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 2427 2005-05-01 12:27:54Z jowenn $
40 #include "mm/memory.h"
41 #include "vm/exceptions.h"
42 #include "vm/options.h"
43 #include "vm/statistics.h"
44 #include "vm/tables.h"
48 hashtable utf_hash; /* hashtable for utf8-symbols */
51 /* utf-symbols for pointer comparison of frequently used strings **************/
53 utf *utf_java_lang_Object; /* java/lang/Object */
55 utf *utf_java_lang_Class;
56 utf *utf_java_lang_ClassLoader;
57 utf *utf_java_lang_Cloneable;
58 utf *utf_java_lang_SecurityManager;
59 utf *utf_java_lang_String;
60 utf *utf_java_lang_System;
61 utf *utf_java_lang_ThreadGroup;
62 utf *utf_java_io_Serializable;
64 utf *utf_java_lang_Throwable;
65 utf *utf_java_lang_VMThrowable;
66 utf *utf_java_lang_Error;
67 utf *utf_java_lang_Exception;
68 utf *utf_java_lang_NoClassDefFoundError;
69 utf *utf_java_lang_OutOfMemoryError;
70 utf *utf_java_lang_ClassNotFoundException;
72 utf* utf_java_lang_Void;
73 utf* utf_java_lang_Boolean;
74 utf* utf_java_lang_Byte;
75 utf* utf_java_lang_Character;
76 utf* utf_java_lang_Short;
77 utf* utf_java_lang_Integer;
78 utf* utf_java_lang_Long;
79 utf* utf_java_lang_Float;
80 utf* utf_java_lang_Double;
82 utf *utf_java_util_Vector;
83 utf *utf_java_lang_reflect_Constructor;
84 utf *utf_java_lang_reflect_Method;
87 utf *utf_InnerClasses; /* InnerClasses */
88 utf *utf_ConstantValue; /* ConstantValue */
89 utf *utf_Code; /* Code */
90 utf *utf_Exceptions; /* Exceptions */
91 utf *utf_LineNumberTable; /* LineNumberTable */
92 utf *utf_SourceFile; /* SourceFile */
94 utf *utf_init; /* <init> */
95 utf *utf_clinit; /* <clinit> */
96 utf *utf_finalize; /* finalize */
98 utf *utf_printStackTrace;
99 utf *utf_fillInStackTrace;
102 utf *utf_void__void; /* ()V */
103 utf *utf_boolean__void; /* (Z)V */
104 utf *utf_byte__void; /* (B)V */
105 utf *utf_char__void; /* (C)V */
106 utf *utf_short__void; /* (S)V */
107 utf *utf_int__void; /* (I)V */
108 utf *utf_long__void; /* (J)V */
109 utf *utf_float__void; /* (F)V */
110 utf *utf_double__void; /* (D)V */
111 utf *utf_void__java_lang_Object; /* ()Ljava/lang/Object; */
112 utf *utf_void__java_lang_Throwable; /* ()Ljava/lang/Throwable; */
113 utf *utf_java_lang_String__void; /* (Ljava/lang/String;)V */
114 utf *utf_java_lang_String__java_lang_Class;
115 utf *utf_java_lang_Throwable__void; /* (Ljava/lang/Throwable;)V */
117 utf *utf_not_named_yet; /* special name for unnamed classes */
119 utf *array_packagename;
122 /* utf_init ********************************************************************
124 Initializes the utf8 subsystem.
126 *******************************************************************************/
130 /* create utf-symbols for pointer comparison of frequently used strings */
132 utf_java_lang_Object = utf_new_char("java/lang/Object");
134 utf_java_lang_Class = utf_new_char("java/lang/Class");
135 utf_java_lang_ClassLoader = utf_new_char("java/lang/ClassLoader");
136 utf_java_lang_Cloneable = utf_new_char("java/lang/Cloneable");
137 utf_java_lang_SecurityManager = utf_new_char("java/lang/SecurityManager");
138 utf_java_lang_String = utf_new_char("java/lang/String");
139 utf_java_lang_System = utf_new_char("java/lang/System");
140 utf_java_lang_ThreadGroup = utf_new_char("java/lang/ThreadGroup");
141 utf_java_io_Serializable = utf_new_char("java/io/Serializable");
143 utf_java_lang_Throwable = utf_new_char(string_java_lang_Throwable);
144 utf_java_lang_VMThrowable = utf_new_char(string_java_lang_VMThrowable);
145 utf_java_lang_Error = utf_new_char(string_java_lang_Error);
146 utf_java_lang_Exception = utf_new_char(string_java_lang_Exception);
148 utf_java_lang_NoClassDefFoundError =
149 utf_new_char(string_java_lang_NoClassDefFoundError);
151 utf_java_lang_OutOfMemoryError =
152 utf_new_char(string_java_lang_OutOfMemoryError);
154 utf_java_lang_ClassNotFoundException =
155 utf_new_char(string_java_lang_ClassNotFoundException);
157 utf_java_lang_Void = utf_new_char("java/lang/Void");
158 utf_java_lang_Boolean = utf_new_char("java/lang/Boolean");
159 utf_java_lang_Byte = utf_new_char("java/lang/Byte");
160 utf_java_lang_Character = utf_new_char("java/lang/Character");
161 utf_java_lang_Short = utf_new_char("java/lang/Short");
162 utf_java_lang_Integer = utf_new_char("java/lang/Integer");
163 utf_java_lang_Long = utf_new_char("java/lang/Long");
164 utf_java_lang_Float = utf_new_char("java/lang/Float");
165 utf_java_lang_Double = utf_new_char("java/lang/Double");
167 utf_java_util_Vector = utf_new_char("java/util/Vector");
168 utf_java_lang_reflect_Constructor = utf_new_char("java/lang/reflect/Constructor");
169 utf_java_lang_reflect_Method = utf_new_char("java/lang/reflect/Method");
171 utf_InnerClasses = utf_new_char("InnerClasses");
172 utf_ConstantValue = utf_new_char("ConstantValue");
173 utf_Code = utf_new_char("Code");
174 utf_Exceptions = utf_new_char("Exceptions");
175 utf_LineNumberTable = utf_new_char("LineNumberTable");
176 utf_SourceFile = utf_new_char("SourceFile");
178 utf_init = utf_new_char("<init>");
179 utf_clinit = utf_new_char("<clinit>");
180 utf_finalize = utf_new_char("finalize");
182 utf_printStackTrace = utf_new_char("printStackTrace");
183 utf_fillInStackTrace = utf_new_char("fillInStackTrace");
184 utf_loadClass = utf_new_char("loadClass");
186 utf_void__void = utf_new_char("()V");
187 utf_boolean__void = utf_new_char("(Z)V");
188 utf_byte__void = utf_new_char("(B)V");
189 utf_char__void = utf_new_char("(C)V");
190 utf_short__void = utf_new_char("(S)V");
191 utf_int__void = utf_new_char("(I)V");
192 utf_long__void = utf_new_char("(J)V");
193 utf_float__void = utf_new_char("(F)V");
194 utf_double__void = utf_new_char("(D)V");
195 utf_void__java_lang_Object = utf_new_char("()Ljava/lang/Object;");
196 utf_void__java_lang_Throwable = utf_new_char("()Ljava/lang/Throwable;");
197 utf_java_lang_String__void = utf_new_char("(Ljava/lang/String;)V");
199 utf_java_lang_String__java_lang_Class =
200 utf_new_char("(Ljava/lang/String;)Ljava/lang/Class;");
202 utf_java_lang_Throwable__void = utf_new_char("(Ljava/lang/Throwable;)V");
204 utf_not_named_yet = utf_new_char("\t<not_named_yet>");
206 array_packagename = utf_new_char("\t<the array package>");
210 /* utf_hashkey *****************************************************************
212 The hashkey is computed from the utf-text by using up to 8
213 characters. For utf-symbols longer than 15 characters 3 characters
214 are taken from the beginning and the end, 2 characters are taken
217 *******************************************************************************/
219 #define nbs(val) ((u4) *(++text) << val) /* get next byte, left shift by val */
220 #define fbs(val) ((u4) *( text) << val) /* get first byte, left shift by val */
222 u4 utf_hashkey(const char *text, u4 length)
224 const char *start_pos = text; /* pointer to utf text */
228 case 0: /* empty string */
231 case 1: return fbs(0);
232 case 2: return fbs(0) ^ nbs(3);
233 case 3: return fbs(0) ^ nbs(3) ^ nbs(5);
234 case 4: return fbs(0) ^ nbs(2) ^ nbs(4) ^ nbs(6);
235 case 5: return fbs(0) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(6);
236 case 6: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(5) ^ nbs(6);
237 case 7: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6);
238 case 8: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7);
245 return a ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7) ^ nbs(8);
254 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9);
263 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9) ^ nbs(10);
275 return a ^ nbs(9) ^ nbs(10);
287 return a ^ nbs(9) ^ nbs(10);
298 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
309 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
311 default: /* 3 characters from beginning */
317 /* 2 characters from middle */
318 text = start_pos + (length / 2);
323 /* 3 characters from end */
324 text = start_pos + length - 4;
329 return a ^ nbs(10) ^ nbs(11);
334 /* utf_hashkey *****************************************************************
336 Compute the hashkey of a unicode string.
338 *******************************************************************************/
340 u4 unicode_hashkey(u2 *text, u2 len)
342 return utf_hashkey((char *) text, len);
346 /* utf_new *********************************************************************
348 Creates a new utf-symbol, the text of the symbol is passed as a
349 u1-array. The function searches the utf-hashtable for a utf-symbol
350 with this text. On success the element returned, otherwise a new
351 hashtable element is created.
353 If the number of entries in the hashtable exceeds twice the size of
354 the hashtable slots a reorganization of the hashtable is done and
355 the utf symbols are copied to a new hashtable with doubled size.
357 *******************************************************************************/
359 utf *utf_new_intern(const char *text, u2 length);
361 utf *utf_new(const char *text, u2 length)
365 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
369 r = utf_new_intern(text, length);
371 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
379 utf *utf_new_intern(const char *text, u2 length)
381 u4 key; /* hashkey computed from utf-text */
382 u4 slot; /* slot in hashtable */
383 utf *u; /* hashtable element */
391 key = utf_hashkey(text, length);
392 slot = key & (utf_hash.size - 1);
393 u = utf_hash.ptr[slot];
395 /* search external hash chain for utf-symbol */
397 if (u->blength == length) {
399 /* compare text of hashtable elements */
400 for (i = 0; i < length; i++)
401 if (text[i] != u->text[i]) goto nomatch;
405 count_utf_new_found++;
408 /* symbol found in hashtable */
412 u = u->hashlink; /* next element in external chain */
417 count_utf_len += sizeof(utf) + length;
420 /* location in hashtable found, create new utf element */
422 u->blength = length; /* length in bytes of utfstring */
423 u->hashlink = utf_hash.ptr[slot]; /* link in external hashchain */
424 u->text = mem_alloc(length + 1);/* allocate memory for utf-text */
425 memcpy(u->text, text, length); /* copy utf-text */
426 u->text[length] = '\0';
427 utf_hash.ptr[slot] = u; /* insert symbol into table */
429 utf_hash.entries++; /* update number of entries */
431 if (utf_hash.entries > (utf_hash.size * 2)) {
433 /* reorganization of hashtable, average length of
434 the external chains is approx. 2 */
438 hashtable newhash; /* the new hashtable */
440 /* create new hashtable, double the size */
441 init_hashtable(&newhash, utf_hash.size * 2);
442 newhash.entries = utf_hash.entries;
446 count_utf_len += sizeof(utf*) * utf_hash.size;
449 /* transfer elements to new hashtable */
450 for (i = 0; i < utf_hash.size; i++) {
451 u = (utf *) utf_hash.ptr[i];
453 utf *nextu = u->hashlink;
454 u4 slot = utf_hashkey(u->text, u->blength) & (newhash.size - 1);
456 u->hashlink = (utf *) newhash.ptr[slot];
457 newhash.ptr[slot] = u;
459 /* follow link in external hash chain */
464 /* dispose old table */
465 MFREE(utf_hash.ptr, void*, utf_hash.size);
473 /* utf_new_u2 ******************************************************************
475 Make utf symbol from u2 array, if isclassname is true '.' is
478 *******************************************************************************/
480 utf *utf_new_u2(u2 *unicode_pos, u4 unicode_length, bool isclassname)
482 char *buffer; /* memory buffer for unicode characters */
483 char *pos; /* pointer to current position in buffer */
484 u4 left; /* unicode characters left */
485 u4 buflength; /* utf length in bytes of the u2 array */
486 utf *result; /* resulting utf-string */
489 /* determine utf length in bytes and allocate memory */
491 buflength = u2_utflength(unicode_pos, unicode_length);
492 buffer = MNEW(char, buflength);
497 for (i = 0; i++ < unicode_length; unicode_pos++) {
498 /* next unicode character */
501 if ((c != 0) && (c < 0x80)) {
504 if ((int) left < 0) break;
505 /* convert classname */
506 if (isclassname && c == '.')
511 } else if (c < 0x800) {
513 unsigned char high = c >> 6;
514 unsigned char low = c & 0x3F;
516 if ((int) left < 0) break;
517 *pos++ = high | 0xC0;
523 char mid = (c >> 6) & 0x3F;
526 if ((int) left < 0) break;
527 *pos++ = high | 0xE0;
533 /* insert utf-string into symbol-table */
534 result = utf_new(buffer,buflength);
536 MFREE(buffer, char, buflength);
542 /* utf_new_char ****************************************************************
544 Creates a new utf symbol, the text for this symbol is passed as a
545 c-string ( = char* ).
547 *******************************************************************************/
549 utf *utf_new_char(const char *text)
551 return utf_new(text, strlen(text));
555 /* utf_new_char_classname ******************************************************
557 Creates a new utf symbol, the text for this symbol is passed as a
558 c-string ( = char* ) "." characters are going to be replaced by
559 "/". Since the above function is used often, this is a separte
560 function, instead of an if.
562 *******************************************************************************/
564 utf *utf_new_char_classname(const char *text)
566 if (strchr(text, '.')) {
567 char *txt = strdup(text);
568 char *end = txt + strlen(txt);
572 for (c = txt; c < end; c++)
573 if (*c == '.') *c = '/';
575 tmpRes = utf_new(txt, strlen(txt));
581 return utf_new(text, strlen(text));
585 /* utf_nextu2 ******************************************************************
587 Read the next unicode character from the utf string and increment
588 the utf-string pointer accordingly.
590 *******************************************************************************/
592 u2 utf_nextu2(char **utf_ptr)
594 /* uncompressed unicode character */
596 /* current position in utf text */
597 unsigned char *utf = (unsigned char *) (*utf_ptr);
598 /* bytes representing the unicode character */
599 unsigned char ch1, ch2, ch3;
600 /* number of bytes used to represent the unicode character */
603 switch ((ch1 = utf[0]) >> 4) {
604 default: /* 1 byte */
608 case 0xD: /* 2 bytes */
609 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
610 unsigned char high = ch1 & 0x1F;
611 unsigned char low = ch2 & 0x3F;
612 unicode_char = (high << 6) + low;
617 case 0xE: /* 2 or 3 bytes */
618 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
619 if (((ch3 = utf[2]) & 0xC0) == 0x80) {
620 unsigned char low = ch3 & 0x3f;
621 unsigned char mid = ch2 & 0x3f;
622 unsigned char high = ch1 & 0x0f;
623 unicode_char = (((high << 6) + mid) << 6) + low;
631 /* update position in utf-text */
632 *utf_ptr = (char *) (utf + len);
638 /* utf_strlen ******************************************************************
640 Determine number of unicode characters in the utf string.
642 *******************************************************************************/
644 u4 utf_strlen(utf *u)
646 char *endpos; /* points behind utf string */
647 char *utf_ptr; /* current position in utf text */
648 u4 len = 0; /* number of unicode characters */
651 *exceptionptr = new_nullpointerexception();
658 while (utf_ptr < endpos) {
660 /* next unicode character */
661 utf_nextu2(&utf_ptr);
664 if (utf_ptr != endpos)
665 /* string ended abruptly */
666 throw_cacao_exception_exit(string_java_lang_InternalError,
667 "Illegal utf8 string");
673 /* u2_utflength ****************************************************************
675 Returns the utf length in bytes of a u2 array.
677 *******************************************************************************/
679 u4 u2_utflength(u2 *text, u4 u2_length)
681 u4 result_len = 0; /* utf length in bytes */
682 u2 ch; /* current unicode character */
685 for (len = 0; len < u2_length; len++) {
686 /* next unicode character */
689 /* determine bytes required to store unicode character as utf */
690 if (ch && (ch < 0x80))
702 /* utf_display *****************************************************************
704 Write utf symbol to stdout (for debugging purposes).
706 *******************************************************************************/
708 void utf_display(utf *u)
710 char *endpos; /* points behind utf string */
711 char *utf_ptr; /* current position in utf text */
722 while (utf_ptr < endpos) {
723 /* read next unicode character */
724 u2 c = utf_nextu2(&utf_ptr);
725 if (c >= 32 && c <= 127) printf("%c", c);
733 /* utf_display_classname *******************************************************
735 Write utf symbol to stdout with `/' converted to `.' (for debugging
738 *******************************************************************************/
740 void utf_display_classname(utf *u)
742 char *endpos; /* points behind utf string */
743 char *utf_ptr; /* current position in utf text */
754 while (utf_ptr < endpos) {
755 /* read next unicode character */
756 u2 c = utf_nextu2(&utf_ptr);
757 if (c == '/') c = '.';
758 if (c >= 32 && c <= 127) printf("%c", c);
766 /* utf_sprint ******************************************************************
768 Write utf symbol into c-string (for debugging purposes).
770 *******************************************************************************/
772 void utf_sprint(char *buffer, utf *u)
774 char *endpos; /* points behind utf string */
775 char *utf_ptr; /* current position in utf text */
776 u2 pos = 0; /* position in c-string */
779 strcpy(buffer, "NULL");
786 while (utf_ptr < endpos)
787 /* copy next unicode character */
788 buffer[pos++] = utf_nextu2(&utf_ptr);
790 /* terminate string */
795 /* utf_sprint_classname ********************************************************
797 Write utf symbol into c-string with `/' converted to `.' (for debugging
800 *******************************************************************************/
802 void utf_sprint_classname(char *buffer, utf *u)
804 char *endpos; /* points behind utf string */
805 char *utf_ptr; /* current position in utf text */
806 u2 pos = 0; /* position in c-string */
809 strcpy(buffer, "NULL");
816 while (utf_ptr < endpos) {
817 /* copy next unicode character */
818 u2 c = utf_nextu2(&utf_ptr);
819 if (c == '/') c = '.';
823 /* terminate string */
828 /* utf_strcat ******************************************************************
830 Like libc strcat, but uses an utf8 string.
832 *******************************************************************************/
834 void utf_strcat(char *buffer, utf *u)
836 utf_sprint(buffer + strlen(buffer), u);
840 /* utf_strcat_classname ********************************************************
842 Like libc strcat, but uses an utf8 string.
844 *******************************************************************************/
846 void utf_strcat_classname(char *buffer, utf *u)
848 utf_sprint_classname(buffer + strlen(buffer), u);
852 /* utf_fprint ******************************************************************
854 Write utf symbol into file.
856 *******************************************************************************/
858 void utf_fprint(FILE *file, utf *u)
860 char *endpos; /* points behind utf string */
861 char *utf_ptr; /* current position in utf text */
869 while (utf_ptr < endpos) {
870 /* read next unicode character */
871 u2 c = utf_nextu2(&utf_ptr);
873 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
874 else fprintf(file, "?");
879 /* utf_fprint_classname ********************************************************
881 Write utf symbol into file with `/' converted to `.'.
883 *******************************************************************************/
885 void utf_fprint_classname(FILE *file, utf *u)
887 char *endpos; /* points behind utf string */
888 char *utf_ptr; /* current position in utf text */
896 while (utf_ptr < endpos) {
897 /* read next unicode character */
898 u2 c = utf_nextu2(&utf_ptr);
899 if (c == '/') c = '.';
901 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
902 else fprintf(file, "?");
907 /* is_valid_utf ****************************************************************
909 Return true if the given string is a valid UTF-8 string.
911 utf_ptr...points to first character
912 end_pos...points after last character
914 *******************************************************************************/
916 static unsigned long min_codepoint[6] = {0,1L<<7,1L<<11,1L<<16,1L<<21,1L<<26};
918 bool is_valid_utf(char *utf_ptr, char *end_pos)
925 if (end_pos < utf_ptr) return false;
926 bytes = end_pos - utf_ptr;
930 if (!c) return false; /* 0x00 is not allowed */
931 if ((c & 0x80) == 0) continue; /* ASCII */
933 if ((c & 0xe0) == 0xc0) len = 1; /* 110x xxxx */
934 else if ((c & 0xf0) == 0xe0) len = 2; /* 1110 xxxx */
935 else if ((c & 0xf8) == 0xf0) len = 3; /* 1111 0xxx */
936 else if ((c & 0xfc) == 0xf8) len = 4; /* 1111 10xx */
937 else if ((c & 0xfe) == 0xfc) len = 5; /* 1111 110x */
938 else return false; /* invalid leading byte */
940 if (len > 2) return false; /* Java limitation */
942 v = (unsigned long)c & (0x3f >> len);
944 if ((bytes -= len) < 0) return false; /* missing bytes */
946 for (i = len; i--; ) {
948 if ((c & 0xc0) != 0x80) /* 10xx xxxx */
950 v = (v << 6) | (c & 0x3f);
954 if (len != 1) return false; /* Java special */
957 /* Sun Java seems to allow overlong UTF-8 encodings */
959 if (v < min_codepoint[len]) { /* overlong UTF-8 */
961 fprintf(stderr,"WARNING: Overlong UTF-8 sequence found.\n");
962 /* XXX change this to panic? */
966 /* surrogates in UTF-8 seem to be allowed in Java classfiles */
967 /* if (v >= 0xd800 && v <= 0xdfff) return false; */ /* surrogates */
969 /* even these seem to be allowed */
970 /* if (v == 0xfffe || v == 0xffff) return false; */ /* invalid codepoints */
977 /* is_valid_name ***************************************************************
979 Return true if the given string may be used as a class/field/method
980 name. (Currently this only disallows empty strings and control
983 NOTE: The string is assumed to have passed is_valid_utf!
985 utf_ptr...points to first character
986 end_pos...points after last character
988 *******************************************************************************/
990 bool is_valid_name(char *utf_ptr, char *end_pos)
992 if (end_pos <= utf_ptr) return false; /* disallow empty names */
994 while (utf_ptr < end_pos) {
995 unsigned char c = *utf_ptr++;
997 if (c < 0x20) return false; /* disallow control characters */
998 if (c == 0xc0 && (unsigned char) *utf_ptr == 0x80) /* disallow zero */
1005 bool is_valid_name_utf(utf *u)
1007 return is_valid_name(u->text,utf_end(u));
1011 /* utf_show ********************************************************************
1013 Writes the utf symbols in the utfhash to stdout and displays the
1014 number of external hash chains grouped according to the chainlength
1015 (for debugging purposes).
1017 *******************************************************************************/
1022 #define CHAIN_LIMIT 20 /* limit for seperated enumeration */
1024 u4 chain_count[CHAIN_LIMIT]; /* numbers of chains */
1025 u4 max_chainlength = 0; /* maximum length of the chains */
1026 u4 sum_chainlength = 0; /* sum of the chainlengths */
1027 u4 beyond_limit = 0; /* number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1030 printf ("UTF-HASH:\n");
1032 /* show element of utf-hashtable */
1033 for (i=0; i<utf_hash.size; i++) {
1034 utf *u = utf_hash.ptr[i];
1036 printf ("SLOT %d: ", (int) i);
1048 printf ("UTF-HASH: %d slots for %d entries\n",
1049 (int) utf_hash.size, (int) utf_hash.entries );
1052 if (utf_hash.entries == 0)
1055 printf("chains:\n chainlength number of chains %% of utfstrings\n");
1057 for (i=0;i<CHAIN_LIMIT;i++)
1060 /* count numbers of hashchains according to their length */
1061 for (i=0; i<utf_hash.size; i++) {
1063 utf *u = (utf*) utf_hash.ptr[i];
1064 u4 chain_length = 0;
1066 /* determine chainlength */
1072 /* update sum of all chainlengths */
1073 sum_chainlength+=chain_length;
1075 /* determine the maximum length of the chains */
1076 if (chain_length>max_chainlength)
1077 max_chainlength = chain_length;
1079 /* update number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1080 if (chain_length>=CHAIN_LIMIT) {
1081 beyond_limit+=chain_length;
1082 chain_length=CHAIN_LIMIT-1;
1085 /* update number of hashchains of current length */
1086 chain_count[chain_length]++;
1089 /* display results */
1090 for (i=1;i<CHAIN_LIMIT-1;i++)
1091 printf(" %2d %17d %18.2f%%\n",i,chain_count[i],(((float) chain_count[i]*i*100)/utf_hash.entries));
1093 printf(" >=%2d %17d %18.2f%%\n",CHAIN_LIMIT-1,chain_count[CHAIN_LIMIT-1],((float) beyond_limit*100)/utf_hash.entries);
1096 printf("max. chainlength:%5d\n",max_chainlength);
1098 /* avg. chainlength = sum of chainlengths / number of chains */
1099 printf("avg. chainlength:%5.2f\n",(float) sum_chainlength / (utf_hash.size-chain_count[0]));
1104 * These are local overrides for various environment variables in Emacs.
1105 * Please do not remove this and leave it at the end of the file, where
1106 * Emacs will automagically detect them.
1107 * ---------------------------------------------------------------------
1110 * indent-tabs-mode: t