1 // © 2016 and later: Unicode, Inc. and others. 2 // License & terms of use: http://www.unicode.org/copyright.html 4 ******************************************************************************* 5 * Copyright (C) 2011-2013, International Business Machines 6 * Corporation and others. All Rights Reserved. 7 ******************************************************************************* 8 * file name: messagepattern.h 10 * tab size: 8 (not used) 13 * created on: 2011mar14 14 * created by: Markus W. Scherer 17 #ifndef __MESSAGEPATTERN_H__ 18 #define __MESSAGEPATTERN_H__ 27 #if U_SHOW_CPLUSPLUS_API 29 #if !UCONFIG_NO_FORMATTING 272 #define UMSGPAT_ARG_TYPE_HAS_PLURAL_STYLE(argType) \ 273 ((argType)==UMSGPAT_ARG_TYPE_PLURAL || (argType)==UMSGPAT_ARG_TYPE_SELECTORDINAL) 299 #define UMSGPAT_NO_NUMERIC_VALUE ((double)(-123456789)) 303 classMessagePatternDoubleList;
304 classMessagePatternPartsList;
554 // Java has package-private jdkAposMode() here. 555 // In C++, this is declared in the MessageImpl class. 580 return hasArgNumbers;
638 return getPart(i).type;
649 return getPart(partIndex).index;
671 return 0==msg.compare(part.index, part.length, s);
699 int32_t limit=getPart(start).limitPartIndex;
816 return ((type*37+index)*37+length)*37+value;
822 staticconst int32_t MAX_LENGTH=0xffff;
823 staticconst int32_t MAX_VALUE=0x7fff;
824 staticconst int32_t MAX_NESTED_LEVELS=0x03ff;
826 // Some fields are not final because they are modified during pattern parsing. 827 // After pattern parsing, the parts are effectively immutable. 832 int32_t limitPartIndex;
840 int32_t parseMessage(int32_t index, int32_t msgStartLength,
844 int32_t parseArg(int32_t index, int32_t argStartLength, int32_t nestingLevel,
849 int32_t parseChoiceStyle(int32_t index, int32_t nestingLevel,
863 static int32_t parseArgNumber(
constUnicodeString &s, int32_t start, int32_t limit);
865 int32_t parseArgNumber(int32_t start, int32_t limit) {
866 return parseArgNumber(msg, start, limit);
877 void parseDouble(int32_t start, int32_t limit,
UBool allowInfinity,
880 // Java has package-private appendReducedApostrophes() here. 881 // In C++, this is declared in the MessageImpl class. 883 int32_t skipWhiteSpace(int32_t index);
885 int32_t skipIdentifier(int32_t index);
891 int32_t skipDouble(int32_t index);
895 UBool isChoice(int32_t index);
897 UBool isPlural(int32_t index);
899 UBool isSelect(int32_t index);
901 UBool isOrdinal(int32_t index);
907 UBool inMessageFormatPattern(int32_t nestingLevel);
918 void addLimitPart(int32_t start,
922 void addArgDoublePart(
double numericValue, int32_t start, int32_t length,
UErrorCode &errorCode);
924 void setParseError(
UParseError *parseError, int32_t index);
931 // ArrayList<Part> parts=new ArrayList<Part>(); 932 MessagePatternPartsList *partsList;
935 // ArrayList<Double> numericValues; 936 MessagePatternDoubleList *numericValuesList;
937 double *numericValues;
938 int32_t numericValuesLength;
941 UBool needsAutoQuoting;
946 #endif// !UCONFIG_NO_FORMATTING 948 #endif/* U_SHOW_CPLUSPLUS_API */ 950 #endif// __MESSAGEPATTERN_H__ A message pattern "part", representing a pattern parsing event.
bool operator==(const Part &other) const
UMessagePatternArgType getArgType() const
Returns the argument type if this part is of type ARG_START or ARG_LIMIT, otherwise UMSGPAT_ARG_TYPE_...
int32_t getIndex() const
Returns the pattern string index associated with this Part.
int32_t getValue() const
Returns a value associated with this part.
bool operator!=(const Part &other) const
UMessagePatternPartType getType() const
Returns the type of this part.
int32_t getLength() const
Returns the length of the pattern substring associated with this Part.
int32_t getLimit() const
Returns the pattern string limit (exclusive-end) index associated with this Part.
static UBool hasNumericValue(UMessagePatternPartType type)
Indicates whether the Part type has a numeric value.
Part()
Default constructor, do not use.
Parses and represents ICU MessageFormat patterns.
void clear()
Clears this MessagePattern.
bool operator!=(const MessagePattern &other) const
MessagePattern & parseSelectStyle(const UnicodeString &pattern, UParseError *parseError, UErrorCode &errorCode)
Parses a SelectFormat pattern string.
int32_t getLimitPartIndex(int32_t start) const
Returns the index of the ARG|MSG_LIMIT part corresponding to the ARG|MSG_START at start.
MessagePattern & operator=(const MessagePattern &other)
Assignment operator.
UnicodeString getSubstring(const Part &part) const
Returns the substring of the pattern string indicated by the Part.
double getNumericValue(const Part &part) const
Returns the numeric value associated with an ARG_INT or ARG_DOUBLE.
int32_t countParts() const
Returns the number of "parts" created by parsing the pattern string.
MessagePattern & parseChoiceStyle(const UnicodeString &pattern, UParseError *parseError, UErrorCode &errorCode)
Parses a ChoiceFormat pattern string.
UnicodeString autoQuoteApostropheDeep() const
Returns a version of the parsed pattern string where each ASCII apostrophe is doubled (escaped) if it...
UMessagePatternPartType getPartType(int32_t i) const
Returns the UMessagePatternPartType of the i-th pattern "part".
void clearPatternAndSetApostropheMode(UMessagePatternApostropheMode mode)
Clears this MessagePattern and sets the UMessagePatternApostropheMode.
MessagePattern & parse(const UnicodeString &pattern, UParseError *parseError, UErrorCode &errorCode)
Parses a MessageFormat pattern string.
UBool hasNamedArguments() const
Does the parsed pattern have named arguments like {first_name}?
MessagePattern(UErrorCode &errorCode)
Constructs an empty MessagePattern with default UMessagePatternApostropheMode.
MessagePattern(const MessagePattern &other)
Copy constructor.
static int32_t validateArgumentName(const UnicodeString &name)
Validates and parses an argument name or argument number string.
MessagePattern(const UnicodeString &pattern, UParseError *parseError, UErrorCode &errorCode)
Constructs a MessagePattern with default UMessagePatternApostropheMode and parses the MessageFormat p...
double getPluralOffset(int32_t pluralStart) const
Returns the "offset:" value of a PluralFormat argument, or 0 if none is specified.
int32_t getPatternIndex(int32_t partIndex) const
Returns the pattern index of the specified pattern "part".
MessagePattern(UMessagePatternApostropheMode mode, UErrorCode &errorCode)
Constructs an empty MessagePattern.
bool operator==(const MessagePattern &other) const
UBool hasNumberedArguments() const
Does the parsed pattern have numbered arguments like {2}?
MessagePattern & parsePluralStyle(const UnicodeString &pattern, UParseError *parseError, UErrorCode &errorCode)
Parses a PluralFormat pattern string.
const Part & getPart(int32_t i) const
Gets the i-th pattern "part".
UMessagePatternApostropheMode getApostropheMode() const
virtual ~MessagePattern()
Destructor.
UBool partSubstringMatches(const Part &part, const UnicodeString &s) const
Compares the part's substring with the input string s.
const UnicodeString & getPatternString() const
UMemory is the common ICU base class.
UObject is the common ICU "boilerplate" class.
UnicodeString is a string class that stores Unicode characters directly and provides similar function...
UnicodeString tempSubString(int32_t start=0, int32_t length=INT32_MAX) const
Create a temporary substring for the specified range.
UMessagePatternPartType
MessagePattern::Part type constants.
@ UMSGPAT_PART_TYPE_INSERT_CHAR
Indicates that a syntax character needs to be inserted for auto-quoting.
@ UMSGPAT_PART_TYPE_ARG_TYPE
The argument type.
@ UMSGPAT_PART_TYPE_MSG_START
Start of a message pattern (main or nested).
@ UMSGPAT_PART_TYPE_ARG_SELECTOR
A selector substring in a "complex" argument style.
@ UMSGPAT_PART_TYPE_ARG_LIMIT
End of an argument.
@ UMSGPAT_PART_TYPE_ARG_NUMBER
The argument number, provided by the value.
@ UMSGPAT_PART_TYPE_MSG_LIMIT
End of a message pattern (main or nested).
@ UMSGPAT_PART_TYPE_ARG_DOUBLE
A numeric value, for example the offset or an explicit selector value in a PluralFormat style.
@ UMSGPAT_PART_TYPE_ARG_NAME
The argument name.
@ UMSGPAT_PART_TYPE_ARG_START
Start of an argument.
@ UMSGPAT_PART_TYPE_REPLACE_NUMBER
Indicates a syntactic (non-escaped) # symbol in a plural variant.
@ UMSGPAT_PART_TYPE_ARG_STYLE
The argument style text.
@ UMSGPAT_PART_TYPE_ARG_INT
An integer value, for example the offset or an explicit selector value in a PluralFormat style.
@ UMSGPAT_PART_TYPE_SKIP_SYNTAX
Indicates a substring of the pattern string which is to be skipped when formatting.
UMessagePatternArgType
Argument type constants.
@ UMSGPAT_ARG_TYPE_SELECT
The argument is a SelectFormat with one or more (ARG_SELECTOR, message) pairs.
@ UMSGPAT_ARG_TYPE_NONE
The argument has no specified type.
@ UMSGPAT_ARG_TYPE_PLURAL
The argument is a cardinal-number PluralFormat with an optional ARG_INT or ARG_DOUBLE offset (e....
@ UMSGPAT_ARG_TYPE_SIMPLE
The argument has a "simple" type which is provided by the ARG_TYPE part.
@ UMSGPAT_ARG_TYPE_SELECTORDINAL
The argument is an ordinal-number PluralFormat with the same style parts sequence and semantics as UM...
@ UMSGPAT_ARG_TYPE_CHOICE
The argument is a ChoiceFormat with one or more ((ARG_INT | ARG_DOUBLE), ARG_SELECTOR,...
@ UMSGPAT_ARG_NAME_NOT_VALID
Return value from MessagePattern.validateArgumentName() for when the string is invalid.
@ UMSGPAT_ARG_NAME_NOT_NUMBER
Return value from MessagePattern.validateArgumentName() for when the string is a valid "pattern ident...
UMessagePatternApostropheMode
Mode for when an apostrophe starts quoted literal text for MessageFormat output.
@ UMSGPAT_APOS_DOUBLE_OPTIONAL
A literal apostrophe is represented by either a single or a double apostrophe pattern character.
@ UMSGPAT_APOS_DOUBLE_REQUIRED
A literal apostrophe must be represented by a double apostrophe pattern character.
U_EXPORT UBool operator==(const StringPiece &x, const StringPiece &y)
Global operator == for StringPiece.
C API: Parse Error Information.
A UParseError struct is used to returned detailed information about parsing errors.
int32_t UChar32
Define UChar32 as a type for single Unicode code points.
int8_t UBool
The ICU boolean type, a signed-byte integer.
Basic definitions for ICU, for both C and C++ APIs.
UErrorCode
Standard ICU4C error code type, a substitute for exceptions.
#define U_COMMON_API
Set to export library symbols from inside the common library, and to import them from outside.
