"
COPYRIGHT (c) 1994 by Claus Gittinger
All Rights Reserved
This software is furnished under a license and may be used
only in accordance with the terms of that license and with the
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.
"
ByteArray subclass:#CharacterArray
instanceVariableNames:''
classVariableNames:'PreviousMatch DecoderTables EncoderTables'
poolDictionaries:''
category:'Collections-Text'
!
!CharacterArray class methodsFor:'documentation'!
copyright
"
COPYRIGHT (c) 1994 by Claus Gittinger
All Rights Reserved
This software is furnished under a license and may be used
only in accordance with the terms of that license and with the
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.
"
!
documentation
"
CharacterArray is a superclass for all kinds of Strings (i.e.
(singleByte-)Strings, TwoByteStrings and whatever comes in the future.
This class is abstract, meaning that there are no instances of it.
All this class does is provide common protocol for concrete subclasses.
[author:]
Claus Gittinger
[see also:]
String TwoByteString
StringCollection
"
! !
!CharacterArray class methodsFor:'instance creation'!
basicNew
"return a new empty string"
^ self basicNew:0
!
fromString:aString
"return a copy of the argument, aString"
^ (self basicNew:(aString size)) replaceFrom:1 with:aString
"TwoByteString fromString:'hello'"
!
fromStringCollection:aCollectionOfStrings
"return new string formed by concatenating a copy of the argument, aString"
^ self fromStringCollection:aCollectionOfStrings separatedBy:''
"
String fromStringCollection:#('hello' 'world' 'how' 'about' 'this')
"
"Created: 20.11.1995 / 15:26:59 / cg"
!
fromStringCollection:aCollectionOfStrings separatedBy:aSeparatorString
"return new string formed by concatenating a copy of the argument, aString"
|newString first|
newString := ''.
first := true.
aCollectionOfStrings do:[:s |
first ifFalse:[
newString := newString , aSeparatorString
] ifTrue:[
first := false
].
newString := newString , s
].
^ newString
"
String fromStringCollection:#('hello' 'world' 'how' 'about' 'this') separatedBy:' '
"
"Created: 20.11.1995 / 15:32:17 / cg"
!
new
"return a new empty string"
^ self basicNew:0
! !
!CharacterArray class methodsFor:'cleanup'!
lowSpaceCleanup
"cleanup in low-memory situations"
DecoderTables := EncoderTables := nil
"
CharacterArray lowSpaceCleanup
"
"Created: 22.2.1996 / 16:30:30 / cg"
"Modified: 22.2.1996 / 17:58:05 / cg"
! !
!CharacterArray class methodsFor:'code tables'!
classForEncoding:encodingSymbol
"return a class capable of representing text encoded
by encodingSymbol"
encodingSymbol == #jis ifTrue:[^ JISEncodedString].
encodingSymbol == #big5 ifTrue:[^ BIG5EncodedString].
encodingSymbol == #gb ifTrue:[^ GBEncodedString].
"/
"/ assume its an 8-bit code
"/
^ String
"Modified: 17.4.1996 / 16:18:53 / cg"
!
decoderTableFor:encoding
"return a table to decode from an alien encoding into
the internal (iso8859, ansi) format.
This is an experimental interface - unfinished"
|table unrepresentableCharacterCode x80Table|
DecoderTables notNil ifTrue:[
table := DecoderTables at:encoding ifAbsent:nil.
table notNil ifTrue:[^ table].
] ifFalse:[
DecoderTables := IdentityDictionary new
].
unrepresentableCharacterCode := 16rBF.
"/
"/ setup as identityTranslation
"/
table := (0 to:255) asByteArray.
"/
"/ mac -> iso8859
"/
"/ You will loose the following characters:
"/ - all of them will be replaced by 8F
"/
"/ A0 (dagger)
"/ A5 (dot-dot)
"/ AA (tm - trademark)
"/ AD (not equal)
"/ B0 (infinity)
"/ B3 (less-double-equal <=)
"/ B4 (greater-equal >-)
"/ B6 (math lowercase delta)
"/ B7 (math sum)
"/ B8 (math uppercase pi)
"/ B9 (math lowercase pi)
"/ BA (math integral)
"/ BD (math omega)
"/ C3 (math union)
"/ C4 (ext latin f)
"/ C5 (math almost equal)
"/ C6 (math uppercase delta)
"/ C9 (dot-dot-dot)
"/ CE (OE ligature)
"/ CF (oe ligature)
"/ D1 (hyphen)
"/ D2 (opening top dquote)
"/ D3 (closing top dquote)
"/ D4 (opening top quote)
"/ D5 (closing top quote)
"/ D7 (geometric: )
"/ D9 (uppercase Y diacrit )
"/ DA (slash2 )
"/ DC (single oldStyle opening quote (<) )
"/ DD (single oldStyle closing quote (>) )
"/ DE (? )
"/ DF (? )
"/ E0 (double dagger )
"/ E2 (opening bottom quote )
"/ E3 (opening bottom dquote )
"/ F0 (? )
"/ F5 (latin l )
"/ F6 (diacrit circumflex)
"/ F7 (diacrit tilde)
"/ F8 (diacrit top line)
"/ F9 (diacrit inverse circumflex)
"/ FA (diacrit dot)
"/ FB (diacrit ring)
"/ FC (diacrit cedille left)
"/ FD (diacrit dquote)
"/ FE (diacrit cedille right)
"/ FF (diacrit circumflex2 ?)
"/
encoding == #mac ifTrue:[
x80Table :=
#( 16r00c4 16r00c5 16r00c7 16r00c9 16r00d1 16r00d6 16r00dc 16r00e1 "/ 80
16r00e0 16r00e2 16r00e4 16r00e3 16r00e5 16r00e7 16r00e9 16r00e8
16r00ea 16r00eb 16r00ed 16r00ec 16r00ee 16r00ef 16r00f1 16r00f3 "/ 90
16r00f2 16r00f4 16r00f6 16r00f5 16r00fa 16r00f9 16r00fb 16r00fc
16rFFFF 16r00b0 16r00a2 16r00a3 16r00a7 16rFFFF 16r00b6 16r00df "/ a0
16r00ae 16r00a9 16rFFFF 16r00b4 16r00a8 16rFFFF 16r00c6 16r00d8
16rFFFF 16r00b1 16rFFFF 16rFFFF 16r00a5 16r00b5 16rFFFF 16rFFFF "/ b0
16rFFFF 16rFFFF 16rFFFF 16r00aa 16r00ba 16rFFFF 16r00e6 16r00f8
16r00bf 16r00a1 16r00ac 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r00ab "/ c0
16r00bb 16rFFFF 16r00a0 16r00c1 16r00c3 16r00d5 16rFFFF 16rFFFF
16r00ad 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r00f7 16rFFFF "/ d0
16rFFFF 16rFFFF 16rFFFF 16r00a4 16rFFFF 16rFFFF 16rFFFF 16rFFFF
16rFFFF 16r00b7 16rFFFF 16rFFFF 16rFFFF 16r00c2 16r00ca 16r00c1 "/ e0
16r00cb 16r00c8 16r00cd 16r00ce 16r00cf 16r00cc 16r00d3 16r00d4
16rFFFF 16r00d2 16r00da 16r00db 16r00d9 16rFFFF 16rFFFF 16rFFFF "/ f0
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF
).
].
"/
"/ next -> iso8859
"/
"/ You will loose the following characters:
"/ - all of them will be replaced by 8F
"/
"/ A4 (diacrit /)
"/ A9 (diacrit quote)
"/ AA (top opening dquote)
"/ AC (single oldStyle opening quote (<))
"/ AD (single oldStyle closing quote (>))
"/ AE (?)
"/ AF (?)
"/ B2 (dagger)
"/ B3 (double dagger)
"/ B7 (enter dot)
"/ B8 (opening bottom quote)
"/ B9 (opening bottom dquote)
"/ BA (closing top dquote)
"/ BC (dot-dot-dot)
"/ BD (per-mille)
"/ C0 (subscript 1)
"/ C1 (accent grave)
"/ C2 (accent degu)
"/ C3 (accent circonflex)
"/ C4 (accent tilde)
"/ C5 (accent line)
"/ C6 (accent circonflex reverse)
"/ C7 (accent dot)
"/ C8 (accent dot-dot)
"/ D0 (hline)
"/ E8 (engl. pound upper case L)
"/ EA (OE ligature)
"/ F5 (latin l)
"/ F8 (engl. pound lower case l)
"/ FE (?)
"/ FF (?)
"/
encoding == #next ifTrue:[
unrepresentableCharacterCode := 16rBF.
x80Table :=
#( 16r0080 16r00c0 16r00c1 16r00c2 16r00c3 16r00c4 16r00c5 16r00c7 "/ 80
16r00c8 16r00c9 16r00ca 16r00cb 16r00cc 16r00cd 16r00ce 16r00cf
16r00d0 16r00d1 16r00d2 16r00d3 16r00d4 16r00d5 16r00d6 16r00d9 "/ 90
16r00da 16r00db 16r00dc 16r00dd 16r00de 16r00b5 16r00d7 16r00f7
16r00a9 16r00a1 16r00a2 16r00a3 16rFFFF 16r00a5 16rFFFF 16r00a7 "/ a0
16r00a4 16rFFFF 16rFFFF 16r00ab 16rFFFF 16rFFFF 16rFFFF 16rFFFF
16r00ae 16r00ad 16rFFFF 16rFFFF 16r00b7 16r00a6 16r00b6 16rFFFF "/ b0
16rFFFF 16rFFFF 16rFFFF 16r00bb 16rFFFF 16rFFFF 16r00ac 16r00bf
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF "/ c0
16rFFFF 16r00b2 16r00b0 16r00b8 16r00b3 16rFFFF 16rFFFF 16rFFFF
16rFFFF 16r00b1 16r00bc 16r00bd 16r00be 16r00e0 16r00e1 16r00e2 "/ d0
16r00e3 16r00e4 16r00e5 16r00e7 16r00e8 16r00e9 16r00ea 16r00eb
16r00ec 16r00c6 16r00ed 16r00aa 16r00ee 16r00ef 16r00f0 16r00f1 "/ e0
16rFFFF 16r00d8 16rFFFF 16r00ba 16r00f2 16r00f3 16r00f4 16r00f5
16r00f6 16r00e6 16r00f9 16r00fa 16r00fb 16rFFFF 16r00fc 16r00fd "/ f0
16rFFFF 16r00f8 16rFFFF 16r00df 16r00fe 16r00ff 16rFFFF 16rFFFF
).
].
"/
"/ msdos (codePage 437 [=US]) -> iso8859
"/
"/
"/ You will loose the following characters:
"/ - all of them will be replaced by 8F
"/
"/ 9E (pesetas)
"/ 9F (latin f)
"/ A9 (technical not)
"/ B0 - BF (block graphic)
"/ C0 - CF (block graphic)
"/ D0 - DF (block graphic)
"/ E0 (greek alpha)
"/ E2 - E5 (greek)
"/ E7 - EC (greek & math)
"/ EE - EF (greek & math)
"/ F0 (math)
"/ F2 - F5 (math & technical)
"/ F7 (math)
"/ F9 (center dot)
"/ FB (math sqrt)
"/ FC (super n)
"/ FE (block)
"/
encoding == #msdos437 ifTrue:[
unrepresentableCharacterCode := 16rBF.
x80Table :=
#(
"/ Ccedil uuml eacute acirc auml agrave aring ccedil
16r00c7 16r00fc 16r00e9 16r00e2 16r00e4 16r00e0 16r00e5 16r00e7 "/ 80
"/ ecirc euml egrave iuml icirc igrave Auml Aring
16r00ea 16r00eb 16r00e8 16r00ef 16r00ee 16r00ec 16r00c4 16r00c5
"/ Eacute ae AE ocirc ouml ograve ucirc ugrave
16r00c9 16r00e6 16r00c6 16r00f4 16r00f6 16r00f2 16r00fb 16r00f9 "/ 90
"/ yuml Ouml Uuml cent pound Yen psetas latin-f
16r00ff 16r00d6 16r00dc 16r00a2 16r00a3 16r00a5 16rFFFF 16rFFFF
"/ aacute iacute oacute uacute ntilde Ntilde asup 0sup
16r00e1 16r00ed 16r00f3 16r00fa 16r00f1 16r00d1 16r00aa 16r00b0 "/ a0
"/ iquest lognot 1/2 1/4 iexcla << >>
16r00bf 16rFFFF 16r00ac 16r00bd 16r00bc 16r00a1 16r00ab 16r00bb
"/ block graphics
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF "/ b0
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF
"/ block graphics
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF "/ c0
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF
"/ block graphics
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF "/ d0
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF
"/ alpha beta ... mu
16rFFFF 16r00df 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r00b5 16rFFFF "/ e0
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r00f8 16rFFFF
"/ plusmin gr-eq lt-eq integr integr div
16rFFFF 16r00b1 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r00f7 16rFFFF "/ f0
"/ 0sup dot dot sqr sup2 space
16r00b0 16rFFFF 16r00b7 16rFFFF 16rFFFF 16r00b2 16rFFFF 16r00a0
).
].
"/
"/ msdos (codePage 850 [=latin1]) -> iso8859
"/
(encoding == #msdos850
or:[encoding == #msdos]) ifTrue:[
unrepresentableCharacterCode := 16rBF.
x80Table :=
#(
"/ Ccedil uuml eacute acirc auml agrave aring ccedil
16r00c7 16r00fc 16r00e9 16r00e2 16r00e4 16r00e0 16r00e5 16r00e7 "/ 80
"/ ecirc euml egrave iuml icirc igrave Auml Aring
16r00ea 16r00eb 16r00e8 16r00ef 16r00ee 16r00ec 16r00c4 16r00c5
"/ Eacute ae AE ocirc ouml ograve ucirc ugrave
16r00c9 16r00e6 16r00c6 16r00f4 16r00f6 16r00f2 16r00fb 16r00f9 "/ 90
"/ yuml Ouml Uuml oslash pound Oslash x latin-f
16r00ff 16r00d6 16r00dc 16r00f8 16r00a3 16r00d8 16rFFFF 16rFFFF
"/ aacute iacute oacute uacute ntilde Ntilde asup 0sup
16r00e1 16r00ed 16r00f3 16r00fa 16r00f1 16r00d1 16r00aa 16r00da "/ a0
"/ iquest reg lognot 1/2 1/4 iexcla << >>
16r00bf 16r00ae 16r00ac 16r00bd 16r00bc 16r00a1 16r00ab 16r00bb
"/ Aacute Acirc Agrave
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r00c1 16r00c2 16r00c0 "/ b0
"/ cpyr cent Yen
16r00a9 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r00a2 16r00a5 16rFFFF
"/ atilde Atilde
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r00e3 16r00c3 "/ c0
"/
16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16rFFFF
"/ Ecirc Euml Egrave Iacute Icirc
16rFFFF 16rFFFF 16r00ca 16r00cb 16r00c8 16rFFFF 16r00dc 16r00ce "/ d0
"/ Iuml Igrave
16r00cf 16rFFFF 16rFFFF 16rFFFF 16rFFFF 16r007c 16r00cc 16rFFFF
"/ Oacute sz Ocirc Ograve otilde Otilde mu
16r00d3 16r00df 16r00d4 16r00d2 16rFFFF 16r00f5 16r00b5 16rFFFF "/ e0
"/ Uacute Ucirc Ugrav yacute Yacute '
16rFFFF 16r00da 16r00db 16r00d9 16r00fd 16r00dd 16rFFFF 16r0027
"/ plusmin = 3/4 para sect div "/ f0
16rFFFF 16r00b1 16r003d 16r00be 16r00b6 16r00a7 16r00f7 16rFFFF
"/ sup0 sup3 sup2
16r00b0 16rFFFF 16rFFFF 16rFFFF 16r00b3 16r00b2 16rFFFF 16r00a0
).
].
"/ more encodings needed here ....
"/
"/ x80Table should contain the translation for
"/ 0x80..0xFF characters
"/
x80Table notNil ifTrue:[
x80Table keysAndValuesDo:[:idx :repl |
|ch|
repl == 16rFFFF ifTrue:[
ch := unrepresentableCharacterCode
] ifFalse:[
ch := repl
].
table at:(16r80 + idx) put:ch
].
].
DecoderTables at:encoding put:table.
^ table
"
DecoderTables := nil
"
"Created: 20.10.1995 / 23:04:43 / cg"
"Modified: 31.10.1996 / 16:45:43 / cg"
!
encoderTableFor:encoding
"return a table to encode from the internal (iso8859, ansi) format
into an alien encoding.
This is an experimental interface - unfinished"
|table decoderTable unrepresentableCharacterCode
unrepresentableISOCharacterCode|
EncoderTables notNil ifTrue:[
table := EncoderTables at:encoding ifAbsent:nil.
table notNil ifTrue:[^ table].
] ifFalse:[
EncoderTables := IdentityDictionary new
].
"/
"/ iso8859 -> mac
"/
encoding == #mac ifTrue:[
decoderTable := self decoderTableFor:encoding.
unrepresentableCharacterCode := 16rFF.
unrepresentableISOCharacterCode := 16rBF.
].
"/
"/ iso8859 -> msdos
"/
encoding == #msdos ifTrue:[
decoderTable := self decoderTableFor:encoding.
unrepresentableCharacterCode := 16rFF.
unrepresentableISOCharacterCode := 16rBF.
].
"/ more encodings needed here ....
decoderTable notNil ifTrue:[
table := Array new:256 withAll:unrepresentableCharacterCode.
0 to:16rFF do:[:code |
|isoCode destIdex|
isoCode := decoderTable at:(code+1).
table at:isoCode+1 put:code.
]
].
table isNil ifTrue:[
"/
"/ setup as identityTranslation
"/
table := (0 to:255) asByteArray.
].
EncoderTables at:encoding put:table.
^ table
"
EncoderTables := nil
DecoderTables := nil
CharacterArray decoderTableFor:#mac
CharacterArray encoderTableFor:#mac
"
"Created: 22.2.1996 / 16:17:58 / cg"
"Modified: 23.4.1996 / 20:20:42 / cg"
!
supportedEncodings
"return an array containing symbolic names of supported encodings.
These are internally visible supported ones only"
^ self supportedExternalEncodings at:2.
"/ ^ #(
"/ (
"/ #'iso8859' "/ ansi = ascii
"/ #'iso8859-1' "/ latin1
"/ #'iso8859-2' "/ latin2
"/ #'iso8859-3' "/ latin3
"/ #'iso8859-4' "/ latin4
"/ #'iso8859-5' "/ cyrillic
"/ #'iso8859-6' "/ arabic
"/ #'iso8859-7' "/ greek
"/ #'iso8859-8' "/ hebrew
"/ #'ascii'
"/ #'msdos'
"/ #'mac'
"/ #'next'
"/ #'euc'
"/ #'jis'
"/ #'jis7'
"/ #'gb' "/ china
"/ #'big5' "/ china
"/ #'ksc' "/ korean
"/ )
"/ )
"Created: 22.2.1996 / 16:19:20 / cg"
"Modified: 17.3.1997 / 12:29:06 / cg"
!
supportedExternalEncodings
"return an array of two arrays containing the names of supported
encodings which are supported for external resources (i.e. files).
The first array contains user-readable strings (descriptions),
the second contains the internally used symbolic names.
More than one external name may be mapped onto the same symbolic."
^ #(
(
'iso8859 (ansi)'
'iso8859-1 (latin1)'
'iso8859-2 (latin2)'
'iso8859-3 (latin3)'
'iso8859-4 (latin4)'
'iso8859-5 (cyrillic)'
'iso8859-6 (arabic)'
'iso8859-7 (greek)'
'iso8859-8 (hebrew)'
'msdos US (codepage 437)'
'msdos Latin1 (codepage 850)'
'macintosh (8 bit)'
'NEXT (8 bit)'
'EUC (extended unix code japanese)'
'JIS7 (jis 7bit escape codes japanese)'
'GB (mainland china)'
'BIG5 (taiwan)'
'KSC (korean)'
)
(
#'iso8859'
#'iso8859-1' "/ latin1
#'iso8859-2' "/ latin2
#'iso8859-3' "/ latin3
#'iso8859-4' "/ latin4
#'iso8859-5' "/ cyrillic
#'iso8859-6' "/ arabic
#'iso8859-7' "/ greek
#'iso8859-8' "/ hebrew
#'msdos437'
#'msdos850'
#'mac'
#'next'
#'euc'
#'jis7'
#'gb'
#'big5'
#'ksc' "/ korean
)
)
"Created: 22.4.1996 / 14:39:39 / cg"
"Modified: 17.3.1997 / 12:27:44 / cg"
! !
!CharacterArray class methodsFor:'encoding / decoding'!
decodeFromBIG5:aString
"return a new string containing the characters from aString,
which is interpreted as a BIG5 encoded singleByte string.
The result is a Big5EncodedString (you need a BIG5 font to display that ...)."
|newString|
newString := BIG5EncodedString new:aString size.
^ self decodeFromBIG5_or_GB:aString into:newString.
"Modified: 17.4.1996 / 18:58:29 / cg"
!
decodeFromBIG5_or_GB:aString into:a16BitString
"return a new string containing the characters from aString,
which is interpreted as a BIG5 encoded singleByte string.
The result is a Big5EncodedString (you need a BIG5 font to display that ...)."
|sz "{ Class: SmallInteger }"
dstIdx "{ Class: SmallInteger }"
srcIdx "{ Class: SmallInteger }"
b1 "{ Class: SmallInteger }"
b2 "{ Class: SmallInteger }"
val "{ Class: SmallInteger }"
c|
sz := aString size.
sz ~~ 0 ifTrue:[
dstIdx := 1.
srcIdx := 1.
[true] whileTrue:[
c := aString at:srcIdx.
"/
"/ characters below 16rA1 are left untranslated
"/
b1 := c asciiValue.
b1 >= 16rA1 ifTrue:[
srcIdx := srcIdx + 1.
srcIdx <= sz ifTrue:[
b2 := (aString at:srcIdx) asciiValue.
val := (b1 bitShift:8) bitOr:b2.
c := Character value:val.
]
].
a16BitString at:dstIdx put:c.
dstIdx := dstIdx + 1.
srcIdx := srcIdx + 1.
srcIdx > sz ifTrue:[
^ a16BitString copyFrom:1 to:dstIdx-1.
]
]
].
^ a16BitString
"Created: 17.4.1996 / 16:55:54 / cg"
"Modified: 17.4.1996 / 18:28:31 / cg"
!
decodeFromEUC:aString
"return a new string containing the characters from aString,
which is interpreted as an EUC encoded singleByte string.
There are various JIS encodings around (New-JIS, Old-JIS and NEC-JIS);
this one only understands New-JIS.
The result is a JISEncodedString (you need a JIS font to display that ...)."
|newString
sz "{ Class: SmallInteger }"
dstIdx "{ Class: SmallInteger }"
start "{ Class: SmallInteger }"
b1 "{ Class: SmallInteger }"
b2 "{ Class: SmallInteger }"
val "{ Class: SmallInteger }"
romans c|
romans := JISEncodedString romanJISDecoderTable.
sz := aString size.
newString := JISEncodedString new:sz.
sz ~~ 0 ifTrue:[
dstIdx := 1.
start := 1.
[true] whileTrue:[
c := aString at:start.
b1 := c asciiValue.
b1 < 33 ifTrue:[
newString at:dstIdx put:c.
dstIdx := dstIdx + 1.
] ifFalse:[
b1 < 161 ifTrue:[
newString at:dstIdx put:c.
dstIdx := dstIdx + 1.
"/ b1 < 127 ifTrue:[
"/ newString at:dstIdx put:(Character value:(romans at:(b1 - 32))).
"/ dstIdx := dstIdx + 1.
"/ ] ifFalse:[
"/ self halt:'unsupported codeSet'
"/ ]
] ifFalse:[
start := start + 1.
b1 := b1 - 161 + 33.
b2 := (aString at:start) asciiValue.
b2 := b2 - 161 + 33.
val := (b1 bitShift:8) bitOr:b2.
newString at:dstIdx put:(Character value:val).
dstIdx := dstIdx + 1.
]
].
start := start + 1.
start > sz ifTrue:[
^ newString copyFrom:1 to:dstIdx-1.
]
]
].
^ newString
"simple:
'hello' decodeFromEUC
"
"Created: 17.4.1996 / 16:10:22 / cg"
"Modified: 17.4.1996 / 16:48:30 / cg"
!
decodeFromGB:aString
"return a new string containing the characters from aString,
which is interpreted as a GB encoded singleByte string.
The result is a Big5EncodedString (you need a GB font to display that ...)."
|newString|
newString := GBEncodedString new:aString size.
^ self decodeFromBIG5_or_GB:aString into:newString
"Created: 17.4.1996 / 16:56:33 / cg"
!
decodeFromJIS7:aString
"return a new string containing the aStrings characters,
which are interpreted as a JIS7 encoded singleByte string.
There are various JIS encodings around (New-JIS, Old-JIS and NEC-JIS);
this one understands New-JIS and treats Old-JIS just the same.
The result is a JISEncodedString (you need a JIS font to display that ...)."
|newString
sz "{ Class: SmallInteger }"
dstIdx "{ Class: SmallInteger }"
start "{ Class: SmallInteger }"
stop "{ Class: SmallInteger }"
n1 n2 n3
b1 "{ Class: SmallInteger }"
b2 "{ Class: SmallInteger }"
val "{ Class: SmallInteger }"
singleBytes romans|
romans := JISEncodedString romanJISDecoderTable.
sz := aString size.
newString := JISEncodedString new:sz.
sz ~~ 0 ifTrue:[
dstIdx := 1.
start := 1.
singleBytes := true.
[true] whileTrue:[
"/
"/ scan for next escape"
"/
stop := aString indexOf:(Character esc) startingAt:start.
stop == 0 ifTrue:[
stop := sz + 1.
] ifFalse:[
(stop + 2) > sz ifTrue:[
stop := sz + 1.
]
].
singleBytes ifTrue:[
start to:(stop - 1) do:[:i |
b1 := (aString at:i) asciiValue.
newString at:dstIdx put:(Character value:b1).
"/ (b1 < 33 or:[b1 > 191]) ifTrue:[
"/ newString at:dstIdx put:(Character value:b1)
"/ ] ifFalse:[
"/ newString at:dstIdx put:(Character value:(romans at:(b1 - 32)))
"/ ].
dstIdx := dstIdx + 1.
].
] ifFalse:[
start to:(stop - 2) by:2 do:[:i |
b1 := (aString at:i) asciiValue.
b2 := (aString at:i+1) asciiValue.
val := (b1 bitShift:8) bitOr:b2.
newString at:dstIdx put:(Character value:val).
dstIdx := dstIdx + 1.
]
].
stop > sz ifTrue:[
^ newString copyFrom:1 to:dstIdx - 1.
].
start := stop.
"/
"/ found an escape (at start)
"/ - check for KI (<ESC> '$' 'B') or OLD-JIS-KI (<ESC> '$' '@')
"/ and KO(ASCII) (<ESC> '(' 'B') or KO(ROMAN) (<ESC> '(' 'J')
"/
n1 := aString at:start.
n2 := aString at:(start + 1).
n3 := aString at:(start + 2).
(n2 == $$ and:[n3 == $B]) ifTrue:[
singleBytes := false.
] ifFalse:[
(n2 == $$ and:[n3 == $@]) ifTrue:[
singleBytes := false.
] ifFalse:[
(n2 == $( and:[n3 == $B]) ifTrue:[
singleBytes := true.
] ifFalse:[
(n2 == $( and:[n3 == $J]) ifTrue:[
singleBytes := true.
] ifFalse:[
singleBytes ifTrue:[
newString at:dstIdx put:n1.
newString at:(dstIdx + 1) put:n2.
newString at:(dstIdx + 2) put:n3.
dstIdx := dstIdx + 3.
] ifFalse:[
self halt.
]
]
]
]
].
start := start + 3.
start > sz ifTrue:[
^ newString copyFrom:1 to:dstIdx-1.
]
]
].
^ newString
"simple:
'hello' decodeFromJIS7
ending with a crippled escape:
|s|
s := 'hello' copyWith:Character esc.
s decodeFromJIS7
|s|
s := 'hello' copyWith:Character esc.
s := s copyWith:$A.
s decodeFromJIS7
|s|
s := 'hello' copyWith:Character esc.
s := s copyWith:$$.
s decodeFromJIS7
|s|
s := 'hello' copyWith:Character esc.
s := s copyWith:$$.
s := s copyWith:$A.
s decodeFromJIS7
ending with a KANJI-in, but no more chars:
|s|
s := 'hello' copyWith:Character esc.
s := s copyWith:$$.
s := s copyWith:$B.
s decodeFromJIS7
ending with a KANJI-in, followed by $3 (KO):
|s|
s := 'hello' copyWith:Character esc.
s := s copyWith:$$.
s := s copyWith:$B.
s := s , '$3'.
s decodeFromJIS7
ending with a KANJI-in, followed by $3$l$OF| (KO RE HA NI):
|s|
s := 'hello' copyWith:Character esc.
s := s copyWith:$$.
s := s copyWith:$B.
s := s , '$3$l$OF|'.
s decodeFromJIS7
a KO in between:
|s|
s := 'hello' copyWith:Character esc.
s := s copyWith:$$.
s := s copyWith:$B.
s := s , '$3'.
s := s copyWith:Character esc.
s := s copyWith:$(.
s := s copyWith:$B.
s := s , 'hello'.
s decodeFromJIS7
I dont know what that means ;-):
|s t l|
s := 'kterm ' copyWith:Character esc.
s := s copyWith:$$.
s := s copyWith:$B.
s := s , '$N4A;zC<Kv%(%_%e%l!!<%?'.
s := s copyWith:Character esc.
s := s copyWith:$(.
s := s copyWith:$B.
s := s , ' kterm'.
t := s decodeFromJIS7.
l := Label new.
l label:t.
l font:(Font family:'k14' face:nil style:nil size:nil).
l font:(Font family:'gothic' size:17).
l font:(Font family:'mincho' size:23).
l realize
"
"Modified: 2.3.1996 / 13:39:08 / cg"
"Created: 17.4.1996 / 16:11:57 / cg"
!
encodeIntoBIG5withRomans:aBIG5String
"return a new string with aBIG5Strings characters as BIG5 encoded 16bit string,
The argument must be a BIG5String.
This is used translate 8bit strings and pseudoBIG5 strings
(romans are left untranslated) into a real BIG5 encoding for drawing.
Assuming, that the ascii codePage is at 16r2100.
Doing this allows single-byte roman strings to be displayed in a BIG5 font."
|sz "{ Class: SmallInteger }"
b "{ Class: SmallInteger }"
c romans newString|
romans := BIG5EncodedString romanBIG5DecoderTable.
sz := aBIG5String size.
newString := BIG5EncodedString new:sz.
1 to:sz do:[:index |
c := aBIG5String at:index.
b := c asciiValue.
(b >= 33 and:[b <= (159+32)]) ifTrue:[
"/
"/ a roman character
"/ map to corresponding BIG5 code
"/
c := Character value:(romans at:b - 32).
] ifFalse:[
"/
"/ control or a big5 character
"/
].
newString at:index put:c
].
^ newString
"Created: 17.4.1996 / 17:17:50 / cg"
"Modified: 17.4.1996 / 18:55:40 / cg"
!
encodeIntoEUC:aJISString
"return a new string with aJISStrings characters as EUC encoded 8bit string.
The argument must be a JIS 16 bit character string.
The resulting string is only useful to be stored on some external file,
not for being displayed in an ST/X view."
|sz "{ Class: SmallInteger }"
b1 "{ Class: SmallInteger }"
val "{ Class: SmallInteger }"
romans out|
romans := JISEncodedString romanJISDecoderTable.
sz := aJISString size.
sz == 0 ifTrue:[^ ''].
out := WriteStream on:(String new:(sz * 2)).
1 to:sz do:[:srcIndex |
b1 := (aJISString at:srcIndex) asciiValue.
b1 < 33 ifTrue:[
"/ a control character
out nextPut:(Character value:b1).
] ifFalse:[
"/ check for a roman character
"/ the two numbers below are romanTable min and romanTable max
"/
(b1 between:16r2121 and:16r2573) ifTrue:[
val := romans indexOf:b1.
val ~~ 0 ifTrue:[
out nextPut:(Character value:(val - 1 + 33))
] ifFalse:[
out nextPut:(Character value:(val bitShift:-8) - 33 + 161).
out nextPut:(Character value:(val bitAnd:16rFF) - 33 + 161).
].
] ifFalse:[
out nextPut:(Character value:(val bitShift:-8) - 33 + 161).
out nextPut:(Character value:(val bitAnd:16rFF) - 33 + 161).
]
].
].
^ out contents
"simple:
'hello' decodeFromEUC encodeIntoEUC
"
"Modified: 2.3.1996 / 13:40:12 / cg"
"Created: 17.4.1996 / 16:13:33 / cg"
!
encodeIntoGBwithRomans:aGBString
"return a new string with aGBStrings characters as GB encoded 16bit string,
The argument must be a GBString.
This is used translate 8bit strings and pseudoGB5 strings
(romans are left untranslated) into a real GB encoding for drawing.
Assuming, that the ascii codePage is at 16r2100.
Doing this allows single-byte roman strings to be displayed in a GB font."
|sz "{ Class: SmallInteger }"
b "{ Class: SmallInteger }"
c newString|
sz := aGBString size.
newString := GBEncodedString new:sz.
1 to:sz do:[:index |
c := aGBString at:index.
b := c asciiValue.
(b >= 33 and:[b <= (159+32)]) ifTrue:[
"/
"/ a roman character
"/ map to corresponding GB code
"/
c := Character value:(16r2100 + b).
] ifFalse:[
"/
"/ control or a gb character
"/
].
newString at:index put:c
].
^ newString
"Modified: 17.4.1996 / 17:30:36 / cg"
"Created: 17.4.1996 / 17:31:48 / cg"
!
encodeIntoJIS7:aJISString
"return a new string with aJISStrings characters as JIS7 encoded 7bit string,
The receiver must be a 16bit JIS character string.
The resulting string is only useful to be stored on some external file,
not for being displayed in an ST/X view."
|sz "{ Class:SmallInteger }"
b1 "{ Class:SmallInteger }"
val romans out inSingleByteMode c kanji roman|
inSingleByteMode := true.
kanji := JISEncodedString jis7KanjiEscapeSequence.
roman := JISEncodedString jis7RomanEscapeSequence.
romans := JISEncodedString romanJISDecoderTable.
sz := aJISString size.
sz == 0 ifTrue:[^ ''].
out := WriteStream on:(String new:(sz * 2)).
1 to:sz do:[:srcIndex |
c := aJISString at:srcIndex.
b1 := c asciiValue.
b1 < 33 ifTrue:[
"/ a control character
inSingleByteMode ifFalse:[
out nextPutAll:roman.
inSingleByteMode := true
].
out nextPut:c.
] ifFalse:[
"/ check for a roman character
"/ the two numbers below are romanTable min and romanTable max
"/
(b1 between:16r2121 and:16r2573) ifTrue:[
val := romans indexOf:b1.
val ~~ 0 ifTrue:[
inSingleByteMode ifFalse:[
out nextPutAll:roman.
inSingleByteMode := true
].
out nextPut:(Character value:(val - 1 + 33))
] ifFalse:[
inSingleByteMode ifTrue:[
out nextPutAll:kanji.
inSingleByteMode := false
].
out nextPut:(Character value:(b1 bitShift:-8)).
out nextPut:(Character value:(b1 bitAnd:16rFF)).
].
] ifFalse:[
b1 <= 255 ifTrue:[
"/ mhmh - unrepresentable roman (national chars)
"/ b1 >= 160 ifTrue:[
"/ ('no rep for ' , b1 printString) printNL.
"/ ].
inSingleByteMode ifFalse:[
out nextPutAll:roman.
inSingleByteMode := true
].
out nextPut:c
] ifFalse:[
inSingleByteMode ifTrue:[
out nextPutAll:kanji.
inSingleByteMode := false
].
out nextPut:(Character value:(b1 bitShift:-8)).
out nextPut:(Character value:(b1 bitAnd:16rFF)).
]
]
].
].
inSingleByteMode ifFalse:[
out nextPutAll:roman.
].
^ out contents
"simple:
'hello' decodeFromJIS7 encodeIntoJIS7
"
"Modified: 2.3.1996 / 13:40:31 / cg"
"Created: 17.4.1996 / 16:17:40 / cg"
!
encodeIntoJISwithRomans:aJISString
"return a new string with aJISStrings characters as JIS encoded 16bit string,
The argument must be a JISString.
This is used translate 8bit strings and pseudoJIS strings
(romans are left untranslated) into a real JIS encoding for drawing.
Doing this allows single-byte roman strings to be displayed in a JIS font."
|sz "{ Class: SmallInteger }"
b "{ Class: SmallInteger }"
c romans newString|
romans := JISEncodedString romanJISDecoderTable.
sz := aJISString size.
newString := JISEncodedString new:sz.
1 to:sz do:[:index |
c := aJISString at:index.
b := c asciiValue.
(b >= 33 and:[b <= (159+32)]) ifTrue:[
"/
"/ a roman character
"/ map to corresponding JIS code
"/
c := Character value:(romans at:b - 32).
] ifFalse:[
"/
"/ control or a kanji character
"/
].
newString at:index put:c
].
^ newString
"simple:
'hello' encodeIntoJIS
"
"Created: 17.4.1996 / 17:15:23 / cg"
"Modified: 17.4.1996 / 17:30:59 / cg"
! !
!CharacterArray class methodsFor:'pattern matching'!
matchScan:matchScanArray from:matchStart to:matchStop with:aString from:start to:stop ignoreCase:ignoreCase
"helper for match; return true if the characters from start to stop in
aString are matching the scan in matchScan from matchStart to matchStop.
The matchScan is as created by asMatchScanArray.
This algorithm is not at all the most efficient;
for heavy duty pattern matching, an interface (primitive) to the regex
pattern matching package should be added."
|matchEntry
mStart "{ Class: SmallInteger }"
mStop "{ Class: SmallInteger }"
sStart "{ Class: SmallInteger }"
sStop "{ Class: SmallInteger }"
mSize "{ Class: SmallInteger }"
sSize "{ Class: SmallInteger }"
index "{ Class: SmallInteger }"
quickCheck matchLast nextMatchEntry
checkChar included|
mStart := matchStart.
mStop := matchStop.
sStart := start.
sStop := stop.
[true] whileTrue:[
mSize := mStop - mStart + 1.
sSize := sStop - sStart + 1.
"empty strings match"
(mSize == 0) ifTrue:[^ (sSize == 0)].
matchEntry := matchScanArray at:mStart.
"/ the most common case first:
(sSize ~~ 0
and:[(checkChar := (aString at:sStart)) == matchEntry]) ifTrue:[
"advance by one and continue"
mStart := mStart + 1.
sStart := sStart + 1
] ifFalse:[
(matchEntry == #any) ifTrue:[
"restString empty -> no match"
(sSize == 0) ifTrue:[^ false].
"# matches single character"
((sSize == 1) and:[mSize == 1]) ifTrue:[^ true].
"advance by one and continue"
mStart := mStart + 1.
sStart := sStart + 1
] ifFalse:[
(matchEntry == #anyString) ifTrue:[
"* alone matches anything"
(mSize == 1) ifTrue:[^ true].
"restString empty & matchString not empty -> no match"
(sSize == 0) ifTrue:[^ false].
"
try to avoid some of the recursion by checking last
character and continue with shortened strings if possible
"
quickCheck := false.
(mStop >= mStart) ifTrue:[
matchLast := matchScanArray at:mStop.
(matchLast ~~ #anyString) ifTrue:[
(matchLast == #any) ifTrue:[
quickCheck := true
] ifFalse:[
matchLast == (aString at:sStop) ifTrue:[
quickCheck := true
] ifFalse:[
matchLast isString ifTrue:[
quickCheck := matchLast includes:(aString at:sStop)
]
]
]
]
].
quickCheck ifFalse:[
"/ no quick check;
"/ look for the next character(s)
"/ and trye matching there
"/ (to avoid recursion)
mStart < mStop ifTrue:[
nextMatchEntry := matchScanArray at:mStart+1.
nextMatchEntry isCharacter ifTrue:[
sStart <= sStop ifTrue:[
[true] whileTrue:[
ignoreCase ifFalse:[
index := aString indexOf:nextMatchEntry startingAt:sStart
] ifTrue:[
index := aString findFirst:[:c | c asLowercase == nextMatchEntry asLowercase]
startingAt:sStart.
].
index == 0 ifTrue:[
^ false
].
(self matchScan:matchScanArray
from:(mStart + 1)
to:mStop
with:aString
from:index
to:stop
ignoreCase:ignoreCase
) ifTrue:[
^ true
].
sStart := index + 1.
]
]
]
].
"
no quick check possible;
loop over all possible substrings
"
index := sStart.
[index <= sStop] whileTrue:[
(self matchScan:matchScanArray
from:(mStart + 1)
to:mStop
with:aString
from:index
to:stop
ignoreCase:ignoreCase
) ifTrue:[
^ true
].
index := index + 1
].
^ false
].
"
quickCheck ok, advance from the right
"
mStop := mStop - 1.
sStop := sStop - 1
] ifFalse:[
(matchEntry isString) ifTrue:[
"testString empty -> no match"
(sSize == 0) ifTrue:[^ false].
included := false.
"/ checkChar := aString at:sStart.
included := matchEntry includes:checkChar.
included ifFalse:[
ignoreCase ifTrue:[
checkChar isUppercase ifTrue:[
included := matchEntry includes:checkChar asLowercase.
] ifFalse:[
included := matchEntry includes:checkChar asUppercase.
]
].
].
mStart := mStart + 1.
mSize := mSize - 1.
included ifFalse:[^ false].
((sSize == 1) and:[mSize == 0]) ifTrue:[^ true].
"cut off 1st char and continue"
sStart := sStart + 1
] ifFalse:[
"/ must be single character
"testString empty ?"
(sSize == 0) ifTrue:[^ false].
"first characters equal ?"
"/ checkChar := aString at:sStart.
ignoreCase ifFalse:[^ false].
(checkChar asUppercase ~~ matchEntry asUppercase) ifTrue:[^ false].
"advance and continue"
mStart := mStart + 1.
sStart := sStart + 1
]
]
]
]
].
"
|scanArray s|
scanArray := self matchScanArrayFrom:'*hello'.
s := 'foo bar hello world'.
CharacterArray
matchScan:scanArray
from:1
to:scanArray size
with:s
from:1
to:s size
ignoreCase:false
"
"
|scanArray s|
scanArray := self matchScanArrayFrom:'*hello*'.
s := 'foo bar hello world'.
CharacterArray
matchScan:scanArray
from:1
to:scanArray size
with:s
from:1
to:s size
ignoreCase:false
"
"Modified: 2.4.1997 / 16:35:48 / cg"
!
matchScanArrayFrom:aString
"scan a pattern string and decompose it into a scanArray.
This is processed faster (especially with character ranges), and
can also be reused later. (if the same pattern is to be searched again)"
|coll
idx "{ Class: SmallInteger }"
end c1 c2 matchSet previous|
coll := OrderedCollection new.
idx := 1. end := aString size.
[idx <= end] whileTrue:[
|char this|
char := aString at:idx.
char == $* ifTrue:[
previous ~~ #anyString ifTrue:[
this := #anyString
]
] ifFalse:[
char == $# ifTrue:[
previous ~~ #anyString ifTrue:[
this := #any
]
] ifFalse:[
char == $[ ifTrue:[
matchSet := IdentitySet new.
idx := idx + 1.
idx > end ifTrue:[^ nil].
char := aString at:idx.
c1 := nil.
[char ~~ $]] whileTrue:[
((char == $-) and:[c1 notNil]) ifTrue:[
idx := idx + 1.
idx > end ifTrue:[^ nil].
c2 := aString at:idx.
c1 to:c2 do:[:c | matchSet add:c].
c1 := nil.
idx := idx + 1.
] ifFalse:[
(char ~~ $]) ifTrue:[
matchSet add:char.
c1 := char.
idx := idx + 1
]
].
idx > end ifTrue:[^ nil].
char := aString at:idx
].
this := matchSet asString
] ifFalse:[
char == $\ ifTrue:[
idx := idx + 1.
idx > end ifTrue:[
"/ mhmh - what should we do here ?
this := char
] ifFalse:[
this := aString at:idx.
]
] ifFalse:[
this := char
]
]
]
].
this notNil ifTrue:[coll add:this. previous := this].
idx := idx + 1
].
^ coll asArray
"
String matchScanArrayFrom:'*ute*'
String matchScanArrayFrom:'**ute**'
String matchScanArrayFrom:'*uter'
String matchScanArrayFrom:'\*uter'
String matchScanArrayFrom:'[cC]#mpute[rR]'
String matchScanArrayFrom:'[abcd]*'
String matchScanArrayFrom:'[a-k]*'
String matchScanArrayFrom:'*some*compl*ern*'
String matchScanArrayFrom:'[a-'
String matchScanArrayFrom:'[a-zA-Z]'
String matchScanArrayFrom:'[a-z01234A-Z]'
"
"Modified: 2.4.1997 / 16:20:29 / cg"
! !
!CharacterArray methodsFor:'Compatibility - ST/V'!
byteAt:index put:aByte
"store a byte at given index.
This is an ST/V compatibility method."
"/ (aByte == 0) ifTrue:[
"/ "store a space instead"
"/ ^ super basicAt:index put:(Character space)
"/ ].
^ super at:index put:(Character value:aByte)
"
'hello' copy at:1 put:$H asciiValue; yourself
'hello' copy byteAt:1 put:72; yourself
'hello' copy byteAt:1 put:0; yourself
"
"Modified: 6.5.1996 / 10:35:26 / cg"
!
replChar:oldChar with:newChar
"return a copy of the receiver, with all oldChars replaced
by newChar.
This is an ST/V compatibility method."
^ self copy replaceAll:oldChar by:newChar
"
'12345678901234567890' replChar:$0 with:$*
"
!
replChar:oldChar withString:newString
"return a copy of the receiver, with all oldChars replaced
by newString (i.e. slice in the newString in place of the oldChar).
This is an ST/V compatibility method."
|tmpStream|
tmpStream := WriteStream on:(self class new).
self do:[:element |
element = oldChar ifTrue:[
tmpStream nextPutAll:newString
] ifFalse:[
tmpStream nextPut:element
].
].
^ tmpStream contents
"
'12345678901234567890' replChar:$0 withString:'foo'
'a string with spaces' replChar:$ withString:' foo '
"
!
trimBlanks
"return a copy of the receiver without leading
and trailing spaces.
This is an ST/V compatibility method."
^ self withoutSpaces
"
' spaces at beginning' trimBlanks
'spaces at end ' trimBlanks
' spaces at beginning and end ' trimBlanks
'no spaces' trimBlanks
"
! !
!CharacterArray methodsFor:'Compatibility - V''Age'!
addLineDelimiter
"replace all '\'-characters by line delimiter (cr) - characters.
This has been added for VisualAge compatibility."
^ self withCRs
!
bindWith:aString
"return a copy of the receiver, where a '%1' escape is
replaced by aString.
This has been added for VisualAge compatibility."
^ self expandPlaceholdersWith:(Array with:aString)
"
'do you like %1 ?' bindWith:'smalltalk'
"
!
bindWith:string1 with:string2
"return a copy of the receiver, where a '%1' escape is
replaced by string1 and '%2' is replaced by string2.
This has been added for VisualAge compatibility."
^ self expandPlaceholdersWith:(Array with:string1 with:string2)
"
'do you prefer %1 or rather %2 ?'
bindWith:'smalltalk' with:'c++'
"
!
bindWith:str1 with:str2 with:str3
"return a copy of the receiver, where a '%1', '%2' and '%3' escapes
are replaced by str1, str2 and str3 respectively.
This has been added for VisualAge compatibility."
^ self expandPlaceholdersWith:(Array with:str1 with:str2 with:str3)
"
'do you prefer %1 or rather %2 (not talking about %3) ?'
bindWith:'smalltalk' with:'c++' with:'c'
"
!
bindWith:str1 with:str2 with:str3 with:str4
"return a copy of the receiver, where a '%1', '%2', '%3' and '%4' escapes
are replaced by str1, str2, str3 and str4 respectively.
This has been added for VisualAge compatibility."
^ self expandPlaceholdersWith:(Array with:str1 with:str2 with:str3 with:str4)
"
'do you prefer %1 or rather %2 (not talking about %3 or even %4) ?'
bindWith:'smalltalk' with:'c++' with:'c' with:'assembler'
"
!
bindWith:str1 with:str2 with:str3 with:str4 with:str5
"return a copy of the receiver, where a '%1' .. '%5' escapes
are replaced by str1 .. str5 respectively.
This has been added for VisualAge compatibility."
^ self expandPlaceholdersWith:(Array with:str1 with:str2 with:str3 with:str4 with:str5)
"Created: 31.1.1997 / 16:25:42 / cg"
!
bindWithArguments:anArrayOfStrings
"return a copy of the receiver, where a '%i' escape
is replaced by the coresponding string from the argument array.
'i' may be between 1 and 9 (i.e. a maximum of 9 placeholders is allowed).
This has been added for VisualAge compatibility."
^ self expandPlaceholdersWith:anArrayOfStrings
"
'do you prefer %1 or rather %2 (not talking about %3) ?'
bindWithArguments:#('smalltalk' 'c++' 'c')
"
!
subStrings
"return an array consisting of all words contained in the receiver.
Words are separated by whitespace.
This has been added for VisualAge compatibility."
^ self asCollectionOfWords
"
'hello world, this is smalltalk' subStrings
"
!
subStrings:separatorCharacter
"return an array consisting of all words contained in the receiver.
Words are separated by separatorCharacter.
This has been added for VisualAge compatibility."
^ self asCollectionOfSubstringsSeparatedBy:separatorCharacter
"
'foo:bar:baz:smalltalk' subStrings:$:
"
!
trimSeparators
"return a copy of the receiver without leading and trailing whiteSpace"
^ self withoutSeparators
! !
!CharacterArray methodsFor:'character searching'!
includesMatchCharacters
"return true if the receiver includes any meta characters (i.e. $* or $#)
for match operations; false if not.
Here, do not care for $\ escapes"
^ self includesAny:'*#['
"
'*foo' includesMatchCharacters
'\*foo' includesMatchCharacters
'\*foo' includesUnescapedMatchCharacters
'*foo' includesMatchCharacters
'\\*foo' includesMatchCharacters
'foo*' includesMatchCharacters
'foo\*' includesMatchCharacters
'foo\' includesMatchCharacters
"
"Modified: 2.4.1997 / 18:12:34 / cg"
!
includesUnescapedMatchCharacters
"return true if the receiver really includes any meta characters (i.e. $* or $#)
for match operations; false if not.
Here, care for $\ escapes"
|idx sz|
idx := 1.
sz := self size.
[true] whileTrue:[
idx := self indexOfAny:'*#[\' startingAt:idx.
idx == 0 ifTrue:[^ false].
(self at:idx) == $\ ifFalse:[^ true].
idx := idx + 2.
idx > sz ifTrue:[^ false].
].
"
'*foo' includesUnescapedMatchCharacters
'\*foo' includesUnescapedMatchCharacters
'*foo' includesUnescapedMatchCharacters
'\\*foo' includesUnescapedMatchCharacters
'foo*' includesUnescapedMatchCharacters
'foo\*' includesUnescapedMatchCharacters
'foo\' includesUnescapedMatchCharacters
"
"Modified: 2.4.1997 / 17:08:52 / cg"
"Created: 2.4.1997 / 17:23:26 / cg"
!
indexOfNonSeparatorStartingAt:startIndex
"return the index of the next non-whitespace character"
|start "{ Class: SmallInteger }"
mySize "{ Class: SmallInteger }"|
start := startIndex.
mySize := self size.
start to:mySize do:[:index |
(self at:index) isSeparator ifFalse:[^ index]
].
^ 0
"
' hello world' indexOfNonSeparatorStartingAt:1
"
"
|s index1 index2|
s := ' foo bar baz'.
index1 := s indexOfNonSeparatorStartingAt:1.
index2 := s indexOfSeparatorStartingAt:index1.
s copyFrom:index1 to:index2 - 1
"
!
indexOfSeparator
"return the index of the first whitespace character"
^ self indexOfSeparatorStartingAt:1
"'hello world' indexOfSeparator"
!
indexOfSeparatorStartingAt:startIndex
"return the index of the next whitespace character"
|start "{ Class: SmallInteger }"
mySize "{ Class: SmallInteger }"|
start := startIndex.
mySize := self size.
start to:mySize do:[:index |
(self at:index) isSeparator ifTrue:[^ index]
].
^ 0
"'hello world' indexOfSeparatorStartingAt:3"
!
withMatchEscapes
"return a copy of the receiver with all match characters escaped
by $\ characters (to be usable as a match string).
Return the receiver, if there are none."
|in out c|
in := self readStream.
out := '' writeStream.
[in atEnd] whileFalse:[
c := in next.
('*[#\' includes:c) ifTrue:[
out nextPut:$\.
].
out nextPut:c.
].
^ out contents.
"
'*foo' withMatchEscapes
'\*foo' withMatchEscapes
'*foo' withMatchEscapes
'\\*foo' withMatchEscapes
'foo*' withMatchEscapes
'foo\*' withMatchEscapes
'foo\' withMatchEscapes
'f*o*o' withMatchEscapes
"
"Modified: 2.4.1997 / 18:13:04 / cg"
!
withoutMatchEscapes
"return a copy of the receiver with all $\ removed or
the receiver, if there are none."
|in out c|
in := self readStream.
out := '' writeStream.
[in atEnd] whileFalse:[
c := in next.
c == $\ ifTrue:[
in atEnd ifFalse:[
c := in next.
]
].
out nextPut:c.
].
^ out contents.
"
'*foo' withoutMatchEscapes
'\*foo' withoutMatchEscapes
'*foo' withoutMatchEscapes
'\\*foo' withoutMatchEscapes
'foo*' withoutMatchEscapes
'foo\*' withoutMatchEscapes
'foo\' withoutMatchEscapes
'f\*o\*o' withoutMatchEscapes
"
"Modified: 2.4.1997 / 17:17:37 / cg"
! !
!CharacterArray methodsFor:'comparing'!
< something
"Compare the receiver with the argument and return true if the
receiver is less than the argument. Otherwise return false.
This comparison is based on the elements ascii code -
i.e. upper/lowercase & national characters are NOT treated specially."
^ (something > self)
"Modified: 22.4.1996 / 15:54:54 / cg"
!
<= something
"Compare the receiver with the argument and return true if the
receiver is less than or equal to the argument. Otherwise return false.
This comparison is based on the elements ascii code -
i.e. upper/lowercase & national characters are NOT treated specially."
^ (self > something) not
"Modified: 22.4.1996 / 15:54:56 / cg"
!
= aString
"Compare the receiver with the argument and return true if the
receiver is equal to the argument. Otherwise return false.
This compare does NOT ignore case differences,
therefore 'foo' = 'Foo' will return false.
Since this is incompatible to ST-80 (at least, V2.x) , this may change."
|mySize "{ Class: SmallInteger }"
otherSize |
aString species == self species ifFalse:[^ false].
mySize := self size.
otherSize := aString size.
mySize == otherSize ifFalse:[^ false].
1 to:mySize do:[:index |
(self at:index) = (aString at:index) ifFalse:[^ false].
].
^ true
"
'foo' = 'Foo'
'foo' = 'bar'
'foo' = 'foo'
"
"Modified: 22.4.1996 / 15:53:58 / cg"
!
> aString
"Compare the receiver with the argument and return true if the
receiver is greater than the argument. Otherwise return false.
This comparison is based on the elements ascii code -
i.e. upper/lowercase & upper/lowercase & national characters are NOT treated specially."
|mySize "{ Class: SmallInteger }"
otherSize "{ Class: SmallInteger }"
n "{ Class: SmallInteger }"
c1 c2|
mySize := self size.
otherSize := aString size.
n := mySize min:otherSize.
1 to:n do:[:index |
c1 := self at:index.
c2 := aString at:index.
c1 > c2 ifTrue:[^ true].
c1 < c2 ifTrue:[^ false].
].
^ mySize > otherSize
"Modified: 22.4.1996 / 15:55:00 / cg"
!
>= something
"Compare the receiver with the argument and return true if the
receiver is greater than or equal to the argument.
Otherwise return false.
This comparison is based on the elements ascii code -
i.e. upper/lowercase & national characters are NOT treated specially."
^ (something > self) not
"Modified: 22.4.1996 / 15:55:03 / cg"
!
compareWith:aString
"Compare the receiver with the argument and return 1 if the receiver is
greater, 0 if equal and -1 if less than the argument.
This comparison is based on the elements ascii code -
i.e. upper/lowercase & national characters are NOT treated specially.
'foo' compareWith: 'Foo' will return 1.
while 'foo' sameAs:'Foo' will return true"
|mySize "{ Class: SmallInteger }"
otherSize "{ Class: SmallInteger }"
n "{ Class: SmallInteger }"
c1 c2|
mySize := self size.
otherSize := aString size.
n := mySize min:otherSize.
1 to:n do:[:index |
c1 := self at:index.
c2 := aString at:index.
c1 > c2 ifTrue:[^ 1].
c1 < c2 ifTrue:[^ -1].
].
mySize > otherSize ifTrue:[^ 1].
mySize < otherSize ifTrue:[^ -1].
^ 0
"Modified: 22.4.1996 / 15:56:07 / cg"
!
hash
"return an integer useful as a hash-key"
%{ /* NOCONTEXT */
REGISTER int g, val;
REGISTER unsigned char *cp, *cp0;
int l;
cp = __stringVal(self);
l = __stringSize(self);
if (__qClass(self) != @global(String)) {
int n = __OBJS2BYTES__(__intVal(__ClassInstPtr(__qClass(self))->c_ninstvars));
cp += n;
l -= n;
}
/*
* this is the dragon-book algorithm with a funny start
* value (to give short strings a number above 8192)
*/
val = 12345;
for (cp0 = cp, cp += l - 1; cp >= cp0; cp--) {
val = (val << 5) + (*cp & 0x1F);
if (g = (val & 0x3E000000))
val ^= g >> 25 /* 23 */ /* 25 */;
val &= 0x3FFFFFFF;
}
if (l) {
l |= 1;
val = (val * l) & 0x3FFFFFFF;
}
RETURN ( __MKSMALLINT(val) );
%}
!
sameAs:aString
"Compare the receiver with the argument like =, but ignore
case differences.
Return true or false."
|mySize "{ Class: SmallInteger }"
otherSize c1 c2|
self == aString ifTrue:[^ self].
mySize := self size.
otherSize := aString size.
mySize == otherSize ifFalse:[^ false].
1 to:mySize do:[:index |
c1 := self at:index.
c2 := aString at:index.
c1 == c2 ifFalse:[
c1 asLowercase = c2 asLowercase ifFalse:[^ false].
]
].
^ true
"
'foo' sameAs: 'Foo'
'foo' sameAs: 'bar'
'foo' sameAs: 'foo'
"
"Modified: 22.4.1996 / 15:56:17 / cg"
!
sameCharacters:aString
"count & return the number of characters which are the same
(ignoring case) in the receiver and the argument, aString."
|n "{ Class: SmallInteger }"
c1 c2 cnt|
n := self size.
n := n min:(aString size).
cnt := 0.
1 to:n do:[:index |
c1 := self at:index.
c2 := aString at:index.
((c1 == c2)
or:[c1 asLowercase = c2 asLowercase]) ifTrue:[
cnt := cnt + 1
]
].
^ cnt
"
'foobarbaz' sameCharacters: 'foo'
'foobarbaz' sameCharacters: 'Foo'
'foobarbaz' sameCharacters: 'baz'
"
! !
!CharacterArray methodsFor:'converting'!
asArrayOfSubstrings
"return an array of substrings from the receiver, interpreting
separators (i.e. spaces & newlines) as word-delimiters.
This is a compatibility method - the actual work is done in
asCollectionOfWords."
^ self asCollectionOfWords asArray
"
'1 one two three four 5 five' asArrayOfSubstrings
"
!
asCollectionOfLines
"return a collection containing the lines (separated by cr)
of the receiver. If multiple cr's occur in a row, the result will
contain empty strings."
^ self asCollectionOfSubstringsSeparatedBy:Character cr
"
'1 one\2 two\3 three\4 four\5 five' withCRs asCollectionOfLines
'1 one\\\\2 two\3 three' withCRs asCollectionOfLines
"
!
asCollectionOfSubstringsSeparatedBy:aCharacter
"return a collection containing the lines (separated by aCharacter)
of the receiver. If aCharacter occurs multiple times in a row,
the result will contain empty strings."
|lines myClass
numberOfLines "{ Class:SmallInteger }"
startIndex "{ Class:SmallInteger }"
stopIndex "{ Class:SmallInteger }" |
"
count first, to avoid regrowing of the OC
"
numberOfLines := (self occurrencesOf:aCharacter) + 1.
lines := OrderedCollection new:numberOfLines.
myClass := self species.
startIndex := 1.
1 to:numberOfLines do:[:lineNr |
stopIndex := self indexOf:aCharacter startingAt:startIndex.
stopIndex == 0 ifTrue:[
stopIndex := self size
] ifFalse: [
stopIndex := stopIndex - 1.
].
(stopIndex < startIndex) ifTrue: [
lines add:(myClass new:0)
] ifFalse: [
lines add:(self copyFrom:startIndex to:stopIndex)
].
startIndex := stopIndex + 2
].
^ lines
"
'1 one:2 two:3 three:4 four:5 five' withCRs asCollectionOfSubstringsSeparatedBy:$:
'1 one 2 two 3 three 4 four 5 five' withCRs asCollectionOfSubstringsSeparatedBy:Character space
"
!
asCollectionOfSubstringsSeparatedByAny:aCollectionOfSeparators
"return a collection containing the words (separated by any character
from aCollectionOfSeparators) of the receiver.
This allows breaking up strings using any character as separator."
|words
start "{ Class:SmallInteger }"
stop "{ Class:SmallInteger }"
mySize "{ Class:SmallInteger }"|
words := OrderedCollection new.
start := 1.
mySize := self size.
[start <= mySize] whileTrue:[
"skip multiple separators"
[aCollectionOfSeparators includes:(self at:start)] whileTrue:[
start := start + 1 .
start > mySize ifTrue:[
^ words
].
].
stop := self indexOfAny:aCollectionOfSeparators startingAt:start.
stop == 0 ifTrue:[
words add:(self copyFrom:start to:mySize).
^ words
].
words add:(self copyFrom:start to:(stop - 1)).
start := stop
].
^ words
"
'hello:world:isnt:this nice' asCollectionOfSubstringsSeparatedByAny:#($:)
'hello:world:isnt:this nice' asCollectionOfSubstringsSeparatedByAny:':'
'hello:world:isnt:this nice' asCollectionOfSubstringsSeparatedByAny:(Array with:$: with:Character space)
'hello:world:isnt:this nice' asCollectionOfSubstringsSeparatedByAny:': '
'h1e2l3l4o' asCollectionOfSubstringsSeparatedByAny:($1 to: $9)
"
!
asCollectionOfWords
"return a collection containing the words (separated by whitespace)
of the receiver. Multiple occurences of whitespace characters will
be treated like one - i.e. whitespace is skipped."
|words
start "{ Class:SmallInteger }"
stop "{ Class:SmallInteger }"
mySize "{ Class:SmallInteger }"|
words := OrderedCollection new.
start := 1.
mySize := self size.
[start <= mySize] whileTrue:[
start := self indexOfNonSeparatorStartingAt:start.
start == 0 ifTrue:[
^ words
].
stop := self indexOfSeparatorStartingAt:start.
stop == 0 ifTrue:[
words add:(self copyFrom:start to:mySize).
^ words
].
words add:(self copyFrom:start to:(stop - 1)).
start := stop
].
^ words
"
'hello world isnt this nice' asCollectionOfWords
' hello world isnt this nice ' asCollectionOfWords
'hello' asCollectionOfWords
'' asCollectionOfWords
' ' asCollectionOfWords
"
!
asComposedText
"ST-80 compatibility
- ST/X does not (as today) support composedTexts."
^ ComposedText fromString:self
"Modified: 27.4.1996 / 13:30:30 / cg"
!
asFilename
"return a Filename with pathname taken from the receiver"
^ Filename named:self string
"Modified: 20.5.1996 / 09:38:15 / cg"
!
asFloat
"read a float number from the receiver.
Notice, that errors may occur during the read, so you better
setup some signal handler when using this method."
^ (Number readFromString:self) asFloat
"
'0.123' asFloat
'12345' asFloat
'(1/5)' asFloat
Object errorSignal handle:[:ex | ex returnWith:0] do:['foo' asFloat]
"
!
asInteger
"read an integer from the receiver.
Notice, that errors may occur during the read, so you better
setup some signal handler when using this method."
^ Integer readFromString:self
"
'12345678901234567890' asInteger
'-1234' asInteger
'0.123' asInteger <- reader stops at ., returning 0 here
'0.123' asNumber <- returns what you expect
Object errorSignal handle:[:ex | ex returnWith:0] do:['foo' asInteger]
"
!
asLowercase
"return a copy of myself in lowercase letters"
|newStr
mySize "{ Class: SmallInteger }" |
mySize := self size.
newStr := self species new:mySize.
1 to:mySize do:[:i |
newStr at:i put:(self at:i) asLowercase
].
^ newStr
"
'HelloWorld' asLowercase
'HelloWorld' asLowercaseFirst
"
!
asLowercaseFirst
"return a copy of myself where the first character is
converted to lowercase."
|newString sz|
sz := self size.
newString := self copyFrom:1 to:sz.
sz > 0 ifTrue:[
newString at:1 put:(newString at:1) asLowercase
].
^ newString
"
'HelloWorld' asLowercase
'HelloWorld' asLowercaseFirst
"
!
asNumber
"read a number from the receiver.
Notice, that (in contrast to ST-80) errors may occur during the read,
so you better setup some signal handler when using this method.
This may change if ANSI specifies it."
"/ ST-80 behavior:
"/ ^ Number readFromString:self onError:0
^ Number readFromString:self
"
'123' asNumber
'123.567' asNumber
'(5/6)' asNumber
'foo' asNumber
Object errorSignal handle:[:ex | ex returnWith:0] do:['foo' asNumber]
"
!
asNumberFromFormatString:ignored
"read a number from the receiver, ignoring any nonDigit characters.
This is typically used to convert from strings which include
dollar-signs or millenium digits. However, this method also ignores
the decimal point (if any) and therefore should be used with care."
|tempString|
tempString := self collect:[:char | char isDigit].
^ Number readFromString:tempString onError:0
"
'USD 123' asNumberFromFormatString:'foo'
'DM 123' asNumberFromFormatString:'foo'
'123' asNumberFromFormatString:'foo'
'123.567' asNumberFromFormatString:'foo'
'(5/6)' asNumberFromFormatString:'foo'
'foo' asNumberFromFormatString:'foo'
"
!
asSingleByteString
"return the receiver converted to a 'normal' string"
^ String fromString:self
!
asString
"return myself - I am a string"
^ self
!
asStringCollection
"return a collection of lines from myself."
^ StringCollection from:self "string"
"Modified: 13.5.1996 / 20:36:59 / cg"
!
asSymbol
"return a unique symbol with the name taken from my characters.
The receiver must be a singleByte-String; twoByteSymbols are
(currently) not allowed."
^ self string asSymbol
"Created: 22.5.1996 / 15:56:11 / cg"
!
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]
but slightly faster, since the symbol lookup operation is only
performed once.
The receiver must be a singleByte-String; twoByteSymbols are
(currently) not allowed."
^ self string asSymbolIfInterned
"Created: 22.5.1996 / 16:37:04 / cg"
!
asText
"return a Text-object (collection of lines) from myself."
^ Text fromString:self
"Created: 12.5.1996 / 10:41:14 / cg"
!
asTwoByteString
"return the receiver converted to a two-byte string"
^ TwoByteString fromString:self
!
asUppercase
"return a copy of myself in uppercase letters"
|newStr
mySize "{ Class: SmallInteger }" |
mySize := self size.
newStr := self species new:mySize.
1 to:mySize do:[:i |
newStr at:i put:(self at:i) asUppercase
].
^ newStr
"
'helloWorld' asUppercase
'helloWorld' asUppercaseFirst
"
!
asUppercaseFirst
"return a copy of myself where the first character is
converted to uppercase."
|newString sz|
sz := self size.
newString := self copyFrom:1 to:sz.
sz > 0 ifTrue:[
newString at:1 put:(newString at:1) asUppercase
].
^ newString
"
'helloWorld' asUppercase
'helloWorld' asUppercaseFirst
'HelloWorld' asUppercaseFirst
"
!
string
"return the receiver - for ST-80 compatibility"
^ self
"Modified: 27.4.1996 / 13:29:30 / cg"
!
tokensBasedOn:aCharacter
"this is an ST-80 alias for the ST/X method
asCollectionOfSubstringsSeparatedBy:"
^ self asCollectionOfSubstringsSeparatedBy:aCharacter
"
'hello:world:isnt:this nice' tokensBasedOn:$:
'foo,bar,baz' tokensBasedOn:$,
'/etc/passwd' asFilename readStream nextLine tokensBasedOn:$:
"
! !
!CharacterArray methodsFor:'copying'!
, aStringOrCharacter
"redefined to allow characters to be appended.
This is nonStandard, but convenient"
aStringOrCharacter isCharacter ifTrue:[
^ self , aStringOrCharacter asString
].
aStringOrCharacter isText ifTrue:[
^ aStringOrCharacter concatenateFromString:self
].
^ super , aStringOrCharacter
"
'hello' , $1
'hello' , '1'
'hello' , (' world' asText allBold)
Transcript showCR:
(Text string:'hello' emphasis:#italic) , (Text string:' world' emphasis:#bold)
"
"Modified: 18.5.1996 / 12:29:30 / cg"
!
concatenate:string1 and:string2
"return the concatenation of myself and the arguments, string1 and string2.
This is equivalent to self , string1 , string2
- generated by compiler when such a construct is detected and the receiver
is known to be a string."
^ self , string1 , string2
!
concatenate:string1 and:string2 and:string3
"return the concatenation of myself and the string arguments.
This is equivalent to self , string1 , string2 , string3
- generated by compiler when such a construct is detected and the receiver
is known to be a string."
^ self , string1 , string2 , string3
! !
!CharacterArray methodsFor:'displaying'!
displayOn:aGc x:x y:y
"display the receiver in a graphicsContext - this method allows
strings to be used like DisplayObjects."
^ aGc displayString:self string x:x y:y.
"Modified: 11.5.1996 / 14:42:48 / cg"
!
displayOpaqueOn:aGc x:x y:y
"display the receiver in a graphicsContext - this method allows
strings to be used like DisplayObjects."
^ aGc displayOpaqueString:self string x:x y:y.
"Modified: 11.5.1996 / 14:42:48 / cg"
"Created: 12.5.1996 / 12:28:40 / cg"
! !
!CharacterArray methodsFor:'emphasis'!
emphasis
"return the emphasis.
Since characterArrays do not hold any emphasis information,
nil (no emphasis) is returned here."
^ RunArray new:self size withAll:nil
"Created: 14.5.1996 / 13:58:58 / cg"
!
emphasisAt:characterIndex
"return the emphasis at some index.
Since characterArrays do not hold any emphasis information,
nil (no emphasis) is returned here."
^ nil
"Created: 11.5.1996 / 14:13:27 / cg"
!
emphasisCollection
"return the emphasis.
Since characterArrays do not hold any emphasis information,
nil (no emphasis) is returned here."
^ RunArray new:(self size)
"Created: 14.5.1996 / 13:58:58 / cg"
"Modified: 14.5.1996 / 15:02:29 / cg"
! !
!CharacterArray methodsFor:'encoding/decoding'!
decodeFrom:encodingSymbol
"given the receiver encoded as described by encodingSymbol,
convert it into internal ST/X encoding and return a corresponding CharacterArray.
Currently, only a few encodings are supported
(and those are untested/incomplete):
#euc
#jis7
#mac
#msdos
"
|newString|
encodingSymbol == #jis7 ifTrue:[
^ self class decodeFromJIS7:self
].
encodingSymbol == #euc ifTrue:[
^ self class decodeFromEUC:self
].
encodingSymbol == #big5 ifTrue:[
^ self class decodeFromBIG5:self
].
encodingSymbol == #gb ifTrue:[
^ self class decodeFromGB:self
].
newString := self class new:self size.
newString replaceFrom:self decode:encodingSymbol.
newString = self ifTrue:[^ self].
^ newString
"Created: 22.2.1996 / 15:06:49 / cg"
"Modified: 17.4.1996 / 18:25:17 / cg"
!
encodeInto:encodingSymbol
"given the receiver in internal ST/X (ISO8859) encoding,
convert it into the format as described by encodingSymbol
and return a corresponding CharacterArray.
Currently, only a few encodings are supported
(and those are untested/incomplete):
#euc
#jis7
#mac
#msdos
"
|newString|
"/ self encoding storeString print. ' -> ' print. encodingSymbol storeString printNL.
encodingSymbol isNil ifTrue:[^ self].
encodingSymbol == self encoding ifTrue:[^ self].
encodingSymbol == #jis7 ifTrue:[
^ self class encodeIntoJIS7:self
].
(encodingSymbol startsWith:'jis') ifTrue:[
^ self class encodeIntoJISwithRomans:self
].
encodingSymbol == #euc ifTrue:[
^ self class encodeIntoEUC:self
].
encodingSymbol == #big5 ifTrue:[
^ self class encodeIntoBIG5:self
].
(encodingSymbol startsWith:'big5') ifTrue:[
^ self class encodeIntoBIG5withRomans:self
].
encodingSymbol == #gb ifTrue:[
^ self class encodeIntoGB:self
].
(encodingSymbol startsWith:'gb') ifTrue:[
^ self class encodeIntoGBwithRomans:self
].
newString := self class new:self size.
newString replaceFrom:self encode:encodingSymbol.
newString = self ifTrue:[^ self].
^ newString
"Created: 22.2.1996 / 15:07:31 / cg"
"Modified: 17.4.1996 / 18:59:12 / cg"
!
replaceFrom:aString decode:encoding
"this is an experimental interface - unfinished"
^ self replaceFrom:aString translateBy:(self class decoderTableFor:encoding)
"Created: 20.10.1995 / 23:00:09 / cg"
"Modified: 22.2.1996 / 16:08:26 / cg"
!
replaceFrom:aString encode:encoding
"this is an experimental interface - unfinished"
^ self replaceFrom:aString translateBy:(self class encoderTableFor:encoding)
"Modified: 20.10.1995 / 23:08:16 / cg"
"Created: 22.2.1996 / 16:08:34 / cg"
!
replaceFrom:aString translateBy:encodingTable
"replace the receivers characters by translations coming from
an encoding table."
|mySize "{ Class: SmallInteger }"
maxCode "{ Class: SmallInteger }"
char
oldCode "{ Class: SmallInteger }"
newCode|
self replaceFrom:1 with:aString.
encodingTable isNil ifTrue:[^ self].
maxCode := encodingTable size.
mySize := self size.
1 to:mySize do:[:index |
char := aString at:index.
oldCode := char asciiValue.
oldCode >= maxCode ifFalse:[
newCode := encodingTable at:(oldCode + 1).
newCode ~~ oldCode ifTrue:[
self at:index put:(Character value:newCode)
]
]
].
"Created: 22.2.1996 / 16:07:26 / cg"
"Modified: 2.4.1997 / 21:35:41 / cg"
! !
!CharacterArray methodsFor:'padded copying'!
centerPaddedTo:newSize
"return a new string consisting of the receivers characters,
plus spaces up to length and center the receivers characters in
the resulting string.
If the receivers size is equal or greater than the length argument,
the original receiver is returned unchanged."
^ self centerPaddedTo:newSize with:(Character space)
"
'foo' centerPaddedTo:10
123 printString centerPaddedTo:10
"
"Created: 25.11.1995 / 10:53:57 / cg"
!
centerPaddedTo:size with:padCharacter
"return a new string of length size, which contains the receiver
centered (i.e. padded on both sides).
Characters are filled with padCharacter.
If the receivers size is equal or greater than the length argument,
the original receiver is returned unchanged."
|len s|
len := self size.
(len < size) ifTrue:[
s := self species new:size withAll:padCharacter.
s replaceFrom:(size - len) // 2 + 1 with:self.
^ s
]
"
'foo' centerPaddedTo:11 with:$.
'fooBar' centerPaddedTo:5 with:$.
123 printString centerPaddedTo:10 with:$.
(' ' , 123 printString) centerPaddedTo:10 with:$.
(Float pi printString) centerPaddedTo:15 with:(Character space)
(Float pi printString) centerPaddedTo:15 with:$-
(' ' , Float pi class name) centerPaddedTo:15 with:$.
"
!
decimalPaddedTo:size and:afterPeriod at:decimalCharacter
"return a new string of overall length size, which contains the receiver
aligned at the decimal-period column and afterPeriod characters to the right
of the period. The periodCharacter is passed as arguments (allowing for US and European formats
to be padded).
If the receivers size is equal or greater than the length argument,
the original receiver is returned unchanged.
(sounds complicated ? -> see examples below)."
^ self
decimalPaddedTo:size
and:afterPeriod
at:decimalCharacter
withLeft:(Character space)
right:$0
"
'123' decimalPaddedTo:10 and:3 at:$. -> ' 123 '
'123' decimalPaddedTo:10 and:3 at:$. -> ' 123.000'
'123.' decimalPaddedTo:10 and:3 at:$. -> ' 123.000'
'123.1' decimalPaddedTo:10 and:3 at:$. -> ' 123.100'
'123.1' decimalPaddedTo:10 and:3 at:$. -> ' 123.1 '
'123.123' decimalPaddedTo:10 and:3 at:$. -> ' 123.123'
"
"Created: 23.12.1995 / 13:11:52 / cg"
!
decimalPaddedTo:size and:afterPeriod at:decimalCharacter withLeft:leftPadChar right:rightPadChar
"return a new string of overall length size, which contains the receiver
aligned at the decimal-period column and afterPeriod characters to the right
of the period.
Characters on the left are filled with leftPadChar.
If rightPadChar is nil, characters on the right are filled with leftPadCharacter too;
otherwise, if missing, a decimal point is added and right characters filled with this.
If the receivers size is equal or greater than the length argument,
the original receiver is returned unchanged.
(sounds complicated ? -> see examples below)."
|s idx n|
idx := self indexOf:decimalCharacter.
idx == 0 ifTrue:[
"/
"/ no decimal point found; adjust string to the left of the period column
"/
rightPadChar isNil ifTrue:[
s := self , (self species new:afterPeriod + 1 withAll:leftPadChar)
] ifFalse:[
s:= self , decimalCharacter asString , (self species new:afterPeriod withAll:rightPadChar).
].
] ifFalse:[
"/ the number of after-decimalPoint characters
n := self size - idx.
rightPadChar isNil ifTrue:[
s := self , (self species new:afterPeriod - n withAll:leftPadChar).
] ifFalse:[
s := self , (self species new:afterPeriod - n withAll:rightPadChar).
].
].
^ s leftPaddedTo:size with:leftPadChar
"
'123' decimalPaddedTo:10 and:3 at:$. withLeft:(Character space) right:nil -> ' 123 '
'123' decimalPaddedTo:10 and:3 at:$. withLeft:(Character space) right:$0 -> ' 123.000'
'123.' decimalPaddedTo:10 and:3 at:$. withLeft:(Character space) right:$0 -> ' 123.000'
'123.1' decimalPaddedTo:10 and:3 at:$. withLeft:(Character space) right:$0 -> ' 123.100'
'123.1' decimalPaddedTo:10 and:3 at:$. withLeft:(Character space) right:nil -> ' 123.1 '
'123.123' decimalPaddedTo:10 and:3 at:$. withLeft:(Character space) right:$0 -> ' 123.123'
"
"Modified: 23.12.1995 / 13:08:18 / cg"
!
leftPaddedTo:size
"return a new string of length size, which contains the receiver
right-adjusted (i.e. padded on the left).
Characters on the left are filled with spaces.
If the receivers size is equal or greater than the length argument,
the original receiver is returned unchanged."
^ self leftPaddedTo:size with:(Character space)
"
'foo' leftPaddedTo:10
'fooBar' leftPaddedTo:5
123 printString leftPaddedTo:10
"
!
leftPaddedTo:size with:padCharacter
"return a new string of length size, which contains the receiver
right-adjusted (i.e. padded on the left).
Characters on the left are filled with padCharacter.
If the receivers size is equal or greater than the length argument,
the original receiver is returned unchanged."
|len s|
len := self size.
(len < size) ifTrue:[
s := self species new:size withAll:padCharacter.
s replaceFrom:(size - len + 1) with:self.
^ s
]
"
'foo' leftPaddedTo:10 with:$.
'fooBar' leftPaddedTo:5 with:$.
123 printString leftPaddedTo:10 with:$.
(' ' , 123 printString) leftPaddedTo:10 with:$.
(Float pi printString) leftPaddedTo:15 with:(Character space)
(Float pi printString) leftPaddedTo:15 with:$-
(' ' , Float pi class name) leftPaddedTo:15 with:$.
"
!
paddedTo:newSize
"return a new string consisting of the receivers characters,
plus spaces up to length.
If the receivers size is equal or greater than the length argument,
the original receiver is returned unchanged."
^ self paddedTo:newSize with:(Character space)
"
'foo' paddedTo:10
123 printString paddedTo:10
"
!
paddedTo:newSize with:padCharacter
"return a new string consisting of the receivers characters,
plus pad characters up to length.
If the receivers size is equal or greater than the length argument,
the original receiver is returned unchanged."
|s len|
len := self size.
len < newSize ifTrue:[
s := self species new:newSize withAll:padCharacter.
s replaceFrom:1 to:len with:self.
^ s
]
"
'foo' paddedTo:10 with:$.
123 printString paddedTo:10 with:$*
(Float pi printString) paddedTo:15 with:(Character space)
(Float pi printString) paddedTo:15 with:$-
(Float pi class name , ' ') paddedTo:15 with:$.
"
! !
!CharacterArray methodsFor:'pattern matching'!
compoundMatch:aString
"like match, but the receiver may be a compount match pattern,
consisting of multiple simple patterns, separated by semicolons.
This is usable with fileName pattern fields."
|matchers|
matchers := self asCollectionOfSubstringsSeparatedBy:$;.
matchers do:[:aPattern |
(aPattern match:aString) ifTrue:[^ true].
].
^ false.
"
'f*' match:'foo'
'b*' match:'foo'
'f*;b*' match:'foo'
'f*;b*' match:'bar'
'f*;b*' compoundMatch:'foo'
'f*;b*' compoundMatch:'bar'
"
"Modified: 15.4.1997 / 15:50:33 / cg"
!
findMatchString:matchString
"like findString/indexOfSubCollection, but allowing match patterns.
find matchstring; if found, return the index;
if not found, return 0."
^ self findMatchString:matchString startingAt:1 ignoreCase:false ifAbsent:0
!
findMatchString:matchString startingAt:index
"like findString, but allowing match patterns.
find matchstring, starting at index. if found, return the index;
if not found, return 0."
^ self findMatchString:matchString startingAt:index ignoreCase:false ifAbsent:0
!
findMatchString:matchString startingAt:index ignoreCase:ignoreCase ifAbsent:exceptionBlock
"like findString, but allowing match patterns.
find matchstring, starting at index. if found, return the index;
if not found, return the result of evaluating exceptionBlock.
This is a q&d hack - not very efficient"
|firstChar
startIndex "{ Class: SmallInteger }"
matchSize "{ Class: SmallInteger }"
mySize "{ Class: SmallInteger }"
realMatchString|
matchSize := matchString size.
matchSize == 0 ifTrue:[^ index]. "empty string matches"
realMatchString := matchString.
(realMatchString endsWith:$*) ifFalse:[
realMatchString := realMatchString , '*'.
matchSize := matchSize + 1
].
mySize := self size.
firstChar := realMatchString at:1.
firstChar == $\ ifTrue:[
firstChar := realMatchString at:2.
].
firstChar asString includesMatchCharacters ifTrue:[
index to:mySize do:[:col |
(realMatchString match:self from:col to:mySize ignoreCase:ignoreCase)
ifTrue:[^ col]
].
^ exceptionBlock value.
].
startIndex := self indexOf:firstChar startingAt:index.
[startIndex == 0] whileFalse:[
(realMatchString match:self from:startIndex to:mySize ignoreCase:ignoreCase)
ifTrue:[^ startIndex].
startIndex := self indexOf:firstChar startingAt:(startIndex + 1)
].
^ exceptionBlock value
"
'one two three four' findMatchString:'o[nu]'
'one two three four' findMatchString:'o[nu]' startingAt:3
"
"Modified: 2.4.1997 / 17:26:39 / cg"
!
includesMatchString:matchString
"like includesString, but allowing match patterns.
find matchstring; if found, return true, otherwise return false"
^ (self findMatchString:matchString) ~~ 0
"
'hello world' includesMatchString:'h*'
'hello world' includesMatchString:'h[aeiou]llo'
'hello world' includesMatchString:'wor*'
'hello world' includesMatchString:'woR*'
"
!
match:aString
"return true if aString matches self, where self may contain meta-match
characters $* (to match any string) or $# (to match any character).
or [...] to match a set of characters.
Lower/uppercase are considered different.
NOTICE: match-meta character interpretation is like in unix-matching,
NOT the ST-80 meaning."
^ self match:aString from:1 to:aString size ignoreCase:false
"
'\*f*' match:'f'
'\*f*' match:'*f'
'*\*f*' match:'*f'
'*f*' match:'*f'
'*ute*' match:'computer'
'*uter' match:'computer'
'uter*' match:'computer'
'*ute*' match:''
'[abcd]*' match:'computer'
'[abcd]*' match:'komputer'
'*some*compl*ern*' match:'this is some more complicated pattern match'
'*some*compl*ern*' match:'this is another complicated pattern match'
"
"Modified: 1.4.1997 / 18:38:02 / cg"
!
match:aString from:start to:stop ignoreCase:ignoreCase
"return true if part of aString matches myself,
where self may contain meta-match
characters $* (to match any string) or $# (to match any character)
or [...] to match a set of characters.
If ignoreCase is true, lower/uppercase are considered the same.
NOTICE: match-meta character interpretation is like in unix-matching,
NOT the ST-80 meaning."
|matchScanArray|
"
keep the matchScanArray from the most recent match -
avoids parsing the pattern over-and over if multiple searches
are done with the same pattern.
"
(PreviousMatch notNil
and:[PreviousMatch key = self]) ifTrue:[
matchScanArray := PreviousMatch value
] ifFalse:[
matchScanArray := self class matchScanArrayFrom:self.
matchScanArray isNil ifTrue:[
'CharacterArray [info]: invalid matchpattern:' infoPrint. self infoPrintCR.
^ false
].
PreviousMatch := self -> matchScanArray.
].
^ self class
matchScan:matchScanArray
from:1 to:matchScanArray size
with:aString
from:start to:stop
ignoreCase:ignoreCase
"
'*ute*' match:'12345COMPUTER' from:1 to:5 ignoreCase:true
'*ute*' match:'12345COMPUTER' from:6 to:13 ignoreCase:true
"
"Modified: 10.1.1997 / 15:36:02 / cg"
!
match:aString ignoreCase:ignoreCase
"return true if aString matches self, where self may contain meta-match
characters $* (to match any string) or $# (to match any character)
or [...] to match a set of characters.
If ignoreCase is true, lower/uppercase are considered the same.
NOTICE: match-meta character interpretation is like in unix-matching,
NOT the ST-80 meaning."
^ self match:aString from:1 to:aString size ignoreCase:ignoreCase
"
'*ute*' match:'COMPUTER' ignoreCase:true
'*uter' match:'COMPUTER' ignoreCase:false
'[abcd]*' match:'computer' ignoreCase:false
'[abcd]*' match:'Computer' ignoreCase:false
'[a-k]*' match:'komputer' ignoreCase:false
'[a-k]*' match:'zomputer' ignoreCase:false
'[a-k]*' match:'Komputer' ignoreCase:false
'[a-k]*' match:'Komputer' ignoreCase:true
'*some*compl*ern*' match:'this is some more complicated pattern match' ignoreCase:true
'*some*compl*ern*' match:'this is another complicated pattern match' ignoreCase:true
Time millisecondsToRun:[
Symbol allInstancesDo:[:sym |
'[ab]*' match:sym ignoreCase:false
]
].
Time millisecondsToRun:[
Symbol allInstancesDo:[:sym |
'*at:*' match:sym ignoreCase:false
]
].
"
"Modified: 2.4.1997 / 17:28:58 / cg"
! !
!CharacterArray methodsFor:'printing & storing'!
article
"return an article string for the receiver."
|firstChar|
firstChar := (self at:1) asLowercase.
(firstChar isVowel or:[firstChar == $x]) ifTrue:[
firstChar ~~ $u ifTrue:[
^ 'an'
]
].
^ 'a'
!
displayString
"return a string to display the receiver - use storeString to have
quotes around."
^ self storeString
!
printOn:aStream
"print the receiver on aStream"
aStream nextPutAll:self
!
printString
"return a string for printing - thats myself"
^ self
! !
!CharacterArray methodsFor:'queries'!
encoding
"return the strings encoding, as a symbol.
Here, by default, we assume #iso8859 encoding.
This has to be redefined in XXXEncodedString subclasses."
^ #iso8859
"Modified: 27.4.1996 / 13:23:49 / cg"
!
hasChangeOfEmphasis
^ false
"Created: 12.5.1996 / 12:31:39 / cg"
!
heightOn:aGC
"return the size of the recevier in device units if displayed on aGC"
^ aGC font heightOf:self
"
'hello world' heightOn:(View new)
"
"Created: 12.5.1996 / 20:09:29 / cg"
"Modified: 12.5.1996 / 20:32:05 / cg"
!
isString
"return true, if the receiver is some kind of string;
true is returned here - redefinition of Object>>isString."
^ true
!
leftIndent
"if the receiver starts with spaces, return the number of spaces
at the left - otherwise, return 0.
If the receiver consists of spaces only, return the receivers size."
|index "{Class: SmallInteger }"
end "{Class: SmallInteger }"|
index := 1.
end := self size.
[index <= end] whileTrue:[
(self at:index) isSeparator ifFalse:[^ index - 1].
index := index + 1
].
^ end
"
' hello' leftIndent
'foo ' leftIndent
' ' leftIndent
"
"Modified: 20.4.1996 / 19:28:43 / cg"
!
widthOn:aGC
"return ths size of the recevier in device units if displayed on aGC"
^ (aGC font on:aGC device) widthOf:self
"
'hello world' widthOn:(View new)
"
"Created: 12.5.1996 / 20:09:29 / cg"
"Modified: 17.4.1997 / 12:50:23 / cg"
! !
!CharacterArray methodsFor:'special string converting'!
chopTo:maxLen
"if the receivers size is less or equal to maxLen, return it.
Otherwise, return a copy of the receiver, where some characters
in the middle have been removed for a total string length
of maxLen."
|sz n1 n2|
(sz := self size) > maxLen ifTrue:[
n1 := n2 := maxLen // 2.
maxLen odd ifTrue:[
n2 := n1 + 1
].
^ (self copyFrom:1 to:n1) , (self copyFrom:sz - n2 + 1)
]
"
'12345678901234' chopTo:15
'123456789012345' chopTo:15
'1234567890123456' chopTo:15
'aShortString' chopTo:15
'aVeryLongNameForAStringThatShouldBeShortened' chopTo:15
"
!
contractAtBeginningTo:maxLen
"if the receivers size is less or equal to maxLen, return it.
Otherwise, return a copy of the receiver, where some characters
at the beginning have been replaced by '...' for a total string length
of maxLen. Can be used to abbreviate long entries in tables."
|sz|
(sz := self size) > maxLen ifTrue:[
^ '...' , (self copyFrom:(sz - (maxLen - 4)))
]
"
'12345678901234' contractAtBeginningTo:15
'123456789012345' contractAtBeginningTo:15
'1234567890123456' contractAtBeginningTo:15
'aShortString' contractAtBeginningTo:15
'aVeryLongNameForAStringThatShouldBeShortened' contractAtBeginningTo:15
"
!
contractAtEndTo:maxLen
"if the receivers size is less or equal to maxLen, return it.
Otherwise, return a copy of the receiver, where some characters
at the end have been replaced by '...' for a total string length
of maxLen. Can be used to abbreviate long entries in tables."
|sz|
(sz := self size) > maxLen ifTrue:[
^ self copyReplaceFrom:maxLen - 3
with:'...'
]
"
'12345678901234' contractAtEndTo:15
'123456789012345' contractAtEndTo:15
'1234567890123456' contractAtEndTo:15
'aShortString' contractAtEndTo:15
'aVeryLongNameForAStringThatShouldBeShortened' contractAtEndTo:15
"
!
contractTo:maxLen
"if the receivers size is less or equal to maxLen, return it.
Otherwise, return a copy of the receiver, where some characters
in the middle have been replaced by '...' for a total string length
of maxLen. Can be used to abbreviate long entries in tables."
|sz "{ SmallInteger }"
halfSize "{ SmallInteger }"|
(sz := self size) > maxLen ifTrue:[
halfSize := maxLen // 2.
^ self copyReplaceFrom:halfSize - 1
to:sz - maxLen + halfSize + 1
with:'...'
]
"
'12345678901234' contractTo:15
'123456789012345' contractTo:15
'1234567890123456' contractTo:15
'aShortString' contractTo:15
'aVeryLongNameForAStringThatShouldBeShortened' contractTo:15
"
!
expandPlaceholdersWith:argArray
"return a copy of the receiver, where all %i escapes are
replaced by corresponding arguments' printStrings from the argArray.
I.e. 'hello %1; how is %2' expandPlaceholdersWith:#('world' 'this') results
in the new string 'hello world; how is this'.
To get a '%' character, use a '%%'-escape.
See also bindWith:... for VisualAge compatibility."
|expandedString next
idx "{ SmallInteger }"
start "{ SmallInteger }"
stop "{ SmallInteger }"|
expandedString := self species new:0.
stop := self size.
start := 1.
[start <= stop] whileTrue:[
idx := self indexOf:$% startingAt:start.
idx == 0 ifTrue:[
^ expandedString , (self copyFrom:start to:stop)
].
"found a %"
expandedString := expandedString , (self copyFrom:start to:(idx - 1)).
next := self at:(idx + 1).
(next == $%) ifTrue:[
expandedString := expandedString , '%'
] ifFalse:[
expandedString := expandedString , (argArray at:(next digitValue)) printString
].
start := idx + 2
].
^ expandedString
"
'hello %1' expandPlaceholdersWith:#('world')
'hello %1; how is %2' expandPlaceholdersWith:#('world' 'this')
'hello %2; how is %1' expandPlaceholdersWith:#('world' 'this')
'%1 plus %2 gives %3 ' expandPlaceholdersWith:#(4 5 9)
"
"Modified: 27.4.1996 / 11:58:36 / cg"
!
withCRs
"return a new string consisting of receivers characters
with all \-characters replaced by cr-characters."
^ self copy replaceAll:$\ by:(Character cr)
"
'hello\world' withCRs
"
!
withEscapes
"return a new string consisting of receivers characters
with all \X-character escapes replaced by corresponding-characters.
(similar to the way C-language Strings are converted).
The following escapes are supported:
\r return character
\n newline character
\b backspace character
\f formfeed character
\t tab character
\e escape character
\\ the \ character itself
\nnn three digit octal number defining the characters ascii value
\other other
Notice, that \' is NOT a valid escape, since the general syntax of
string constants is not affected by this method.
Although easily implementable, this is NOT done automatically
by the compiler (due to a lack of a language standard for this).
However, the compiler may detect sends ot #withEscapes to string literals
and place a modified string constant into the binary/byte-code.
Therefore, no runtime penalty will be payed for using these escapes.
(not in pre 2.11 versions)
"
|sz "{ SmallInteger }"
newSize "{ SmallInteger }"
srcIdx "{ SmallInteger }"
dstIdx "{ SmallInteger }"
val "{ SmallInteger }"
newString next hasEmphasis e|
"
first, count the number of escapes, to allow preallocation
of the new string ...
(it is faster to scan the string twice than to reallocate it multiple
times in a WriteStream)
"
sz := newSize := self size.
srcIdx := 1.
[(srcIdx := self indexOf:$\ startingAt:srcIdx) ~~ 0] whileTrue:[
srcIdx == sz ifFalse:[
newSize := newSize - 1.
srcIdx := srcIdx + 1.
next := self at:srcIdx.
next == $0 ifTrue:[
[(self at:srcIdx) isDigit] whileTrue:[
newSize := newSize - 1. srcIdx := srcIdx + 1.
]
].
].
srcIdx := srcIdx + 1.
].
newSize == sz ifTrue:[
^ self
].
newString := self species new:newSize.
hasEmphasis := self hasChangeOfEmphasis.
"
copy over, replace escapes
"
srcIdx := dstIdx := 1.
[srcIdx <= sz] whileTrue:[
next := self at:srcIdx.
hasEmphasis ifTrue:[
e := self emphasisAt:srcIdx
].
srcIdx := srcIdx + 1.
next == $\ ifTrue:[
srcIdx <= sz ifTrue:[
next := self at:srcIdx.
srcIdx := srcIdx + 1.
next == $r ifTrue:[
next := Character return
] ifFalse:[
next == $n ifTrue:[
next := Character nl
] ifFalse:[
next == $b ifTrue:[
next := Character backspace
] ifFalse:[
next == $f ifTrue:[
next := Character newPage
] ifFalse:[
next == $t ifTrue:[
next := Character tab
] ifFalse:[
next == $e ifTrue:[
next := Character esc
] ifFalse:[
next == $0 ifTrue:[
val := 0.
[next isDigit] whileTrue:[
val := val * 8 + next digitValue.
next := self at:srcIdx.
srcIdx := srcIdx + 1.
].
next := Character value:val.
]
]
]
]
]
]
].
].
].
newString at:dstIdx put:next.
hasEmphasis ifTrue:[
newString emphasisAt:dstIdx put:e
].
dstIdx := dstIdx + 1.
].
^ newString
"
'hello world' withEscapes
'hello\world' withEscapes
'hello\world\' withEscapes
'hello world\' withEscapes
'hello\tworld' withEscapes
'hello\nworld\na\n\tnice\n\t\tstring' withEscapes
'hello\tworld\n' withEscapes
'hello\010world' withEscapes
'hello\r\nworld' withEscapes
"
"Modified: 12.5.1996 / 12:53:34 / cg"
!
withTabs
"return a string consisting of the receivers characters
where leading spaces are replaced by tabulator characters (assuming 8-col tabs).
Notice: if the receiver does not contain any tabs, it is returned unchanged;
otherwise a new string is returned.
Limitation: only the very first spaces are replaced
(i.e. if the receiver contains newLine characters,
no tabs are inserted after those lineBreaks)"
|idx "{ SmallInteger }"
nTabs "{ SmallInteger }"
newString|
idx := self findFirst:[:c | (c ~~ Character space)].
nTabs := (idx-1) // 8.
nTabs == 0 ifTrue:[^ self].
"any tabs"
newString := self class new:(self size - (nTabs * 7)).
newString atAll:(1 to:nTabs) put:(Character tab).
newString replaceFrom:(nTabs + 1) with:self startingAt:(nTabs * 8 + 1).
^ newString
"
'12345678901234567890' withTabs
' 8901234567890' withTabs
' 901234567890' withTabs
' 67890' withTabs
' 7890' withTabs
' 890' withTabs
"
!
withTabsExpanded
"return a string consisting of the receivers characters,
where all tabulator characters are expanded into spaces (assuming 8-col tabs).
Notice: if the receiver does not contain any tabs, it is returned unchanged;
otherwise a new string is returned.
This does handle multiline strings."
|col "{ SmallInteger }"
str ch
dstIdx "{ SmallInteger }"
newSz "{ SmallInteger }"
sz "{ SmallInteger }"
hasEmphasis e|
(self includes:(Character tab)) ifFalse:[^ self].
sz := self size.
"/ count the new size first, instead of
"/ multiple resizing (better for large strings)
col := 1. newSz := 0.
1 to:sz do:[:srcIdx |
ch := self at:srcIdx.
ch == Character tab ifFalse:[
col := col + 1.
newSz := newSz + 1.
ch == Character cr ifTrue:[
col := 1
].
] ifTrue:[
(col \\ 8) to:8 do:[:ii |
newSz := newSz + 1.
col := col + 1
].
]
].
str := self species new:newSz.
hasEmphasis := self hasChangeOfEmphasis.
col := 1. dstIdx := 1.
1 to:sz do:[:srcIdx |
ch := self at:srcIdx.
ch == Character tab ifFalse:[
col := col + 1.
ch == Character cr ifTrue:[
col := 1
].
hasEmphasis ifTrue:[
e := self emphasisAt:srcIdx.
str emphasisAt:dstIdx put:e
].
str at:dstIdx put:ch.
dstIdx := dstIdx + 1
] ifTrue:[
(col \\ 8) to:8 do:[:ii |
str at:dstIdx put:Character space.
dstIdx := dstIdx + 1.
col := col + 1
].
]
].
^ str
"
('1' , Character tab asString , 'x') withTabsExpanded
('12345' , Character tab asString , 'x') withTabsExpanded
('123456' , Character tab asString , 'x') withTabsExpanded
('1234567' , Character tab asString , 'x') withTabsExpanded
('12345678' , Character tab asString , 'x') withTabsExpanded
('123456789' , Character tab asString , 'x') withTabsExpanded
(String with:Character tab
with:Character tab
with:$1) withTabsExpanded
(String with:Character tab
with:$1
with:Character tab
with:$2) withTabsExpanded
(String with:Character tab
with:$1
with:Character cr
with:Character tab
with:$2) withTabsExpanded
"
"Modified: 12.5.1996 / 13:05:10 / cg"
!
withoutCRs
"return a new collection consisting of receivers elements
with all cr-characters replaced by \-characters.
This is the reverse operation of withCRs."
^ self copy replaceAll:(Character cr) by:$\
"
'hello
world' withoutCRs
"
!
withoutLeadingSeparators
"return a copy of myself without leading separators.
Notice: this does remove tabs, newline or any other whitespace.
Returns an empty string, if the receiver consist only of whitespace."
|index|
index := self indexOfNonSeparatorStartingAt:1.
index ~~ 0 ifTrue:[
index == 1 ifTrue:[
^ self
].
^ self copyFrom:index
].
^ ''
"
' foo ' withoutLeadingSeparators
'foo ' withoutLeadingSeparators
' foo' withoutLeadingSeparators
' ' withoutLeadingSeparators
'foo' withoutLeadingSeparators
(' ' , Character tab asString , ' foo ') withoutLeadingSeparators inspect
"
!
withoutSeparators
"return a copy of myself without leading and trailing whitespace.
Whitespace is space, tab, newline, formfeed.
Use withoutSpaces, if you want to remove spaces only."
|startIndex "{ Class: SmallInteger }"
endIndex "{ Class: SmallInteger }"
sz|
sz := self size.
startIndex := 1.
endIndex := sz.
[(startIndex < endIndex) and:[(self at:startIndex) isSeparator]] whileTrue:[
startIndex := startIndex + 1
].
[(endIndex > 1) and:[(self at:endIndex) isSeparator]] whileTrue:[
endIndex := endIndex - 1
].
startIndex > endIndex ifTrue:[
^ ''
].
((startIndex == 1) and:[endIndex == sz]) ifTrue:[
^ self
].
^ self copyFrom:startIndex to:endIndex
"
' foo ' withoutSeparators
' foo' withoutSeparators
'foo ' withoutSeparators
' ' withoutSeparators
(' foo' , Character tab asString , ' ') withoutSeparators inspect
"
!
withoutSpaces
"return a copy of myself without leading and trailing spaces.
Notice: this does NOT remove tabs, newline or any other whitespace.
Use withoutSeparators for this."
|startIndex "{ Class: SmallInteger }"
endIndex "{ Class: SmallInteger }"
sz|
sz := self size.
startIndex := 1.
endIndex := sz.
[(startIndex < endIndex) and:[(self at:startIndex) == Character space]] whileTrue:[
startIndex := startIndex + 1
].
[(endIndex > 1) and:[(self at:endIndex) == Character space]] whileTrue:[
endIndex := endIndex - 1
].
startIndex > endIndex ifTrue:[
^ ''
].
((startIndex == 1) and:[endIndex == sz]) ifTrue:[
^ self
].
^ self copyFrom:startIndex to:endIndex
"
' foo ' withoutSpaces
'foo ' withoutSpaces
' foo' withoutSpaces
' ' withoutSpaces
(' foo' , Character tab asString , ' ') withoutSpaces inspect
"
!
withoutTrailingSeparators
"return a copy of myself without trailing separators.
Notice: this does remove tabs, newline or any other whitespace.
Returns an empty string, if the receiver consist only of whitespace."
|index|
index := self size.
[index ~~ 0] whileTrue:[
(self at:index) isSeparator ifFalse:[
^ self copyTo:index
].
index := index - 1
].
^ ''
"
' foo ' withoutTrailingSeparators
'foo ' withoutTrailingSeparators
' foo' withoutTrailingSeparators
' ' withoutTrailingSeparators
'foo' withoutTrailingSeparators
(' ' , Character tab asString , ' foo ') withoutTrailingSeparators inspect
(' foo' , Character tab asString) withoutTrailingSeparators inspect
"
! !
!CharacterArray methodsFor:'substring searching'!
findString:subString
"find a substring. if found, return the index;
if not found, return 0."
^ self indexOfSubCollection:subString startingAt:1 ifAbsent:0
"
'hello world' findString:'llo'
'hello world' findString:'ole'
"
!
findString:subString ifAbsent:exceptionBlock
"find a substring. If found, return the index;
if not found, return the result of evaluating exceptionBlock."
^ self indexOfSubCollection:subString startingAt:1 ifAbsent:exceptionBlock
!
findString:subString startingAt:index
"find a substring, starting at index. if found, return the index;
if not found, return 0."
^ self indexOfSubCollection:subString startingAt:index ifAbsent:0
"
'hello yello' findString:'llo' startingAt:1
'hello yello' findString:'llo' startingAt:5
'hello yello' findString:'llo' startingAt:15
"
!
findString:subString startingAt:index ifAbsent:exceptionBlock
"find a substring, starting at index. if found, return the index;
if not found, return the result of evaluating exceptionBlock."
^ self indexOfSubCollection:subString startingAt:index ifAbsent:exceptionBlock
!
includesString:aString
"return true, if a substring is contained in the receiver"
^ (self indexOfSubCollection:aString startingAt:1 ifAbsent:0) ~~ 0
"
'hello world' includesString:'hel'
'hello world' includesString:'rld'
'hello world' includesString:'llo'
'hello world' includesString:'LLO'
"
!
indexOfSubCollection:subString startingAt:index ifAbsent:exceptionBlock
"find a substring, starting at index. if found, return the index;
if not found, return the result of evaluating exceptionBlock.
This is a q&d hack - not very efficient"
|firstChar found
startIndex "{ Class: SmallInteger }"
subSize "{ Class: SmallInteger }"
mySize "{ Class: SmallInteger }"
runIdx "{ Class: SmallInteger }" |
subSize := subString size.
subSize == 0 ifTrue:[^ index]. "empty string matches"
mySize := self size.
firstChar := subString at:1.
startIndex := self indexOf:firstChar startingAt:index.
[startIndex == 0] whileFalse:[
runIdx := startIndex.
found := true.
1 to:subSize do:[:i |
runIdx > mySize ifTrue:[
found := false
] ifFalse:[
(subString at:i) ~= (self at:runIdx) ifTrue:[
found := false
]
].
runIdx := runIdx + 1
].
found ifTrue:[
^ startIndex
].
startIndex := self indexOf:firstChar startingAt:(startIndex + 1)
].
^ exceptionBlock value
"Modified: 23.2.1996 / 15:35:15 / cg"
!
restAfter:keyword withoutSeparators:strip
"compare the left of the receiver with keyword,
if it matches return the right.
Finally, if strip is true, remove whiteSpace.
This method is used to match and extract lines of the form:
something: rest
where we are interested in rest, but only if the receiver string
begins with something.
You may wonder why such a specialized method exists here
- this is so common when processing mailboxes,
rcs files, nntp/pop3 responses, that is was considered worth
a special method here to avoid having the code below a hundred
times in variuos places."
|rest|
(self startsWith:keyword) ifTrue:[
rest := self copyFrom:(keyword size + 1).
strip ifTrue:[
rest := rest withoutSeparators
].
^ rest
].
^ nil
"
'foo: hello world' restAfter:'foo:' withoutSeparators:true
'funny: something' restAfter:'foo:' withoutSeparators:true
'foo: hello world ' restAfter:'foo:' withoutSeparators:true
'foo: hello world ' restAfter:'foo:' withoutSeparators:false
"
"Created: 25.11.1995 / 11:04:18 / cg"
! !
!CharacterArray methodsFor:'testing'!
continuesWith:aString startingAt:startIndex
"return true, if the receiver beginning at startIndex
contains the characters in aString."
|sz "{Class: SmallInteger }"
idx "{Class: SmallInteger }"|
sz := aString size.
idx := startIndex.
1 to:sz do:[:i |
(self at:idx) ~~ (aString at:i) ifTrue:[^ false].
idx := idx + 1
].
^ true
"
'hello world' continuesWith:'world' startingAt:6
'hello world' continuesWith:'world' startingAt:7
"
"Created: 12.5.1996 / 15:46:40 / cg"
"Modified: 26.7.1996 / 19:08:36 / cg"
!
countWords
"return the number of words, which are separated by separators"
|tally "{ Class: SmallInteger }"
start "{ Class: SmallInteger }"
mySize "{ Class: SmallInteger }"
stop ch|
tally := 0.
start := 1.
mySize := self size.
[start <= mySize] whileTrue:[
ch := self at:start.
ch isSeparator ifTrue:[
start := start + 1
] ifFalse:[
stop := self indexOfSeparatorStartingAt:start.
(stop == 0) ifTrue:[
stop := mySize + 1
].
tally := tally + 1.
start := stop
]
].
^ tally
"
'hello world isnt this nice' countWords'
"
!
endsWith:aString
"return true, if the receiver ends with something, aString."
|s|
(s := self string) ~~ self ifTrue:[
^ s endsWith:aString
].
^ super endsWith:aString
"
'hello world' endsWith:'world'
'hello world' asText allBold endsWith:'world'
"
"Modified: 12.5.1996 / 15:49:18 / cg"
!
isAlphaNumeric
"return true, if the receiver is some alphanumeric word;
i.e. consists of a letter followed by letters or digits."
self size == 0 ifTrue:[
"mhmh what is this ?"
^ false
].
(self at:1) isLetter ifFalse:[^ false].
self do:[:char |
char isLetterOrDigit ifFalse:[^ false].
].
^ true
"
'helloWorld' isAlphaNumeric
'foo1234' isAlphaNumeric
'f1234' isAlphaNumeric
'1234' isAlphaNumeric
'+' isAlphaNumeric
"
!
isBlank
"return true, if the receiver contains spaces only"
self do:[:char |
char ~~ Character space ifTrue:[^ false].
].
^ true
!
levenshteinTo:aString
"return the levenshtein distance to the argument, aString;
this value corrensponds to the number of replacements that have to be
made to get aString from the receiver.
See IEEE transactions on Computers 1976 Pg 172 ff."
"
in the following, we assum that ommiting a character
is less of an error than inserting an extra character.
Therefore the different insertion (i) and deletion (d)
values.
"
^ self levenshteinTo:aString s:4 c:1 i:2 d:6
"
'ocmprt' levenshteinTo:'computer'
'computer' levenshteinTo:'computer'
'ocmputer' levenshteinTo:'computer'
'cmputer' levenshteinTo:'computer'
'computer' levenshteinTo:'cmputer'
'Computer' levenshteinTo:'computer'
"
!
levenshteinTo:aString s:substWeight c:caseWeight i:insrtWeight d:deleteWeight
"parametrized levenshtein.
return the levenshtein distance to the argument, aString;
this value corrensponds to the number of replacements that have to be
made to get aString from the receiver.
The arguments are the costs for substitution, case-change, insertion and
deletion of a character.
See IEEE transactions on Computers 1976 Pg 172 ff"
|d "delta matrix"
len1 "{ Class: SmallInteger }"
len2 "{ Class: SmallInteger }"
dim "{ Class: SmallInteger }"
prevRow row col
dimPlus1 "{ Class: SmallInteger }"
min pp c1 c2|
len1 := self size.
len2 := aString size.
"create the help-matrix"
dim := len1 max:len2.
dimPlus1 := dim + 1.
d := Array new:dimPlus1.
1 to:dimPlus1 do:[:i |
d at:i put:(Array new:dimPlus1)
].
"init help-matrix"
(d at:1) at:1 put:0.
row := d at:1.
1 to:dim do:[:j |
row at:(j + 1) put:( (row at:j) + insrtWeight )
].
1 to:dim do:[:i |
(d at:(i + 1)) at:1 put:( ((d at:i) at:1) + deleteWeight )
].
1 to:len1 do:[:i |
c1 := self at:i.
1 to:len2 do:[:j |
c2 := aString at:j.
(c1 == c2) ifTrue:[
pp := 0
] ifFalse:[
(c1 asLowercase == c2 asLowercase) ifTrue:[
pp := caseWeight
] ifFalse:[
pp := substWeight
]
].
prevRow := d at:i.
row := d at:(i + 1).
col := j + 1.
min := (prevRow at:j) + pp.
min := min min:( (row at:j) + insrtWeight).
min := min min:( (prevRow at:col) + deleteWeight).
row at:col put: min
]
].
^ (d at:(len1 + 1)) at:(len2 + 1)
!
numArgs
"treating the receiver as a message selector, return how many arguments would it take"
|binopChars firstChar|
(self size > 2) ifFalse:[
binopChars := Scanner binarySelectorCharacters.
firstChar := self at:1.
(self size == 1) ifTrue:[
((binopChars occurrencesOf:firstChar) == 0) ifTrue:[^ 0].
^ 1
].
((binopChars occurrencesOf:firstChar) == 0) ifFalse:[
((binopChars occurrencesOf:(self at:2)) == 0) ifFalse:[^ 1]
]
].
^ self occurrencesOf:$:
"
'foo:bar:' numArgs
#foo:bar: numArgs
'hello' numArgs
'+' numArgs
'|' numArgs
'?' numArgs
"
"Modified: 4.1.1997 / 14:16:14 / cg"
!
partsIfSelector
"treat the receiver as a message selector, return a collection of parts."
|idx1 "{ Class: SmallInteger }"
coll idx2 sz|
coll := OrderedCollection new.
idx1 := 1.
sz := self size.
[true] whileTrue:[
idx2 := self indexOf:$: startingAt:idx1 + 1.
(idx2 == 0 or:[idx2 == sz]) ifTrue:[
coll add:(self copyFrom:idx1).
^ coll
].
coll add:(self copyFrom:idx1 to:idx2).
idx1 := idx2 + 1
].
"
'foo:bar:' partsIfSelector
#foo:bar: partsIfSelector
'hello' partsIfSelector
'+' partsIfSelector
"
!
spellAgainst: aString
"return an integer between 0 and 100 indicating how similar
the argument is to the receiver. No case conversion is done.
This algorithm is much simpler (but also less exact) than the
levenshtein distance. Experiment which is better for your
application."
| i1 "{ Class: SmallInteger }"
i2 "{ Class: SmallInteger }"
next1 "{ Class: SmallInteger }"
next2 "{ Class: SmallInteger }"
size1 "{ Class: SmallInteger }"
size2 "{ Class: SmallInteger }"
score "{ Class: SmallInteger }"
maxLen "{ Class: SmallInteger }" |
size1 := self size.
size2 := aString size.
maxLen := size1 max:size2.
score := 0.
i1 := i2 := 1.
[i1 <= size1 and: [i2 <= size2]] whileTrue:[
next1 := i1 + 1.
next2 := i2 + 1.
(self at:i1) == (aString at:i2) ifTrue: [
score := score+1.
i1 := next1.
i2 := next2
] ifFalse: [
(i2 < size2 and: [(self at:i1) == (aString at:next2)]) ifTrue: [
i2 := next2
] ifFalse: [
(i1 < size1 and: [(self at:next1) == (aString at:i2)]) ifTrue: [
i1 := next1
] ifFalse: [
i1 := next1.
i2 := next2
]
]
]
].
score == maxLen ifTrue: [^ 100].
^ 100 * score // maxLen
"
'Smalltalk' spellAgainst: 'Smalltlak'
'Smalltalk' spellAgainst: 'smalltlak'
'Smalltalk' spellAgainst: 'smalltalk'
'Smalltalk' spellAgainst: 'smalltlk'
'Smalltalk' spellAgainst: 'Smalltolk'
"
!
startsWith:aString
"return true, if the receiver starts with something, aString."
|s|
(s := self string) ~~ self ifTrue:[
^ s startsWith:aString
].
^ super startsWith:aString
"
'hello world' startsWith:'hello'
'hello world' asText allBold startsWith:'hello'
"
"Created: 12.5.1996 / 15:46:40 / cg"
"Modified: 12.5.1996 / 15:49:24 / cg"
! !
!CharacterArray class methodsFor:'documentation'!
version
^ '$Header: /cvs/stx/stx/libbasic/CharacterArray.st,v 1.110 1997-04-22 21:51:26 cg Exp $'
! !