static void
demoUnicodeStringInit() {
// *** Make sure to read about invariant characters in utypes.h! ***
// Initialization of Unicode strings from C literals works _only_ for
// invariant characters!
printf("\n* demoUnicodeStringInit() ---------- ***\n\n");
// the string literal is 32 chars long - this must be counted for the macro
UnicodeString invariantOnly=UNICODE_STRING("such characters are safe 123 %-.", 32);
/*
* In C, we need two macros: one to declare the UChar[] array, and
* one to populate it; the second one is a noop on platforms where
* wchar_t is compatible with UChar and ASCII-based.
* The length of the string literal must be counted for both macros.
*/
/* declare the invString array for the string */
U_STRING_DECL(invString, "such characters are safe 123 %-.", 32);
/* populate it with the characters */
U_STRING_INIT(invString, "such characters are safe 123 %-.", 32);
// compare the C and C++ strings
printf("C and C++ Unicode strings are equal: %d\n", invariantOnly==UnicodeString(TRUE, invString, 32));
/*
* convert between char * and UChar * strings that
* contain only invariant characters
*/
static const char *cs1="such characters are safe 123 %-.";
static UChar us1[40];
static char cs2[40];
u_charsToUChars(cs1, us1, 33); /* include the terminating NUL */
u_UCharsToChars(us1, cs2, 33);
printf("char * -> UChar * -> char * with only "
"invariant characters: \"%s\"\n",
cs2);
// initialize a UnicodeString from a string literal that contains
// escape sequences written with invariant characters
// do not forget to duplicate the backslashes for ICU to see them
// then, count each double backslash only once!
UnicodeString german=UNICODE_STRING(
"Sch\\u00f6nes Auto: \\u20ac 11240.\\fPrivates Zeichen: \\U00102345\\n", 64).
unescape();
printUnicodeString("german UnicodeString from unescaping:\n ", german);
/*
* C: convert and unescape a char * string with only invariant
* characters to fill a UChar * string
*/
UChar buffer[200];
int32_t length;
length=u_unescape(
"Sch\\u00f6nes Auto: \\u20ac 11240.\\fPrivates Zeichen: \\U00102345\\n",
buffer, UPRV_LENGTHOF(buffer));
printf("german C Unicode string from char * unescaping: (length %d)\n ", length);
printUnicodeString("", UnicodeString(buffer));
}
static void demoCaseMapInCPlusPlus() {
/*
* input=
* "aB<capital sigma>"
* "iI<small dotless i><capital dotted I> "
* "<sharp s> <small lig. ffi>"
* "<small final sigma><small sigma><capital sigma>"
*/
static const UChar input[]={
0x61, 0x42, 0x3a3,
0x69, 0x49, 0x131, 0x130, 0x20,
0xdf, 0x20, 0xfb03,
0x3c2, 0x3c3, 0x3a3, 0
};
printf("\n* demoCaseMapInCPlusPlus() --------- ***\n\n");
UnicodeString s(input), t;
const Locale &en=Locale::getEnglish();
Locale tr("tr");
/*
* Full case mappings as in demoCaseMapInC(), using UnicodeString functions.
* These functions modify the string object itself.
* Since we want to keep the input string around, we copy it each time
* and case-map the copy.
*/
printUnicodeString("input string: ", s);
/* lowercase/English */
printUnicodeString("full-lowercased/en: ", (t=s).toLower(en));
/* lowercase/Turkish */
printUnicodeString("full-lowercased/tr: ", (t=s).toLower(tr));
/* uppercase/English */
printUnicodeString("full-uppercased/en: ", (t=s).toUpper(en));
/* uppercase/Turkish */
printUnicodeString("full-uppercased/tr: ", (t=s).toUpper(tr));
/* titlecase/English */
printUnicodeString("full-titlecased/en: ", (t=s).toTitle(NULL, en));
/* titlecase/Turkish */
printUnicodeString("full-titlecased/tr: ", (t=s).toTitle(NULL, tr));
/* case-folde/default */
printUnicodeString("full-case-folded/default: ", (t=s).foldCase(U_FOLD_CASE_DEFAULT));
/* case-folde/Turkic */
printUnicodeString("full-case-folded/Turkic: ", (t=s).foldCase(U_FOLD_CASE_EXCLUDE_SPECIAL_I));
}
int DoConvert
(
char * lpInBuffer,
int nInLen,
char * lpOutBuffer,
int& rnOutLen
)
{
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "IcuTranslitEC.CppDoConvert() BEGIN\n");
#endif
int hr = 0;
UnicodeString sInOut;
#ifdef _MSC_VER
sInOut.setTo((UChar *)lpInBuffer, nInLen / 2);
#else
sInOut.setTo(lpInBuffer);
#endif
printUnicodeString("Will transliterate: ", sInOut);
if( m_bLTR )
{
if( m_pTForwards )
{
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "Doing forwards transliteration...\n");
#endif
m_pTForwards->transliterate(sInOut);
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "Did forwards transliteration.\n");
#endif
}
else
{
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "There is no forward transliterator.\n");
#endif
hr = -1;
}
}
else // !m_bLTR
{
if( m_pTBackwards )
{
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "Doing backwards transliteration...\n");
#endif
m_pTBackwards->transliterate(sInOut);
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "Did backwards transliteration.\n");
#endif
}
else
{
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "There is no backwards transliterator.\n");
#endif
hr = -1;
}
}
if (hr == 0)
{
printUnicodeString("Result of transliteration: ", sInOut);
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "sInOut.length %d\n", sInOut.length());
#endif
#ifdef _MSC_VER
UErrorCode err = U_ZERO_ERROR;
int nLen = sInOut.extract((UChar *)lpOutBuffer, rnOutLen/sizeof(UChar) , err);
if (nLen >= (int)rnOutLen || U_FAILURE(err))
#else
int nLen = sInOut.extract(0, sInOut.length(), (char *)NULL); // "preflight" to get size
if( nLen >= (int)rnOutLen )
#endif
{
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "Length %d more than output buffer size %d\n",
nLen, rnOutLen);
#endif
hr = -1;
}
else
{
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "nLen %d < original rnOutLen %d\n", nLen, rnOutLen);
#endif
#ifdef _MSC_VER
rnOutLen = nLen * sizeof(UChar);
#else
nLen = sInOut.extract(0, sInOut.length(), lpOutBuffer);
rnOutLen = nLen;
#endif
#ifdef VERBOSE_DEBUGGING
fprintf(stderr, "lpOutBuffer length = %u (should be %d)\n",
(unsigned)strlen(lpOutBuffer), rnOutLen);
fprintf(stderr, "lpOutBuffer: '%s'\n", lpOutBuffer);
#endif
}
}
return hr;
}
int main( void )
{
UFILE *out;
UErrorCode status = U_ZERO_ERROR;
out = u_finit(stdout, NULL, NULL);
if(!out) {
fprintf(stderr, "Could not initialize (finit()) over stdout! \n");
return 1;
}
ucnv_setFromUCallBack(u_fgetConverter(out), UCNV_FROM_U_CALLBACK_ESCAPE,
NULL, NULL, NULL, &status);
if(U_FAILURE(status)) {
u_fprintf(out, "Warning- couldn't set the substitute callback - err %s\n", u_errorName(status));
}
/* End Demo boilerplate */
u_fprintf(out,"ICU Case Mapping Sample Program\n\n");
u_fprintf(out, "C++ Case Mapping\n\n");
UnicodeString string("This is a test");
/* lowercase = "istanbul" */
UChar lowercase[] = {0x69, 0x73, 0x74, 0x61, 0x6e, 0x62, 0x75, 0x6c, 0};
/* uppercase = "LATIN CAPITAL I WITH DOT ABOVE STANBUL" */
UChar uppercase[] = {0x0130, 0x53, 0x54, 0x41, 0x4e, 0x42, 0x55, 0x4C, 0};
UnicodeString upper(uppercase);
UnicodeString lower(lowercase);
u_fprintf(out, "\nstring: ");
printUnicodeString(out, string);
string.toUpper(); /* string = "THIS IS A TEST" */
u_fprintf(out, "\ntoUpper(): ");
printUnicodeString(out, string);
string.toLower(); /* string = "this is a test" */
u_fprintf(out, "\ntoLower(): ");
printUnicodeString(out, string);
u_fprintf(out, "\n\nlowercase=%S, uppercase=%S\n", lowercase, uppercase);
string = upper;
string.toLower(Locale("tr", "TR")); /* Turkish lower case map string =
lowercase */
u_fprintf(out, "\nupper.toLower: ");
printUnicodeString(out, string);
string = lower;
string.toUpper(Locale("tr", "TR")); /* Turkish upper case map string =
uppercase */
u_fprintf(out, "\nlower.toUpper: ");
printUnicodeString(out, string);
u_fprintf(out, "\nEnd C++ sample\n\n");
// Call the C version
int rc = c_main(out);
u_fclose(out);
return rc;
}
static void
demoUnicodeStringStorage() {
// These sample code lines illustrate how to use UnicodeString, and the
// comments tell what happens internally. There are no APIs to observe
// most of this programmatically, except for stepping into the code
// with a debugger.
// This is by design to hide such details from the user.
int32_t i;
printf("\n* demoUnicodeStringStorage() ------- ***\n\n");
// * UnicodeString with internally stored contents
// instantiate a UnicodeString from a single code point
// the few (2) UChars will be stored in the object itself
UnicodeString one((UChar32)0x24001);
// this copies the few UChars into the "two" object
UnicodeString two=one;
printf("length of short string copy: %d\n", two.length());
// set "one" to contain the 3 UChars from readonly
// this setTo() variant copies the characters
one.setTo(readonly, UPRV_LENGTHOF(readonly));
// * UnicodeString with allocated contents
// build a longer string that will not fit into the object's buffer
one+=UnicodeString(writeable, UPRV_LENGTHOF(writeable));
one+=one;
one+=one;
printf("length of longer string: %d\n", one.length());
// copying will use the same allocated buffer and increment the reference
// counter
two=one;
printf("length of longer string copy: %d\n", two.length());
// * UnicodeString using readonly-alias to a const UChar array
// construct a string that aliases a readonly buffer
UnicodeString three(FALSE, readonly, UPRV_LENGTHOF(readonly));
printUnicodeString("readonly-alias string: ", three);
// copy-on-write: any modification to the string results in
// a copy to either the internal buffer or to a newly allocated one
three.setCharAt(1, 0x39);
printUnicodeString("readonly-aliasing string after modification: ", three);
// the aliased array is not modified
for(i=0; i<three.length(); ++i) {
printf("readonly buffer[%d] after modifying its string: 0x%lx\n",
i, readonly[i]);
}
// setTo() readonly alias
one.setTo(FALSE, writeable, UPRV_LENGTHOF(writeable));
// copying the readonly-alias object with fastCopyFrom() (new in ICU 2.4)
// will readonly-alias the same buffer
two.fastCopyFrom(one);
printUnicodeString("fastCopyFrom(readonly alias of \"writeable\" array): ", two);
printf("verify that a fastCopyFrom(readonly alias) uses the same buffer pointer: %d (should be 1)\n",
one.getBuffer()==two.getBuffer());
// a normal assignment will clone the contents (new in ICU 2.4)
two=one;
printf("verify that a regular copy of a readonly alias uses a different buffer pointer: %d (should be 0)\n",
one.getBuffer()==two.getBuffer());
// * UnicodeString using writeable-alias to a non-const UChar array
UnicodeString four(writeable, UPRV_LENGTHOF(writeable), UPRV_LENGTHOF(writeable));
printUnicodeString("writeable-alias string: ", four);
// a modification writes through to the buffer
four.setCharAt(1, 0x39);
for(i=0; i<four.length(); ++i) {
printf("writeable-alias backing buffer[%d]=0x%lx "
"after modification\n", i, writeable[i]);
}
// a copy will not alias any more;
// instead, it will get a copy of the contents into allocated memory
two=four;
two.setCharAt(1, 0x21);
for(i=0; i<two.length(); ++i) {
printf("writeable-alias backing buffer[%d]=0x%lx after "
"modification of string copy\n", i, writeable[i]);
}
// setTo() writeable alias, capacity==length
one.setTo(writeable, UPRV_LENGTHOF(writeable), UPRV_LENGTHOF(writeable));
// grow the string - it will not fit into the backing buffer any more
// and will get copied before modification
one.append((UChar)0x40);
// shrink it back so it would fit
one.truncate(one.length()-1);
// we still operate on the copy
one.setCharAt(1, 0x25);
printf("string after growing too much and then shrinking[1]=0x%lx\n"
" backing store for this[1]=0x%lx\n",
one.charAt(1), writeable[1]);
// if we need it in the original buffer, then extract() to it
// extract() does not do anything if the string aliases that same buffer
// i=min(one.length(), length of array)
if(one.length()<UPRV_LENGTHOF(writeable)) {
i=one.length();
} else {
i=UPRV_LENGTHOF(writeable);
}
one.extract(0, i, writeable);
for(i=0; i<UPRV_LENGTHOF(writeable); ++i) {
printf("writeable-alias backing buffer[%d]=0x%lx after re-extract\n",
i, writeable[i]);
}
}
