/src/icu/source/common/charstr.cpp

Source (jump to first uncovered line)
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*   Copyright (C) 2010-2015, International Business Machines
*   Corporation and others.  All Rights Reserved.
*******************************************************************************
*   file name:  charstr.cpp
*   encoding:   UTF-8
*   tab size:   8 (not used)
*   indentation:4
*
*   created on: 2010may19
*   created by: Markus W. Scherer
*/

#include <cstdlib>

#include "unicode/utypes.h"
#include "unicode/putil.h"
#include "charstr.h"
#include "cmemory.h"
#include "cstring.h"
#include "uinvchar.h"
#include "ustr_imp.h"

U_NAMESPACE_BEGIN

CharString::CharString(CharString&& src) U_NOEXCEPT
        : buffer(std::move(src.buffer)), len(src.len) {
    src.len = 0;  // not strictly necessary because we make no guarantees on the source string
}

CharString& CharString::operator=(CharString&& src) U_NOEXCEPT {
    buffer = std::move(src.buffer);
    len = src.len;
    src.len = 0;  // not strictly necessary because we make no guarantees on the source string
    return *this;
}

char *CharString::cloneData(UErrorCode &errorCode) const {
    if (U_FAILURE(errorCode)) { return nullptr; }
    char *p = static_cast<char *>(uprv_malloc(len + 1));
    if (p == nullptr) {
        errorCode = U_MEMORY_ALLOCATION_ERROR;
        return nullptr;
    }
    uprv_memcpy(p, buffer.getAlias(), len + 1);
    return p;
}

int32_t CharString::extract(char *dest, int32_t capacity, UErrorCode &errorCode) const {
    if (U_FAILURE(errorCode)) { return len; }
    if (capacity < 0 || (capacity > 0 && dest == nullptr)) {
        errorCode = U_ILLEGAL_ARGUMENT_ERROR;
        return len;
    }
    const char *src = buffer.getAlias();
    if (0 < len && len <= capacity && src != dest) {
        uprv_memcpy(dest, src, len);
    }
    return u_terminateChars(dest, capacity, len, &errorCode);
}

CharString &CharString::copyFrom(const CharString &s, UErrorCode &errorCode) {
    if(U_SUCCESS(errorCode) && this!=&s && ensureCapacity(s.len+1, 0, errorCode)) {
        len=s.len;
        uprv_memcpy(buffer.getAlias(), s.buffer.getAlias(), len+1);
    }
    return *this;
}

int32_t CharString::lastIndexOf(char c) const {
    for(int32_t i=len; i>0;) {
        if(buffer[--i]==c) {
            return i;
        }
    }
    return -1;
}

bool CharString::contains(StringPiece s) const {
    if (s.empty()) { return false; }
    const char *p = buffer.getAlias();
    int32_t lastStart = len - s.length();
    for (int32_t i = 0; i <= lastStart; ++i) {
        if (uprv_memcmp(p + i, s.data(), s.length()) == 0) {
            return true;
        }
    }
    return false;
}

CharString &CharString::truncate(int32_t newLength) {
    if(newLength<0) {
        newLength=0;
    }
    if(newLength<len) {
        buffer[len=newLength]=0;
    }
    return *this;
}

CharString &CharString::append(char c, UErrorCode &errorCode) {
    if(ensureCapacity(len+2, 0, errorCode)) {
        buffer[len++]=c;
        buffer[len]=0;
    }
    return *this;
}

CharString &CharString::append(const char *s, int32_t sLength, UErrorCode &errorCode) {
    if(U_FAILURE(errorCode)) {
        return *this;
    }
    if(sLength<-1 || (s==NULL && sLength!=0)) {
        errorCode=U_ILLEGAL_ARGUMENT_ERROR;
        return *this;
    }
    if(sLength<0) {
        sLength= static_cast<int32_t>(uprv_strlen(s));
    }
    if(sLength>0) {
        if(s==(buffer.getAlias()+len)) {
            // The caller wrote into the getAppendBuffer().
            if(sLength>=(buffer.getCapacity()-len)) {
                // The caller wrote too much.
                errorCode=U_INTERNAL_PROGRAM_ERROR;
            } else {
                buffer[len+=sLength]=0;
            }
        } else if(buffer.getAlias()<=s && s<(buffer.getAlias()+len) &&
                  sLength>=(buffer.getCapacity()-len)
        ) {
            // (Part of) this string is appended to itself which requires reallocation,
            // so we have to make a copy of the substring and append that.
            return append(CharString(s, sLength, errorCode), errorCode);
        } else if(ensureCapacity(len+sLength+1, 0, errorCode)) {
            uprv_memcpy(buffer.getAlias()+len, s, sLength);
            buffer[len+=sLength]=0;
        }
    }
    return *this;
}

CharString &CharString::appendNumber(int32_t number, UErrorCode &status) {
    if (number < 0) {
        this->append('-', status);
        if (U_FAILURE(status)) {
            return *this;
        }
    }

    if (number == 0) {
        this->append('0', status);
        return *this;
    }

    int32_t numLen = 0;
    while (number != 0) {
        int32_t residue = number % 10;
        number /= 10;
        this->append(std::abs(residue) + '0', status);
        numLen++;
        if (U_FAILURE(status)) {
            return *this;
        }
    }

    int32_t start = this->length() - numLen, end = this->length() - 1;
    while(start < end) {
        std::swap(this->data()[start++], this->data()[end--]);
    }

    return *this;
}

char *CharString::getAppendBuffer(int32_t minCapacity,
                                  int32_t desiredCapacityHint,
                                  int32_t &resultCapacity,
                                  UErrorCode &errorCode) {
    if(U_FAILURE(errorCode)) {
        resultCapacity=0;
        return NULL;
    }
    int32_t appendCapacity=buffer.getCapacity()-len-1;  // -1 for NUL
    if(appendCapacity>=minCapacity) {
        resultCapacity=appendCapacity;
        return buffer.getAlias()+len;
    }
    if(ensureCapacity(len+minCapacity+1, len+desiredCapacityHint+1, errorCode)) {
        resultCapacity=buffer.getCapacity()-len-1;
        return buffer.getAlias()+len;
    }
    resultCapacity=0;
    return NULL;
}

CharString &CharString::appendInvariantChars(const UnicodeString &s, UErrorCode &errorCode) {
    return appendInvariantChars(s.getBuffer(), s.length(), errorCode);
}

CharString &CharString::appendInvariantChars(const UChar* uchars, int32_t ucharsLen, UErrorCode &errorCode) {
    if(U_FAILURE(errorCode)) {
        return *this;
    }
    if (!uprv_isInvariantUString(uchars, ucharsLen)) {
        errorCode = U_INVARIANT_CONVERSION_ERROR;
        return *this;
    }
    if(ensureCapacity(len+ucharsLen+1, 0, errorCode)) {
        u_UCharsToChars(uchars, buffer.getAlias()+len, ucharsLen);
        len += ucharsLen;
        buffer[len] = 0;
    }
    return *this;
}

UBool CharString::ensureCapacity(int32_t capacity,
                                 int32_t desiredCapacityHint,
                                 UErrorCode &errorCode) {
    if(U_FAILURE(errorCode)) {
        return FALSE;
    }
    if(capacity>buffer.getCapacity()) {
        if(desiredCapacityHint==0) {
            desiredCapacityHint=capacity+buffer.getCapacity();
        }
        if( (desiredCapacityHint<=capacity || buffer.resize(desiredCapacityHint, len+1)==NULL) &&
            buffer.resize(capacity, len+1)==NULL
        ) {
            errorCode=U_MEMORY_ALLOCATION_ERROR;
            return FALSE;
        }
    }
    return TRUE;
}

CharString &CharString::appendPathPart(StringPiece s, UErrorCode &errorCode) {
    if(U_FAILURE(errorCode)) {
        return *this;
    }
    if(s.length()==0) {
        return *this;
    }
    char c;
    if(len>0 && (c=buffer[len-1])!=U_FILE_SEP_CHAR && c!=U_FILE_ALT_SEP_CHAR) {
        append(getDirSepChar(), errorCode);
    }
    append(s, errorCode);
    return *this;
}

CharString &CharString::ensureEndsWithFileSeparator(UErrorCode &errorCode) {
    char c;
    if(U_SUCCESS(errorCode) && len>0 &&
            (c=buffer[len-1])!=U_FILE_SEP_CHAR && c!=U_FILE_ALT_SEP_CHAR) {
        append(getDirSepChar(), errorCode);
    }
    return *this;
}

char CharString::getDirSepChar() const {
    char dirSepChar = U_FILE_SEP_CHAR;
#if (U_FILE_SEP_CHAR != U_FILE_ALT_SEP_CHAR)
    // We may need to return a different directory separator when building for Cygwin or MSYS2.
    if(len>0 && !uprv_strchr(data(), U_FILE_SEP_CHAR) && uprv_strchr(data(), U_FILE_ALT_SEP_CHAR))
        dirSepChar = U_FILE_ALT_SEP_CHAR;
#endif
    return dirSepChar;
}

U_NAMESPACE_END

Coverage Report

Created: 2025-06-24 06:43

Line	Count	Source (jump to first uncovered line)
1		// © 2016 and later: Unicode, Inc. and others.
2		// License & terms of use: http://www.unicode.org/copyright.html
3		/*
4		*******************************************************************************
5		* Copyright (C) 2010-2015, International Business Machines
6		* Corporation and others. All Rights Reserved.
7		*******************************************************************************
8		* file name: charstr.cpp
9		* encoding: UTF-8
10		* tab size: 8 (not used)
11		* indentation:4
12		*
13		* created on: 2010may19
14		* created by: Markus W. Scherer
15		*/
16
17		#include <cstdlib>
18
19		#include "unicode/utypes.h"
20		#include "unicode/putil.h"
21		#include "charstr.h"
22		#include "cmemory.h"
23		#include "cstring.h"
24		#include "uinvchar.h"
25		#include "ustr_imp.h"
26
27		U_NAMESPACE_BEGIN
28
29		CharString::CharString(CharString&& src) U_NOEXCEPT
30	0	: buffer(std::move(src.buffer)), len(src.len) {
31	0	src.len = 0; // not strictly necessary because we make no guarantees on the source string
32	0	}
33
34	0	CharString& CharString::operator=(CharString&& src) U_NOEXCEPT {
35	0	buffer = std::move(src.buffer);
36	0	len = src.len;
37	0	src.len = 0; // not strictly necessary because we make no guarantees on the source string
38	0	return *this;
39	0	}
40
41	0	char *CharString::cloneData(UErrorCode &errorCode) const {
42	0	if (U_FAILURE(errorCode)) { return nullptr; }
43	0	char p = static_cast<char >(uprv_malloc(len + 1));
44	0	if (p == nullptr) {
45	0	errorCode = U_MEMORY_ALLOCATION_ERROR;
46	0	return nullptr;
47	0	}
48	0	uprv_memcpy(p, buffer.getAlias(), len + 1);
49	0	return p;
50	0	}
51
52	0	int32_t CharString::extract(char *dest, int32_t capacity, UErrorCode &errorCode) const {
53	0	if (U_FAILURE(errorCode)) { return len; }
54	0	if (capacity < 0 \|\| (capacity > 0 && dest == nullptr)) {
55	0	errorCode = U_ILLEGAL_ARGUMENT_ERROR;
56	0	return len;
57	0	}
58	0	const char *src = buffer.getAlias();
59	0	if (0 < len && len <= capacity && src != dest) {
60	0	uprv_memcpy(dest, src, len);
61	0	}
62	0	return u_terminateChars(dest, capacity, len, &errorCode);
63	0	}
64
65	0	CharString &CharString::copyFrom(const CharString &s, UErrorCode &errorCode) {
66	0	if(U_SUCCESS(errorCode) && this!=&s && ensureCapacity(s.len+1, 0, errorCode)) {
67	0	len=s.len;
68	0	uprv_memcpy(buffer.getAlias(), s.buffer.getAlias(), len+1);
69	0	}
70	0	return *this;
71	0	}
72
73	0	int32_t CharString::lastIndexOf(char c) const {
74	0	for(int32_t i=len; i>0;) {
75	0	if(buffer[--i]==c) {
76	0	return i;
77	0	}
78	0	}
79	0	return -1;
80	0	}
81
82	0	bool CharString::contains(StringPiece s) const {
83	0	if (s.empty()) { return false; }
84	0	const char *p = buffer.getAlias();
85	0	int32_t lastStart = len - s.length();
86	0	for (int32_t i = 0; i <= lastStart; ++i) {
87	0	if (uprv_memcmp(p + i, s.data(), s.length()) == 0) {
88	0	return true;
89	0	}
90	0	}
91	0	return false;
92	0	}
93
94	0	CharString &CharString::truncate(int32_t newLength) {
95	0	if(newLength<0) {
96	0	newLength=0;
97	0	}
98	0	if(newLength<len) {
99	0	buffer[len=newLength]=0;
100	0	}
101	0	return *this;
102	0	}
103
104	0	CharString &CharString::append(char c, UErrorCode &errorCode) {
105	0	if(ensureCapacity(len+2, 0, errorCode)) {
106	0	buffer[len++]=c;
107	0	buffer[len]=0;
108	0	}
109	0	return *this;
110	0	}
111
112	0	CharString &CharString::append(const char *s, int32_t sLength, UErrorCode &errorCode) {
113	0	if(U_FAILURE(errorCode)) {
114	0	return *this;
115	0	}
116	0	if(sLength<-1 \|\| (s==NULL && sLength!=0)) {
117	0	errorCode=U_ILLEGAL_ARGUMENT_ERROR;
118	0	return *this;
119	0	}
120	0	if(sLength<0) {
121	0	sLength= static_cast<int32_t>(uprv_strlen(s));
122	0	}
123	0	if(sLength>0) {
124	0	if(s==(buffer.getAlias()+len)) {
125		// The caller wrote into the getAppendBuffer().
126	0	if(sLength>=(buffer.getCapacity()-len)) {
127		// The caller wrote too much.
128	0	errorCode=U_INTERNAL_PROGRAM_ERROR;
129	0	} else {
130	0	buffer[len+=sLength]=0;
131	0	}
132	0	} else if(buffer.getAlias()<=s && s<(buffer.getAlias()+len) &&
133	0	sLength>=(buffer.getCapacity()-len)
134	0	) {
135		// (Part of) this string is appended to itself which requires reallocation,
136		// so we have to make a copy of the substring and append that.
137	0	return append(CharString(s, sLength, errorCode), errorCode);
138	0	} else if(ensureCapacity(len+sLength+1, 0, errorCode)) {
139	0	uprv_memcpy(buffer.getAlias()+len, s, sLength);
140	0	buffer[len+=sLength]=0;
141	0	}
142	0	}
143	0	return *this;
144	0	}
145
146	0	CharString &CharString::appendNumber(int32_t number, UErrorCode &status) {
147	0	if (number < 0) {
148	0	this->append('-', status);
149	0	if (U_FAILURE(status)) {
150	0	return *this;
151	0	}
152	0	}
153
154	0	if (number == 0) {
155	0	this->append('0', status);
156	0	return *this;
157	0	}
158
159	0	int32_t numLen = 0;
160	0	while (number != 0) {
161	0	int32_t residue = number % 10;
162	0	number /= 10;
163	0	this->append(std::abs(residue) + '0', status);
164	0	numLen++;
165	0	if (U_FAILURE(status)) {
166	0	return *this;
167	0	}
168	0	}
169
170	0	int32_t start = this->length() - numLen, end = this->length() - 1;
171	0	while(start < end) {
172	0	std::swap(this->data()[start++], this->data()[end--]);
173	0	}
174
175	0	return *this;
176	0	}
177
178		char *CharString::getAppendBuffer(int32_t minCapacity,
179		int32_t desiredCapacityHint,
180		int32_t &resultCapacity,
181	0	UErrorCode &errorCode) {
182	0	if(U_FAILURE(errorCode)) {
183	0	resultCapacity=0;
184	0	return NULL;
185	0	}
186	0	int32_t appendCapacity=buffer.getCapacity()-len-1; // -1 for NUL
187	0	if(appendCapacity>=minCapacity) {
188	0	resultCapacity=appendCapacity;
189	0	return buffer.getAlias()+len;
190	0	}
191	0	if(ensureCapacity(len+minCapacity+1, len+desiredCapacityHint+1, errorCode)) {
192	0	resultCapacity=buffer.getCapacity()-len-1;
193	0	return buffer.getAlias()+len;
194	0	}
195	0	resultCapacity=0;
196	0	return NULL;
197	0	}
198
199	0	CharString &CharString::appendInvariantChars(const UnicodeString &s, UErrorCode &errorCode) {
200	0	return appendInvariantChars(s.getBuffer(), s.length(), errorCode);
201	0	}
202
203	0	CharString &CharString::appendInvariantChars(const UChar* uchars, int32_t ucharsLen, UErrorCode &errorCode) {
204	0	if(U_FAILURE(errorCode)) {
205	0	return *this;
206	0	}
207	0	if (!uprv_isInvariantUString(uchars, ucharsLen)) {
208	0	errorCode = U_INVARIANT_CONVERSION_ERROR;
209	0	return *this;
210	0	}
211	0	if(ensureCapacity(len+ucharsLen+1, 0, errorCode)) {
212	0	u_UCharsToChars(uchars, buffer.getAlias()+len, ucharsLen);
213	0	len += ucharsLen;
214	0	buffer[len] = 0;
215	0	}
216	0	return *this;
217	0	}
218
219		UBool CharString::ensureCapacity(int32_t capacity,
220		int32_t desiredCapacityHint,
221	0	UErrorCode &errorCode) {
222	0	if(U_FAILURE(errorCode)) {
223	0	return FALSE;
224	0	}
225	0	if(capacity>buffer.getCapacity()) {
226	0	if(desiredCapacityHint==0) {
227	0	desiredCapacityHint=capacity+buffer.getCapacity();
228	0	}
229	0	if( (desiredCapacityHint<=capacity \|\| buffer.resize(desiredCapacityHint, len+1)==NULL) &&
230	0	buffer.resize(capacity, len+1)==NULL
231	0	) {
232	0	errorCode=U_MEMORY_ALLOCATION_ERROR;
233	0	return FALSE;
234	0	}
235	0	}
236	0	return TRUE;
237	0	}
238
239	0	CharString &CharString::appendPathPart(StringPiece s, UErrorCode &errorCode) {
240	0	if(U_FAILURE(errorCode)) {
241	0	return *this;
242	0	}
243	0	if(s.length()==0) {
244	0	return *this;
245	0	}
246	0	char c;
247	0	if(len>0 && (c=buffer[len-1])!=U_FILE_SEP_CHAR && c!=U_FILE_ALT_SEP_CHAR) {
248	0	append(getDirSepChar(), errorCode);
249	0	}
250	0	append(s, errorCode);
251	0	return *this;
252	0	}
253
254	0	CharString &CharString::ensureEndsWithFileSeparator(UErrorCode &errorCode) {
255	0	char c;
256	0	if(U_SUCCESS(errorCode) && len>0 &&
257	0	(c=buffer[len-1])!=U_FILE_SEP_CHAR && c!=U_FILE_ALT_SEP_CHAR) {
258	0	append(getDirSepChar(), errorCode);
259	0	}
260	0	return *this;
261	0	}
262
263	0	char CharString::getDirSepChar() const {
264	0	char dirSepChar = U_FILE_SEP_CHAR;
265		#if (U_FILE_SEP_CHAR != U_FILE_ALT_SEP_CHAR)
266		// We may need to return a different directory separator when building for Cygwin or MSYS2.
267		if(len>0 && !uprv_strchr(data(), U_FILE_SEP_CHAR) && uprv_strchr(data(), U_FILE_ALT_SEP_CHAR))
268		dirSepChar = U_FILE_ALT_SEP_CHAR;
269		#endif
270	0	return dirSepChar;
271	0	}
272
273		U_NAMESPACE_END