AuroraMiddleware/scuffed-code
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

ICU-3280 add RuleCharacterIterator

Browse Source 
X-SVN-Rev: 13232
This commit is contained in:
Alan Liu 2003-09-29 20:36:26 +00:00
parent 212715d5fb
commit 375490bb0c
4 changed files with 391 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
icu4c/source/common/Makefile.in
View File

@ -68,7 +68,7 @@ unistr.o utf_impl.o ustring.o ustrcase.o cstring.o ustrfmt.o ustrtrns.o \
normlzr.o unorm.o unorm_it.o chariter.o schriter.o uchriter.o uiter.o \
uchar.o uprops.o propname.o ubidi.o ubidiwrt.o ubidiln.o ushape.o unames.o \
ucln_cmn.o uscript.o usc_impl.o uvector.o ustack.o uvectr32.o ucmp8.o \
uarrsort.o utrie.o uset.o uniset.o caniter.o unifilt.o unifunct.o usetiter.o \
uarrsort.o utrie.o uset.o uniset.o ruleiter.o caniter.o unifilt.o unifunct.o usetiter.o \
brkiter.o brkdict.o ubrk.o dbbi.o dbbi_tbl.o \
rbbi.o rbbidata.o rbbinode.o rbbirb.o rbbiscan.o rbbisetb.o rbbistbl.o rbbitblb.o \
icuserv.o iculserv.o icunotif.o uenum.o ustrenum.o \

8
icu4c/source/common/common.dsp
View File

@ -2306,6 +2306,14 @@ SOURCE=.\propname.h
# End Source File
# Begin Source File
SOURCE=.\ruleiter.cpp
# End Source File
# Begin Source File
SOURCE=.\ruleiter.h
# End Source File
# Begin Source File
SOURCE=.\symtable.h
# End Source File
# Begin Source File

151
icu4c/source/common/ruleiter.cpp Normal file
View File

@ -0,0 +1,151 @@
/*
**********************************************************************
* Copyright (c) 2003, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* Author: Alan Liu
* Created: September 24 2003
* Since: ICU 2.8
**********************************************************************
*/
#include "ruleiter.h"
#include "unicode/parsepos.h"
#include "unicode/unistr.h"
#include "symtable.h" // TODO => unicode/symtable.h
#include "uprops.h"
U_NAMESPACE_BEGIN
RuleCharacterIterator::RuleCharacterIterator(const UnicodeString& theText, const SymbolTable* theSym,
ParsePosition& thePos) :
text(theText),
sym(theSym),
pos(thePos),
buf(0)
{}
UBool RuleCharacterIterator::atEnd() const {
return buf == 0 && pos.getIndex() == text.length();
}
UChar32 RuleCharacterIterator::next(int32_t options, UBool& isEscaped, UErrorCode& ec) {
if (U_FAILURE(ec)) return DONE;
UChar32 c = DONE;
isEscaped = FALSE;
for (;;) {
c = _current();
_advance(UTF_CHAR_LENGTH(c));
if (c == SymbolTable::SYMBOL_REF && buf == 0 &&
(options & PARSE_VARIABLES) != 0 && sym != 0) {
UnicodeString name = sym->parseReference(text, pos, text.length());
// If name is empty there was an isolated SYMBOL_REF;
// return it. Caller must be prepared for this.
if (name.length() == 0) {
break;
}
bufPos = 0;
buf = sym->lookup(name);
if (buf == 0) {
ec = U_UNDEFINED_VARIABLE;
return DONE;
}
// Handle empty variable value
if (buf->length() == 0) {
buf = 0;
}
continue;
}
if ((options & SKIP_WHITESPACE) != 0 &&
uprv_isRuleWhiteSpace(c)) {
continue;
}
if (c == 0x5C /*'\\'*/ && (options & PARSE_ESCAPES) != 0) {
UnicodeString s;
int32_t offset = 0;
c = lookahead(s).unescapeAt(offset);
jumpahead(offset);
isEscaped = TRUE;
if (c < 0) {
ec = U_MALFORMED_UNICODE_ESCAPE;
return DONE;
}
}
break;
}
return c;
}
void RuleCharacterIterator::getPos(RuleCharacterIterator::Pos& p) const {
p.buf = buf;
p.pos = pos.getIndex();
p.bufPos = bufPos;
}
void RuleCharacterIterator::setPos(const RuleCharacterIterator::Pos& p) {
buf = p.buf;
pos.setIndex(p.pos);
bufPos = p.bufPos;
}
void RuleCharacterIterator::skipIgnored(int32_t options) {
if ((options & SKIP_WHITESPACE) != 0) {
for (;;) {
UChar32 a = _current();
if (!uprv_isRuleWhiteSpace(a)) break;
_advance(UTF_CHAR_LENGTH(a));
}
}
}
UnicodeString& RuleCharacterIterator::lookahead(UnicodeString& result) const {
if (buf != 0) {
buf->extract(bufPos, 0x7FFFFFFF, result);
} else {
text.extract(pos.getIndex(), 0x7FFFFFFF, result);
}
return result;
}
void RuleCharacterIterator::jumpahead(int32_t count) {
_advance(count);
}
UnicodeString& RuleCharacterIterator::toString(UnicodeString& result) const {
int32_t b = pos.getIndex();
text.extract(0, b, result);
return result.append((UChar) 0x7C /*'|'*/).append(text, b, 0x7FFFFFFF);
}
UChar32 RuleCharacterIterator::_current() const {
if (buf != 0) {
return buf->char32At(bufPos);
} else {
int i = pos.getIndex();
return (i < text.length()) ? text.char32At(i) : DONE;
}
}
void RuleCharacterIterator::_advance(int32_t count) {
if (buf != 0) {
bufPos += count;
if (bufPos == buf->length()) {
buf = 0;
}
} else {
pos.setIndex(pos.getIndex() + count);
if (pos.getIndex() > text.length()) {
pos.setIndex(text.length());
}
}
}
U_NAMESPACE_END
//eof

231
icu4c/source/common/ruleiter.h Normal file
View File

@ -0,0 +1,231 @@
/*
**********************************************************************
* Copyright (c) 2003, International Business Machines
* Corporation and others. All Rights Reserved.
**********************************************************************
* Author: Alan Liu
* Created: September 24 2003
* Since: ICU 2.8
**********************************************************************
*/
#ifndef _RULEITER_H_
#define _RULEITER_H_
#include "unicode/utypes.h"
U_NAMESPACE_BEGIN
class UnicodeString;
class ParsePosition;
class SymbolTable;
/**
* An iterator that returns 32-bit code points. This class is deliberately
* <em>not</em> related to any of the ICU character iterator classes
* in order to minimize complexity.
* @author Alan Liu
* @since ICU 2.8
*/
class U_COMMON_API RuleCharacterIterator {
// TODO: Ideas for later. (Do not implement if not needed, lest the
// code coverage numbers go down due to unused methods.)
// 1. Add a copy constructor, operator==() method.
// 2. Rather than return DONE, throw an exception if the end
// is reached -- this is an alternate usage model, probably not useful.
private:
/**
* Text being iterated.
*/
const UnicodeString& text;
/**
* Position of iterator.
*/
ParsePosition& pos;
/**
* Symbol table used to parse and dereference variables. May be 0.
*/
const SymbolTable* sym;
/**
* Current variable expansion, or 0 if none.
*/
const UnicodeString* buf;
/**
* Position within buf. Meaningless if buf == 0.
*/
int32_t bufPos;
public:
/**
* Value returned when there are no more characters to iterate.
*/
enum { DONE = -1 };
/**
* Bitmask option to enable parsing of variable names. If (options &
* PARSE_VARIABLES) != 0, then an embedded variable will be expanded to
* its value. Variables are parsed using the SymbolTable API.
*/
enum { PARSE_VARIABLES = 1 };
/**
* Bitmask option to enable parsing of escape sequences. If (options &
* PARSE_ESCAPES) != 0, then an embedded escape sequence will be expanded
* to its value. Escapes are parsed using Utility.unescapeAt().
*/
enum { PARSE_ESCAPES = 2 };
/**
* Bitmask option to enable skipping of whitespace. If (options &
* SKIP_WHITESPACE) != 0, then whitespace characters will be silently
* skipped, as if they were not present in the input. Whitespace
* characters are defined by UCharacterProperty.isRuleWhiteSpace().
*/
enum { SKIP_WHITESPACE = 4 };
/**
* Constructs an iterator over the given text, starting at the given
* position.
* @param text the text to be iterated
* @param sym the symbol table, or null if there is none. If sym is null,
* then variables will not be deferenced, even if the PARSE_VARIABLES
* option is set.
* @param pos upon input, the index of the next character to return. If a
* variable has been dereferenced, then pos will <em>not</em> increment as
* characters of the variable value are iterated.
*/
RuleCharacterIterator(const UnicodeString& text, const SymbolTable* sym,
ParsePosition& pos);
/**
* Returns true if this iterator has no more characters to return.
*/
UBool atEnd() const;
/**
* Returns the next character using the given options, or DONE if there
* are no more characters, and advance the position to the next
* character.
* @param options one or more of the following options, bitwise-OR-ed
* together: PARSE_VARIABLES, PARSE_ESCAPES, SKIP_WHITESPACE.
* @param isEscaped output parameter set to TRUE if the character
* was escaped
* @param ec input-output error code. An error will only be set by
* this routing if options includes PARSE_VARIABLES and an unknown
* variable name is seen, or if options includes PARSE_ESCAPES and
* an invalid escape sequence is seen.
* @return the current 32-bit code point, or DONE
*/
UChar32 next(int32_t options, UBool& isEscaped, UErrorCode& ec);
/**
* Returns true if this iterator is currently within a variable expansion.
*/
inline UBool inVariable() const;
/**
* An opaque object representing the position of a RuleCharacterIterator.
*/
struct Pos {
private:
const UnicodeString* buf;
int32_t pos;
int32_t bufPos;
friend class RuleCharacterIterator;
};
/**
* Sets an object which, when later passed to setPos(), will
* restore this iterator's position. Usage idiom:
*
* RuleCharacterIterator iterator = ...;
* RuleCharacterIterator::Pos pos;
* iterator.getPos(pos);
* for (;;) {
* iterator.getPos(pos);
* int c = iterator.next(...);
* ...
* }
* iterator.setPos(pos);
*
* @param p a position object to be set to this iterator's
* current position.
*/
void getPos(Pos& p) const;
/**
* Restores this iterator to the position it had when getPos()
* set the given object.
* @param p a position object previously set by getPos()
*/
void setPos(const Pos& p);
/**
* Skips ahead past any ignored characters, as indicated by the given
* options. This is useful in conjunction with the lookahead() method.
*
* Currently, this only has an effect for SKIP_WHITESPACE.
* @param options one or more of the following options, bitwise-OR-ed
* together: PARSE_VARIABLES, PARSE_ESCAPES, SKIP_WHITESPACE.
*/
void skipIgnored(int32_t options);
/**
* Returns a string containing the remainder of the characters to be
* returned by this iterator, without any option processing. If the
* iterator is currently within a variable expansion, this will only
* extend to the end of the variable expansion. This method is provided
* so that iterators may interoperate with string-based APIs. The typical
* sequence of calls is to call skipIgnored(), then call lookahead(), then
* parse the string returned by lookahead(), then call jumpahead() to
* resynchronize the iterator.
* @param result a string to receive the characters to be returned
* by future calls to next()
* @return a reference to result
*/
UnicodeString& lookahead(UnicodeString& result) const;
/**
* Advances the position by the given number of 16-bit code units.
* This is useful in conjunction with the lookahead() method.
* @param count the number of 16-bit code units to jump over
*/
void jumpahead(int32_t count);
/**
* Returns a string representation of this object, consisting of the
* characters being iterated, with a '|' marking the current position.
* Position within an expanded variable is <em>not</em> indicated.
* @param result output parameter to receive a string
* representation of this object
*/
UnicodeString& toString(UnicodeString& result) const;
private:
/**
* Returns the current 32-bit code point without parsing escapes, parsing
* variables, or skipping whitespace.
* @return the current 32-bit code point
*/
UChar32 _current() const;
/**
* Advances the position by the given amount.
* @param count the number of 16-bit code units to advance past
*/
void _advance(int32_t count);
};
inline UBool RuleCharacterIterator::inVariable() const {
return buf != 0;
}
U_NAMESPACE_END
#endif // _RULEITER_H_
//eof