1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/common/util.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,239 @@
1.4 +/*
1.5 + **********************************************************************
1.6 + * Copyright (c) 2001-2011, International Business Machines
1.7 + * Corporation and others. All Rights Reserved.
1.8 + **********************************************************************
1.9 + * Date Name Description
1.10 + * 11/19/2001 aliu Creation.
1.11 + **********************************************************************
1.12 + */
1.13 +
1.14 +#ifndef ICU_UTIL_H
1.15 +#define ICU_UTIL_H
1.16 +
1.17 +#include "unicode/utypes.h"
1.18 +#include "unicode/uobject.h"
1.19 +#include "unicode/unistr.h"
1.20 +
1.21 +//--------------------------------------------------------------------
1.22 +// class ICU_Utility
1.23 +// i18n utility functions, scoped into the class ICU_Utility.
1.24 +//--------------------------------------------------------------------
1.25 +
1.26 +U_NAMESPACE_BEGIN
1.27 +
1.28 +class UnicodeMatcher;
1.29 +
1.30 +class U_COMMON_API ICU_Utility /* not : public UObject because all methods are static */ {
1.31 + public:
1.32 +
1.33 + /**
1.34 + * Append a number to the given UnicodeString in the given radix.
1.35 + * Standard digits '0'-'9' are used and letters 'A'-'Z' for
1.36 + * radices 11 through 36.
1.37 + * @param result the digits of the number are appended here
1.38 + * @param n the number to be converted to digits; may be negative.
1.39 + * If negative, a '-' is prepended to the digits.
1.40 + * @param radix a radix from 2 to 36 inclusive.
1.41 + * @param minDigits the minimum number of digits, not including
1.42 + * any '-', to produce. Values less than 2 have no effect. One
1.43 + * digit is always emitted regardless of this parameter.
1.44 + * @return a reference to result
1.45 + */
1.46 + static UnicodeString& appendNumber(UnicodeString& result, int32_t n,
1.47 + int32_t radix = 10,
1.48 + int32_t minDigits = 1);
1.49 +
1.50 + /**
1.51 + * Return true if the character is NOT printable ASCII.
1.52 + *
1.53 + * This method should really be in UnicodeString (or similar). For
1.54 + * now, we implement it here and share it with friend classes.
1.55 + */
1.56 + static UBool isUnprintable(UChar32 c);
1.57 +
1.58 + /**
1.59 + * Escape unprintable characters using \uxxxx notation for U+0000 to
1.60 + * U+FFFF and \Uxxxxxxxx for U+10000 and above. If the character is
1.61 + * printable ASCII, then do nothing and return FALSE. Otherwise,
1.62 + * append the escaped notation and return TRUE.
1.63 + */
1.64 + static UBool escapeUnprintable(UnicodeString& result, UChar32 c);
1.65 +
1.66 + /**
1.67 + * Returns the index of a character, ignoring quoted text.
1.68 + * For example, in the string "abc'hide'h", the 'h' in "hide" will not be
1.69 + * found by a search for 'h'.
1.70 + * @param text text to be searched
1.71 + * @param start the beginning index, inclusive; <code>0 <= start
1.72 + * <= limit</code>.
1.73 + * @param limit the ending index, exclusive; <code>start <= limit
1.74 + * <= text.length()</code>.
1.75 + * @param c character to search for
1.76 + * @return Offset of the first instance of c, or -1 if not found.
1.77 + */
1.78 +//?FOR FUTURE USE. DISABLE FOR NOW for coverage reasons.
1.79 +// static int32_t quotedIndexOf(const UnicodeString& text,
1.80 +// int32_t start, int32_t limit,
1.81 +// UChar c);
1.82 +
1.83 + /**
1.84 + * Skip over a sequence of zero or more white space characters at pos.
1.85 + * @param advance if true, advance pos to the first non-white-space
1.86 + * character at or after pos, or str.length(), if there is none.
1.87 + * Otherwise leave pos unchanged.
1.88 + * @return the index of the first non-white-space character at or
1.89 + * after pos, or str.length(), if there is none.
1.90 + */
1.91 + static int32_t skipWhitespace(const UnicodeString& str, int32_t& pos,
1.92 + UBool advance = FALSE);
1.93 +
1.94 + /**
1.95 + * Skip over Pattern_White_Space in a Replaceable.
1.96 + * Skipping may be done in the forward or
1.97 + * reverse direction. In either case, the leftmost index will be
1.98 + * inclusive, and the rightmost index will be exclusive. That is,
1.99 + * given a range defined as [start, limit), the call
1.100 + * skipWhitespace(text, start, limit) will advance start past leading
1.101 + * whitespace, whereas the call skipWhitespace(text, limit, start),
1.102 + * will back up limit past trailing whitespace.
1.103 + * @param text the text to be analyzed
1.104 + * @param pos either the start or limit of a range of 'text', to skip
1.105 + * leading or trailing whitespace, respectively
1.106 + * @param stop either the limit or start of a range of 'text', to skip
1.107 + * leading or trailing whitespace, respectively
1.108 + * @return the new start or limit, depending on what was passed in to
1.109 + * 'pos'
1.110 + */
1.111 +//?FOR FUTURE USE. DISABLE FOR NOW for coverage reasons.
1.112 +//? static int32_t skipWhitespace(const Replaceable& text,
1.113 +//? int32_t pos, int32_t stop);
1.114 +
1.115 + /**
1.116 + * Parse a single non-whitespace character 'ch', optionally
1.117 + * preceded by whitespace.
1.118 + * @param id the string to be parsed
1.119 + * @param pos INPUT-OUTPUT parameter. On input, pos[0] is the
1.120 + * offset of the first character to be parsed. On output, pos[0]
1.121 + * is the index after the last parsed character. If the parse
1.122 + * fails, pos[0] will be unchanged.
1.123 + * @param ch the non-whitespace character to be parsed.
1.124 + * @return true if 'ch' is seen preceded by zero or more
1.125 + * whitespace characters.
1.126 + */
1.127 + static UBool parseChar(const UnicodeString& id, int32_t& pos, UChar ch);
1.128 +
1.129 + /**
1.130 + * Parse a pattern string starting at offset pos. Keywords are
1.131 + * matched case-insensitively. Spaces may be skipped and may be
1.132 + * optional or required. Integer values may be parsed, and if
1.133 + * they are, they will be returned in the given array. If
1.134 + * successful, the offset of the next non-space character is
1.135 + * returned. On failure, -1 is returned.
1.136 + * @param pattern must only contain lowercase characters, which
1.137 + * will match their uppercase equivalents as well. A space
1.138 + * character matches one or more required spaces. A '~' character
1.139 + * matches zero or more optional spaces. A '#' character matches
1.140 + * an integer and stores it in parsedInts, which the caller must
1.141 + * ensure has enough capacity.
1.142 + * @param parsedInts array to receive parsed integers. Caller
1.143 + * must ensure that parsedInts.length is >= the number of '#'
1.144 + * signs in 'pattern'.
1.145 + * @return the position after the last character parsed, or -1 if
1.146 + * the parse failed
1.147 + */
1.148 + static int32_t parsePattern(const UnicodeString& rule, int32_t pos, int32_t limit,
1.149 + const UnicodeString& pattern, int32_t* parsedInts);
1.150 +
1.151 + /**
1.152 + * Parse a pattern string within the given Replaceable and a parsing
1.153 + * pattern. Characters are matched literally and case-sensitively
1.154 + * except for the following special characters:
1.155 + *
1.156 + * ~ zero or more Pattern_White_Space chars
1.157 + *
1.158 + * If end of pattern is reached with all matches along the way,
1.159 + * pos is advanced to the first unparsed index and returned.
1.160 + * Otherwise -1 is returned.
1.161 + * @param pat pattern that controls parsing
1.162 + * @param text text to be parsed, starting at index
1.163 + * @param index offset to first character to parse
1.164 + * @param limit offset after last character to parse
1.165 + * @return index after last parsed character, or -1 on parse failure.
1.166 + */
1.167 + static int32_t parsePattern(const UnicodeString& pat,
1.168 + const Replaceable& text,
1.169 + int32_t index,
1.170 + int32_t limit);
1.171 +
1.172 + /**
1.173 + * Parse an integer at pos, either of the form \d+ or of the form
1.174 + * 0x[0-9A-Fa-f]+ or 0[0-7]+, that is, in standard decimal, hex,
1.175 + * or octal format.
1.176 + * @param pos INPUT-OUTPUT parameter. On input, the first
1.177 + * character to parse. On output, the character after the last
1.178 + * parsed character.
1.179 + */
1.180 + static int32_t parseInteger(const UnicodeString& rule, int32_t& pos, int32_t limit);
1.181 +
1.182 + /**
1.183 + * Parse a Unicode identifier from the given string at the given
1.184 + * position. Return the identifier, or an empty string if there
1.185 + * is no identifier.
1.186 + * @param str the string to parse
1.187 + * @param pos INPUT-OUPUT parameter. On INPUT, pos is the
1.188 + * first character to examine. It must be less than str.length(),
1.189 + * and it must not point to a whitespace character. That is, must
1.190 + * have pos < str.length() and
1.191 + * !UCharacter::isWhitespace(str.char32At(pos)). On
1.192 + * OUTPUT, the position after the last parsed character.
1.193 + * @return the Unicode identifier, or an empty string if there is
1.194 + * no valid identifier at pos.
1.195 + */
1.196 + static UnicodeString parseUnicodeIdentifier(const UnicodeString& str, int32_t& pos);
1.197 +
1.198 + /**
1.199 + * Parse an unsigned 31-bit integer at the given offset. Use
1.200 + * UCharacter.digit() to parse individual characters into digits.
1.201 + * @param text the text to be parsed
1.202 + * @param pos INPUT-OUTPUT parameter. On entry, pos is the
1.203 + * offset within text at which to start parsing; it should point
1.204 + * to a valid digit. On exit, pos is the offset after the last
1.205 + * parsed character. If the parse failed, it will be unchanged on
1.206 + * exit. Must be >= 0 on entry.
1.207 + * @param radix the radix in which to parse; must be >= 2 and <=
1.208 + * 36.
1.209 + * @return a non-negative parsed number, or -1 upon parse failure.
1.210 + * Parse fails if there are no digits, that is, if pos does not
1.211 + * point to a valid digit on entry, or if the number to be parsed
1.212 + * does not fit into a 31-bit unsigned integer.
1.213 + */
1.214 + static int32_t parseNumber(const UnicodeString& text,
1.215 + int32_t& pos, int8_t radix);
1.216 +
1.217 + static void appendToRule(UnicodeString& rule,
1.218 + UChar32 c,
1.219 + UBool isLiteral,
1.220 + UBool escapeUnprintable,
1.221 + UnicodeString& quoteBuf);
1.222 +
1.223 + static void appendToRule(UnicodeString& rule,
1.224 + const UnicodeString& text,
1.225 + UBool isLiteral,
1.226 + UBool escapeUnprintable,
1.227 + UnicodeString& quoteBuf);
1.228 +
1.229 + static void appendToRule(UnicodeString& rule,
1.230 + const UnicodeMatcher* matcher,
1.231 + UBool escapeUnprintable,
1.232 + UnicodeString& quoteBuf);
1.233 +
1.234 +private:
1.235 + // do not instantiate
1.236 + ICU_Utility();
1.237 +};
1.238 +
1.239 +U_NAMESPACE_END
1.240 +
1.241 +#endif
1.242 +//eof
