ICU-1292 remove unirange; no longer needed due to pragma use variable range xxxx xxxx
X-SVN-Rev: 6677
This commit is contained in:
Alan Liu
parent
f312dc5a1c
commit
1657c4c4c9
@ -66,7 +66,7 @@ ucal.o calendar.o gregocal.o timezone.o simpletz.o \
|
||||
sortkey.o bocsu.o coleitr.o coll.o ucoleitr.o \
|
||||
ucol.o ucol_bld.o ucol_cnt.o ucol_elm.o ucol_tok.o ucol_wgt.o tblcoll.o \
|
||||
strmatch.o usearch.o search.o stsearch.o \
|
||||
uniset.o unifltlg.o unirange.o translit.o utrans.o \
|
||||
uniset.o unifltlg.o translit.o utrans.o \
|
||||
cpdtrans.o hextouni.o rbt.o rbt_data.o rbt_pars.o rbt_rule.o rbt_set.o \
|
||||
dbbi.o dbbi_tbl.o rbbi.o rbbi_tbl.o nultrans.o \
|
||||
remtrans.o titletrn.o tolowtrn.o toupptrn.o xformtrn.o \
|
||||
|
||||
@ -354,10 +354,6 @@ SOURCE=.\unifltlg.cpp
|
||||
# End Source File
|
||||
# Begin Source File
|
||||
|
||||
SOURCE=.\unirange.cpp
|
||||
# End Source File
|
||||
# Begin Source File
|
||||
|
||||
SOURCE=.\uniset.cpp
|
||||
# End Source File
|
||||
# Begin Source File
|
||||
@ -1670,10 +1666,6 @@ InputPath=.\unicode\unimatch.h
|
||||
# End Source File
|
||||
# Begin Source File
|
||||
|
||||
SOURCE=.\unirange.h
|
||||
# End Source File
|
||||
# Begin Source File
|
||||
|
||||
SOURCE=.\unicode\uniset.h
|
||||
|
||||
!IF "$(CFG)" == "i18n - Win32 Release"
|
||||
|
||||
@ -15,7 +15,6 @@
|
||||
#include "rbt_rule.h"
|
||||
#include "strmatch.h"
|
||||
#include "symtable.h"
|
||||
#include "unirange.h"
|
||||
#include "uvector.h"
|
||||
#include "unicode/parseerr.h"
|
||||
#include "unicode/parsepos.h"
|
||||
|
||||
@ -1,120 +0,0 @@
|
||||
/*
|
||||
**********************************************************************
|
||||
* Copyright (C) 1999, International Business Machines
|
||||
* Corporation and others. All Rights Reserved.
|
||||
**********************************************************************
|
||||
* Date Name Description
|
||||
* 11/17/99 aliu Creation.
|
||||
**********************************************************************
|
||||
*/
|
||||
#include "unirange.h"
|
||||
#include "uvector.h"
|
||||
#include "unicode/unistr.h"
|
||||
|
||||
// For UVector of UnicodeRange* objects
|
||||
U_CDECL_BEGIN
|
||||
static void U_CALLCONV
|
||||
UnicodeRange_deleter(void* e) {
|
||||
delete (UnicodeRange*) e;
|
||||
}
|
||||
U_CDECL_END
|
||||
|
||||
U_NAMESPACE_BEGIN
|
||||
|
||||
UnicodeRange::UnicodeRange(UChar theStart, int32_t theLength) {
|
||||
start = theStart;
|
||||
length = theLength;
|
||||
}
|
||||
|
||||
UnicodeRange* UnicodeRange::clone() const {
|
||||
return new UnicodeRange(start, length);
|
||||
}
|
||||
|
||||
/**
|
||||
* CALLER OWNS RESULT.
|
||||
*/
|
||||
UBool UnicodeRange::contains(UChar c) const {
|
||||
return c >= start && (c - start) < length;
|
||||
}
|
||||
|
||||
/**
|
||||
* Assume that contains(c) is true. Split this range into two new
|
||||
* ranges around the character c. Make this range one of the new ranges
|
||||
* (modify it in place) and return the other new range. The character
|
||||
* itself is not included in either range. If the split results in an
|
||||
* empty range (that is, if c == start or c == start + length - 1) then
|
||||
* return null.
|
||||
*
|
||||
* MODIFIES THIS RANGE IN PLACE.
|
||||
*
|
||||
* CALLER OWNS RESULT.
|
||||
*/
|
||||
UnicodeRange* UnicodeRange::split(UChar c) {
|
||||
if (c == start) {
|
||||
++start;
|
||||
--length;
|
||||
return 0;
|
||||
} else if (c - start == length - 1) {
|
||||
--length;
|
||||
return 0;
|
||||
} else {
|
||||
++c;
|
||||
UnicodeRange* r = new UnicodeRange(c, start + length - c);
|
||||
length = --c - start;
|
||||
return r;
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Finds the largest unused subrange by the given string. A
|
||||
* subrange is unused by a string if the string contains no
|
||||
* characters in that range. If the given string contains no
|
||||
* characters in this range, then this range itself is
|
||||
* returned.
|
||||
*
|
||||
* CALLER OWNS RESULT.
|
||||
*/
|
||||
UnicodeRange*
|
||||
UnicodeRange::largestUnusedSubrange(const UnicodeString& str, UErrorCode &status) const {
|
||||
int32_t n = str.length();
|
||||
|
||||
UVector v(status);
|
||||
if (U_FAILURE(status)) {
|
||||
return NULL;
|
||||
}
|
||||
v.setDeleter(UnicodeRange_deleter);
|
||||
v.addElement(clone(), status);
|
||||
for (int32_t i=0; i<n; ++i) {
|
||||
UChar c = str.charAt(i);
|
||||
if (contains(c)) {
|
||||
for (int32_t j=0; j<v.size(); ++j) {
|
||||
UnicodeRange* r = (UnicodeRange*) v.elementAt(j);
|
||||
if (r->contains(c)) {
|
||||
r = r->split(c);
|
||||
if (r != 0) {
|
||||
v.addElement(r, status);
|
||||
}
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
UnicodeRange* bestRange = 0;
|
||||
int32_t ibest = -1;
|
||||
for (int32_t j=0; j<v.size(); ++j) {
|
||||
UnicodeRange* r = (UnicodeRange*) v.elementAt(j);
|
||||
if (bestRange == 0 || r->length > bestRange->length) {
|
||||
bestRange = r;
|
||||
ibest = j;
|
||||
}
|
||||
}
|
||||
|
||||
v.orphanElementAt(ibest); // So bestRange doesn't get deleted
|
||||
|
||||
return bestRange;
|
||||
}
|
||||
|
||||
|
||||
U_NAMESPACE_END
|
||||
|
||||
@ -1,77 +0,0 @@
|
||||
/*
|
||||
**********************************************************************
|
||||
* Copyright (C) {1999}, International Business Machines Corporation and others. All Rights Reserved.
|
||||
**********************************************************************
|
||||
* Date Name Description
|
||||
* 11/17/99 aliu Creation.
|
||||
**********************************************************************
|
||||
*/
|
||||
#ifndef UNIRANGE_H
|
||||
#define UNIRANGE_H
|
||||
|
||||
#include "unicode/utypes.h"
|
||||
|
||||
U_NAMESPACE_BEGIN
|
||||
|
||||
class UnicodeString;
|
||||
|
||||
/**
|
||||
* %%% INTERNAL CLASS USED BY RuleBasedTransliterator %%%
|
||||
*
|
||||
* A range of Unicode characters. Support the operations of testing for
|
||||
* inclusion (does this range contain this character?) and splitting.
|
||||
* Splitting involves breaking a range into two smaller ranges around a
|
||||
* character inside the original range. The split character is not included
|
||||
* in either range. If the split character is at either extreme end of the
|
||||
* range, one of the split products is an empty range.
|
||||
*
|
||||
* This class is used internally to determine the largest available private
|
||||
* use character range for variable stand-ins.
|
||||
*/
|
||||
class UnicodeRange {
|
||||
|
||||
public:
|
||||
|
||||
UChar start;
|
||||
|
||||
int32_t length;
|
||||
|
||||
UnicodeRange(UChar start, int32_t length);
|
||||
|
||||
/**
|
||||
* CALLER OWNS RESULT.
|
||||
*/
|
||||
UnicodeRange* clone() const;
|
||||
|
||||
UBool contains(UChar c) const;
|
||||
|
||||
/**
|
||||
* Assume that contains(c) is true. Split this range into two new
|
||||
* ranges around the character c. Make this range one of the new ranges
|
||||
* (modify it in place) and return the other new range. The character
|
||||
* itself is not included in either range. If the split results in an
|
||||
* empty range (that is, if c == start or c == start + length - 1) then
|
||||
* return null.
|
||||
*
|
||||
* MODIFIES THIS RANGE IN PLACE.
|
||||
*
|
||||
* CALLER OWNS RESULT.
|
||||
*/
|
||||
UnicodeRange* split(UChar c);
|
||||
|
||||
/**
|
||||
* Finds the largest subrange of this range that is unused by the
|
||||
* given string. A subrange is unused by a string if the string
|
||||
* contains no characters in that range. If the given string
|
||||
* contains no characters in this range, then this range itself is
|
||||
* returned.
|
||||
*
|
||||
* CALLER OWNS RESULT.
|
||||
*/
|
||||
UnicodeRange* largestUnusedSubrange(const UnicodeString& str, UErrorCode &status) const;
|
||||
|
||||
};
|
||||
|
||||
U_NAMESPACE_END
|
||||
|
||||
#endif
|
||||
Loading…
Reference in New Issue
Block a user