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 2458 2005-05-12 23:02:07Z 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_util_Vector;
84 utf *utf_java_lang_reflect_Constructor;
85 utf *utf_java_lang_reflect_Method;
88 utf *utf_InnerClasses; /* InnerClasses */
89 utf *utf_ConstantValue; /* ConstantValue */
90 utf *utf_Code; /* Code */
91 utf *utf_Exceptions; /* Exceptions */
92 utf *utf_LineNumberTable; /* LineNumberTable */
93 utf *utf_SourceFile; /* SourceFile */
95 utf *utf_init; /* <init> */
96 utf *utf_clinit; /* <clinit> */
97 utf *utf_finalize; /* finalize */
99 utf *utf_printStackTrace;
100 utf *utf_fillInStackTrace;
103 utf *utf_void__void; /* ()V */
104 utf *utf_boolean__void; /* (Z)V */
105 utf *utf_byte__void; /* (B)V */
106 utf *utf_char__void; /* (C)V */
107 utf *utf_short__void; /* (S)V */
108 utf *utf_int__void; /* (I)V */
109 utf *utf_long__void; /* (J)V */
110 utf *utf_float__void; /* (F)V */
111 utf *utf_double__void; /* (D)V */
112 utf *utf_void__java_lang_Object; /* ()Ljava/lang/Object; */
113 utf *utf_void__java_lang_Throwable; /* ()Ljava/lang/Throwable; */
114 utf *utf_java_lang_String__void; /* (Ljava/lang/String;)V */
115 utf *utf_java_lang_String__java_lang_Class;
116 utf *utf_java_lang_Throwable__void; /* (Ljava/lang/Throwable;)V */
118 utf *utf_not_named_yet; /* special name for unnamed classes */
120 utf *array_packagename;
123 /* utf_init ********************************************************************
125 Initializes the utf8 subsystem.
127 *******************************************************************************/
131 /* create utf-symbols for pointer comparison of frequently used strings */
133 utf_java_lang_Object = utf_new_char("java/lang/Object");
135 utf_java_lang_Class = utf_new_char("java/lang/Class");
136 utf_java_lang_ClassLoader = utf_new_char("java/lang/ClassLoader");
137 utf_java_lang_Cloneable = utf_new_char("java/lang/Cloneable");
138 utf_java_lang_SecurityManager = utf_new_char("java/lang/SecurityManager");
139 utf_java_lang_String = utf_new_char("java/lang/String");
140 utf_java_lang_System = utf_new_char("java/lang/System");
141 utf_java_lang_ThreadGroup = utf_new_char("java/lang/ThreadGroup");
142 utf_java_io_Serializable = utf_new_char("java/io/Serializable");
144 utf_java_lang_Throwable = utf_new_char(string_java_lang_Throwable);
145 utf_java_lang_VMThrowable = utf_new_char(string_java_lang_VMThrowable);
146 utf_java_lang_Error = utf_new_char(string_java_lang_Error);
147 utf_java_lang_Exception = utf_new_char(string_java_lang_Exception);
149 utf_java_lang_NoClassDefFoundError =
150 utf_new_char(string_java_lang_NoClassDefFoundError);
152 utf_java_lang_OutOfMemoryError =
153 utf_new_char(string_java_lang_OutOfMemoryError);
155 utf_java_lang_ClassNotFoundException =
156 utf_new_char(string_java_lang_ClassNotFoundException);
158 utf_java_lang_Void = utf_new_char("java/lang/Void");
159 utf_java_lang_Boolean = utf_new_char("java/lang/Boolean");
160 utf_java_lang_Byte = utf_new_char("java/lang/Byte");
161 utf_java_lang_Character = utf_new_char("java/lang/Character");
162 utf_java_lang_Short = utf_new_char("java/lang/Short");
163 utf_java_lang_Integer = utf_new_char("java/lang/Integer");
164 utf_java_lang_Long = utf_new_char("java/lang/Long");
165 utf_java_lang_Float = utf_new_char("java/lang/Float");
166 utf_java_lang_Double = utf_new_char("java/lang/Double");
168 utf_java_util_Vector = utf_new_char("java/util/Vector");
169 utf_java_lang_reflect_Constructor = utf_new_char("java/lang/reflect/Constructor");
170 utf_java_lang_reflect_Method = utf_new_char("java/lang/reflect/Method");
172 utf_InnerClasses = utf_new_char("InnerClasses");
173 utf_ConstantValue = utf_new_char("ConstantValue");
174 utf_Code = utf_new_char("Code");
175 utf_Exceptions = utf_new_char("Exceptions");
176 utf_LineNumberTable = utf_new_char("LineNumberTable");
177 utf_SourceFile = utf_new_char("SourceFile");
179 utf_init = utf_new_char("<init>");
180 utf_clinit = utf_new_char("<clinit>");
181 utf_finalize = utf_new_char("finalize");
183 utf_printStackTrace = utf_new_char("printStackTrace");
184 utf_fillInStackTrace = utf_new_char("fillInStackTrace");
185 utf_loadClass = utf_new_char("loadClass");
187 utf_void__void = utf_new_char("()V");
188 utf_boolean__void = utf_new_char("(Z)V");
189 utf_byte__void = utf_new_char("(B)V");
190 utf_char__void = utf_new_char("(C)V");
191 utf_short__void = utf_new_char("(S)V");
192 utf_int__void = utf_new_char("(I)V");
193 utf_long__void = utf_new_char("(J)V");
194 utf_float__void = utf_new_char("(F)V");
195 utf_double__void = utf_new_char("(D)V");
196 utf_void__java_lang_Object = utf_new_char("()Ljava/lang/Object;");
197 utf_void__java_lang_Throwable = utf_new_char("()Ljava/lang/Throwable;");
198 utf_java_lang_String__void = utf_new_char("(Ljava/lang/String;)V");
200 utf_java_lang_String__java_lang_Class =
201 utf_new_char("(Ljava/lang/String;)Ljava/lang/Class;");
203 utf_java_lang_Throwable__void = utf_new_char("(Ljava/lang/Throwable;)V");
205 utf_not_named_yet = utf_new_char("\t<not_named_yet>");
207 array_packagename = utf_new_char("\t<the array package>");
211 /* utf_hashkey *****************************************************************
213 The hashkey is computed from the utf-text by using up to 8
214 characters. For utf-symbols longer than 15 characters 3 characters
215 are taken from the beginning and the end, 2 characters are taken
218 *******************************************************************************/
220 #define nbs(val) ((u4) *(++text) << val) /* get next byte, left shift by val */
221 #define fbs(val) ((u4) *( text) << val) /* get first byte, left shift by val */
223 u4 utf_hashkey(const char *text, u4 length)
225 const char *start_pos = text; /* pointer to utf text */
229 case 0: /* empty string */
232 case 1: return fbs(0);
233 case 2: return fbs(0) ^ nbs(3);
234 case 3: return fbs(0) ^ nbs(3) ^ nbs(5);
235 case 4: return fbs(0) ^ nbs(2) ^ nbs(4) ^ nbs(6);
236 case 5: return fbs(0) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(6);
237 case 6: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(5) ^ nbs(6);
238 case 7: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6);
239 case 8: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7);
246 return a ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7) ^ nbs(8);
255 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9);
264 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9) ^ nbs(10);
276 return a ^ nbs(9) ^ nbs(10);
288 return a ^ nbs(9) ^ nbs(10);
299 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
310 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
312 default: /* 3 characters from beginning */
318 /* 2 characters from middle */
319 text = start_pos + (length / 2);
324 /* 3 characters from end */
325 text = start_pos + length - 4;
330 return a ^ nbs(10) ^ nbs(11);
335 /* utf_hashkey *****************************************************************
337 Compute the hashkey of a unicode string.
339 *******************************************************************************/
341 u4 unicode_hashkey(u2 *text, u2 len)
343 return utf_hashkey((char *) text, len);
347 /* utf_new *********************************************************************
349 Creates a new utf-symbol, the text of the symbol is passed as a
350 u1-array. The function searches the utf-hashtable for a utf-symbol
351 with this text. On success the element returned, otherwise a new
352 hashtable element is created.
354 If the number of entries in the hashtable exceeds twice the size of
355 the hashtable slots a reorganization of the hashtable is done and
356 the utf symbols are copied to a new hashtable with doubled size.
358 *******************************************************************************/
360 utf *utf_new_intern(const char *text, u2 length);
362 utf *utf_new(const char *text, u2 length)
366 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
370 r = utf_new_intern(text, length);
372 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
380 utf *utf_new_intern(const char *text, u2 length)
382 u4 key; /* hashkey computed from utf-text */
383 u4 slot; /* slot in hashtable */
384 utf *u; /* hashtable element */
392 key = utf_hashkey(text, length);
393 slot = key & (utf_hash.size - 1);
394 u = utf_hash.ptr[slot];
396 /* search external hash chain for utf-symbol */
398 if (u->blength == length) {
400 /* compare text of hashtable elements */
401 for (i = 0; i < length; i++)
402 if (text[i] != u->text[i]) goto nomatch;
406 count_utf_new_found++;
409 /* symbol found in hashtable */
413 u = u->hashlink; /* next element in external chain */
418 count_utf_len += sizeof(utf) + length;
421 /* location in hashtable found, create new utf element */
423 u->blength = length; /* length in bytes of utfstring */
424 u->hashlink = utf_hash.ptr[slot]; /* link in external hashchain */
425 u->text = mem_alloc(length + 1);/* allocate memory for utf-text */
426 memcpy(u->text, text, length); /* copy utf-text */
427 u->text[length] = '\0';
428 utf_hash.ptr[slot] = u; /* insert symbol into table */
430 utf_hash.entries++; /* update number of entries */
432 if (utf_hash.entries > (utf_hash.size * 2)) {
434 /* reorganization of hashtable, average length of
435 the external chains is approx. 2 */
439 hashtable newhash; /* the new hashtable */
441 /* create new hashtable, double the size */
442 init_hashtable(&newhash, utf_hash.size * 2);
443 newhash.entries = utf_hash.entries;
447 count_utf_len += sizeof(utf*) * utf_hash.size;
450 /* transfer elements to new hashtable */
451 for (i = 0; i < utf_hash.size; i++) {
452 u = (utf *) utf_hash.ptr[i];
454 utf *nextu = u->hashlink;
455 u4 slot = utf_hashkey(u->text, u->blength) & (newhash.size - 1);
457 u->hashlink = (utf *) newhash.ptr[slot];
458 newhash.ptr[slot] = u;
460 /* follow link in external hash chain */
465 /* dispose old table */
466 MFREE(utf_hash.ptr, void*, utf_hash.size);
474 /* utf_new_u2 ******************************************************************
476 Make utf symbol from u2 array, if isclassname is true '.' is
479 *******************************************************************************/
481 utf *utf_new_u2(u2 *unicode_pos, u4 unicode_length, bool isclassname)
483 char *buffer; /* memory buffer for unicode characters */
484 char *pos; /* pointer to current position in buffer */
485 u4 left; /* unicode characters left */
486 u4 buflength; /* utf length in bytes of the u2 array */
487 utf *result; /* resulting utf-string */
490 /* determine utf length in bytes and allocate memory */
492 buflength = u2_utflength(unicode_pos, unicode_length);
493 buffer = MNEW(char, buflength);
498 for (i = 0; i++ < unicode_length; unicode_pos++) {
499 /* next unicode character */
502 if ((c != 0) && (c < 0x80)) {
505 if ((int) left < 0) break;
506 /* convert classname */
507 if (isclassname && c == '.')
512 } else if (c < 0x800) {
514 unsigned char high = c >> 6;
515 unsigned char low = c & 0x3F;
517 if ((int) left < 0) break;
518 *pos++ = high | 0xC0;
524 char mid = (c >> 6) & 0x3F;
527 if ((int) left < 0) break;
528 *pos++ = high | 0xE0;
534 /* insert utf-string into symbol-table */
535 result = utf_new(buffer,buflength);
537 MFREE(buffer, char, buflength);
543 /* utf_new_char ****************************************************************
545 Creates a new utf symbol, the text for this symbol is passed as a
546 c-string ( = char* ).
548 *******************************************************************************/
550 utf *utf_new_char(const char *text)
552 return utf_new(text, strlen(text));
556 /* utf_new_char_classname ******************************************************
558 Creates a new utf symbol, the text for this symbol is passed as a
559 c-string ( = char* ) "." characters are going to be replaced by
560 "/". Since the above function is used often, this is a separte
561 function, instead of an if.
563 *******************************************************************************/
565 utf *utf_new_char_classname(const char *text)
567 if (strchr(text, '.')) {
568 char *txt = strdup(text);
569 char *end = txt + strlen(txt);
573 for (c = txt; c < end; c++)
574 if (*c == '.') *c = '/';
576 tmpRes = utf_new(txt, strlen(txt));
582 return utf_new(text, strlen(text));
586 /* utf_nextu2 ******************************************************************
588 Read the next unicode character from the utf string and increment
589 the utf-string pointer accordingly.
591 *******************************************************************************/
593 u2 utf_nextu2(char **utf_ptr)
595 /* uncompressed unicode character */
597 /* current position in utf text */
598 unsigned char *utf = (unsigned char *) (*utf_ptr);
599 /* bytes representing the unicode character */
600 unsigned char ch1, ch2, ch3;
601 /* number of bytes used to represent the unicode character */
604 switch ((ch1 = utf[0]) >> 4) {
605 default: /* 1 byte */
609 case 0xD: /* 2 bytes */
610 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
611 unsigned char high = ch1 & 0x1F;
612 unsigned char low = ch2 & 0x3F;
613 unicode_char = (high << 6) + low;
618 case 0xE: /* 2 or 3 bytes */
619 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
620 if (((ch3 = utf[2]) & 0xC0) == 0x80) {
621 unsigned char low = ch3 & 0x3f;
622 unsigned char mid = ch2 & 0x3f;
623 unsigned char high = ch1 & 0x0f;
624 unicode_char = (((high << 6) + mid) << 6) + low;
632 /* update position in utf-text */
633 *utf_ptr = (char *) (utf + len);
639 /* utf_strlen ******************************************************************
641 Determine number of unicode characters in the utf string.
643 *******************************************************************************/
645 u4 utf_strlen(utf *u)
647 char *endpos; /* points behind utf string */
648 char *utf_ptr; /* current position in utf text */
649 u4 len = 0; /* number of unicode characters */
652 *exceptionptr = new_nullpointerexception();
659 while (utf_ptr < endpos) {
661 /* next unicode character */
662 utf_nextu2(&utf_ptr);
665 if (utf_ptr != endpos)
666 /* string ended abruptly */
667 throw_cacao_exception_exit(string_java_lang_InternalError,
668 "Illegal utf8 string");
674 /* u2_utflength ****************************************************************
676 Returns the utf length in bytes of a u2 array.
678 *******************************************************************************/
680 u4 u2_utflength(u2 *text, u4 u2_length)
682 u4 result_len = 0; /* utf length in bytes */
683 u2 ch; /* current unicode character */
686 for (len = 0; len < u2_length; len++) {
687 /* next unicode character */
690 /* determine bytes required to store unicode character as utf */
691 if (ch && (ch < 0x80))
703 /* utf_display *****************************************************************
705 Write utf symbol to stdout (for debugging purposes).
707 *******************************************************************************/
709 void utf_display(utf *u)
711 char *endpos; /* points behind utf string */
712 char *utf_ptr; /* current position in utf text */
723 while (utf_ptr < endpos) {
724 /* read next unicode character */
725 u2 c = utf_nextu2(&utf_ptr);
726 if (c >= 32 && c <= 127) printf("%c", c);
734 /* utf_display_classname *******************************************************
736 Write utf symbol to stdout with `/' converted to `.' (for debugging
739 *******************************************************************************/
741 void utf_display_classname(utf *u)
743 char *endpos; /* points behind utf string */
744 char *utf_ptr; /* current position in utf text */
755 while (utf_ptr < endpos) {
756 /* read next unicode character */
757 u2 c = utf_nextu2(&utf_ptr);
758 if (c == '/') c = '.';
759 if (c >= 32 && c <= 127) printf("%c", c);
767 /* utf_sprint ******************************************************************
769 Write utf symbol into c-string (for debugging purposes).
771 *******************************************************************************/
773 void utf_sprint(char *buffer, utf *u)
775 char *endpos; /* points behind utf string */
776 char *utf_ptr; /* current position in utf text */
777 u2 pos = 0; /* position in c-string */
780 strcpy(buffer, "NULL");
787 while (utf_ptr < endpos)
788 /* copy next unicode character */
789 buffer[pos++] = utf_nextu2(&utf_ptr);
791 /* terminate string */
796 /* utf_sprint_classname ********************************************************
798 Write utf symbol into c-string with `/' converted to `.' (for debugging
801 *******************************************************************************/
803 void utf_sprint_classname(char *buffer, utf *u)
805 char *endpos; /* points behind utf string */
806 char *utf_ptr; /* current position in utf text */
807 u2 pos = 0; /* position in c-string */
810 strcpy(buffer, "NULL");
817 while (utf_ptr < endpos) {
818 /* copy next unicode character */
819 u2 c = utf_nextu2(&utf_ptr);
820 if (c == '/') c = '.';
824 /* terminate string */
829 /* utf_strcat ******************************************************************
831 Like libc strcat, but uses an utf8 string.
833 *******************************************************************************/
835 void utf_strcat(char *buffer, utf *u)
837 utf_sprint(buffer + strlen(buffer), u);
841 /* utf_strcat_classname ********************************************************
843 Like libc strcat, but uses an utf8 string.
845 *******************************************************************************/
847 void utf_strcat_classname(char *buffer, utf *u)
849 utf_sprint_classname(buffer + strlen(buffer), u);
853 /* utf_fprint ******************************************************************
855 Write utf symbol into file.
857 *******************************************************************************/
859 void utf_fprint(FILE *file, utf *u)
861 char *endpos; /* points behind utf string */
862 char *utf_ptr; /* current position in utf text */
870 while (utf_ptr < endpos) {
871 /* read next unicode character */
872 u2 c = utf_nextu2(&utf_ptr);
874 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
875 else fprintf(file, "?");
880 /* utf_fprint_classname ********************************************************
882 Write utf symbol into file with `/' converted to `.'.
884 *******************************************************************************/
886 void utf_fprint_classname(FILE *file, utf *u)
888 char *endpos; /* points behind utf string */
889 char *utf_ptr; /* current position in utf text */
897 while (utf_ptr < endpos) {
898 /* read next unicode character */
899 u2 c = utf_nextu2(&utf_ptr);
900 if (c == '/') c = '.';
902 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
903 else fprintf(file, "?");
908 /* is_valid_utf ****************************************************************
910 Return true if the given string is a valid UTF-8 string.
912 utf_ptr...points to first character
913 end_pos...points after last character
915 *******************************************************************************/
917 static unsigned long min_codepoint[6] = {0,1L<<7,1L<<11,1L<<16,1L<<21,1L<<26};
919 bool is_valid_utf(char *utf_ptr, char *end_pos)
926 if (end_pos < utf_ptr) return false;
927 bytes = end_pos - utf_ptr;
931 if (!c) return false; /* 0x00 is not allowed */
932 if ((c & 0x80) == 0) continue; /* ASCII */
934 if ((c & 0xe0) == 0xc0) len = 1; /* 110x xxxx */
935 else if ((c & 0xf0) == 0xe0) len = 2; /* 1110 xxxx */
936 else if ((c & 0xf8) == 0xf0) len = 3; /* 1111 0xxx */
937 else if ((c & 0xfc) == 0xf8) len = 4; /* 1111 10xx */
938 else if ((c & 0xfe) == 0xfc) len = 5; /* 1111 110x */
939 else return false; /* invalid leading byte */
941 if (len > 2) return false; /* Java limitation */
943 v = (unsigned long)c & (0x3f >> len);
945 if ((bytes -= len) < 0) return false; /* missing bytes */
947 for (i = len; i--; ) {
949 if ((c & 0xc0) != 0x80) /* 10xx xxxx */
951 v = (v << 6) | (c & 0x3f);
955 if (len != 1) return false; /* Java special */
958 /* Sun Java seems to allow overlong UTF-8 encodings */
960 if (v < min_codepoint[len]) { /* overlong UTF-8 */
962 fprintf(stderr,"WARNING: Overlong UTF-8 sequence found.\n");
963 /* XXX change this to panic? */
967 /* surrogates in UTF-8 seem to be allowed in Java classfiles */
968 /* if (v >= 0xd800 && v <= 0xdfff) return false; */ /* surrogates */
970 /* even these seem to be allowed */
971 /* if (v == 0xfffe || v == 0xffff) return false; */ /* invalid codepoints */
978 /* is_valid_name ***************************************************************
980 Return true if the given string may be used as a class/field/method
981 name. (Currently this only disallows empty strings and control
984 NOTE: The string is assumed to have passed is_valid_utf!
986 utf_ptr...points to first character
987 end_pos...points after last character
989 *******************************************************************************/
991 bool is_valid_name(char *utf_ptr, char *end_pos)
993 if (end_pos <= utf_ptr) return false; /* disallow empty names */
995 while (utf_ptr < end_pos) {
996 unsigned char c = *utf_ptr++;
998 if (c < 0x20) return false; /* disallow control characters */
999 if (c == 0xc0 && (unsigned char) *utf_ptr == 0x80) /* disallow zero */
1006 bool is_valid_name_utf(utf *u)
1008 return is_valid_name(u->text,utf_end(u));
1012 /* utf_show ********************************************************************
1014 Writes the utf symbols in the utfhash to stdout and displays the
1015 number of external hash chains grouped according to the chainlength
1016 (for debugging purposes).
1018 *******************************************************************************/
1023 #define CHAIN_LIMIT 20 /* limit for seperated enumeration */
1025 u4 chain_count[CHAIN_LIMIT]; /* numbers of chains */
1026 u4 max_chainlength = 0; /* maximum length of the chains */
1027 u4 sum_chainlength = 0; /* sum of the chainlengths */
1028 u4 beyond_limit = 0; /* number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1031 printf ("UTF-HASH:\n");
1033 /* show element of utf-hashtable */
1034 for (i=0; i<utf_hash.size; i++) {
1035 utf *u = utf_hash.ptr[i];
1037 printf ("SLOT %d: ", (int) i);
1049 printf ("UTF-HASH: %d slots for %d entries\n",
1050 (int) utf_hash.size, (int) utf_hash.entries );
1053 if (utf_hash.entries == 0)
1056 printf("chains:\n chainlength number of chains %% of utfstrings\n");
1058 for (i=0;i<CHAIN_LIMIT;i++)
1061 /* count numbers of hashchains according to their length */
1062 for (i=0; i<utf_hash.size; i++) {
1064 utf *u = (utf*) utf_hash.ptr[i];
1065 u4 chain_length = 0;
1067 /* determine chainlength */
1073 /* update sum of all chainlengths */
1074 sum_chainlength+=chain_length;
1076 /* determine the maximum length of the chains */
1077 if (chain_length>max_chainlength)
1078 max_chainlength = chain_length;
1080 /* update number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1081 if (chain_length>=CHAIN_LIMIT) {
1082 beyond_limit+=chain_length;
1083 chain_length=CHAIN_LIMIT-1;
1086 /* update number of hashchains of current length */
1087 chain_count[chain_length]++;
1090 /* display results */
1091 for (i=1;i<CHAIN_LIMIT-1;i++)
1092 printf(" %2d %17d %18.2f%%\n",i,chain_count[i],(((float) chain_count[i]*i*100)/utf_hash.entries));
1094 printf(" >=%2d %17d %18.2f%%\n",CHAIN_LIMIT-1,chain_count[CHAIN_LIMIT-1],((float) beyond_limit*100)/utf_hash.entries);
1097 printf("max. chainlength:%5d\n",max_chainlength);
1099 /* avg. chainlength = sum of chainlengths / number of chains */
1100 printf("avg. chainlength:%5.2f\n",(float) sum_chainlength / (utf_hash.size-chain_count[0]));
1105 * These are local overrides for various environment variables in Emacs.
1106 * Please do not remove this and leave it at the end of the file, where
1107 * Emacs will automagically detect them.
1108 * ---------------------------------------------------------------------
1111 * indent-tabs-mode: t