ⥸塼)ƤΥåɤѤ̤Τ -Х롼ɤϸꤵ줿¿ŷѾ(Mixin)Ȥ롥 - ** ALIAS ʲηǥåɤ̾Ĥ뤳ȤǤ롥 @@ -1204,6 +1180,9 @@ Rubyˤϸ̩ʰ̣ǤϴؿϤʤKernel饹δؿåɤ(Ƥ бĶѿͤ롥Ķѿ¸ߤ nil֤롥 + KCODE бƤɤɽʸ"EUC","SJIS"ޤ + "NONE" + VERSION rubyΥСʸ * Ȥ߹ߥ饹ȥ⥸塼 @@ -1797,6 +1776,18 @@ Methods: Single Methods: + self[pat] + glob(pat) + + ʸpatshΥ磻ɥɤȤŸ̤ʸ + Ȥ֤ϰʲ̤Ǥ롥 + + * Ǥդʸ(ʸޤ)Ȱ + ? Ǥդ1ʸȰ + [ ] []Τ줫1ʸȰ + {..} {}(ޤǶڤ줿)줫ʸ + + chdir(path) ȥǥ쥯ȥpathѹ롥 @@ -2664,20 +2655,32 @@ Private Methods: Methods: - export(name[, name..]) + include(module[, module..]) - nameǻꤵ줿åɤ̾ǸƤӽФǽˤ롥Ǥ - ̾åɤǤˤϲ⤷ʤ + ǻꤷ⥸塼롼ɤơåɡ + ä롥饹⥸塼̤Υ⥸塼롼ɤ뤳 + Ȥˤäơꤵ줿¿ŷѾ(Mixin)¸Ǥ롥 to_s ⥸塼ʸɽ֤⥸塼ξʸɽϥ 塼̾Ǥ롥 - unexport(name[, name..]) + private(name[, name..]) + + nameǻꤵ줿åɤƱ饹줿åɤ + ؿǤƤӽФǽˤ롥Ǥprivateåɤ + ˤϲ⤷ʤ + + protected(name[, name..]) nameǻꤵ줿åɤؿǤƤӽФǽˤ롥 - Ǥ˴ؿåɤǤˤϲ⤷ʤ + ǤprotectedåɤǤˤϲ⤷ʤ + + public(name[, name..]) + + nameǻꤵ줿åɤ̾ǸƤӽФǽˤ롥Ǥ + publicåɤǤˤϲ⤷ʤ ** Nil(饹) @@ -2741,8 +2744,15 @@ SuperClass: Builtin Methods: + extened(module[, module..]) + + ǻꤷ⥸塼self˥롼ɤ롥⥸塼 + ƤåɤðۥåɤȤɲä롥 + init_object + ֥Ȥ˸ƤӽФ롥ǥեȤϲ⤷ + ֥饹ɬפ˱ƺ뤳ȤԤƤ롥 is_member_of(class) @@ -2955,7 +2965,6 @@ Single Methods: åȤΥڥ֤λ openƱǤ롥 - ** Regexp(饹) ɽΥ饹ɽΥƥ/.../ȤɽưŪ @@ -3733,9 +3742,8 @@ PRIMARY : `(' EXPR `)' | super `(' [CALL_ARGS] `)' | super | PRIMARY `[' [ARGS] `]' - | `[' [ARGS] `]' - | `{' ARGS `}' - | `{' ASSOCS `}' + | `[' [ARGS [`,']] `]' + | `{' [ (ARGS|ASSOCS) [`,'] ] `}' | redo | break | continue @@ -360,8 +360,6 @@ Fstr_cmp(str1, str2) VALUE Freg_match(); -extern VALUE C_Glob; - static VALUE Fstr_match(x, y) struct RString *x, *y; @@ -382,9 +380,6 @@ Fstr_match(x, y) return INT2FIX(start); default: - if (obj_is_kind_of(y, C_Glob)) { - return Fglob_match(y, x); - } Fail("type mismatch"); break; } @@ -684,8 +679,8 @@ str_sub(str, pat, val, once) for (offset=0, n=0; (beg=research(pat, str, offset)) >= 0; - offset=RREGEXP(pat)->ptr->regs.start[0]+STRLEN(val)) { - end = RREGEXP(pat)->ptr->regs.end[0]-1; + offset=BEG(0)+STRLEN(val)) { + end = END(0)-1; sub = re_regsub(val); str_replace2(str, beg, end, sub); n++; @@ -1339,11 +1334,8 @@ Fstr_split(argc, argv, str) int last_null = 0; int idx; -#define LMATCH spat->ptr->regs.start -#define RMATCH spat->ptr->regs.end - while ((end = research(spat, str, start)) >= 0) { - if (start == end && LMATCH[0] == RMATCH[0]) { + if (start == end && BEG(0) == END(0)) { if (last_null == 1) { if (ismbchar(str->ptr[beg])) ary_push(result, str_substr(str, beg, 2)); @@ -1360,17 +1352,17 @@ Fstr_split(argc, argv, str) } else { ary_push(result, str_substr(str, beg, end-beg)); - beg = start = RMATCH[0]; + beg = start = END(0); if (limit && lim <= ++i) break; } last_null = 0; for (idx=1; idx < 10; idx++) { - if (LMATCH[idx] == -1) break; - if (LMATCH[idx] == RMATCH[idx]) + if (BEG(idx) == -1) break; + if (BEG(idx) == END(idx)) tmp = str_new(0, 0); else - tmp = str_subseq(str, LMATCH[idx], RMATCH[idx]-1); + tmp = str_subseq(str, BEG(idx), END(idx)-1); ary_push(result, tmp); if (limit && lim <= ++i) break; } @@ -1380,9 +1372,6 @@ Fstr_split(argc, argv, str) if (str->len > beg) { ary_push(result, str_subseq(str, beg, -1)); } - else if (str->len == beg) { - ary_push(result, str_new(0, 0)); - } return result; } diff --git a/util.c b/util.c new file mode 100644 index 0000000000..62e34f5753 --- /dev/null +++ b/util.c @@ -0,0 +1,53 @@ +/************************************************ + + util.c - + + $Author$ + $Revision$ + $Date$ + created at: Fri Mar 10 17:22:34 JST 1995 + +************************************************/ + +#include "defines.h" +#include "config.h" +#include "util.h" + +unsigned long +scan_oct(start, len, retlen) +char *start; +int len; +int *retlen; +{ + register char *s = start; + register unsigned long retval = 0; + + while (len-- && *s >= '0' && *s <= '7') { + retval <<= 3; + retval |= *s++ - '0'; + } + *retlen = s - start; + return retval; +} + +extern char *strchr(); + +unsigned long +scan_hex(start, len, retlen) +char *start; +int len; +int *retlen; +{ + static char hexdigit[] = "0123456789abcdef0123456789ABCDEFx"; + register char *s = start; + register unsigned long retval = 0; + char *tmp; + + while (len-- && *s && (tmp = strchr(hexdigit, *s))) { + retval <<= 4; + retval |= (tmp - hexdigit) & 15; + s++; + } + *retlen = s - start; + return retval; +} diff --git a/util.h b/util.h new file mode 100644 index 0000000000..e2ada4176b --- /dev/null +++ b/util.h @@ -0,0 +1,16 @@ +/************************************************ + + util.h - + + $Author$ + $Date$ + created at: Thu Mar 9 11:55:53 JST 1995 + +************************************************/ +#ifndef UTIL_H +#define UTIL_H + +unsigned long scan_hex(); +unsigned long scan_oct(); + +#endif /* UTIL_H */ diff --git a/variable.c b/variable.c index b65a1fc05e..34bd8f4de5 100644 --- a/variable.c +++ b/variable.c @@ -203,17 +203,6 @@ rb_readonly_hook(val, id) } VALUE -rb_id2class(id) - ID id; -{ - VALUE class; - - if (st_lookup(class_tbl, id, &class)) - return class; - return Qnil; -} - -VALUE rb_gvar_get(entry) struct global_entry *entry; { @@ -1,2 +1,2 @@ -#define RUBY_VERSION "0.68" -#define VERSION_DATE "95/02/23" +#define RUBY_VERSION "0.69" +#define VERSION_DATE "95/03/15" |