1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/tools/toolutil/uparse.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,151 @@
1.4 +/*
1.5 +*******************************************************************************
1.6 +*
1.7 +* Copyright (C) 2000-2010, International Business Machines
1.8 +* Corporation and others. All Rights Reserved.
1.9 +*
1.10 +*******************************************************************************
1.11 +* file name: uparse.h
1.12 +* encoding: US-ASCII
1.13 +* tab size: 8 (not used)
1.14 +* indentation:4
1.15 +*
1.16 +* created on: 2000apr18
1.17 +* created by: Markus W. Scherer
1.18 +*
1.19 +* This file provides a parser for files that are delimited by one single
1.20 +* character like ';' or TAB. Example: the Unicode Character Properties files
1.21 +* like UnicodeData.txt are semicolon-delimited.
1.22 +*/
1.23 +
1.24 +#ifndef __UPARSE_H__
1.25 +#define __UPARSE_H__
1.26 +
1.27 +#include "unicode/utypes.h"
1.28 +
1.29 +/**
1.30 + * Is c an invariant-character whitespace?
1.31 + * @param c invariant character
1.32 + */
1.33 +#define U_IS_INV_WHITESPACE(c) ((c)==' ' || (c)=='\t' || (c)=='\r' || (c)=='\n')
1.34 +
1.35 +U_CDECL_BEGIN
1.36 +
1.37 +/**
1.38 + * Skip space ' ' and TAB '\t' characters.
1.39 + *
1.40 + * @param s Pointer to characters.
1.41 + * @return Pointer to first character at or after s that is not a space or TAB.
1.42 + */
1.43 +U_CAPI const char * U_EXPORT2
1.44 +u_skipWhitespace(const char *s);
1.45 +
1.46 +/**
1.47 + * Trim whitespace (including line endings) from the end of the string.
1.48 + *
1.49 + * @param s Pointer to the string.
1.50 + * @return Pointer to the new end of the string.
1.51 + */
1.52 +U_CAPI char * U_EXPORT2
1.53 +u_rtrim(char *s);
1.54 +
1.55 +/** Function type for u_parseDelimitedFile(). */
1.56 +typedef void U_CALLCONV
1.57 +UParseLineFn(void *context,
1.58 + char *fields[][2],
1.59 + int32_t fieldCount,
1.60 + UErrorCode *pErrorCode);
1.61 +
1.62 +/**
1.63 + * Parser for files that are similar to UnicodeData.txt:
1.64 + * This function opens the file and reads it line by line. It skips empty lines
1.65 + * and comment lines that start with a '#'.
1.66 + * All other lines are separated into fields with one delimiter character
1.67 + * (semicolon for Unicode Properties files) between two fields. The last field in
1.68 + * a line does not need to be terminated with a delimiter.
1.69 + *
1.70 + * For each line, after segmenting it, a line function is called.
1.71 + * It gets passed the array of field start and limit pointers that is
1.72 + * passed into this parser and filled by it for each line.
1.73 + * For each field i of the line, the start pointer in fields[i][0]
1.74 + * points to the beginning of the field, while the limit pointer in fields[i][1]
1.75 + * points behind the field, i.e., to the delimiter or the line end.
1.76 + *
1.77 + * The context parameter of the line function is
1.78 + * the same as the one for the parse function.
1.79 + *
1.80 + * The line function may modify the contents of the fields including the
1.81 + * limit characters.
1.82 + *
1.83 + * If the file cannot be opened, or there is a parsing error or a field function
1.84 + * sets *pErrorCode, then the parser returns with *pErrorCode set to an error code.
1.85 + */
1.86 +U_CAPI void U_EXPORT2
1.87 +u_parseDelimitedFile(const char *filename, char delimiter,
1.88 + char *fields[][2], int32_t fieldCount,
1.89 + UParseLineFn *lineFn, void *context,
1.90 + UErrorCode *pErrorCode);
1.91 +
1.92 +/**
1.93 + * Parse a string of code points like 0061 0308 0300.
1.94 + * s must end with either ';' or NUL.
1.95 + *
1.96 + * @return Number of code points.
1.97 + */
1.98 +U_CAPI int32_t U_EXPORT2
1.99 +u_parseCodePoints(const char *s,
1.100 + uint32_t *dest, int32_t destCapacity,
1.101 + UErrorCode *pErrorCode);
1.102 +
1.103 +/**
1.104 + * Parse a list of code points like 0061 0308 0300
1.105 + * into a UChar * string.
1.106 + * s must end with either ';' or NUL.
1.107 + *
1.108 + * Set the first code point in *pFirst.
1.109 + *
1.110 + * @param s Input char * string.
1.111 + * @param dest Output string buffer.
1.112 + * @param destCapacity Capacity of dest in numbers of UChars.
1.113 + * @param pFirst If pFirst!=NULL the *pFirst will be set to the first
1.114 + * code point in the string.
1.115 + * @param pErrorCode ICU error code.
1.116 + * @return The length of the string in numbers of UChars.
1.117 + */
1.118 +U_CAPI int32_t U_EXPORT2
1.119 +u_parseString(const char *s,
1.120 + UChar *dest, int32_t destCapacity,
1.121 + uint32_t *pFirst,
1.122 + UErrorCode *pErrorCode);
1.123 +
1.124 +/**
1.125 + * Parse a code point range like
1.126 + * 0085 or
1.127 + * 4E00..9FA5.
1.128 + *
1.129 + * s must contain such a range and end with either ';' or NUL.
1.130 + *
1.131 + * @return Length of code point range, end-start+1
1.132 + */
1.133 +U_CAPI int32_t U_EXPORT2
1.134 +u_parseCodePointRange(const char *s,
1.135 + uint32_t *pStart, uint32_t *pEnd,
1.136 + UErrorCode *pErrorCode);
1.137 +
1.138 +/**
1.139 + * Same as u_parseCodePointRange() but the range may be terminated by
1.140 + * any character. The position of the terminating character is returned via
1.141 + * the *terminator output parameter.
1.142 + */
1.143 +U_CAPI int32_t U_EXPORT2
1.144 +u_parseCodePointRangeAnyTerminator(const char *s,
1.145 + uint32_t *pStart, uint32_t *pEnd,
1.146 + const char **terminator,
1.147 + UErrorCode *pErrorCode);
1.148 +
1.149 +U_CAPI int32_t U_EXPORT2
1.150 +u_parseUTF8(const char *source, int32_t sLen, char *dest, int32_t destCapacity, UErrorCode *status);
1.151 +
1.152 +U_CDECL_END
1.153 +
1.154 +#endif
