util.cpp | Explore in Territory

// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
**********************************************************************
*   Copyright (c) 2001-2011, International Business Machines
*   Corporation and others.  All Rights Reserved.
**********************************************************************
*   Date        Name        Description
*   11/19/2001  aliu        Creation.
**********************************************************************
*/

#include "unicode/unimatch.h"
#include "unicode/utf16.h"
#include "patternprops.h"
#include "util.h"

// Define char16_t constants using hex for EBCDIC compatibility

static const char16_t BACKSLASH  = …; /*\*/
static const char16_t UPPER_U    = …; /*U*/
static const char16_t LOWER_U    = …; /*u*/
static const char16_t APOSTROPHE = …; // '\''
static const char16_t SPACE      = …; // ' '

// "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"
static const char16_t DIGITS[] = …;

U_NAMESPACE_BEGIN

UnicodeString& ICU_Utility::appendNumber(UnicodeString& result, int32_t n,
                                     int32_t radix, int32_t minDigits) { … }

UBool ICU_Utility::isUnprintable(UChar32 c) { … }

UBool ICU_Utility::shouldAlwaysBeEscaped(UChar32 c) { … }

UBool ICU_Utility::escapeUnprintable(UnicodeString& result, UChar32 c) { … }

UnicodeString &ICU_Utility::escape(UnicodeString& result, UChar32 c) { … }

/**
 * Returns the index of a character, ignoring quoted text.
 * For example, in the string "abc'hide'h", the 'h' in "hide" will not be
 * found by a search for 'h'.
 */
// FOR FUTURE USE.  DISABLE FOR NOW for coverage reasons.
/*
int32_t ICU_Utility::quotedIndexOf(const UnicodeString& text,
                               int32_t start, int32_t limit,
                               char16_t charToFind) {
    for (int32_t i=start; i<limit; ++i) {
        char16_t c = text.charAt(i);
        if (c == BACKSLASH) {
            ++i;
        } else if (c == APOSTROPHE) {
            while (++i < limit
                   && text.charAt(i) != APOSTROPHE) {}
        } else if (c == charToFind) {
            return i;
        }
    }
    return -1;
}
*/

/**
 * Skip over a sequence of zero or more white space characters at pos.
 * @param advance if true, advance pos to the first non-white-space
 * character at or after pos, or str.length(), if there is none.
 * Otherwise leave pos unchanged.
 * @return the index of the first non-white-space character at or
 * after pos, or str.length(), if there is none.
 */
int32_t ICU_Utility::skipWhitespace(const UnicodeString& str, int32_t& pos,
                                    UBool advance) { … }

/**
 * Skip over Pattern_White_Space in a Replaceable.
 * Skipping may be done in the forward or
 * reverse direction.  In either case, the leftmost index will be
 * inclusive, and the rightmost index will be exclusive.  That is,
 * given a range defined as [start, limit), the call
 * skipWhitespace(text, start, limit) will advance start past leading
 * whitespace, whereas the call skipWhitespace(text, limit, start),
 * will back up limit past trailing whitespace.
 * @param text the text to be analyzed
 * @param pos either the start or limit of a range of 'text', to skip
 * leading or trailing whitespace, respectively
 * @param stop either the limit or start of a range of 'text', to skip
 * leading or trailing whitespace, respectively
 * @return the new start or limit, depending on what was passed in to
 * 'pos'
 */
//?FOR FUTURE USE.  DISABLE FOR NOW for coverage reasons.
//?int32_t ICU_Utility::skipWhitespace(const Replaceable& text,
//?                                    int32_t pos, int32_t stop) {
//?    UChar32 c;
//?    UBool isForward = (stop >= pos);
//?
//?    if (!isForward) {
//?        --pos; // pos is a limit, so back up by one
//?    }
//?    
//?    while (pos != stop &&
//?           PatternProps::isWhiteSpace(c = text.char32At(pos))) {
//?        if (isForward) {
//?            pos += U16_LENGTH(c);
//?        } else {
//?            pos -= U16_LENGTH(c);
//?        }
//?    }
//?
//?    if (!isForward) {
//?        ++pos; // make pos back into a limit
//?    }
//?
//?    return pos;
//?}

/**
 * Parse a single non-whitespace character 'ch', optionally
 * preceded by whitespace.
 * @param id the string to be parsed
 * @param pos INPUT-OUTPUT parameter.  On input, pos[0] is the
 * offset of the first character to be parsed.  On output, pos[0]
 * is the index after the last parsed character.  If the parse
 * fails, pos[0] will be unchanged.
 * @param ch the non-whitespace character to be parsed.
 * @return true if 'ch' is seen preceded by zero or more
 * whitespace characters.
 */
UBool ICU_Utility::parseChar(const UnicodeString& id, int32_t& pos, char16_t ch) { … }

/**
 * Parse a pattern string within the given Replaceable and a parsing
 * pattern.  Characters are matched literally and case-sensitively
 * except for the following special characters:
 *
 * ~  zero or more Pattern_White_Space chars
 *
 * If end of pattern is reached with all matches along the way,
 * pos is advanced to the first unparsed index and returned.
 * Otherwise -1 is returned.
 * @param pat pattern that controls parsing
 * @param text text to be parsed, starting at index
 * @param index offset to first character to parse
 * @param limit offset after last character to parse
 * @return index after last parsed character, or -1 on parse failure.
 */
int32_t ICU_Utility::parsePattern(const UnicodeString& pat,
                                  const Replaceable& text,
                                  int32_t index,
                                  int32_t limit) { … }

int32_t ICU_Utility::parseAsciiInteger(const UnicodeString& str, int32_t& pos) { … }

/**
 * Append a character to a rule that is being built up.  To flush
 * the quoteBuf to rule, make one final call with isLiteral == true.
 * If there is no final character, pass in (UChar32)-1 as c.
 * @param rule the string to append the character to
 * @param c the character to append, or (UChar32)-1 if none.
 * @param isLiteral if true, then the given character should not be
 * quoted or escaped.  Usually this means it is a syntactic element
 * such as > or $
 * @param escapeUnprintable if true, then unprintable characters
 * should be escaped using \uxxxx or \Uxxxxxxxx.  These escapes will
 * appear outside of quotes.
 * @param quoteBuf a buffer which is used to build up quoted
 * substrings.  The caller should initially supply an empty buffer,
 * and thereafter should not modify the buffer.  The buffer should be
 * cleared out by, at the end, calling this method with a literal
 * character.
 */
void ICU_Utility::appendToRule(UnicodeString& rule,
                               UChar32 c,
                               UBool isLiteral,
                               UBool escapeUnprintable,
                               UnicodeString& quoteBuf) { … }

void ICU_Utility::appendToRule(UnicodeString& rule,
                               const UnicodeString& text,
                               UBool isLiteral,
                               UBool escapeUnprintable,
                               UnicodeString& quoteBuf) { … }

/**
 * Given a matcher reference, which may be null, append its
 * pattern as a literal to the given rule.
 */
void ICU_Utility::appendToRule(UnicodeString& rule,
                               const UnicodeMatcher* matcher,
                               UBool escapeUnprintable,
                               UnicodeString& quoteBuf) { … }

U_NAMESPACE_END
chromium/third_party/icu/source/common/util.cpp
