summaryrefslogtreecommitdiffstats
path: root/intl/icu/source/io/ustream.cpp
blob: fd783b3b2394f40f1cb8e66be79da567773ae2a9 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
**********************************************************************
*   Copyright (C) 2001-2016, International Business Machines
*   Corporation and others.  All Rights Reserved.
**********************************************************************
*  FILE NAME : ustream.cpp
*
*   Modification History:
*
*   Date        Name        Description
*   06/25/2001  grhoten     Move iostream from unistr.h to here
******************************************************************************
*/

#include "unicode/utypes.h"

#if !UCONFIG_NO_CONVERSION

#include "unicode/uobject.h"
#include "unicode/ustream.h"
#include "unicode/ucnv.h"
#include "unicode/uchar.h"
#include "unicode/utf16.h"
#include "ustr_cnv.h"
#include "cmemory.h"
#include <string.h>

// console IO

#define STD_NAMESPACE std::

#define STD_OSTREAM STD_NAMESPACE ostream
#define STD_ISTREAM STD_NAMESPACE istream

U_NAMESPACE_BEGIN

U_IO_API STD_OSTREAM & U_EXPORT2
operator<<(STD_OSTREAM& stream, const UnicodeString& str)
{
    if(str.length() > 0) {
        char buffer[200];
        UConverter *converter;
        UErrorCode errorCode = U_ZERO_ERROR;

        // use the default converter to convert chunks of text
        converter = u_getDefaultConverter(&errorCode);
        if(U_SUCCESS(errorCode)) {
            const char16_t *us = str.getBuffer();
            const char16_t *uLimit = us + str.length();
            char *s, *sLimit = buffer + (sizeof(buffer) - 1);
            do {
                errorCode = U_ZERO_ERROR;
                s = buffer;
                ucnv_fromUnicode(converter, &s, sLimit, &us, uLimit, 0, false, &errorCode);
                *s = 0;

                // write this chunk
                if(s > buffer) {
                    stream << buffer;
                }
            } while(errorCode == U_BUFFER_OVERFLOW_ERROR);
            u_releaseDefaultConverter(converter);
        }
    }

/*    stream.flush();*/
    return stream;
}

U_IO_API STD_ISTREAM & U_EXPORT2
operator>>(STD_ISTREAM& stream, UnicodeString& str)
{
    // This is like ICU status checking.
    if (stream.fail()) {
        return stream;
    }

    /* ipfx should eat whitespace when ios::skipws is set */
    char16_t uBuffer[16];
    char buffer[16];
    int32_t idx = 0;
    UConverter *converter;
    UErrorCode errorCode = U_ZERO_ERROR;

    // use the default converter to convert chunks of text
    converter = u_getDefaultConverter(&errorCode);
    if(U_SUCCESS(errorCode)) {
        char16_t *us = uBuffer;
        const char16_t *uLimit = uBuffer + UPRV_LENGTHOF(uBuffer);
        const char *s, *sLimit;
        char ch;
        char16_t ch32;
        UBool initialWhitespace = true;
        UBool continueReading = true;

        /* We need to consume one byte at a time to see what is considered whitespace. */
        while (continueReading) {
            ch = stream.get();
            if (stream.eof()) {
                // The EOF is only set after the get() of an unavailable byte.
                if (!initialWhitespace) {
                    stream.clear(stream.eofbit);
                }
                continueReading = false;
            }
            sLimit = &ch + (int)continueReading;
            us = uBuffer;
            s = &ch;
            errorCode = U_ZERO_ERROR;
            /*
            Since we aren't guaranteed to see the state before this call,
            this code won't work on stateful encodings like ISO-2022 or an EBCDIC stateful encoding.
            We flush on the last byte to ensure that we output truncated multibyte characters.
            */
            ucnv_toUnicode(converter, &us, uLimit, &s, sLimit, 0, !continueReading, &errorCode);
            if(U_FAILURE(errorCode)) {
                /* Something really bad happened. setstate() isn't always an available API */
                stream.clear(stream.failbit);
                goto STOP_READING;
            }
            /* Was the character consumed? */
            if (us != uBuffer) {
                /* Reminder: ibm-1390 & JISX0213 can output 2 Unicode code points */
                int32_t uBuffSize = static_cast<int32_t>(us-uBuffer);
                int32_t uBuffIdx = 0;
                while (uBuffIdx < uBuffSize) {
                    U16_NEXT(uBuffer, uBuffIdx, uBuffSize, ch32);
                    if (u_isWhitespace(ch32)) {
                        if (!initialWhitespace) {
                            buffer[idx++] = ch;
                            while (idx > 0) {
                                stream.putback(buffer[--idx]);
                            }
                            goto STOP_READING;
                        }
                        /* else skip intialWhitespace */
                    }
                    else {
                        if (initialWhitespace) {
                            /*
                            When initialWhitespace is true, we haven't appended any
                            character yet.  This is where we truncate the string,
                            to avoid modifying the string before we know if we can
                            actually read from the stream.
                            */
                            str.truncate(0);
                            initialWhitespace = false;
                        }
                        str.append(ch32);
                    }
                }
                idx = 0;
            }
            else {
                buffer[idx++] = ch;
            }
        }
STOP_READING:
        u_releaseDefaultConverter(converter);
    }

/*    stream.flush();*/
    return stream;
}

U_NAMESPACE_END

#endif