hg/stx-libbasic: comparison String.st

equal deleted inserted replaced

-:4a6995b61c0e
+:d6729266a95b
 inclusion of the above copyright notice.   This software may not
 be provided or otherwise made available to, or used by, any
 other person.  No title to or ownership of the software is
 hereby transferred.
 "
+"{ Package: 'stx:libbasic' }"
 CharacterArray variableByteSubclass:#String
 	instanceVariableNames:''
 	classVariableNames:''
 	poolDictionaries:''
 %}.
 "
 invalid argument, or out-of-memory:
 use error handling in superclass
 "
-^ (super basicNew:anInteger) atAllPut:(Character space)
+^ (super basicNew:anInteger+1) atAllPut:(Character space)
 !
 new:n
 "return a new empty string with anInteger characters.
 In contrast to other smalltalks, this returns a string filled
 REGISTER int indx;
 REGISTER OBJ slf, cls;
 if (__isSmallInteger(index)) {
-	slf = self;
+slf = self;
-	cls = __qClass(slf);
+cls = __qClass(slf);
-	indx = __intVal(index) - 1;
+indx = __intVal(index) - 1;
-	if (cls != String) {
+if (cls != String) {
-	    if (indx < 0) goto badIndex;
+if (indx < 0) goto badIndex;
-	    indx += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+indx += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
-	}
+}
-	if ((unsigned)indx < (unsigned)(__stringSize(slf))) {
+if ((unsigned)indx < (unsigned)(__stringSize(slf))) {
-	    RETURN ( __MKCHARACTER(__stringVal(slf)[indx] & 0xFF) );
+RETURN ( __MKCHARACTER(__stringVal(slf)[indx] & 0xFF) );
-	}
+}
 }
 badIndex: ;
 %}.
-index isInteger ifFalse:[
+^ self basicAt:index
-	^ self indexNotInteger:index
-].
-^ self subscriptBoundsError:index
 !
 at:index put:aCharacter
 "store the argument, aCharacter at position index, an Integer.
 Return aCharacter (sigh).
 REGISTER OBJ slf;
 slf = self;
 if (__isString(slf)) {
-	if (__isCharacter(aCharacter)) {
+if (__isCharacter(aCharacter)) {
-	    value = __intVal(_characterVal(aCharacter));
+value = __intVal(_characterVal(aCharacter));
-	    if (((unsigned)value <= 0xFF)
+if (((unsigned)value <= 0xFF)
-	     && __isSmallInteger(index)) {
+&& __isSmallInteger(index)) {
-		indx = __intVal(index) - 1;
+indx = __intVal(index) - 1;
-		if ((unsigned)indx < (unsigned)(__stringSize(slf))) {
+if ((unsigned)indx < (unsigned)(__stringSize(slf))) {
-		    __stringVal(slf)[indx] = value;
+__stringVal(slf)[indx] = value;
-		    RETURN ( aCharacter );
+RETURN ( aCharacter );
-		}
+}
-	    }
+}
-	}
+}
 }
 %}.
-(self isMemberOf:String) ifFalse:[
+^ self basicAt:index put:aCharacter
-	^ super at:index put:aCharacter
-].
-(aCharacter isMemberOf:Character) ifFalse:[
-	"
-	 tried to store something which is not a character
-	"
-	^ self elementNotCharacter
-].
-(aCharacter asciiValue between:1 and:255) ifFalse:[
-	"
-	 tried to store a multibyte character
-	"
-	^ self elementBoundsError
-].
-"
-invalid index
-"
-index isInteger ifFalse:[
-	^ self indexNotInteger:index
-].
-^ self subscriptBoundsError:index
-"Modified: 19.4.1996 / 11:40:27 / cg"
 !
 basicAt:index
 "return the character at position index, an Integer
 - reimplemented here since we return characters"
 REGISTER int indx;
 REGISTER OBJ slf, cls;
 if (__isSmallInteger(index)) {
-	slf = self;
+slf = self;
-	cls = __qClass(slf);
+cls = __qClass(slf);
-	indx = __intVal(index) - 1;
+indx = __intVal(index) - 1;
-	if (cls != String) {
+if (cls != String) {
-	    if (indx < 0) goto badIndex;
+if (indx < 0) goto badIndex;
-	    indx += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+indx += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
-	}
+}
-	if ((unsigned)indx < (unsigned)(__stringSize(slf))) {
+if ((unsigned)indx < (unsigned)(__stringSize(slf))) {
-	    RETURN ( __MKCHARACTER(__stringVal(slf)[indx] & 0xFF) );
+RETURN ( __MKCHARACTER(__stringVal(slf)[indx] & 0xFF) );
-	}
+}
 }
 badIndex: ;
 %}.
 index isInteger ifFalse:[
-	^ self indexNotInteger:index
+^ self indexNotInteger:index
 ].
-^ self subscriptBoundsError:index
+index == super basicSize ifTrue:[
+^ self subscriptBoundsError:index
+].
+^ Character value:(super basicAt:index)
 !
 basicAt:index put:aCharacter
 "store the argument, aCharacter at position index, an Integer.
 Returns aCharacter (sigh).
 REGISTER OBJ cls;
 slf = self;
 if (__isCharacter(aCharacter)) {
-	value = __intVal(_characterVal(aCharacter));
+value = __intVal(_characterVal(aCharacter));
-	if (((unsigned)value <= 0xFF)
+if (((unsigned)value <= 0xFF)
-	 && __isSmallInteger(index)) {
+&& __isSmallInteger(index)) {
-	    cls = __qClass(slf);
+cls = __qClass(slf);
-	    indx = __intVal(index) - 1;
+indx = __intVal(index) - 1;
-	    if (cls != String) {
+if (cls != String) {
-		if (indx < 0) goto badIndex;
+if (indx < 0) goto badIndex;
-		indx += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+indx += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
-	    }
+}
-	    if ((unsigned)indx < (unsigned)(__stringSize(slf))) {
+if ((unsigned)indx < (unsigned)(__stringSize(slf))) {
-		__stringVal(slf)[indx] = value;
+__stringVal(slf)[indx] = value;
-		RETURN ( aCharacter );
+RETURN ( aCharacter );
-	    }
+}
-	}
+}
 }
 badIndex: ;
 %}.
 (aCharacter isMemberOf:Character) ifFalse:[
-	"
+"
-	 tried to store something which is not a character
+tried to store something which is not a character
-	"
+"
-	^ self elementNotCharacter
+^ self elementNotCharacter
 ].
 (aCharacter asciiValue between:1 and:255) ifFalse:[
-	"
+"
-	 tried to store a multibyte character
+tried to store a multibyte character
-	"
+"
-	^ self elementBoundsError
+^ self elementBoundsError
 ].
 "
 invalid index
 "
 index isInteger ifFalse:[
-	^ self indexNotInteger:index
+^ self indexNotInteger:index
 ].
-^ self subscriptBoundsError:index
+index == super basicSize ifTrue:[
+^ self subscriptBoundsError:index
-"Modified: 19.4.1996 / 11:15:48 / cg"
+].
+super basicAt:index put:aCharacter asciiValue.
+^ aCharacter
 !
 basicSize
 "return the number of characters in myself.
 Redefined here to exclude the 0-byte at the end."
 slf = self;
 cls = __qClass(slf);
 if (cls == String) {
-	RETURN ( __MKSMALLINT(__stringSize(slf)) );
+RETURN ( __MKSMALLINT(__stringSize(slf)) );
 }
 RETURN ( __MKSMALLINT(__stringSize(slf)
-			  - __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars))));
+- __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars))));
-%}
+%}.
+^ super basicSize - 1
 !
 size
 "return the number of characters in myself.
 Reimplemented here to avoid the additional size->basicSize send
 REGISTER OBJ cls, slf;
 slf = self;
 cls = __qClass(slf);
 if (cls == String) {
-	RETURN ( __MKSMALLINT(__stringSize(slf)) );
+RETURN ( __MKSMALLINT(__stringSize(slf)) );
 }
 RETURN ( __MKSMALLINT(__stringSize(slf)
-			 - __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars))));
+- __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars))));
-%}
+%}.
+^ self basicSize
 ! !
 !String methodsFor:'binary storage'!
 storeBinaryDefinitionOn:stream manager:manager
 char c;
 #endif
 OBJ cls;
 if (__isCharacter(aCharacter)) {
-	cp = __stringVal(self);
+cp = __stringVal(self);
-	if ((cls = __qClass(self)) != String)
+if ((cls = __qClass(self)) != String)
-	    cp += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+cp += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
 #ifdef FAST_STRCHR
-	cp = (unsigned char *) strchr(cp, __intVal(_characterVal(aCharacter)));
+cp = (unsigned char *) strchr(cp, __intVal(_characterVal(aCharacter)));
-	if (cp) {
+if (cp) {
-	    RETURN ( __MKSMALLINT(cp - __stringVal(self) + 1) );
+RETURN ( __MKSMALLINT(cp - __stringVal(self) + 1) );
-	}
+}
 #else
-	byteValue = __intVal(_characterVal(aCharacter));
+byteValue = __intVal(_characterVal(aCharacter));
-	index = 1;
+index = 1;
-	while (c = *cp++) {
+while (c = *cp++) {
-	    if (c == byteValue) { RETURN ( __MKSMALLINT(index) ); }
+if (c == byteValue) { RETURN ( __MKSMALLINT(index) ); }
-	    index++;
+index++;
-	}
+}
 #endif
-	RETURN ( __MKSMALLINT(0));
+RETURN ( __MKSMALLINT(0));
 }
-%}.
+/* with identity compares, only characters can be in myself */
-"/ with identity compares, only characters can be in myself
+RETURN ( __MKSMALLINT(0));
-^ 0
+%}.
+^ self primitiveFailed
 "
 'hello world' indexOf:(Character space)
 'hello world' indexOf:$A
+'hello world' indexOf:1
 "
 !
 includes:aCharacter
 "return true if the argument, aCharacter is included in the receiver
 REGISTER unsigned char *cp;
 REGISTER unsigned char *matchP;
 OBJ cls;
 if (__isString(aCollection)) {
-	matchP = __stringVal(aCollection);
+matchP = __stringVal(aCollection);
-	cp = __stringVal(self);
+cp = __stringVal(self);
-	if ((cls = __qClass(self)) != String)
+if ((cls = __qClass(self)) != String)
-	    cp += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+cp += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
-	while (*cp) {
+while (*cp) {
-	    if (strchr(matchP, *cp)) {
+if (strchr(matchP, *cp)) {
-		RETURN ( true );
+RETURN ( true );
-	    }
+}
-	    cp++;
+cp++;
-	}
+}
-	RETURN ( false );
+RETURN ( false );
 }
-%}
+%}.
-.
 ^ super includesAny:aCollection
 "
 'hello world' includesAny:'abcd'
 'hello world' includesAny:'xyz'
 int len, index;
 OBJ cls;
 index = __intVal(start);
 if (index <= 0) {
-	index = 1;
+index = 1;
 }
 if ((cls = __qClass(self)) != String)
-	index += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+index += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
 len = __stringSize(self);
 if (index > len) {
-	RETURN ( __MKSMALLINT(0) );
+RETURN ( __MKSMALLINT(0) );
 }
 cp = __stringVal(self) + index - 1;
 while (c = *cp++) {
 #ifndef NON_ASCII       /* i.e. EBCDIC ;-) */
-	if (c > ' ')
+if (c > ' ')
 #endif
-	if ((c != ' ') && (c != '\t') && (c != '\n')
+if ((c != ' ') && (c != '\t') && (c != '\n')
-	 && (c != '\r') && (c != '\f')) {
+&& (c != '\r') && (c != '\f')) {
-	    RETURN ( __MKSMALLINT(cp - __stringVal(self)) );
+RETURN ( __MKSMALLINT(cp - __stringVal(self)) );
-	}
+}
 }
-%}.
+RETURN ( __MKSMALLINT(0) );
-^ 0
+%}.
+^ super indexOfNonSeparatorStartingAt:start
 "
 'hello world' indexOfNonWhiteSpaceStartingAt:3
 'hello world' indexOfNonWhiteSpaceStartingAt:7
 "
 int len, index;
 OBJ cls;
 index = __intVal(start);
 if (index <= 0) {
-	index = 1;
+index = 1;
 }
 if ((cls = __qClass(self)) != String)
-	index += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+index += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
 len = __stringSize(self);
 if (index > len) {
-	RETURN ( __MKSMALLINT(0) );
+RETURN ( __MKSMALLINT(0) );
 }
 cp = __stringVal(self) + index - 1;
 while (c = *cp++) {
 #ifndef NON_ASCII       /* i.e. EBCDIC ;-) */
-	if (c <= ' ')
+if (c <= ' ')
 #endif
-	if ((c == ' ') || (c == '\t') || (c == '\n')
+if ((c == ' ') || (c == '\t') || (c == '\n')
-	 || (c == '\r') || (c == '\f')) {
+|| (c == '\r') || (c == '\f')) {
-	    RETURN ( __MKSMALLINT(cp - __stringVal(self)) );
+RETURN ( __MKSMALLINT(cp - __stringVal(self)) );
-	}
+}
 }
-%}
+RETURN ( __MKSMALLINT(0) );
-.
+%}.
-^ 0
+^ super indexOfSeparatorStartingAt:start
 "
 'hello world' indexOfSeparatorStartingAt:3
 'hello world' indexOfSeparatorStartingAt:7
 "
 asSymbolIfInterned
 "if a symbol with the receivers characters is already known, return it.
 Otherwise, return nil. This can be used to query for an existing
 symbol and is the same as
-	self knownAsSymbol ifTrue:[self asSymbol] ifFalse:[nil]
+self knownAsSymbol ifTrue:[self asSymbol] ifFalse:[nil]
 but slightly faster, since the symbol lookup operation is only
 performed once."
 %{  /* NOCONTEXT */
 OBJ cls;
 int indx;
 cls = __qClass(self);
 if (cls != String) {
-	indx = __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+indx = __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
 } else {
-	indx = 0;
+indx = 0;
 }
 RETURN ( __SYMBOL_OR_NIL(__stringVal(self) + indx));
-%}
+%}.
+self primitiveFailed
 "
 'hello' asSymbolIfInterned
 'fooBarBaz' asSymbolIfInterned
 "
 !
 OBJ cls;
 int indx;
 cls = __qClass(self);
 if (cls != String) {
-	indx = __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+indx = __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
 } else {
-	indx = 0;
+indx = 0;
 }
 RETURN ( __KNOWNASSYMBOL(__stringVal(self) + indx) );
-%}
+%}.
+self primitiveFailed
 "
 'hello' knownAsSymbol
 'fooBarBaz' knownAsSymbol
 "
 REGISTER unsigned char c;
 OBJ cls;
 src = __stringVal(self);
 if ((cls = __qClass(self)) != String)
-	src += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
+src += __OBJS2BYTES__(__intVal(__ClassInstPtr(cls)->c_ninstvars));
 #ifndef NON_ASCII
 # ifdef UINT64
 while (*((UINT64 *)src) == 0x2020202020202020L) {
-	src += 8;
+src += 8;
 }
 # endif /* UINT64 */
 while (*((unsigned *)src) == 0x20202020) {
-	src += 4;
+src += 4;
 }
 #endif /* ascii */
 while (c = *src++) {
-	if (c != ' ') {
+if (c != ' ') {
-	    RETURN ( false );
+RETURN ( false );
-	}
+}
 }
-%}.
+RETURN ( true );
-^ true
+%}.
+^ super isBlank
 !
 levenshteinTo:aString s:substWeight c:caseWeight i:insrtWeight d:deleteWeight
 "parametrized levenshtein. arguments are the costs for
 substitution, case-change, insertion and deletion of a character."
 ! !
 !String class methodsFor:'documentation'!
 version
-^ '$Header: /cvs/stx/stx/libbasic/String.st,v 1.151 2000-03-21 12:56:10 cg Exp $'
+^ '$Header: /cvs/stx/stx/libbasic/String.st,v 1.152 2000-06-23 08:20:15 cg Exp $'
 ! !

changeset 5407	d6729266a95b
parent 5312	ac5719fafe43
child 5556	1056cc5d6ce0