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 2277 2005-04-12 19:48:03Z twisti $
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_io_Serializable;
63 utf *utf_java_lang_Throwable;
64 utf *utf_java_lang_VMThrowable;
65 utf *utf_java_lang_Error;
66 utf *utf_java_lang_Exception;
67 utf *utf_java_lang_NoClassDefFoundError;
68 utf *utf_java_lang_OutOfMemoryError;
69 utf *utf_java_lang_ClassNotFoundException;
71 utf* utf_java_lang_Void;
72 utf* utf_java_lang_Boolean;
73 utf* utf_java_lang_Byte;
74 utf* utf_java_lang_Character;
75 utf* utf_java_lang_Short;
76 utf* utf_java_lang_Integer;
77 utf* utf_java_lang_Long;
78 utf* utf_java_lang_Float;
79 utf* utf_java_lang_Double;
81 utf *utf_java_util_Vector;
83 utf *utf_InnerClasses; /* InnerClasses */
84 utf *utf_ConstantValue; /* ConstantValue */
85 utf *utf_Code; /* Code */
86 utf *utf_Exceptions; /* Exceptions */
87 utf *utf_LineNumberTable; /* LineNumberTable */
88 utf *utf_SourceFile; /* SourceFile */
90 utf *utf_init; /* <init> */
91 utf *utf_clinit; /* <clinit> */
92 utf *utf_finalize; /* finalize */
94 utf *utf_printStackTrace;
95 utf *utf_fillInStackTrace;
98 utf *utf_void__void; /* ()V */
99 utf *utf_boolean__void; /* (Z)V */
100 utf *utf_byte__void; /* (B)V */
101 utf *utf_char__void; /* (C)V */
102 utf *utf_short__void; /* (S)V */
103 utf *utf_int__void; /* (I)V */
104 utf *utf_long__void; /* (J)V */
105 utf *utf_float__void; /* (F)V */
106 utf *utf_double__void; /* (D)V */
107 utf *utf_void__java_lang_Object; /* ()Ljava/lang/Object; */
108 utf *utf_void__java_lang_Throwable; /* ()Ljava/lang/Throwable; */
109 utf *utf_java_lang_String__void; /* (Ljava/lang/String;)V */
110 utf *utf_java_lang_String__java_lang_Class;
111 utf *utf_java_lang_Throwable__void; /* (Ljava/lang/Throwable;)V */
113 utf *array_packagename;
116 /* utf_init ********************************************************************
118 Initializes the utf8 subsystem.
120 *******************************************************************************/
124 /* create utf-symbols for pointer comparison of frequently used strings */
126 utf_java_lang_Object = utf_new_char("java/lang/Object");
128 utf_java_lang_Class = utf_new_char("java/lang/Class");
129 utf_java_lang_ClassLoader = utf_new_char("java/lang/ClassLoader");
130 utf_java_lang_Cloneable = utf_new_char("java/lang/Cloneable");
131 utf_java_lang_SecurityManager = utf_new_char("java/lang/SecurityManager");
132 utf_java_lang_String = utf_new_char("java/lang/String");
133 utf_java_lang_System = utf_new_char("java/lang/System");
134 utf_java_io_Serializable = utf_new_char("java/io/Serializable");
136 utf_java_lang_Throwable = utf_new_char(string_java_lang_Throwable);
137 utf_java_lang_VMThrowable = utf_new_char(string_java_lang_VMThrowable);
138 utf_java_lang_Error = utf_new_char(string_java_lang_Error);
139 utf_java_lang_Exception = utf_new_char(string_java_lang_Exception);
141 utf_java_lang_NoClassDefFoundError =
142 utf_new_char(string_java_lang_NoClassDefFoundError);
144 utf_java_lang_OutOfMemoryError =
145 utf_new_char(string_java_lang_OutOfMemoryError);
147 utf_java_lang_ClassNotFoundException =
148 utf_new_char(string_java_lang_ClassNotFoundException);
150 utf_java_lang_Void = utf_new_char("java/lang/Void");
151 utf_java_lang_Boolean = utf_new_char("java/lang/Boolean");
152 utf_java_lang_Byte = utf_new_char("java/lang/Byte");
153 utf_java_lang_Character = utf_new_char("java/lang/Character");
154 utf_java_lang_Short = utf_new_char("java/lang/Short");
155 utf_java_lang_Integer = utf_new_char("java/lang/Integer");
156 utf_java_lang_Long = utf_new_char("java/lang/Long");
157 utf_java_lang_Float = utf_new_char("java/lang/Float");
158 utf_java_lang_Double = utf_new_char("java/lang/Double");
160 utf_java_util_Vector = utf_new_char("java/util/Vector");
162 utf_InnerClasses = utf_new_char("InnerClasses");
163 utf_ConstantValue = utf_new_char("ConstantValue");
164 utf_Code = utf_new_char("Code");
165 utf_Exceptions = utf_new_char("Exceptions");
166 utf_LineNumberTable = utf_new_char("LineNumberTable");
167 utf_SourceFile = utf_new_char("SourceFile");
169 utf_init = utf_new_char("<init>");
170 utf_clinit = utf_new_char("<clinit>");
171 utf_finalize = utf_new_char("finalize");
173 utf_printStackTrace = utf_new_char("printStackTrace");
174 utf_fillInStackTrace = utf_new_char("fillInStackTrace");
175 utf_loadClass = utf_new_char("loadClass");
177 utf_void__void = utf_new_char("()V");
178 utf_boolean__void = utf_new_char("(Z)V");
179 utf_byte__void = utf_new_char("(B)V");
180 utf_char__void = utf_new_char("(C)V");
181 utf_short__void = utf_new_char("(S)V");
182 utf_int__void = utf_new_char("(I)V");
183 utf_long__void = utf_new_char("(J)V");
184 utf_float__void = utf_new_char("(F)V");
185 utf_double__void = utf_new_char("(D)V");
186 utf_void__java_lang_Object = utf_new_char("()Ljava/lang/Object;");
187 utf_void__java_lang_Throwable = utf_new_char("()Ljava/lang/Throwable;");
188 utf_java_lang_String__void = utf_new_char("(Ljava/lang/String;)V");
190 utf_java_lang_String__java_lang_Class =
191 utf_new_char("(Ljava/lang/String;)Ljava/lang/Class;");
193 utf_java_lang_Throwable__void = utf_new_char("(Ljava/lang/Throwable;)V");
195 array_packagename = utf_new_char("\t<the array package>");
199 /* utf_hashkey *****************************************************************
201 The hashkey is computed from the utf-text by using up to 8
202 characters. For utf-symbols longer than 15 characters 3 characters
203 are taken from the beginning and the end, 2 characters are taken
206 *******************************************************************************/
208 #define nbs(val) ((u4) *(++text) << val) /* get next byte, left shift by val */
209 #define fbs(val) ((u4) *( text) << val) /* get first byte, left shift by val */
211 u4 utf_hashkey(const char *text, u4 length)
213 const char *start_pos = text; /* pointer to utf text */
217 case 0: /* empty string */
220 case 1: return fbs(0);
221 case 2: return fbs(0) ^ nbs(3);
222 case 3: return fbs(0) ^ nbs(3) ^ nbs(5);
223 case 4: return fbs(0) ^ nbs(2) ^ nbs(4) ^ nbs(6);
224 case 5: return fbs(0) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(6);
225 case 6: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(5) ^ nbs(6);
226 case 7: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6);
227 case 8: return fbs(0) ^ nbs(1) ^ nbs(2) ^ nbs(3) ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7);
234 return a ^ nbs(4) ^ nbs(5) ^ nbs(6) ^ nbs(7) ^ nbs(8);
243 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9);
252 return a ^ nbs(6) ^ nbs(7) ^ nbs(8) ^ nbs(9) ^ nbs(10);
264 return a ^ nbs(9) ^ nbs(10);
276 return a ^ nbs(9) ^ nbs(10);
287 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
298 return a ^ nbs(9) ^ nbs(10) ^ nbs(11);
300 default: /* 3 characters from beginning */
306 /* 2 characters from middle */
307 text = start_pos + (length / 2);
312 /* 3 characters from end */
313 text = start_pos + length - 4;
318 return a ^ nbs(10) ^ nbs(11);
323 /* utf_hashkey *****************************************************************
325 Compute the hashkey of a unicode string.
327 *******************************************************************************/
329 u4 unicode_hashkey(u2 *text, u2 len)
331 return utf_hashkey((char *) text, len);
335 /* utf_new *********************************************************************
337 Creates a new utf-symbol, the text of the symbol is passed as a
338 u1-array. The function searches the utf-hashtable for a utf-symbol
339 with this text. On success the element returned, otherwise a new
340 hashtable element is created.
342 If the number of entries in the hashtable exceeds twice the size of
343 the hashtable slots a reorganization of the hashtable is done and
344 the utf symbols are copied to a new hashtable with doubled size.
346 *******************************************************************************/
348 utf *utf_new_intern(const char *text, u2 length);
350 utf *utf_new(const char *text, u2 length)
354 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
358 r = utf_new_intern(text, length);
360 #if defined(USE_THREADS) && defined(NATIVE_THREADS)
368 utf *utf_new_intern(const char *text, u2 length)
370 u4 key; /* hashkey computed from utf-text */
371 u4 slot; /* slot in hashtable */
372 utf *u; /* hashtable element */
380 key = utf_hashkey(text, length);
381 slot = key & (utf_hash.size - 1);
382 u = utf_hash.ptr[slot];
384 /* search external hash chain for utf-symbol */
386 if (u->blength == length) {
388 /* compare text of hashtable elements */
389 for (i = 0; i < length; i++)
390 if (text[i] != u->text[i]) goto nomatch;
394 count_utf_new_found++;
397 /* symbol found in hashtable */
401 u = u->hashlink; /* next element in external chain */
406 count_utf_len += sizeof(utf) + length;
409 /* location in hashtable found, create new utf element */
411 u->blength = length; /* length in bytes of utfstring */
412 u->hashlink = utf_hash.ptr[slot]; /* link in external hashchain */
413 u->text = mem_alloc(length + 1);/* allocate memory for utf-text */
414 memcpy(u->text, text, length); /* copy utf-text */
415 u->text[length] = '\0';
416 utf_hash.ptr[slot] = u; /* insert symbol into table */
418 utf_hash.entries++; /* update number of entries */
420 if (utf_hash.entries > (utf_hash.size * 2)) {
422 /* reorganization of hashtable, average length of
423 the external chains is approx. 2 */
427 hashtable newhash; /* the new hashtable */
429 /* create new hashtable, double the size */
430 init_hashtable(&newhash, utf_hash.size * 2);
431 newhash.entries = utf_hash.entries;
435 count_utf_len += sizeof(utf*) * utf_hash.size;
438 /* transfer elements to new hashtable */
439 for (i = 0; i < utf_hash.size; i++) {
440 u = (utf *) utf_hash.ptr[i];
442 utf *nextu = u->hashlink;
443 u4 slot = utf_hashkey(u->text, u->blength) & (newhash.size - 1);
445 u->hashlink = (utf *) newhash.ptr[slot];
446 newhash.ptr[slot] = u;
448 /* follow link in external hash chain */
453 /* dispose old table */
454 MFREE(utf_hash.ptr, void*, utf_hash.size);
462 /* utf_new_u2 ******************************************************************
464 Make utf symbol from u2 array, if isclassname is true '.' is
467 *******************************************************************************/
469 utf *utf_new_u2(u2 *unicode_pos, u4 unicode_length, bool isclassname)
471 char *buffer; /* memory buffer for unicode characters */
472 char *pos; /* pointer to current position in buffer */
473 u4 left; /* unicode characters left */
474 u4 buflength; /* utf length in bytes of the u2 array */
475 utf *result; /* resulting utf-string */
478 /* determine utf length in bytes and allocate memory */
480 buflength = u2_utflength(unicode_pos, unicode_length);
481 buffer = MNEW(char, buflength);
486 for (i = 0; i++ < unicode_length; unicode_pos++) {
487 /* next unicode character */
490 if ((c != 0) && (c < 0x80)) {
493 if ((int) left < 0) break;
494 /* convert classname */
495 if (isclassname && c == '.')
500 } else if (c < 0x800) {
502 unsigned char high = c >> 6;
503 unsigned char low = c & 0x3F;
505 if ((int) left < 0) break;
506 *pos++ = high | 0xC0;
512 char mid = (c >> 6) & 0x3F;
515 if ((int) left < 0) break;
516 *pos++ = high | 0xE0;
522 /* insert utf-string into symbol-table */
523 result = utf_new(buffer,buflength);
525 MFREE(buffer, char, buflength);
531 /* utf_new_char ****************************************************************
533 Creates a new utf symbol, the text for this symbol is passed as a
534 c-string ( = char* ).
536 *******************************************************************************/
538 utf *utf_new_char(const char *text)
540 return utf_new(text, strlen(text));
544 /* utf_new_char_classname ******************************************************
546 Creates a new utf symbol, the text for this symbol is passed as a
547 c-string ( = char* ) "." characters are going to be replaced by
548 "/". Since the above function is used often, this is a separte
549 function, instead of an if.
551 *******************************************************************************/
553 utf *utf_new_char_classname(const char *text)
555 if (strchr(text, '.')) {
556 char *txt = strdup(text);
557 char *end = txt + strlen(txt);
561 for (c = txt; c < end; c++)
562 if (*c == '.') *c = '/';
564 tmpRes = utf_new(txt, strlen(txt));
570 return utf_new(text, strlen(text));
574 /* utf_nextu2 ******************************************************************
576 Read the next unicode character from the utf string and increment
577 the utf-string pointer accordingly.
579 *******************************************************************************/
581 u2 utf_nextu2(char **utf_ptr)
583 /* uncompressed unicode character */
585 /* current position in utf text */
586 unsigned char *utf = (unsigned char *) (*utf_ptr);
587 /* bytes representing the unicode character */
588 unsigned char ch1, ch2, ch3;
589 /* number of bytes used to represent the unicode character */
592 switch ((ch1 = utf[0]) >> 4) {
593 default: /* 1 byte */
597 case 0xD: /* 2 bytes */
598 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
599 unsigned char high = ch1 & 0x1F;
600 unsigned char low = ch2 & 0x3F;
601 unicode_char = (high << 6) + low;
606 case 0xE: /* 2 or 3 bytes */
607 if (((ch2 = utf[1]) & 0xC0) == 0x80) {
608 if (((ch3 = utf[2]) & 0xC0) == 0x80) {
609 unsigned char low = ch3 & 0x3f;
610 unsigned char mid = ch2 & 0x3f;
611 unsigned char high = ch1 & 0x0f;
612 unicode_char = (((high << 6) + mid) << 6) + low;
620 /* update position in utf-text */
621 *utf_ptr = (char *) (utf + len);
627 /* utf_strlen ******************************************************************
629 Determine number of unicode characters in the utf string.
631 *******************************************************************************/
633 u4 utf_strlen(utf *u)
635 char *endpos; /* points behind utf string */
636 char *utf_ptr; /* current position in utf text */
637 u4 len = 0; /* number of unicode characters */
640 *exceptionptr = new_nullpointerexception();
647 while (utf_ptr < endpos) {
649 /* next unicode character */
650 utf_nextu2(&utf_ptr);
653 if (utf_ptr != endpos)
654 /* string ended abruptly */
655 throw_cacao_exception_exit(string_java_lang_InternalError,
656 "Illegal utf8 string");
662 /* u2_utflength ****************************************************************
664 Returns the utf length in bytes of a u2 array.
666 *******************************************************************************/
668 u4 u2_utflength(u2 *text, u4 u2_length)
670 u4 result_len = 0; /* utf length in bytes */
671 u2 ch; /* current unicode character */
674 for (len = 0; len < u2_length; len++) {
675 /* next unicode character */
678 /* determine bytes required to store unicode character as utf */
679 if (ch && (ch < 0x80))
691 /* utf_display *****************************************************************
693 Write utf symbol to stdout (for debugging purposes).
695 *******************************************************************************/
697 void utf_display(utf *u)
699 char *endpos; /* points behind utf string */
700 char *utf_ptr; /* current position in utf text */
711 while (utf_ptr < endpos) {
712 /* read next unicode character */
713 u2 c = utf_nextu2(&utf_ptr);
714 if (c >= 32 && c <= 127) printf("%c", c);
722 /* utf_display_classname *******************************************************
724 Write utf symbol to stdout with `/' converted to `.' (for debugging
727 *******************************************************************************/
729 void utf_display_classname(utf *u)
731 char *endpos; /* points behind utf string */
732 char *utf_ptr; /* current position in utf text */
743 while (utf_ptr < endpos) {
744 /* read next unicode character */
745 u2 c = utf_nextu2(&utf_ptr);
746 if (c == '/') c = '.';
747 if (c >= 32 && c <= 127) printf("%c", c);
755 /* utf_sprint ******************************************************************
757 Write utf symbol into c-string (for debugging purposes).
759 *******************************************************************************/
761 void utf_sprint(char *buffer, utf *u)
763 char *endpos; /* points behind utf string */
764 char *utf_ptr; /* current position in utf text */
765 u2 pos = 0; /* position in c-string */
768 strcpy(buffer, "NULL");
775 while (utf_ptr < endpos)
776 /* copy next unicode character */
777 buffer[pos++] = utf_nextu2(&utf_ptr);
779 /* terminate string */
784 /* utf_sprint_classname ********************************************************
786 Write utf symbol into c-string with `/' converted to `.' (for debugging
789 *******************************************************************************/
791 void utf_sprint_classname(char *buffer, utf *u)
793 char *endpos; /* points behind utf string */
794 char *utf_ptr; /* current position in utf text */
795 u2 pos = 0; /* position in c-string */
798 strcpy(buffer, "NULL");
805 while (utf_ptr < endpos) {
806 /* copy next unicode character */
807 u2 c = utf_nextu2(&utf_ptr);
808 if (c == '/') c = '.';
812 /* terminate string */
817 /* utf_strcat ******************************************************************
819 Like libc strcat, but uses an utf8 string.
821 *******************************************************************************/
823 void utf_strcat(char *buffer, utf *u)
825 utf_sprint(buffer + strlen(buffer), u);
829 /* utf_strcat_classname ********************************************************
831 Like libc strcat, but uses an utf8 string.
833 *******************************************************************************/
835 void utf_strcat_classname(char *buffer, utf *u)
837 utf_sprint_classname(buffer + strlen(buffer), u);
841 /* utf_fprint ******************************************************************
843 Write utf symbol into file.
845 *******************************************************************************/
847 void utf_fprint(FILE *file, utf *u)
849 char *endpos; /* points behind utf string */
850 char *utf_ptr; /* current position in utf text */
858 while (utf_ptr < endpos) {
859 /* read next unicode character */
860 u2 c = utf_nextu2(&utf_ptr);
862 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
863 else fprintf(file, "?");
868 /* utf_fprint_classname ********************************************************
870 Write utf symbol into file with `/' converted to `.'.
872 *******************************************************************************/
874 void utf_fprint_classname(FILE *file, utf *u)
876 char *endpos; /* points behind utf string */
877 char *utf_ptr; /* current position in utf text */
885 while (utf_ptr < endpos) {
886 /* read next unicode character */
887 u2 c = utf_nextu2(&utf_ptr);
888 if (c == '/') c = '.';
890 if (c >= 32 && c <= 127) fprintf(file, "%c", c);
891 else fprintf(file, "?");
896 /* is_valid_utf ****************************************************************
898 Return true if the given string is a valid UTF-8 string.
900 utf_ptr...points to first character
901 end_pos...points after last character
903 *******************************************************************************/
905 static unsigned long min_codepoint[6] = {0,1L<<7,1L<<11,1L<<16,1L<<21,1L<<26};
907 bool is_valid_utf(char *utf_ptr, char *end_pos)
914 if (end_pos < utf_ptr) return false;
915 bytes = end_pos - utf_ptr;
919 if (!c) return false; /* 0x00 is not allowed */
920 if ((c & 0x80) == 0) continue; /* ASCII */
922 if ((c & 0xe0) == 0xc0) len = 1; /* 110x xxxx */
923 else if ((c & 0xf0) == 0xe0) len = 2; /* 1110 xxxx */
924 else if ((c & 0xf8) == 0xf0) len = 3; /* 1111 0xxx */
925 else if ((c & 0xfc) == 0xf8) len = 4; /* 1111 10xx */
926 else if ((c & 0xfe) == 0xfc) len = 5; /* 1111 110x */
927 else return false; /* invalid leading byte */
929 if (len > 2) return false; /* Java limitation */
931 v = (unsigned long)c & (0x3f >> len);
933 if ((bytes -= len) < 0) return false; /* missing bytes */
935 for (i = len; i--; ) {
937 if ((c & 0xc0) != 0x80) /* 10xx xxxx */
939 v = (v << 6) | (c & 0x3f);
943 if (len != 1) return false; /* Java special */
946 /* Sun Java seems to allow overlong UTF-8 encodings */
948 if (v < min_codepoint[len]) { /* overlong UTF-8 */
950 fprintf(stderr,"WARNING: Overlong UTF-8 sequence found.\n");
951 /* XXX change this to panic? */
955 /* surrogates in UTF-8 seem to be allowed in Java classfiles */
956 /* if (v >= 0xd800 && v <= 0xdfff) return false; */ /* surrogates */
958 /* even these seem to be allowed */
959 /* if (v == 0xfffe || v == 0xffff) return false; */ /* invalid codepoints */
966 /* is_valid_name ***************************************************************
968 Return true if the given string may be used as a class/field/method
969 name. (Currently this only disallows empty strings and control
972 NOTE: The string is assumed to have passed is_valid_utf!
974 utf_ptr...points to first character
975 end_pos...points after last character
977 *******************************************************************************/
979 bool is_valid_name(char *utf_ptr, char *end_pos)
981 if (end_pos <= utf_ptr) return false; /* disallow empty names */
983 while (utf_ptr < end_pos) {
984 unsigned char c = *utf_ptr++;
986 if (c < 0x20) return false; /* disallow control characters */
987 if (c == 0xc0 && (unsigned char) *utf_ptr == 0x80) /* disallow zero */
994 bool is_valid_name_utf(utf *u)
996 return is_valid_name(u->text,utf_end(u));
1000 /* utf_show ********************************************************************
1002 Writes the utf symbols in the utfhash to stdout and displays the
1003 number of external hash chains grouped according to the chainlength
1004 (for debugging purposes).
1006 *******************************************************************************/
1011 #define CHAIN_LIMIT 20 /* limit for seperated enumeration */
1013 u4 chain_count[CHAIN_LIMIT]; /* numbers of chains */
1014 u4 max_chainlength = 0; /* maximum length of the chains */
1015 u4 sum_chainlength = 0; /* sum of the chainlengths */
1016 u4 beyond_limit = 0; /* number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1019 printf ("UTF-HASH:\n");
1021 /* show element of utf-hashtable */
1022 for (i=0; i<utf_hash.size; i++) {
1023 utf *u = utf_hash.ptr[i];
1025 printf ("SLOT %d: ", (int) i);
1037 printf ("UTF-HASH: %d slots for %d entries\n",
1038 (int) utf_hash.size, (int) utf_hash.entries );
1041 if (utf_hash.entries == 0)
1044 printf("chains:\n chainlength number of chains %% of utfstrings\n");
1046 for (i=0;i<CHAIN_LIMIT;i++)
1049 /* count numbers of hashchains according to their length */
1050 for (i=0; i<utf_hash.size; i++) {
1052 utf *u = (utf*) utf_hash.ptr[i];
1053 u4 chain_length = 0;
1055 /* determine chainlength */
1061 /* update sum of all chainlengths */
1062 sum_chainlength+=chain_length;
1064 /* determine the maximum length of the chains */
1065 if (chain_length>max_chainlength)
1066 max_chainlength = chain_length;
1068 /* update number of utf-symbols in chains with length>=CHAIN_LIMIT-1 */
1069 if (chain_length>=CHAIN_LIMIT) {
1070 beyond_limit+=chain_length;
1071 chain_length=CHAIN_LIMIT-1;
1074 /* update number of hashchains of current length */
1075 chain_count[chain_length]++;
1078 /* display results */
1079 for (i=1;i<CHAIN_LIMIT-1;i++)
1080 printf(" %2d %17d %18.2f%%\n",i,chain_count[i],(((float) chain_count[i]*i*100)/utf_hash.entries));
1082 printf(" >=%2d %17d %18.2f%%\n",CHAIN_LIMIT-1,chain_count[CHAIN_LIMIT-1],((float) beyond_limit*100)/utf_hash.entries);
1085 printf("max. chainlength:%5d\n",max_chainlength);
1087 /* avg. chainlength = sum of chainlengths / number of chains */
1088 printf("avg. chainlength:%5.2f\n",(float) sum_chainlength / (utf_hash.size-chain_count[0]));
1093 * These are local overrides for various environment variables in Emacs.
1094 * Please do not remove this and leave it at the end of the file, where
1095 * Emacs will automagically detect them.
1096 * ---------------------------------------------------------------------
1099 * indent-tabs-mode: t