1 /* src/vm/utf.c - utf functions
3 Copyright (C) 1996-2005 R. Grafl, A. Krall, C. Kruegel, C. Oates,
4 R. Obermaisser, M. Platter, M. Probst, S. Ring, E. Steiner,
5 C. Thalinger, D. Thuernbeck, P. Tomsich, C. Ullrich, J. Wenninger,
6 Institut f. Computersprachen - TU Wien
8 This file is part of CACAO.
10 This program is free software; you can redistribute it and/or
11 modify it under the terms of the GNU General Public License as
12 published by the Free Software Foundation; either version 2, or (at
13 your option) any later version.
15 This program is distributed in the hope that it will be useful, but
16 WITHOUT ANY WARRANTY; without even the implied warranty of
17 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
18 General Public License for more details.
20 You should have received a copy of the GNU General Public License
21 along with this program; if not, write to the Free Software
22 Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
25 Contact: cacao@complang.tuwien.ac.at
27 Authors: Reinhard Grafl
33 $Id: utf8.c 3512 2005-10-28 11:29:52Z twisti $
40 #include "mm/memory.h"
41 #include "vm/exceptions.h"
42 #include "vm/options.h"
43 #include "vm/statistics.h"
44 #include "vm/stringlocal.h"
45 #include "vm/tables.h"
49 hashtable utf_hash; /* hashtable for utf8-symbols */
52 /* utf-symbols for pointer comparison of frequently used strings **************/
54 utf *utf_java_lang_Object; /* java/lang/Object */
56 utf *utf_java_lang_Class;
57 utf *utf_java_lang_ClassLoader;
58 utf *utf_java_lang_Cloneable;
59 utf *utf_java_lang_SecurityManager;
60 utf *utf_java_lang_String;
61 utf *utf_java_lang_System;
62 utf *utf_java_lang_ThreadGroup;
63 utf *utf_java_io_Serializable;
65 utf *utf_java_lang_Throwable;
66 utf *utf_java_lang_VMThrowable;
67 utf *utf_java_lang_Error;
68 utf *utf_java_lang_NoClassDefFoundError;
69 utf *utf_java_lang_OutOfMemoryError;
71 utf *utf_java_lang_Exception;
72 utf *utf_java_lang_ClassNotFoundException;
73 utf *utf_java_lang_IllegalArgumentException;
75 utf *utf_java_lang_NullPointerException;
77 utf* utf_java_lang_Void;
78 utf* utf_java_lang_Boolean;
79 utf* utf_java_lang_Byte;
80 utf* utf_java_lang_Character;
81 utf* utf_java_lang_Short;
82 utf* utf_java_lang_Integer;
83 utf* utf_java_lang_Long;
84 utf* utf_java_lang_Float;
85 utf* utf_java_lang_Double;
87 utf *utf_java_lang_StackTraceElement;
88 utf *utf_java_lang_reflect_Constructor;
89 utf *utf_java_lang_reflect_Field;
90 utf *utf_java_lang_reflect_Method;
91 utf *utf_java_util_Vector;
93 utf *utf_InnerClasses; /* InnerClasses */
94 utf *utf_ConstantValue; /* ConstantValue */
95 utf *utf_Code; /* Code */
96 utf *utf_Exceptions; /* Exceptions */
97 utf *utf_LineNumberTable; /* LineNumberTable */
98 utf *utf_SourceFile; /* SourceFile */
100 utf *utf_init; /* <init> */
101 utf *utf_clinit; /* <clinit> */
102 utf *utf_clone; /* clone */
103 utf *utf_finalize; /* finalize */
104 utf *utf_run; /* run */
106 utf *utf_add; /* add */
107 utf *utf_remove; /* remove */
108 utf *utf_put; /* put */
109 utf *utf_get; /* get */
110 utf *utf_value; /* value */
112 utf *utf_fillInStackTrace;
113 utf *utf_getSystemClassLoader;
115 utf *utf_printStackTrace;
126 utf *utf_void__void; /* ()V */
127 utf *utf_boolean__void; /* (Z)V */
128 utf *utf_byte__void; /* (B)V */
129 utf *utf_char__void; /* (C)V */
130 utf *utf_short__void; /* (S)V */
131 utf *utf_int__void; /* (I)V */
132 utf *utf_long__void; /* (J)V */
133 utf *utf_float__void; /* (F)V */
134 utf *utf_double__void; /* (D)V */
136 utf *utf_void__java_lang_ClassLoader; /* ()Ljava/lang/ClassLoader; */
137 utf *utf_void__java_lang_Object; /* ()Ljava/lang/Object; */
138 utf *utf_void__java_lang_Throwable; /* ()Ljava/lang/Throwable; */
139 utf *utf_java_lang_Object__java_lang_Object;
140 utf *utf_java_lang_String__void; /* (Ljava/lang/String;)V */
141 utf *utf_java_lang_String__java_lang_Class;
142 utf *utf_java_lang_Throwable__void; /* (Ljava/lang/Throwable;)V */
144 utf *utf_not_named_yet; /* special name for unnamed classes */
146 utf *array_packagename;
149 /* utf_init ********************************************************************
151 Initializes the utf8 subsystem.
153 *******************************************************************************/
157 /* create utf-symbols for pointer comparison of frequently used strings */
159 utf_java_lang_Object = utf_new_char("java/lang/Object");
161 utf_java_lang_Class = utf_new_char("java/lang/Class");
162 utf_java_lang_ClassLoader = utf_new_char("java/lang/ClassLoader");
163 utf_java_lang_Cloneable = utf_new_char("java/lang/Cloneable");
164 utf_java_lang_SecurityManager = utf_new_char("java/lang/SecurityManager");
165 utf_java_lang_String = utf_new_char("java/lang/String");
166 utf_java_lang_System = utf_new_char("java/lang/System");
167 utf_java_lang_ThreadGroup = utf_new_char("java/lang/ThreadGroup");
168 utf_java_io_Serializable = utf_new_char("java/io/Serializable");
170 utf_java_lang_Throwable = utf_new_char(string_java_lang_Throwable);
171 utf_java_lang_VMThrowable = utf_new_char(string_java_lang_VMThrowable);
172 utf_java_lang_Error = utf_new_char(string_java_lang_Error);
174 utf_java_lang_NoClassDefFoundError =
175 utf_new_char(string_java_lang_NoClassDefFoundError);
177 utf_java_lang_OutOfMemoryError =
178 utf_new_char(string_java_lang_OutOfMemoryError);
180 utf_java_lang_Exception = utf_new_char(string_java_lang_Exception);
182 utf_java_lang_ClassNotFoundException =
183 utf_new_char(string_java_lang_ClassNotFoundException);
185 utf_java_lang_IllegalArgumentException =
186 utf_new_char(string_java_lang_IllegalArgumentException);
188 utf_java_lang_NullPointerException =
189 utf_new_char(string_java_lang_NullPointerException);
191 utf_java_lang_Void = utf_new_char("java/lang/Void");
192 utf_java_lang_Boolean = utf_new_char("java/lang/Boolean");
193 utf_java_lang_Byte = utf_new_char("java/lang/Byte");
194 utf_java_lang_Character = utf_new_char("java/lang/Character");
195 utf_java_lang_Short = utf_new_char("java/lang/Short");
196 utf_java_lang_Integer = utf_new_char("java/lang/Integer");
197 utf_java_lang_Long = utf_new_char("java/lang/Long");
198 utf_java_lang_Float = utf_new_char("java/lang/Float");
199 utf_java_lang_Double = utf_new_char("java/lang/Double");
201 utf_java_lang_StackTraceElement =
202 utf_new_char("java/lang/StackTraceElement");
204 utf_java_lang_reflect_Constructor =
205 utf_new_char("java/lang/reflect/Constructor");
207 utf_java_lang_reflect_Field = utf_new_char("java/lang/reflect/Field");
208 utf_java_lang_reflect_Method = utf_new_char("java/lang/reflect/Method");
209 utf_java_util_Vector = utf_new_char("java/util/Vector");
211 utf_InnerClasses = utf_new_char("InnerClasses");
212 utf_ConstantValue = utf_new_char("ConstantValue");
213 utf_Code = utf_new_char("Code");
214 utf_Exceptions = utf_new_char("Exceptions");
215 utf_LineNumberTable = utf_new_char("LineNumberTable");
216 utf_SourceFile = utf_new_char("SourceFile");
218 utf_init = utf_new_char("<init>");
219 utf_clinit = utf_new_char("<clinit>");
220 utf_clone = utf_new_char("clone");
221 utf_finalize = utf_new_char("finalize");
222 utf_run = utf_new_char("run");
224 utf_add = utf_new_char("add");
225 utf_remove = utf_new_char("remove");
226 utf_put = utf_new_char("put");
227 utf_get = utf_new_char("get");
228 utf_value = utf_new_char("value");
230 utf_printStackTrace = utf_new_char("printStackTrace");
231 utf_fillInStackTrace = utf_new_char("fillInStackTrace");
232 utf_loadClass = utf_new_char("loadClass");
233 utf_getSystemClassLoader = utf_new_char("getSystemClassLoader");
235 utf_Z = utf_new_char("Z");
236 utf_B = utf_new_char("B");
237 utf_C = utf_new_char("C");
238 utf_S = utf_new_char("S");
239 utf_I = utf_new_char("I");
240 utf_J = utf_new_char("J");
241 utf_F = utf_new_char("F");
242 utf_D = utf_new_char("D");
244 utf_void__void = utf_new_char("()V");
245 utf_boolean__void = utf_new_char("(Z)V");
246 utf_byte__void = utf_new_char("(B)V");
247 utf_char__void = utf_new_char("(C)V");
248 utf_short__void = utf_new_char("(S)V");
249 utf_int__void = utf_new_char("(I)V");
250 utf_long__void = utf_new_char("(J)V");
251 utf_float__void = utf_new_char("(F)V");
252 utf_double__void = utf_new_char("(D)V");
253 utf_void__java_lang_Object = utf_new_char("()Ljava/lang/Object;");
254 utf_void__java_lang_Throwable = utf_new_char("()Ljava/lang/Throwable;");
256 utf_void__java_lang_ClassLoader =
257 utf_new_char("()Ljava/lang/ClassLoader;");
259 utf_java_lang_Object__java_lang_Object =
260 utf_new_char("(Ljava/lang/Object;)Ljava/lang/Object;");
262 utf_java_lang_String__void = utf_new_char("(Ljava/lang/String;)V");
264 utf_java_lang_String__java_lang_Class =
265 utf_new_char("(Ljava/lang/String;)Ljava/lang/Class;");
267 utf_java_lang_Throwable__void = utf_new_char("(Ljava/lang/Throwable;)V");
269 utf_not_named_yet = utf_new_char("\t<not_named_yet>");
271 array_packagename = utf_new_char("\t<the array package>");
275 /* utf_hashkey *****************************************************************
277 The hashkey is computed from the utf-text by using up to 8
278 characters. For utf-symbols longer than 15 characters 3 characters
279 are taken from the beginning and the end, 2 characters are taken
282 *******************************************************************************/
284 #define nbs(val) ((u4) *(++text) << val) /* get next byte, left shift by val */
285 #define fbs(val) ((u4) *( text) << val) /* get first byte, left shift by val */
287 u4 utf_hashkey(const char *text, u4 length)
289 const char *start_pos = text; /* pointer to utf text */
293 case 0: /* empty string */
296 case 1: return fbs(0);
297 case 2: return fbs(0) ^ nbs(3);
298 case 3: return fbs(0) ^ nbs(3) ^ nbs(5);
299 case 4: return fbs(0) ^ nbs(2) ^ nbs(4) ^ nbs(6);
300 case 5: return fbs(0) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(6);
301 case 6: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(5) ^ nbs(6);
302 case 7: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6);
303 case 8: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7);
310 return a ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7) ^ nbs(8);
319 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9);
328 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9) ^ nbs(10);
340 return a ^ nbs(9) ^ nbs(10);
352 return a ^ nbs(9) ^ nbs(10);
363 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
374 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
376 default: /* 3 characters from beginning */
382 /* 2 characters from middle */
383 text = start_pos + (length / 2);
388 /* 3 characters from end */
389 text = start_pos + length - 4;
394 return a ^ nbs(10) ^ nbs(11);
399 /* utf_hashkey *****************************************************************
401 Compute the hashkey of a unicode string.
403 *******************************************************************************/
405 u4 unicode_hashkey(u2 *text, u2 len)
407 return utf_hashkey((char *) text, len);
411 /* utf_new *********************************************************************
413 Creates a new utf-symbol, the text of the symbol is passed as a
414 u1-array. The function searches the utf-hashtable for a utf-symbol
415 with this text. On success the element returned, otherwise a new
416 hashtable element is created.
418 If the number of entries in the hashtable exceeds twice the size of
419 the hashtable slots a reorganization of the hashtable is done and
420 the utf symbols are copied to a new hashtable with doubled size.
422 *******************************************************************************/
424 utf *utf_new_intern(const char *text, u2 length);
426 utf *utf_new(const char *text, u2 length)
430 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
434 r = utf_new_intern(text, length);
436 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
444 utf *utf_new_intern(const char *text, u2 length)
446 u4 key; /* hashkey computed from utf-text */
447 u4 slot; /* slot in hashtable */
448 utf *u; /* hashtable element */
456 key = utf_hashkey(text, length);
457 slot = key & (utf_hash.size - 1);
458 u = utf_hash.ptr[slot];
460 /* search external hash chain for utf-symbol */
462 if (u->blength == length) {
464 /* compare text of hashtable elements */
465 for (i = 0; i < length; i++)
466 if (text[i] != u->text[i]) goto nomatch;
470 count_utf_new_found++;
473 /* symbol found in hashtable */
477 u = u->hashlink; /* next element in external chain */
482 count_utf_len += sizeof(utf) + length + 1;
485 /* location in hashtable found, create new utf element */
487 u->blength = length; /* length in bytes of utfstring */
488 u->hashlink = utf_hash.ptr[slot]; /* link in external hashchain */
489 u->text = mem_alloc(length + 1);/* allocate memory for utf-text */
490 memcpy(u->text, text, length); /* copy utf-text */
491 u->text[length] = '\0';
492 utf_hash.ptr[slot] = u; /* insert symbol into table */
494 utf_hash.entries++; /* update number of entries */
496 if (utf_hash.entries > (utf_hash.size * 2)) {
498 /* reorganization of hashtable, average length of
499 the external chains is approx. 2 */
503 hashtable newhash; /* the new hashtable */
505 /* create new hashtable, double the size */
506 init_hashtable(&newhash, utf_hash.size * 2);
507 newhash.entries = utf_hash.entries;
511 count_utf_len += sizeof(utf*) * utf_hash.size;
514 /* transfer elements to new hashtable */
515 for (i = 0; i < utf_hash.size; i++) {
516 u = (utf *) utf_hash.ptr[i];
518 utf *nextu = u->hashlink;
519 u4 slot = utf_hashkey(u->text, u->blength) & (newhash.size - 1);
521 u->hashlink = (utf *) newhash.ptr[slot];
522 newhash.ptr[slot] = u;
524 /* follow link in external hash chain */
529 /* dispose old table */
530 MFREE(utf_hash.ptr, void*, utf_hash.size);
538 /* utf_new_u2 ******************************************************************
540 Make utf symbol from u2 array, if isclassname is true '.' is
543 *******************************************************************************/
545 utf *utf_new_u2(u2 *unicode_pos, u4 unicode_length, bool isclassname)
547 char *buffer; /* memory buffer for unicode characters */
548 char *pos; /* pointer to current position in buffer */
549 u4 left; /* unicode characters left */
550 u4 buflength; /* utf length in bytes of the u2 array */
551 utf *result; /* resulting utf-string */
554 /* determine utf length in bytes and allocate memory */
556 buflength = u2_utflength(unicode_pos, unicode_length);
557 buffer = MNEW(char, buflength);
562 for (i = 0; i++ < unicode_length; unicode_pos++) {
563 /* next unicode character */
566 if ((c != 0) && (c < 0x80)) {
569 if ((int) left < 0) break;
570 /* convert classname */
571 if (isclassname && c == '.')
576 } else if (c < 0x800) {
578 unsigned char high = c >> 6;
579 unsigned char low = c & 0x3F;
581 if ((int) left < 0) break;
582 *pos++ = high | 0xC0;
588 char mid = (c >> 6) & 0x3F;
591 if ((int) left < 0) break;
592 *pos++ = high | 0xE0;
598 /* insert utf-string into symbol-table */
599 result = utf_new(buffer,buflength);
601 MFREE(buffer, char, buflength);
607 /* utf_new_char ****************************************************************
609 Creates a new utf symbol, the text for this symbol is passed as a
610 c-string ( = char* ).
612 *******************************************************************************/
614 utf *utf_new_char(const char *text)
616 return utf_new(text, strlen(text));
620 /* utf_new_char_classname ******************************************************
622 Creates a new utf symbol, the text for this symbol is passed as a
623 c-string ( = char* ) "." characters are going to be replaced by
624 "/". Since the above function is used often, this is a separte
625 function, instead of an if.
627 *******************************************************************************/
629 utf *utf_new_char_classname(const char *text)
631 if (strchr(text, '.')) {
632 char *txt = strdup(text);
633 char *end = txt + strlen(txt);
637 for (c = txt; c < end; c++)
638 if (*c == '.') *c = '/';
640 tmpRes = utf_new(txt, strlen(txt));
646 return utf_new(text, strlen(text));
650 /* utf_nextu2 ******************************************************************
652 Read the next unicode character from the utf string and increment
653 the utf-string pointer accordingly.
655 *******************************************************************************/
657 u2 utf_nextu2(char **utf_ptr)
659 /* uncompressed unicode character */
661 /* current position in utf text */
662 unsigned char *utf = (unsigned char *) (*utf_ptr);
663 /* bytes representing the unicode character */
664 unsigned char ch1, ch2, ch3;
665 /* number of bytes used to represent the unicode character */
668 switch ((ch1 = utf[0]) >> 4) {
669 default: /* 1 byte */
673 case 0xD: /* 2 bytes */
674 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
675 unsigned char high = ch1 & 0x1F;
676 unsigned char low = ch2 & 0x3F;
677 unicode_char = (high << 6) + low;
682 case 0xE: /* 2 or 3 bytes */
683 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
684 if (((ch3 = utf[2]) & 0xC0) == 0x80) {
685 unsigned char low = ch3 & 0x3f;
686 unsigned char mid = ch2 & 0x3f;
687 unsigned char high = ch1 & 0x0f;
688 unicode_char = (((high << 6) + mid) << 6) + low;
696 /* update position in utf-text */
697 *utf_ptr = (char *) (utf + len);
703 /* utf_strlen ******************************************************************
705 Determine number of unicode characters in the utf string.
707 *******************************************************************************/
709 u4 utf_strlen(utf *u)
711 char *endpos; /* points behind utf string */
712 char *utf_ptr; /* current position in utf text */
713 u4 len = 0; /* number of unicode characters */
716 *exceptionptr = new_nullpointerexception();
723 while (utf_ptr < endpos) {
725 /* next unicode character */
726 utf_nextu2(&utf_ptr);
729 if (utf_ptr != endpos)
730 /* string ended abruptly */
731 throw_cacao_exception_exit(string_java_lang_InternalError,
732 "Illegal utf8 string");
738 /* u2_utflength ****************************************************************
740 Returns the utf length in bytes of a u2 array.
742 *******************************************************************************/
744 u4 u2_utflength(u2 *text, u4 u2_length)
746 u4 result_len = 0; /* utf length in bytes */
747 u2 ch; /* current unicode character */
750 for (len = 0; len < u2_length; len++) {
751 /* next unicode character */
754 /* determine bytes required to store unicode character as utf */
755 if (ch && (ch < 0x80))
767 /* utf_display *****************************************************************
769 Write utf symbol to stdout (for debugging purposes).
771 *******************************************************************************/
773 void utf_display(utf *u)
775 char *endpos; /* points behind utf string */
776 char *utf_ptr; /* current position in utf text */
787 while (utf_ptr < endpos) {
788 /* read next unicode character */
789 u2 c = utf_nextu2(&utf_ptr);
790 if (c >= 32 && c <= 127) printf("%c", c);
798 /* utf_display_classname *******************************************************
800 Write utf symbol to stdout with `/' converted to `.' (for debugging
803 *******************************************************************************/
805 void utf_display_classname(utf *u)
807 char *endpos; /* points behind utf string */
808 char *utf_ptr; /* current position in utf text */
819 while (utf_ptr < endpos) {
820 /* read next unicode character */
821 u2 c = utf_nextu2(&utf_ptr);
822 if (c == '/') c = '.';
823 if (c >= 32 && c <= 127) printf("%c", c);
831 /* utf_sprint ******************************************************************
833 Write utf symbol into c-string (for debugging purposes).
835 *******************************************************************************/
837 void utf_sprint(char *buffer, utf *u)
839 char *endpos; /* points behind utf string */
840 char *utf_ptr; /* current position in utf text */
841 u2 pos = 0; /* position in c-string */
844 strcpy(buffer, "NULL");
851 while (utf_ptr < endpos)
852 /* copy next unicode character */
853 buffer[pos++] = utf_nextu2(&utf_ptr);
855 /* terminate string */
860 /* utf_sprint_classname ********************************************************
862 Write utf symbol into c-string with `/' converted to `.' (for debugging
865 *******************************************************************************/
867 void utf_sprint_classname(char *buffer, utf *u)
869 char *endpos; /* points behind utf string */
870 char *utf_ptr; /* current position in utf text */
871 u2 pos = 0; /* position in c-string */
874 strcpy(buffer, "NULL");
881 while (utf_ptr < endpos) {
882 /* copy next unicode character */
883 u2 c = utf_nextu2(&utf_ptr);
884 if (c == '/') c = '.';
888 /* terminate string */
893 /* utf_strcat ******************************************************************
895 Like libc strcat, but uses an utf8 string.
897 *******************************************************************************/
899 void utf_strcat(char *buffer, utf *u)
901 utf_sprint(buffer + strlen(buffer), u);
905 /* utf_strcat_classname ********************************************************
907 Like libc strcat, but uses an utf8 string.
909 *******************************************************************************/
911 void utf_strcat_classname(char *buffer, utf *u)
913 utf_sprint_classname(buffer + strlen(buffer), u);
917 /* utf_fprint ******************************************************************
919 Write utf symbol into file.
921 *******************************************************************************/
923 void utf_fprint(FILE *file, utf *u)
925 char *endpos; /* points behind utf string */
926 char *utf_ptr; /* current position in utf text */
934 while (utf_ptr < endpos) {
935 /* read next unicode character */
936 u2 c = utf_nextu2(&utf_ptr);
938 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
939 else fprintf(file, "?");
944 /* utf_fprint_classname ********************************************************
946 Write utf symbol into file with `/' converted to `.'.
948 *******************************************************************************/
950 void utf_fprint_classname(FILE *file, utf *u)
952 char *endpos; /* points behind utf string */
953 char *utf_ptr; /* current position in utf text */
961 while (utf_ptr < endpos) {
962 /* read next unicode character */
963 u2 c = utf_nextu2(&utf_ptr);
964 if (c == '/') c = '.';
966 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
967 else fprintf(file, "?");
972 /* is_valid_utf ****************************************************************
974 Return true if the given string is a valid UTF-8 string.
976 utf_ptr...points to first character
977 end_pos...points after last character
979 *******************************************************************************/
981 /* static unsigned long min_codepoint[6] = {0,1L<<7,1L<<11,1L<<16,1L<<21,1L<<26}; */
983 bool is_valid_utf(char *utf_ptr, char *end_pos)
990 if (end_pos < utf_ptr) return false;
991 bytes = end_pos - utf_ptr;
995 if (!c) return false; /* 0x00 is not allowed */
996 if ((c & 0x80) == 0) continue; /* ASCII */
998 if ((c & 0xe0) == 0xc0) len = 1; /* 110x xxxx */
999 else if ((c & 0xf0) == 0xe0) len = 2; /* 1110 xxxx */
1000 else if ((c & 0xf8) == 0xf0) len = 3; /* 1111 0xxx */
1001 else if ((c & 0xfc) == 0xf8) len = 4; /* 1111 10xx */
1002 else if ((c & 0xfe) == 0xfc) len = 5; /* 1111 110x */
1003 else return false; /* invalid leading byte */
1005 if (len > 2) return false; /* Java limitation */
1007 v = (unsigned long)c & (0x3f >> len);
1009 if ((bytes -= len) < 0) return false; /* missing bytes */
1011 for (i = len; i--; ) {
1013 if ((c & 0xc0) != 0x80) /* 10xx xxxx */
1015 v = (v << 6) | (c & 0x3f);
1019 if (len != 1) return false; /* Java special */
1022 /* Sun Java seems to allow overlong UTF-8 encodings */
1024 /* if (v < min_codepoint[len]) */
1025 /* XXX throw exception? */
1028 /* surrogates in UTF-8 seem to be allowed in Java classfiles */
1029 /* if (v >= 0xd800 && v <= 0xdfff) return false; */ /* surrogates */
1031 /* even these seem to be allowed */
1032 /* if (v == 0xfffe || v == 0xffff) return false; */ /* invalid codepoints */
1039 /* is_valid_name ***************************************************************
1041 Return true if the given string may be used as a class/field/method
1042 name. (Currently this only disallows empty strings and control
1045 NOTE: The string is assumed to have passed is_valid_utf!
1047 utf_ptr...points to first character
1048 end_pos...points after last character
1050 *******************************************************************************/
1052 bool is_valid_name(char *utf_ptr, char *end_pos)
1054 if (end_pos <= utf_ptr) return false; /* disallow empty names */
1056 while (utf_ptr < end_pos) {
1057 unsigned char c = *utf_ptr++;
1059 if (c < 0x20) return false; /* disallow control characters */
1060 if (c == 0xc0 && (unsigned char) *utf_ptr == 0x80) /* disallow zero */
1067 bool is_valid_name_utf(utf *u)
1069 return is_valid_name(u->text, UTF_END(u));
1073 /* utf_show ********************************************************************
1075 Writes the utf symbols in the utfhash to stdout and displays the
1076 number of external hash chains grouped according to the chainlength
1077 (for debugging purposes).
1079 *******************************************************************************/
1084 #define CHAIN_LIMIT 20 /* limit for seperated enumeration */
1086 u4 chain_count[CHAIN_LIMIT]; /* numbers of chains */
1087 u4 max_chainlength = 0; /* maximum length of the chains */
1088 u4 sum_chainlength = 0; /* sum of the chainlengths */
1089 u4 beyond_limit = 0; /* number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1092 printf ("UTF-HASH:\n");
1094 /* show element of utf-hashtable */
1095 for (i=0; i<utf_hash.size; i++) {
1096 utf *u = utf_hash.ptr[i];
1098 printf ("SLOT %d: ", (int) i);
1110 printf ("UTF-HASH: %d slots for %d entries\n",
1111 (int) utf_hash.size, (int) utf_hash.entries );
1114 if (utf_hash.entries == 0)
1117 printf("chains:\n chainlength number of chains %% of utfstrings\n");
1119 for (i=0;i<CHAIN_LIMIT;i++)
1122 /* count numbers of hashchains according to their length */
1123 for (i=0; i<utf_hash.size; i++) {
1125 utf *u = (utf*) utf_hash.ptr[i];
1126 u4 chain_length = 0;
1128 /* determine chainlength */
1134 /* update sum of all chainlengths */
1135 sum_chainlength+=chain_length;
1137 /* determine the maximum length of the chains */
1138 if (chain_length>max_chainlength)
1139 max_chainlength = chain_length;
1141 /* update number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1142 if (chain_length>=CHAIN_LIMIT) {
1143 beyond_limit+=chain_length;
1144 chain_length=CHAIN_LIMIT-1;
1147 /* update number of hashchains of current length */
1148 chain_count[chain_length]++;
1151 /* display results */
1152 for (i=1;i<CHAIN_LIMIT-1;i++)
1153 printf(" %2d %17d %18.2f%%\n",i,chain_count[i],(((float) chain_count[i]*i*100)/utf_hash.entries));
1155 printf(" >=%2d %17d %18.2f%%\n",CHAIN_LIMIT-1,chain_count[CHAIN_LIMIT-1],((float) beyond_limit*100)/utf_hash.entries);
1158 printf("max. chainlength:%5d\n",max_chainlength);
1160 /* avg. chainlength = sum of chainlengths / number of chains */
1161 printf("avg. chainlength:%5.2f\n",(float) sum_chainlength / (utf_hash.size-chain_count[0]));
1166 * These are local overrides for various environment variables in Emacs.
1167 * Please do not remove this and leave it at the end of the file, where
1168 * Emacs will automagically detect them.
1169 * ---------------------------------------------------------------------
1172 * indent-tabs-mode: t