• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 /*
2 *******************************************************************************
3 *
4 *   Copyright (C) 2000-2010, International Business Machines
5 *   Corporation and others.  All Rights Reserved.
6 *
7 *******************************************************************************
8 *   file name:  uparse.h
9 *   encoding:   US-ASCII
10 *   tab size:   8 (not used)
11 *   indentation:4
12 *
13 *   created on: 2000apr18
14 *   created by: Markus W. Scherer
15 *
16 *   This file provides a parser for files that are delimited by one single
17 *   character like ';' or TAB. Example: the Unicode Character Properties files
18 *   like UnicodeData.txt are semicolon-delimited.
19 */
20 
21 #ifndef __UPARSE_H__
22 #define __UPARSE_H__
23 
24 #include "unicode/utypes.h"
25 
26 /**
27  * Is c an invariant-character whitespace?
28  * @param c invariant character
29  */
30 #define U_IS_INV_WHITESPACE(c) ((c)==' ' || (c)=='\t' || (c)=='\r' || (c)=='\n')
31 
32 U_CDECL_BEGIN
33 
34 /**
35  * Skip space ' ' and TAB '\t' characters.
36  *
37  * @param s Pointer to characters.
38  * @return Pointer to first character at or after s that is not a space or TAB.
39  */
40 U_CAPI const char * U_EXPORT2
41 u_skipWhitespace(const char *s);
42 
43 /**
44  * Trim whitespace (including line endings) from the end of the string.
45  *
46  * @param s Pointer to the string.
47  * @return Pointer to the new end of the string.
48  */
49 U_CAPI char * U_EXPORT2
50 u_rtrim(char *s);
51 
52 /** Function type for u_parseDelimitedFile(). */
53 typedef void U_CALLCONV
54 UParseLineFn(void *context,
55               char *fields[][2],
56               int32_t fieldCount,
57               UErrorCode *pErrorCode);
58 
59 /**
60  * Parser for files that are similar to UnicodeData.txt:
61  * This function opens the file and reads it line by line. It skips empty lines
62  * and comment lines that start with a '#'.
63  * All other lines are separated into fields with one delimiter character
64  * (semicolon for Unicode Properties files) between two fields. The last field in
65  * a line does not need to be terminated with a delimiter.
66  *
67  * For each line, after segmenting it, a line function is called.
68  * It gets passed the array of field start and limit pointers that is
69  * passed into this parser and filled by it for each line.
70  * For each field i of the line, the start pointer in fields[i][0]
71  * points to the beginning of the field, while the limit pointer in fields[i][1]
72  * points behind the field, i.e., to the delimiter or the line end.
73  *
74  * The context parameter of the line function is
75  * the same as the one for the parse function.
76  *
77  * The line function may modify the contents of the fields including the
78  * limit characters.
79  *
80  * If the file cannot be opened, or there is a parsing error or a field function
81  * sets *pErrorCode, then the parser returns with *pErrorCode set to an error code.
82  */
83 U_CAPI void U_EXPORT2
84 u_parseDelimitedFile(const char *filename, char delimiter,
85                      char *fields[][2], int32_t fieldCount,
86                      UParseLineFn *lineFn, void *context,
87                      UErrorCode *pErrorCode);
88 
89 /**
90  * Parse a string of code points like 0061 0308 0300.
91  * s must end with either ';' or NUL.
92  *
93  * @return Number of code points.
94  */
95 U_CAPI int32_t U_EXPORT2
96 u_parseCodePoints(const char *s,
97                   uint32_t *dest, int32_t destCapacity,
98                   UErrorCode *pErrorCode);
99 
100 /**
101  * Parse a list of code points like 0061 0308 0300
102  * into a UChar * string.
103  * s must end with either ';' or NUL.
104  *
105  * Set the first code point in *pFirst.
106  *
107  * @param s Input char * string.
108  * @param dest Output string buffer.
109  * @param destCapacity Capacity of dest in numbers of UChars.
110  * @param pFirst If pFirst!=NULL the *pFirst will be set to the first
111  *               code point in the string.
112  * @param pErrorCode ICU error code.
113  * @return The length of the string in numbers of UChars.
114  */
115 U_CAPI int32_t U_EXPORT2
116 u_parseString(const char *s,
117               UChar *dest, int32_t destCapacity,
118               uint32_t *pFirst,
119               UErrorCode *pErrorCode);
120 
121 /**
122  * Parse a code point range like
123  * 0085 or
124  * 4E00..9FA5.
125  *
126  * s must contain such a range and end with either ';' or NUL.
127  *
128  * @return Length of code point range, end-start+1
129  */
130 U_CAPI int32_t U_EXPORT2
131 u_parseCodePointRange(const char *s,
132                       uint32_t *pStart, uint32_t *pEnd,
133                       UErrorCode *pErrorCode);
134 
135 /**
136  * Same as u_parseCodePointRange() but the range may be terminated by
137  * any character. The position of the terminating character is returned via
138  * the *terminator output parameter.
139  */
140 U_CAPI int32_t U_EXPORT2
141 u_parseCodePointRangeAnyTerminator(const char *s,
142                                    uint32_t *pStart, uint32_t *pEnd,
143                                    const char **terminator,
144                                    UErrorCode *pErrorCode);
145 
146 U_CAPI int32_t U_EXPORT2
147 u_parseUTF8(const char *source, int32_t sLen, char *dest, int32_t destCapacity, UErrorCode *status);
148 
149 U_CDECL_END
150 
151 #endif
152