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 2741 2005-06-20 09:57:14Z 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_Exception;
69 utf *utf_java_lang_NoClassDefFoundError;
70 utf *utf_java_lang_OutOfMemoryError;
71 utf *utf_java_lang_ClassNotFoundException;
73 utf* utf_java_lang_Void;
74 utf* utf_java_lang_Boolean;
75 utf* utf_java_lang_Byte;
76 utf* utf_java_lang_Character;
77 utf* utf_java_lang_Short;
78 utf* utf_java_lang_Integer;
79 utf* utf_java_lang_Long;
80 utf* utf_java_lang_Float;
81 utf* utf_java_lang_Double;
83 utf *utf_java_lang_StackTraceElement;
84 utf *utf_java_lang_reflect_Constructor;
85 utf *utf_java_lang_reflect_Field;
86 utf *utf_java_lang_reflect_Method;
87 utf *utf_java_util_Vector;
89 utf *utf_InnerClasses; /* InnerClasses */
90 utf *utf_ConstantValue; /* ConstantValue */
91 utf *utf_Code; /* Code */
92 utf *utf_Exceptions; /* Exceptions */
93 utf *utf_LineNumberTable; /* LineNumberTable */
94 utf *utf_SourceFile; /* SourceFile */
96 utf *utf_init; /* <init> */
97 utf *utf_clinit; /* <clinit> */
98 utf *utf_clone; /* clone */
99 utf *utf_finalize; /* finalize */
100 utf *utf_run; /* run */
102 utf *utf_fillInStackTrace;
103 utf *utf_getSystemClassLoader;
105 utf *utf_printStackTrace;
107 utf *utf_void__void; /* ()V */
108 utf *utf_boolean__void; /* (Z)V */
109 utf *utf_byte__void; /* (B)V */
110 utf *utf_char__void; /* (C)V */
111 utf *utf_short__void; /* (S)V */
112 utf *utf_int__void; /* (I)V */
113 utf *utf_long__void; /* (J)V */
114 utf *utf_float__void; /* (F)V */
115 utf *utf_double__void; /* (D)V */
117 utf *utf_void__java_lang_ClassLoader; /* ()Ljava/lang/ClassLoader; */
118 utf *utf_void__java_lang_Object; /* ()Ljava/lang/Object; */
119 utf *utf_void__java_lang_Throwable; /* ()Ljava/lang/Throwable; */
120 utf *utf_java_lang_String__void; /* (Ljava/lang/String;)V */
121 utf *utf_java_lang_String__java_lang_Class;
122 utf *utf_java_lang_Throwable__void; /* (Ljava/lang/Throwable;)V */
124 utf *utf_not_named_yet; /* special name for unnamed classes */
126 utf *array_packagename;
129 /* utf_init ********************************************************************
131 Initializes the utf8 subsystem.
133 *******************************************************************************/
137 /* create utf-symbols for pointer comparison of frequently used strings */
139 utf_java_lang_Object = utf_new_char("java/lang/Object");
141 utf_java_lang_Class = utf_new_char("java/lang/Class");
142 utf_java_lang_ClassLoader = utf_new_char("java/lang/ClassLoader");
143 utf_java_lang_Cloneable = utf_new_char("java/lang/Cloneable");
144 utf_java_lang_SecurityManager = utf_new_char("java/lang/SecurityManager");
145 utf_java_lang_String = utf_new_char("java/lang/String");
146 utf_java_lang_System = utf_new_char("java/lang/System");
147 utf_java_lang_ThreadGroup = utf_new_char("java/lang/ThreadGroup");
148 utf_java_io_Serializable = utf_new_char("java/io/Serializable");
150 utf_java_lang_Throwable = utf_new_char(string_java_lang_Throwable);
151 utf_java_lang_VMThrowable = utf_new_char(string_java_lang_VMThrowable);
152 utf_java_lang_Error = utf_new_char(string_java_lang_Error);
153 utf_java_lang_Exception = utf_new_char(string_java_lang_Exception);
155 utf_java_lang_NoClassDefFoundError =
156 utf_new_char(string_java_lang_NoClassDefFoundError);
158 utf_java_lang_OutOfMemoryError =
159 utf_new_char(string_java_lang_OutOfMemoryError);
161 utf_java_lang_ClassNotFoundException =
162 utf_new_char(string_java_lang_ClassNotFoundException);
164 utf_java_lang_Void = utf_new_char("java/lang/Void");
165 utf_java_lang_Boolean = utf_new_char("java/lang/Boolean");
166 utf_java_lang_Byte = utf_new_char("java/lang/Byte");
167 utf_java_lang_Character = utf_new_char("java/lang/Character");
168 utf_java_lang_Short = utf_new_char("java/lang/Short");
169 utf_java_lang_Integer = utf_new_char("java/lang/Integer");
170 utf_java_lang_Long = utf_new_char("java/lang/Long");
171 utf_java_lang_Float = utf_new_char("java/lang/Float");
172 utf_java_lang_Double = utf_new_char("java/lang/Double");
174 utf_java_lang_StackTraceElement =
175 utf_new_char("java/lang/StackTraceElement");
177 utf_java_lang_reflect_Constructor =
178 utf_new_char("java/lang/reflect/Constructor");
180 utf_java_lang_reflect_Field = utf_new_char("java/lang/reflect/Field");
181 utf_java_lang_reflect_Method = utf_new_char("java/lang/reflect/Method");
182 utf_java_util_Vector = utf_new_char("java/util/Vector");
184 utf_InnerClasses = utf_new_char("InnerClasses");
185 utf_ConstantValue = utf_new_char("ConstantValue");
186 utf_Code = utf_new_char("Code");
187 utf_Exceptions = utf_new_char("Exceptions");
188 utf_LineNumberTable = utf_new_char("LineNumberTable");
189 utf_SourceFile = utf_new_char("SourceFile");
191 utf_init = utf_new_char("<init>");
192 utf_clinit = utf_new_char("<clinit>");
193 utf_clone = utf_new_char("clone");
194 utf_finalize = utf_new_char("finalize");
195 utf_run = utf_new_char("run");
197 utf_printStackTrace = utf_new_char("printStackTrace");
198 utf_fillInStackTrace = utf_new_char("fillInStackTrace");
199 utf_loadClass = utf_new_char("loadClass");
200 utf_getSystemClassLoader = utf_new_char("getSystemClassLoader");
202 utf_void__void = utf_new_char("()V");
203 utf_boolean__void = utf_new_char("(Z)V");
204 utf_byte__void = utf_new_char("(B)V");
205 utf_char__void = utf_new_char("(C)V");
206 utf_short__void = utf_new_char("(S)V");
207 utf_int__void = utf_new_char("(I)V");
208 utf_long__void = utf_new_char("(J)V");
209 utf_float__void = utf_new_char("(F)V");
210 utf_double__void = utf_new_char("(D)V");
211 utf_void__java_lang_Object = utf_new_char("()Ljava/lang/Object;");
212 utf_void__java_lang_Throwable = utf_new_char("()Ljava/lang/Throwable;");
214 utf_void__java_lang_ClassLoader =
215 utf_new_char("()Ljava/lang/ClassLoader;");
217 utf_java_lang_String__void = utf_new_char("(Ljava/lang/String;)V");
219 utf_java_lang_String__java_lang_Class =
220 utf_new_char("(Ljava/lang/String;)Ljava/lang/Class;");
222 utf_java_lang_Throwable__void = utf_new_char("(Ljava/lang/Throwable;)V");
224 utf_not_named_yet = utf_new_char("\t<not_named_yet>");
226 array_packagename = utf_new_char("\t<the array package>");
230 /* utf_hashkey *****************************************************************
232 The hashkey is computed from the utf-text by using up to 8
233 characters. For utf-symbols longer than 15 characters 3 characters
234 are taken from the beginning and the end, 2 characters are taken
237 *******************************************************************************/
239 #define nbs(val) ((u4) *(++text) << val) /* get next byte, left shift by val */
240 #define fbs(val) ((u4) *( text) << val) /* get first byte, left shift by val */
242 u4 utf_hashkey(const char *text, u4 length)
244 const char *start_pos = text; /* pointer to utf text */
248 case 0: /* empty string */
251 case 1: return fbs(0);
252 case 2: return fbs(0) ^ nbs(3);
253 case 3: return fbs(0) ^ nbs(3) ^ nbs(5);
254 case 4: return fbs(0) ^ nbs(2) ^ nbs(4) ^ nbs(6);
255 case 5: return fbs(0) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(6);
256 case 6: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(5) ^ nbs(6);
257 case 7: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6);
258 case 8: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7);
265 return a ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7) ^ nbs(8);
274 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9);
283 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9) ^ nbs(10);
295 return a ^ nbs(9) ^ nbs(10);
307 return a ^ nbs(9) ^ nbs(10);
318 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
329 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
331 default: /* 3 characters from beginning */
337 /* 2 characters from middle */
338 text = start_pos + (length / 2);
343 /* 3 characters from end */
344 text = start_pos + length - 4;
349 return a ^ nbs(10) ^ nbs(11);
354 /* utf_hashkey *****************************************************************
356 Compute the hashkey of a unicode string.
358 *******************************************************************************/
360 u4 unicode_hashkey(u2 *text, u2 len)
362 return utf_hashkey((char *) text, len);
366 /* utf_new *********************************************************************
368 Creates a new utf-symbol, the text of the symbol is passed as a
369 u1-array. The function searches the utf-hashtable for a utf-symbol
370 with this text. On success the element returned, otherwise a new
371 hashtable element is created.
373 If the number of entries in the hashtable exceeds twice the size of
374 the hashtable slots a reorganization of the hashtable is done and
375 the utf symbols are copied to a new hashtable with doubled size.
377 *******************************************************************************/
379 utf *utf_new_intern(const char *text, u2 length);
381 utf *utf_new(const char *text, u2 length)
385 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
389 r = utf_new_intern(text, length);
391 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
399 utf *utf_new_intern(const char *text, u2 length)
401 u4 key; /* hashkey computed from utf-text */
402 u4 slot; /* slot in hashtable */
403 utf *u; /* hashtable element */
411 key = utf_hashkey(text, length);
412 slot = key & (utf_hash.size - 1);
413 u = utf_hash.ptr[slot];
415 /* search external hash chain for utf-symbol */
417 if (u->blength == length) {
419 /* compare text of hashtable elements */
420 for (i = 0; i < length; i++)
421 if (text[i] != u->text[i]) goto nomatch;
425 count_utf_new_found++;
428 /* symbol found in hashtable */
432 u = u->hashlink; /* next element in external chain */
437 count_utf_len += sizeof(utf) + length;
440 /* location in hashtable found, create new utf element */
442 u->blength = length; /* length in bytes of utfstring */
443 u->hashlink = utf_hash.ptr[slot]; /* link in external hashchain */
444 u->text = mem_alloc(length + 1);/* allocate memory for utf-text */
445 memcpy(u->text, text, length); /* copy utf-text */
446 u->text[length] = '\0';
447 utf_hash.ptr[slot] = u; /* insert symbol into table */
449 utf_hash.entries++; /* update number of entries */
451 if (utf_hash.entries > (utf_hash.size * 2)) {
453 /* reorganization of hashtable, average length of
454 the external chains is approx. 2 */
458 hashtable newhash; /* the new hashtable */
460 /* create new hashtable, double the size */
461 init_hashtable(&newhash, utf_hash.size * 2);
462 newhash.entries = utf_hash.entries;
466 count_utf_len += sizeof(utf*) * utf_hash.size;
469 /* transfer elements to new hashtable */
470 for (i = 0; i < utf_hash.size; i++) {
471 u = (utf *) utf_hash.ptr[i];
473 utf *nextu = u->hashlink;
474 u4 slot = utf_hashkey(u->text, u->blength) & (newhash.size - 1);
476 u->hashlink = (utf *) newhash.ptr[slot];
477 newhash.ptr[slot] = u;
479 /* follow link in external hash chain */
484 /* dispose old table */
485 MFREE(utf_hash.ptr, void*, utf_hash.size);
493 /* utf_new_u2 ******************************************************************
495 Make utf symbol from u2 array, if isclassname is true '.' is
498 *******************************************************************************/
500 utf *utf_new_u2(u2 *unicode_pos, u4 unicode_length, bool isclassname)
502 char *buffer; /* memory buffer for unicode characters */
503 char *pos; /* pointer to current position in buffer */
504 u4 left; /* unicode characters left */
505 u4 buflength; /* utf length in bytes of the u2 array */
506 utf *result; /* resulting utf-string */
509 /* determine utf length in bytes and allocate memory */
511 buflength = u2_utflength(unicode_pos, unicode_length);
512 buffer = MNEW(char, buflength);
517 for (i = 0; i++ < unicode_length; unicode_pos++) {
518 /* next unicode character */
521 if ((c != 0) && (c < 0x80)) {
524 if ((int) left < 0) break;
525 /* convert classname */
526 if (isclassname && c == '.')
531 } else if (c < 0x800) {
533 unsigned char high = c >> 6;
534 unsigned char low = c & 0x3F;
536 if ((int) left < 0) break;
537 *pos++ = high | 0xC0;
543 char mid = (c >> 6) & 0x3F;
546 if ((int) left < 0) break;
547 *pos++ = high | 0xE0;
553 /* insert utf-string into symbol-table */
554 result = utf_new(buffer,buflength);
556 MFREE(buffer, char, buflength);
562 /* utf_new_char ****************************************************************
564 Creates a new utf symbol, the text for this symbol is passed as a
565 c-string ( = char* ).
567 *******************************************************************************/
569 utf *utf_new_char(const char *text)
571 return utf_new(text, strlen(text));
575 /* utf_new_char_classname ******************************************************
577 Creates a new utf symbol, the text for this symbol is passed as a
578 c-string ( = char* ) "." characters are going to be replaced by
579 "/". Since the above function is used often, this is a separte
580 function, instead of an if.
582 *******************************************************************************/
584 utf *utf_new_char_classname(const char *text)
586 if (strchr(text, '.')) {
587 char *txt = strdup(text);
588 char *end = txt + strlen(txt);
592 for (c = txt; c < end; c++)
593 if (*c == '.') *c = '/';
595 tmpRes = utf_new(txt, strlen(txt));
601 return utf_new(text, strlen(text));
605 /* utf_nextu2 ******************************************************************
607 Read the next unicode character from the utf string and increment
608 the utf-string pointer accordingly.
610 *******************************************************************************/
612 u2 utf_nextu2(char **utf_ptr)
614 /* uncompressed unicode character */
616 /* current position in utf text */
617 unsigned char *utf = (unsigned char *) (*utf_ptr);
618 /* bytes representing the unicode character */
619 unsigned char ch1, ch2, ch3;
620 /* number of bytes used to represent the unicode character */
623 switch ((ch1 = utf[0]) >> 4) {
624 default: /* 1 byte */
628 case 0xD: /* 2 bytes */
629 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
630 unsigned char high = ch1 & 0x1F;
631 unsigned char low = ch2 & 0x3F;
632 unicode_char = (high << 6) + low;
637 case 0xE: /* 2 or 3 bytes */
638 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
639 if (((ch3 = utf[2]) & 0xC0) == 0x80) {
640 unsigned char low = ch3 & 0x3f;
641 unsigned char mid = ch2 & 0x3f;
642 unsigned char high = ch1 & 0x0f;
643 unicode_char = (((high << 6) + mid) << 6) + low;
651 /* update position in utf-text */
652 *utf_ptr = (char *) (utf + len);
658 /* utf_strlen ******************************************************************
660 Determine number of unicode characters in the utf string.
662 *******************************************************************************/
664 u4 utf_strlen(utf *u)
666 char *endpos; /* points behind utf string */
667 char *utf_ptr; /* current position in utf text */
668 u4 len = 0; /* number of unicode characters */
671 *exceptionptr = new_nullpointerexception();
678 while (utf_ptr < endpos) {
680 /* next unicode character */
681 utf_nextu2(&utf_ptr);
684 if (utf_ptr != endpos)
685 /* string ended abruptly */
686 throw_cacao_exception_exit(string_java_lang_InternalError,
687 "Illegal utf8 string");
693 /* u2_utflength ****************************************************************
695 Returns the utf length in bytes of a u2 array.
697 *******************************************************************************/
699 u4 u2_utflength(u2 *text, u4 u2_length)
701 u4 result_len = 0; /* utf length in bytes */
702 u2 ch; /* current unicode character */
705 for (len = 0; len < u2_length; len++) {
706 /* next unicode character */
709 /* determine bytes required to store unicode character as utf */
710 if (ch && (ch < 0x80))
722 /* utf_display *****************************************************************
724 Write utf symbol to stdout (for debugging purposes).
726 *******************************************************************************/
728 void utf_display(utf *u)
730 char *endpos; /* points behind utf string */
731 char *utf_ptr; /* current position in utf text */
742 while (utf_ptr < endpos) {
743 /* read next unicode character */
744 u2 c = utf_nextu2(&utf_ptr);
745 if (c >= 32 && c <= 127) printf("%c", c);
753 /* utf_display_classname *******************************************************
755 Write utf symbol to stdout with `/' converted to `.' (for debugging
758 *******************************************************************************/
760 void utf_display_classname(utf *u)
762 char *endpos; /* points behind utf string */
763 char *utf_ptr; /* current position in utf text */
774 while (utf_ptr < endpos) {
775 /* read next unicode character */
776 u2 c = utf_nextu2(&utf_ptr);
777 if (c == '/') c = '.';
778 if (c >= 32 && c <= 127) printf("%c", c);
786 /* utf_sprint ******************************************************************
788 Write utf symbol into c-string (for debugging purposes).
790 *******************************************************************************/
792 void utf_sprint(char *buffer, utf *u)
794 char *endpos; /* points behind utf string */
795 char *utf_ptr; /* current position in utf text */
796 u2 pos = 0; /* position in c-string */
799 strcpy(buffer, "NULL");
806 while (utf_ptr < endpos)
807 /* copy next unicode character */
808 buffer[pos++] = utf_nextu2(&utf_ptr);
810 /* terminate string */
815 /* utf_sprint_classname ********************************************************
817 Write utf symbol into c-string with `/' converted to `.' (for debugging
820 *******************************************************************************/
822 void utf_sprint_classname(char *buffer, utf *u)
824 char *endpos; /* points behind utf string */
825 char *utf_ptr; /* current position in utf text */
826 u2 pos = 0; /* position in c-string */
829 strcpy(buffer, "NULL");
836 while (utf_ptr < endpos) {
837 /* copy next unicode character */
838 u2 c = utf_nextu2(&utf_ptr);
839 if (c == '/') c = '.';
843 /* terminate string */
848 /* utf_strcat ******************************************************************
850 Like libc strcat, but uses an utf8 string.
852 *******************************************************************************/
854 void utf_strcat(char *buffer, utf *u)
856 utf_sprint(buffer + strlen(buffer), u);
860 /* utf_strcat_classname ********************************************************
862 Like libc strcat, but uses an utf8 string.
864 *******************************************************************************/
866 void utf_strcat_classname(char *buffer, utf *u)
868 utf_sprint_classname(buffer + strlen(buffer), u);
872 /* utf_fprint ******************************************************************
874 Write utf symbol into file.
876 *******************************************************************************/
878 void utf_fprint(FILE *file, utf *u)
880 char *endpos; /* points behind utf string */
881 char *utf_ptr; /* current position in utf text */
889 while (utf_ptr < endpos) {
890 /* read next unicode character */
891 u2 c = utf_nextu2(&utf_ptr);
893 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
894 else fprintf(file, "?");
899 /* utf_fprint_classname ********************************************************
901 Write utf symbol into file with `/' converted to `.'.
903 *******************************************************************************/
905 void utf_fprint_classname(FILE *file, utf *u)
907 char *endpos; /* points behind utf string */
908 char *utf_ptr; /* current position in utf text */
916 while (utf_ptr < endpos) {
917 /* read next unicode character */
918 u2 c = utf_nextu2(&utf_ptr);
919 if (c == '/') c = '.';
921 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
922 else fprintf(file, "?");
927 /* is_valid_utf ****************************************************************
929 Return true if the given string is a valid UTF-8 string.
931 utf_ptr...points to first character
932 end_pos...points after last character
934 *******************************************************************************/
936 static unsigned long min_codepoint[6] = {0,1L<<7,1L<<11,1L<<16,1L<<21,1L<<26};
938 bool is_valid_utf(char *utf_ptr, char *end_pos)
945 if (end_pos < utf_ptr) return false;
946 bytes = end_pos - utf_ptr;
950 if (!c) return false; /* 0x00 is not allowed */
951 if ((c & 0x80) == 0) continue; /* ASCII */
953 if ((c & 0xe0) == 0xc0) len = 1; /* 110x xxxx */
954 else if ((c & 0xf0) == 0xe0) len = 2; /* 1110 xxxx */
955 else if ((c & 0xf8) == 0xf0) len = 3; /* 1111 0xxx */
956 else if ((c & 0xfc) == 0xf8) len = 4; /* 1111 10xx */
957 else if ((c & 0xfe) == 0xfc) len = 5; /* 1111 110x */
958 else return false; /* invalid leading byte */
960 if (len > 2) return false; /* Java limitation */
962 v = (unsigned long)c & (0x3f >> len);
964 if ((bytes -= len) < 0) return false; /* missing bytes */
966 for (i = len; i--; ) {
968 if ((c & 0xc0) != 0x80) /* 10xx xxxx */
970 v = (v << 6) | (c & 0x3f);
974 if (len != 1) return false; /* Java special */
977 /* Sun Java seems to allow overlong UTF-8 encodings */
979 if (v < min_codepoint[len]) { /* overlong UTF-8 */
981 fprintf(stderr,"WARNING: Overlong UTF-8 sequence found.\n");
982 /* XXX change this to exception? */
986 /* surrogates in UTF-8 seem to be allowed in Java classfiles */
987 /* if (v >= 0xd800 && v <= 0xdfff) return false; */ /* surrogates */
989 /* even these seem to be allowed */
990 /* if (v == 0xfffe || v == 0xffff) return false; */ /* invalid codepoints */
997 /* is_valid_name ***************************************************************
999 Return true if the given string may be used as a class/field/method
1000 name. (Currently this only disallows empty strings and control
1003 NOTE: The string is assumed to have passed is_valid_utf!
1005 utf_ptr...points to first character
1006 end_pos...points after last character
1008 *******************************************************************************/
1010 bool is_valid_name(char *utf_ptr, char *end_pos)
1012 if (end_pos <= utf_ptr) return false; /* disallow empty names */
1014 while (utf_ptr < end_pos) {
1015 unsigned char c = *utf_ptr++;
1017 if (c < 0x20) return false; /* disallow control characters */
1018 if (c == 0xc0 && (unsigned char) *utf_ptr == 0x80) /* disallow zero */
1025 bool is_valid_name_utf(utf *u)
1027 return is_valid_name(u->text, UTF_END(u));
1031 /* utf_show ********************************************************************
1033 Writes the utf symbols in the utfhash to stdout and displays the
1034 number of external hash chains grouped according to the chainlength
1035 (for debugging purposes).
1037 *******************************************************************************/
1042 #define CHAIN_LIMIT 20 /* limit for seperated enumeration */
1044 u4 chain_count[CHAIN_LIMIT]; /* numbers of chains */
1045 u4 max_chainlength = 0; /* maximum length of the chains */
1046 u4 sum_chainlength = 0; /* sum of the chainlengths */
1047 u4 beyond_limit = 0; /* number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1050 printf ("UTF-HASH:\n");
1052 /* show element of utf-hashtable */
1053 for (i=0; i<utf_hash.size; i++) {
1054 utf *u = utf_hash.ptr[i];
1056 printf ("SLOT %d: ", (int) i);
1068 printf ("UTF-HASH: %d slots for %d entries\n",
1069 (int) utf_hash.size, (int) utf_hash.entries );
1072 if (utf_hash.entries == 0)
1075 printf("chains:\n chainlength number of chains %% of utfstrings\n");
1077 for (i=0;i<CHAIN_LIMIT;i++)
1080 /* count numbers of hashchains according to their length */
1081 for (i=0; i<utf_hash.size; i++) {
1083 utf *u = (utf*) utf_hash.ptr[i];
1084 u4 chain_length = 0;
1086 /* determine chainlength */
1092 /* update sum of all chainlengths */
1093 sum_chainlength+=chain_length;
1095 /* determine the maximum length of the chains */
1096 if (chain_length>max_chainlength)
1097 max_chainlength = chain_length;
1099 /* update number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1100 if (chain_length>=CHAIN_LIMIT) {
1101 beyond_limit+=chain_length;
1102 chain_length=CHAIN_LIMIT-1;
1105 /* update number of hashchains of current length */
1106 chain_count[chain_length]++;
1109 /* display results */
1110 for (i=1;i<CHAIN_LIMIT-1;i++)
1111 printf(" %2d %17d %18.2f%%\n",i,chain_count[i],(((float) chain_count[i]*i*100)/utf_hash.entries));
1113 printf(" >=%2d %17d %18.2f%%\n",CHAIN_LIMIT-1,chain_count[CHAIN_LIMIT-1],((float) beyond_limit*100)/utf_hash.entries);
1116 printf("max. chainlength:%5d\n",max_chainlength);
1118 /* avg. chainlength = sum of chainlengths / number of chains */
1119 printf("avg. chainlength:%5.2f\n",(float) sum_chainlength / (utf_hash.size-chain_count[0]));
1124 * These are local overrides for various environment variables in Emacs.
1125 * Please do not remove this and leave it at the end of the file, where
1126 * Emacs will automagically detect them.
1127 * ---------------------------------------------------------------------
1130 * indent-tabs-mode: t