1 /* 2 ****************************************************************************** 3 * 4 * Copyright (C) 1998-2010, International Business Machines 5 * Corporation and others. All Rights Reserved. 6 * 7 ****************************************************************************** 8 * 9 * File ustdio.h 10 * 11 * Modification History: 12 * 13 * Date Name Description 14 * 10/16/98 stephen Creation. 15 * 11/06/98 stephen Modified per code review. 16 * 03/12/99 stephen Modified for new C API. 17 * 07/19/99 stephen Minor doc update. 18 * 02/01/01 george Added sprintf & sscanf with all of its variants 19 ****************************************************************************** 20 */ 21 22 #ifndef USTDIO_H 23 #define USTDIO_H 24 25 #include <stdio.h> 26 #include <stdarg.h> 27 28 #include "unicode/utypes.h" 29 #include "unicode/ucnv.h" 30 #include "unicode/utrans.h" 31 #include "unicode/localpointer.h" 32 33 /* 34 TODO 35 The following is a small list as to what is currently wrong/suggestions for 36 ustdio. 37 38 * Make sure that * in the scanf format specification works for all formats. 39 * Each UFILE takes up at least 2KB. 40 Look into adding setvbuf() for configurable buffers. 41 * This library does buffering. The OS should do this for us already. Check on 42 this, and remove it from this library, if this is the case. Double buffering 43 wastes a lot of time and space. 44 * Test stdin and stdout with the u_f* functions 45 * Testing should be done for reading and writing multi-byte encodings, 46 and make sure that a character that is contained across buffer boundries 47 works even for incomplete characters. 48 * Make sure that the last character is flushed when the file/string is closed. 49 * snprintf should follow the C99 standard for the return value, which is 50 return the number of characters (excluding the trailing '\0') 51 which would have been written to the destination string regardless 52 of available space. This is like pre-flighting. 53 * Everything that uses %s should do what operator>> does for UnicodeString. 54 It should convert one byte at a time, and once a character is 55 converted then check to see if it's whitespace or in the scanset. 56 If it's whitespace or in the scanset, put all the bytes back (do nothing 57 for sprintf/sscanf). 58 * If bad string data is encountered, make sure that the function fails 59 without memory leaks and the unconvertable characters are valid 60 substitution or are escaped characters. 61 * u_fungetc() can't unget a character when it's at the beginning of the 62 internal conversion buffer. For example, read the buffer size # of 63 characters, and then ungetc to get the previous character that was 64 at the end of the last buffer. 65 * u_fflush() and u_fclose should return an int32_t like C99 functions. 66 0 is returned if the operation was successful and EOF otherwise. 67 * u_fsettransliterator does not support U_READ side of transliteration. 68 * The format specifier should limit the size of a format or honor it in 69 order to prevent buffer overruns. (e.g. %256.256d). 70 * u_fread and u_fwrite don't exist. They're needed for reading and writing 71 data structures without any conversion. 72 * u_file_read and u_file_write are used for writing strings. u_fgets and 73 u_fputs or u_fread and u_fwrite should be used to do this. 74 * The width parameter for all scanf formats, including scanset, needs 75 better testing. This prevents buffer overflows. 76 * Figure out what is suppose to happen when a codepage is changed midstream. 77 Maybe a flush or a rewind are good enough. 78 * Make sure that a UFile opened with "rw" can be used after using 79 u_fflush with a u_frewind. 80 * scanf(%i) should detect what type of number to use. 81 * Add more testing of the alternate format, %# 82 * Look at newline handling of fputs/puts 83 * Think more about codeunit/codepoint error handling/support in %S,%s,%C,%c,%[] 84 * Complete the file documentation with proper doxygen formatting. 85 See http://oss.software.ibm.com/pipermail/icu/2003-July/005647.html 86 */ 87 88 /** 89 * \file 90 * \brief C API: Unicode stdio-like API 91 * 92 * <h2>Unicode stdio-like C API</h2> 93 * 94 * <p>This API provides an stdio-like API wrapper around ICU's other 95 * formatting and parsing APIs. It is meant to ease the transition of adding 96 * Unicode support to a preexisting applications using stdio. The following 97 * is a small list of noticable differences between stdio and ICU I/O's 98 * ustdio implementation.</p> 99 * 100 * <ul> 101 * <li>Locale specific formatting and parsing is only done with file IO.</li> 102 * <li>u_fstropen can be used to simulate file IO with strings. 103 * This is similar to the iostream API, and it allows locale specific 104 * formatting and parsing to be used.</li> 105 * <li>This API provides uniform formatting and parsing behavior between 106 * platforms (unlike the standard stdio implementations found on various 107 * platforms).</li> 108 * <li>This API is better suited for text data handling than binary data 109 * handling when compared to the typical stdio implementation.</li> 110 * <li>You can specify a Transliterator while using the file IO.</li> 111 * <li>You can specify a file's codepage separately from the default 112 * system codepage.</li> 113 * </ul> 114 * 115 * <h2>Formatting and Parsing Specification</h2> 116 * 117 * General printf format:<br> 118 * %[format modifier][width][.precision][type modifier][format] 119 * 120 * General scanf format:<br> 121 * %[*][format modifier][width][type modifier][format] 122 * 123 <table cellspacing="3"> 124 <tr><td>format</td><td>default<br>printf<br>type</td><td>default<br>scanf<br>type</td><td>description</td></tr> 125 <tr><td>%E</td><td>double</td><td>float</td><td>Scientific with an uppercase exponent</td></tr> 126 <tr><td>%e</td><td>double</td><td>float</td><td>Scientific with a lowercase exponent</td></tr> 127 <tr><td>%G</td><td>double</td><td>float</td><td>Use %E or %f for best format</td></tr> 128 <tr><td>%g</td><td>double</td><td>float</td><td>Use %e or %f for best format</td></tr> 129 <tr><td>%f</td><td>double</td><td>float</td><td>Simple floating point without the exponent</td></tr> 130 <tr><td>%X</td><td>int32_t</td><td>int32_t</td><td>ustdio special uppercase hex radix formatting</td></tr> 131 <tr><td>%x</td><td>int32_t</td><td>int32_t</td><td>ustdio special lowercase hex radix formatting</td></tr> 132 <tr><td>%d</td><td>int32_t</td><td>int32_t</td><td>Decimal format</td></tr> 133 <tr><td>%i</td><td>int32_t</td><td>int32_t</td><td>Same as %d</td></tr> 134 <tr><td>%n</td><td>int32_t</td><td>int32_t</td><td>count (write the number of UTF-16 codeunits read/written)</td></tr> 135 <tr><td>%o</td><td>int32_t</td><td>int32_t</td><td>ustdio special octal radix formatting</td></tr> 136 <tr><td>%u</td><td>uint32_t</td><td>uint32_t</td><td>Decimal format</td></tr> 137 <tr><td>%p</td><td>void *</td><td>void *</td><td>Prints the pointer value</td></tr> 138 <tr><td>%s</td><td>char *</td><td>char *</td><td>Use default converter or specified converter from fopen</td></tr> 139 <tr><td>%c</td><td>char</td><td>char</td><td>Use default converter or specified converter from fopen<br> 140 When width is specified for scanf, this acts like a non-NULL-terminated char * string.<br> 141 By default, only one char is written.</td></tr> 142 <tr><td>%S</td><td>UChar *</td><td>UChar *</td><td>Null terminated UTF-16 string</td></tr> 143 <tr><td>%C</td><td>UChar</td><td>UChar</td><td>16-bit Unicode code unit<br> 144 When width is specified for scanf, this acts like a non-NULL-terminated UChar * string<br> 145 By default, only one codepoint is written.</td></tr> 146 <tr><td>%[]</td><td> </td><td>UChar *</td><td>Null terminated UTF-16 string which contains the filtered set of characters specified by the UnicodeSet</td></tr> 147 <tr><td>%%</td><td> </td><td> </td><td>Show a percent sign</td></tr> 148 </table> 149 150 Format modifiers 151 <table> 152 <tr><td>modifier</td><td>formats</td><td>type</td><td>comments</td></tr> 153 <tr><td>%h</td><td>%d, %i, %o, %x</td><td>int16_t</td><td>short format</td></tr> 154 <tr><td>%h</td><td>%u</td><td>uint16_t</td><td>short format</td></tr> 155 <tr><td>%h</td><td>c</td><td>char</td><td><b>(Unimplemented)</b> Use invariant converter</td></tr> 156 <tr><td>%h</td><td>s</td><td>char *</td><td><b>(Unimplemented)</b> Use invariant converter</td></tr> 157 <tr><td>%h</td><td>C</td><td>char</td><td><b>(Unimplemented)</b> 8-bit Unicode code unit</td></tr> 158 <tr><td>%h</td><td>S</td><td>char *</td><td><b>(Unimplemented)</b> Null terminated UTF-8 string</td></tr> 159 <tr><td>%l</td><td>%d, %i, %o, %x</td><td>int32_t</td><td>long format (no effect)</td></tr> 160 <tr><td>%l</td><td>%u</td><td>uint32_t</td><td>long format (no effect)</td></tr> 161 <tr><td>%l</td><td>c</td><td>N/A</td><td><b>(Unimplemented)</b> Reserved for future implementation</td></tr> 162 <tr><td>%l</td><td>s</td><td>N/A</td><td><b>(Unimplemented)</b> Reserved for future implementation</td></tr> 163 <tr><td>%l</td><td>C</td><td>UChar32</td><td><b>(Unimplemented)</b> 32-bit Unicode code unit</td></tr> 164 <tr><td>%l</td><td>S</td><td>UChar32 *</td><td><b>(Unimplemented)</b> Null terminated UTF-32 string</td></tr> 165 <tr><td>%ll</td><td>%d, %i, %o, %x</td><td>int64_t</td><td>long long format</td></tr> 166 <tr><td>%ll</td><td>%u</td><td>uint64_t</td><td><b>(Unimplemented)</b> long long format</td></tr> 167 <tr><td>%-</td><td><i>all</i></td><td>N/A</td><td>Left justify</td></tr> 168 <tr><td>%+</td><td>%d, %i, %o, %x, %e, %f, %g, %E, %G</td><td>N/A</td><td>Always show the plus or minus sign. Needs data for plus sign.</td></tr> 169 <tr><td>% </td><td>%d, %i, %o, %x, %e, %f, %g, %E, %G</td><td>N/A</td><td>Instead of a "+" output a blank character for positive numbers.</td></tr> 170 <tr><td>%#</td><td>%d, %i, %o, %x, %e, %f, %g, %E, %G</td><td>N/A</td><td>Precede octal value with 0, hex with 0x and show the 171 decimal point for floats.</td></tr> 172 <tr><td>%<i>n</i></td><td><i>all</i></td><td>N/A</td><td>Width of input/output. num is an actual number from 0 to 173 some large number.</td></tr> 174 <tr><td>%.<i>n</i></td><td>%e, %f, %g, %E, %F, %G</td><td>N/A</td><td>Significant digits precision. num is an actual number from 175 0 to some large number.<br>If * is used in printf, then the precision is passed in as an argument before the number to be formatted.</td></tr> 176 </table> 177 178 printf modifier 179 %* int32_t Next argument after this one specifies the width 180 181 scanf modifier 182 %* N/A This field is scanned, but not stored 183 184 <p>If you are using this C API instead of the ustream.h API for C++, 185 you can use one of the following u_fprintf examples to display a UnicodeString.</p> 186 187 <pre><code> 188 UFILE *out = u_finit(stdout, NULL, NULL); 189 UnicodeString string1("string 1"); 190 UnicodeString string2("string 2"); 191 u_fprintf(out, "%S\n", string1.getTerminatedBuffer()); 192 u_fprintf(out, "%.*S\n", string2.length(), string2.getBuffer()); 193 u_fclose(out); 194 </code></pre> 195 196 */ 197 198 199 /** 200 * When an end of file is encountered, this value can be returned. 201 * @see u_fgetc 202 * @stable 3.0 203 */ 204 #define U_EOF 0xFFFF 205 206 /** Forward declaration of a Unicode-aware file @stable 3.0 */ 207 typedef struct UFILE UFILE; 208 209 /** 210 * Enum for which direction of stream a transliterator applies to. 211 * @see u_fsettransliterator 212 * @stable ICU 3.0 213 */ 214 typedef enum { 215 U_READ = 1, 216 U_WRITE = 2, 217 U_READWRITE =3 /* == (U_READ | U_WRITE) */ 218 } UFileDirection; 219 220 /** 221 * Open a UFILE. 222 * A UFILE is a wrapper around a FILE* that is locale and codepage aware. 223 * That is, data written to a UFILE will be formatted using the conventions 224 * specified by that UFILE's Locale; this data will be in the character set 225 * specified by that UFILE's codepage. 226 * @param filename The name of the file to open. 227 * @param perm The read/write permission for the UFILE; one of "r", "w", "rw" 228 * @param locale The locale whose conventions will be used to format 229 * and parse output. If this parameter is NULL, the default locale will 230 * be used. 231 * @param codepage The codepage in which data will be written to and 232 * read from the file. If this paramter is NULL the system default codepage 233 * will be used. 234 * @return A new UFILE, or NULL if an error occurred. 235 * @stable ICU 3.0 236 */ 237 U_STABLE UFILE* U_EXPORT2 238 u_fopen(const char *filename, 239 const char *perm, 240 const char *locale, 241 const char *codepage); 242 243 /** 244 * Open a UFILE on top of an existing FILE* stream. The FILE* stream 245 * ownership remains with the caller. To have the UFILE take over 246 * ownership and responsibility for the FILE* stream, use the 247 * function u_fadopt. 248 * @param f The FILE* to which this UFILE will attach and use. 249 * @param locale The locale whose conventions will be used to format 250 * and parse output. If this parameter is NULL, the default locale will 251 * be used. 252 * @param codepage The codepage in which data will be written to and 253 * read from the file. If this paramter is NULL, data will be written and 254 * read using the default codepage for <TT>locale</TT>, unless <TT>locale</TT> 255 * is NULL, in which case the system default codepage will be used. 256 * @return A new UFILE, or NULL if an error occurred. 257 * @stable ICU 3.0 258 */ 259 U_STABLE UFILE* U_EXPORT2 260 u_finit(FILE *f, 261 const char *locale, 262 const char *codepage); 263 264 /** 265 * Open a UFILE on top of an existing FILE* stream. The FILE* stream 266 * ownership is transferred to the new UFILE. It will be closed when the 267 * UFILE is closed. 268 * @param f The FILE* which this UFILE will take ownership of. 269 * @param locale The locale whose conventions will be used to format 270 * and parse output. If this parameter is NULL, the default locale will 271 * be used. 272 * @param codepage The codepage in which data will be written to and 273 * read from the file. If this paramter is NULL, data will be written and 274 * read using the default codepage for <TT>locale</TT>, unless <TT>locale</TT> 275 * is NULL, in which case the system default codepage will be used. 276 * @return A new UFILE, or NULL if an error occurred. If an error occurs 277 * the ownership of the FILE* stream remains with the caller. 278 * @stable ICU 4.4 279 */ 280 U_STABLE UFILE* U_EXPORT2 281 u_fadopt(FILE *f, 282 const char *locale, 283 const char *codepage); 284 285 /** 286 * Create a UFILE that can be used for localized formatting or parsing. 287 * The u_sprintf and u_sscanf functions do not read or write numbers for a 288 * specific locale. The ustdio.h file functions can be used on this UFILE. 289 * The string is usable once u_fclose or u_fflush has been called on the 290 * returned UFILE. 291 * @param stringBuf The string used for reading or writing. 292 * @param capacity The number of code units available for use in stringBuf 293 * @param locale The locale whose conventions will be used to format 294 * and parse output. If this parameter is NULL, the default locale will 295 * be used. 296 * @return A new UFILE, or NULL if an error occurred. 297 * @stable ICU 3.0 298 */ 299 U_STABLE UFILE* U_EXPORT2 300 u_fstropen(UChar *stringBuf, 301 int32_t capacity, 302 const char *locale); 303 304 /** 305 * Close a UFILE. Implies u_fflush first. 306 * @param file The UFILE to close. 307 * @stable ICU 3.0 308 * @see u_fflush 309 */ 310 U_STABLE void U_EXPORT2 311 u_fclose(UFILE *file); 312 313 #if U_SHOW_CPLUSPLUS_API 314 315 U_NAMESPACE_BEGIN 316 317 /** 318 * \class LocalUFILEPointer 319 * "Smart pointer" class, closes a UFILE via u_fclose(). 320 * For most methods see the LocalPointerBase base class. 321 * 322 * @see LocalPointerBase 323 * @see LocalPointer 324 * @stable ICU 4.4 325 */ 326 U_DEFINE_LOCAL_OPEN_POINTER(LocalUFILEPointer, UFILE, u_fclose); 327 328 U_NAMESPACE_END 329 330 #endif 331 332 /** 333 * Tests if the UFILE is at the end of the file stream. 334 * @param f The UFILE from which to read. 335 * @return Returns TRUE after the first read operation that attempts to 336 * read past the end of the file. It returns FALSE if the current position is 337 * not end of file. 338 * @stable ICU 3.0 339 */ 340 U_STABLE UBool U_EXPORT2 341 u_feof(UFILE *f); 342 343 /** 344 * Flush output of a UFILE. Implies a flush of 345 * converter/transliterator state. (That is, a logical break is 346 * made in the output stream - for example if a different type of 347 * output is desired.) The underlying OS level file is also flushed. 348 * Note that for a stateful encoding, the converter may write additional 349 * bytes to return the stream to default state. 350 * @param file The UFILE to flush. 351 * @stable ICU 3.0 352 */ 353 U_STABLE void U_EXPORT2 354 u_fflush(UFILE *file); 355 356 /** 357 * Rewind the file pointer to the beginning of the file. 358 * @param file The UFILE to rewind. 359 * @stable ICU 3.0 360 */ 361 U_STABLE void 362 u_frewind(UFILE *file); 363 364 /** 365 * Get the FILE* associated with a UFILE. 366 * @param f The UFILE 367 * @return A FILE*, owned by the UFILE. The FILE <EM>must not</EM> be closed. 368 * @stable ICU 3.0 369 */ 370 U_STABLE FILE* U_EXPORT2 371 u_fgetfile(UFILE *f); 372 373 #if !UCONFIG_NO_FORMATTING 374 375 /** 376 * Get the locale whose conventions are used to format and parse output. 377 * This is the same locale passed in the preceding call to<TT>u_fsetlocale</TT> 378 * or <TT>u_fopen</TT>. 379 * @param file The UFILE to set. 380 * @return The locale whose conventions are used to format and parse output. 381 * @stable ICU 3.0 382 */ 383 U_STABLE const char* U_EXPORT2 384 u_fgetlocale(UFILE *file); 385 386 /** 387 * Set the locale whose conventions will be used to format and parse output. 388 * @param locale The locale whose conventions will be used to format 389 * and parse output. 390 * @param file The UFILE to query. 391 * @return NULL if successful, otherwise a negative number. 392 * @stable ICU 3.0 393 */ 394 U_STABLE int32_t U_EXPORT2 395 u_fsetlocale(UFILE *file, 396 const char *locale); 397 398 #endif 399 400 /** 401 * Get the codepage in which data is written to and read from the UFILE. 402 * This is the same codepage passed in the preceding call to 403 * <TT>u_fsetcodepage</TT> or <TT>u_fopen</TT>. 404 * @param file The UFILE to query. 405 * @return The codepage in which data is written to and read from the UFILE, 406 * or NULL if an error occurred. 407 * @stable ICU 3.0 408 */ 409 U_STABLE const char* U_EXPORT2 410 u_fgetcodepage(UFILE *file); 411 412 /** 413 * Set the codepage in which data will be written to and read from the UFILE. 414 * All Unicode data written to the UFILE will be converted to this codepage 415 * before it is written to the underlying FILE*. It it generally a bad idea to 416 * mix codepages within a file. This should only be called right 417 * after opening the <TT>UFile</TT>, or after calling <TT>u_frewind</TT>. 418 * @param codepage The codepage in which data will be written to 419 * and read from the file. For example <TT>"latin-1"</TT> or <TT>"ibm-943"</TT>. 420 * A value of NULL means the default codepage for the UFILE's current 421 * locale will be used. 422 * @param file The UFILE to set. 423 * @return 0 if successful, otherwise a negative number. 424 * @see u_frewind 425 * @stable ICU 3.0 426 */ 427 U_STABLE int32_t U_EXPORT2 428 u_fsetcodepage(const char *codepage, 429 UFILE *file); 430 431 432 /** 433 * Returns an alias to the converter being used for this file. 434 * @param f The UFILE to get the value from 435 * @return alias to the converter 436 * @stable ICU 3.0 437 */ 438 U_STABLE UConverter* U_EXPORT2 u_fgetConverter(UFILE *f); 439 440 #if !UCONFIG_NO_FORMATTING 441 442 /* Output functions */ 443 444 /** 445 * Write formatted data to a UFILE. 446 * @param f The UFILE to which to write. 447 * @param patternSpecification A pattern specifying how <TT>u_fprintf</TT> will 448 * interpret the variable arguments received and format the data. 449 * @return The number of Unicode characters written to <TT>f</TT>. 450 * @stable ICU 3.0 451 */ 452 U_STABLE int32_t U_EXPORT2 453 u_fprintf(UFILE *f, 454 const char *patternSpecification, 455 ... ); 456 457 /** 458 * Write formatted data to a UFILE. 459 * This is identical to <TT>u_fprintf</TT>, except that it will 460 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 461 * @param f The UFILE to which to write. 462 * @param patternSpecification A pattern specifying how <TT>u_fprintf</TT> will 463 * interpret the variable arguments received and format the data. 464 * @param ap The argument list to use. 465 * @return The number of Unicode characters written to <TT>f</TT>. 466 * @see u_fprintf 467 * @stable ICU 3.0 468 */ 469 U_STABLE int32_t U_EXPORT2 470 u_vfprintf(UFILE *f, 471 const char *patternSpecification, 472 va_list ap); 473 474 /** 475 * Write formatted data to a UFILE. 476 * @param f The UFILE to which to write. 477 * @param patternSpecification A pattern specifying how <TT>u_fprintf</TT> will 478 * interpret the variable arguments received and format the data. 479 * @return The number of Unicode characters written to <TT>f</TT>. 480 * @stable ICU 3.0 481 */ 482 U_STABLE int32_t U_EXPORT2 483 u_fprintf_u(UFILE *f, 484 const UChar *patternSpecification, 485 ... ); 486 487 /** 488 * Write formatted data to a UFILE. 489 * This is identical to <TT>u_fprintf_u</TT>, except that it will 490 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 491 * @param f The UFILE to which to write. 492 * @param patternSpecification A pattern specifying how <TT>u_fprintf</TT> will 493 * interpret the variable arguments received and format the data. 494 * @param ap The argument list to use. 495 * @return The number of Unicode characters written to <TT>f</TT>. 496 * @see u_fprintf_u 497 * @stable ICU 3.0 498 */ 499 U_STABLE int32_t U_EXPORT2 500 u_vfprintf_u(UFILE *f, 501 const UChar *patternSpecification, 502 va_list ap); 503 #endif 504 /** 505 * Write a Unicode to a UFILE. The null (U+0000) terminated UChar* 506 * <TT>s</TT> will be written to <TT>f</TT>, excluding the NULL terminator. 507 * A newline will be added to <TT>f</TT>. 508 * @param s The UChar* to write. 509 * @param f The UFILE to which to write. 510 * @return A non-negative number if successful, EOF otherwise. 511 * @see u_file_write 512 * @stable ICU 3.0 513 */ 514 U_STABLE int32_t U_EXPORT2 515 u_fputs(const UChar *s, 516 UFILE *f); 517 518 /** 519 * Write a UChar to a UFILE. 520 * @param uc The UChar to write. 521 * @param f The UFILE to which to write. 522 * @return The character written if successful, EOF otherwise. 523 * @stable ICU 3.0 524 */ 525 U_STABLE UChar32 U_EXPORT2 526 u_fputc(UChar32 uc, 527 UFILE *f); 528 529 /** 530 * Write Unicode to a UFILE. 531 * The ustring passed in will be converted to the UFILE's underlying 532 * codepage before it is written. 533 * @param ustring A pointer to the Unicode data to write. 534 * @param count The number of Unicode characters to write 535 * @param f The UFILE to which to write. 536 * @return The number of Unicode characters written. 537 * @see u_fputs 538 * @stable ICU 3.0 539 */ 540 U_STABLE int32_t U_EXPORT2 541 u_file_write(const UChar *ustring, 542 int32_t count, 543 UFILE *f); 544 545 546 /* Input functions */ 547 #if !UCONFIG_NO_FORMATTING 548 549 /** 550 * Read formatted data from a UFILE. 551 * @param f The UFILE from which to read. 552 * @param patternSpecification A pattern specifying how <TT>u_fscanf</TT> will 553 * interpret the variable arguments received and parse the data. 554 * @return The number of items successfully converted and assigned, or EOF 555 * if an error occurred. 556 * @stable ICU 3.0 557 */ 558 U_STABLE int32_t U_EXPORT2 559 u_fscanf(UFILE *f, 560 const char *patternSpecification, 561 ... ); 562 563 /** 564 * Read formatted data from a UFILE. 565 * This is identical to <TT>u_fscanf</TT>, except that it will 566 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 567 * @param f The UFILE from which to read. 568 * @param patternSpecification A pattern specifying how <TT>u_fscanf</TT> will 569 * interpret the variable arguments received and parse the data. 570 * @param ap The argument list to use. 571 * @return The number of items successfully converted and assigned, or EOF 572 * if an error occurred. 573 * @see u_fscanf 574 * @stable ICU 3.0 575 */ 576 U_STABLE int32_t U_EXPORT2 577 u_vfscanf(UFILE *f, 578 const char *patternSpecification, 579 va_list ap); 580 581 /** 582 * Read formatted data from a UFILE. 583 * @param f The UFILE from which to read. 584 * @param patternSpecification A pattern specifying how <TT>u_fscanf</TT> will 585 * interpret the variable arguments received and parse the data. 586 * @return The number of items successfully converted and assigned, or EOF 587 * if an error occurred. 588 * @stable ICU 3.0 589 */ 590 U_STABLE int32_t U_EXPORT2 591 u_fscanf_u(UFILE *f, 592 const UChar *patternSpecification, 593 ... ); 594 595 /** 596 * Read formatted data from a UFILE. 597 * This is identical to <TT>u_fscanf_u</TT>, except that it will 598 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 599 * @param f The UFILE from which to read. 600 * @param patternSpecification A pattern specifying how <TT>u_fscanf</TT> will 601 * interpret the variable arguments received and parse the data. 602 * @param ap The argument list to use. 603 * @return The number of items successfully converted and assigned, or EOF 604 * if an error occurred. 605 * @see u_fscanf_u 606 * @stable ICU 3.0 607 */ 608 U_STABLE int32_t U_EXPORT2 609 u_vfscanf_u(UFILE *f, 610 const UChar *patternSpecification, 611 va_list ap); 612 #endif 613 614 /** 615 * Read one line of text into a UChar* string from a UFILE. The newline 616 * at the end of the line is read into the string. The string is always 617 * null terminated 618 * @param f The UFILE from which to read. 619 * @param n The maximum number of characters - 1 to read. 620 * @param s The UChar* to receive the read data. Characters will be 621 * stored successively in <TT>s</TT> until a newline or EOF is 622 * reached. A null character (U+0000) will be appended to <TT>s</TT>. 623 * @return A pointer to <TT>s</TT>, or NULL if no characters were available. 624 * @stable ICU 3.0 625 */ 626 U_STABLE UChar* U_EXPORT2 627 u_fgets(UChar *s, 628 int32_t n, 629 UFILE *f); 630 631 /** 632 * Read a UChar from a UFILE. It is recommended that <TT>u_fgetcx</TT> 633 * used instead for proper parsing functions, but sometimes reading 634 * code units is needed instead of codepoints. 635 * 636 * @param f The UFILE from which to read. 637 * @return The UChar value read, or U+FFFF if no character was available. 638 * @stable ICU 3.0 639 */ 640 U_STABLE UChar U_EXPORT2 641 u_fgetc(UFILE *f); 642 643 /** 644 * Read a UChar32 from a UFILE. 645 * 646 * @param f The UFILE from which to read. 647 * @return The UChar32 value read, or U_EOF if no character was 648 * available, or U+FFFFFFFF if an ill-formed character was 649 * encountered. 650 * @see u_unescape() 651 * @stable ICU 3.0 652 */ 653 U_STABLE UChar32 U_EXPORT2 654 u_fgetcx(UFILE *f); 655 656 /** 657 * Unget a UChar from a UFILE. 658 * If this function is not the first to operate on <TT>f</TT> after a call 659 * to <TT>u_fgetc</TT>, the results are undefined. 660 * If this function is passed a character that was not recieved from the 661 * previous <TT>u_fgetc</TT> or <TT>u_fgetcx</TT> call, the results are undefined. 662 * @param c The UChar to put back on the stream. 663 * @param f The UFILE to receive <TT>c</TT>. 664 * @return The UChar32 value put back if successful, U_EOF otherwise. 665 * @stable ICU 3.0 666 */ 667 U_STABLE UChar32 U_EXPORT2 668 u_fungetc(UChar32 c, 669 UFILE *f); 670 671 /** 672 * Read Unicode from a UFILE. 673 * Bytes will be converted from the UFILE's underlying codepage, with 674 * subsequent conversion to Unicode. The data will not be NULL terminated. 675 * @param chars A pointer to receive the Unicode data. 676 * @param count The number of Unicode characters to read. 677 * @param f The UFILE from which to read. 678 * @return The number of Unicode characters read. 679 * @stable ICU 3.0 680 */ 681 U_STABLE int32_t U_EXPORT2 682 u_file_read(UChar *chars, 683 int32_t count, 684 UFILE *f); 685 686 #if !UCONFIG_NO_TRANSLITERATION 687 688 /** 689 * Set a transliterator on the UFILE. The transliterator will be owned by the 690 * UFILE. 691 * @param file The UFILE to set transliteration on 692 * @param adopt The UTransliterator to set. Can be NULL, which will 693 * mean that no transliteration is used. 694 * @param direction either U_READ, U_WRITE, or U_READWRITE - sets 695 * which direction the transliterator is to be applied to. If 696 * U_READWRITE, the "Read" transliteration will be in the inverse 697 * direction. 698 * @param status ICU error code. 699 * @return The previously set transliterator, owned by the 700 * caller. If U_READWRITE is specified, only the WRITE transliterator 701 * is returned. In most cases, the caller should call utrans_close() 702 * on the result of this function. 703 * @stable ICU 3.0 704 */ 705 U_STABLE UTransliterator* U_EXPORT2 706 u_fsettransliterator(UFILE *file, UFileDirection direction, 707 UTransliterator *adopt, UErrorCode *status); 708 709 #endif 710 711 712 /* Output string functions */ 713 #if !UCONFIG_NO_FORMATTING 714 715 716 /** 717 * Write formatted data to a Unicode string. 718 * 719 * @param buffer The Unicode String to which to write. 720 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will 721 * interpret the variable arguments received and format the data. 722 * @return The number of Unicode code units written to <TT>buffer</TT>. This 723 * does not include the terminating null character. 724 * @stable ICU 3.0 725 */ 726 U_STABLE int32_t U_EXPORT2 727 u_sprintf(UChar *buffer, 728 const char *patternSpecification, 729 ... ); 730 731 /** 732 * Write formatted data to a Unicode string. When the number of code units 733 * required to store the data exceeds <TT>count</TT>, then <TT>count</TT> code 734 * units of data are stored in <TT>buffer</TT> and a negative value is 735 * returned. When the number of code units required to store the data equals 736 * <TT>count</TT>, the string is not null terminated and <TT>count</TT> is 737 * returned. 738 * 739 * @param buffer The Unicode String to which to write. 740 * @param count The number of code units to read. 741 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will 742 * interpret the variable arguments received and format the data. 743 * @return The number of Unicode characters that would have been written to 744 * <TT>buffer</TT> had count been sufficiently large. This does not include 745 * the terminating null character. 746 * @stable ICU 3.0 747 */ 748 U_STABLE int32_t U_EXPORT2 749 u_snprintf(UChar *buffer, 750 int32_t count, 751 const char *patternSpecification, 752 ... ); 753 754 /** 755 * Write formatted data to a Unicode string. 756 * This is identical to <TT>u_sprintf</TT>, except that it will 757 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 758 * 759 * @param buffer The Unicode string to which to write. 760 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will 761 * interpret the variable arguments received and format the data. 762 * @param ap The argument list to use. 763 * @return The number of Unicode characters written to <TT>buffer</TT>. 764 * @see u_sprintf 765 * @stable ICU 3.0 766 */ 767 U_STABLE int32_t U_EXPORT2 768 u_vsprintf(UChar *buffer, 769 const char *patternSpecification, 770 va_list ap); 771 772 /** 773 * Write formatted data to a Unicode string. 774 * This is identical to <TT>u_snprintf</TT>, except that it will 775 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.<br><br> 776 * When the number of code units required to store the data exceeds 777 * <TT>count</TT>, then <TT>count</TT> code units of data are stored in 778 * <TT>buffer</TT> and a negative value is returned. When the number of code 779 * units required to store the data equals <TT>count</TT>, the string is not 780 * null terminated and <TT>count</TT> is returned. 781 * 782 * @param buffer The Unicode string to which to write. 783 * @param count The number of code units to read. 784 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will 785 * interpret the variable arguments received and format the data. 786 * @param ap The argument list to use. 787 * @return The number of Unicode characters that would have been written to 788 * <TT>buffer</TT> had count been sufficiently large. 789 * @see u_sprintf 790 * @stable ICU 3.0 791 */ 792 U_STABLE int32_t U_EXPORT2 793 u_vsnprintf(UChar *buffer, 794 int32_t count, 795 const char *patternSpecification, 796 va_list ap); 797 798 /** 799 * Write formatted data to a Unicode string. 800 * 801 * @param buffer The Unicode string to which to write. 802 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will 803 * interpret the variable arguments received and format the data. 804 * @return The number of Unicode characters written to <TT>buffer</TT>. 805 * @stable ICU 3.0 806 */ 807 U_STABLE int32_t U_EXPORT2 808 u_sprintf_u(UChar *buffer, 809 const UChar *patternSpecification, 810 ... ); 811 812 /** 813 * Write formatted data to a Unicode string. When the number of code units 814 * required to store the data exceeds <TT>count</TT>, then <TT>count</TT> code 815 * units of data are stored in <TT>buffer</TT> and a negative value is 816 * returned. When the number of code units required to store the data equals 817 * <TT>count</TT>, the string is not null terminated and <TT>count</TT> is 818 * returned. 819 * 820 * @param buffer The Unicode string to which to write. 821 * @param count The number of code units to read. 822 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will 823 * interpret the variable arguments received and format the data. 824 * @return The number of Unicode characters that would have been written to 825 * <TT>buffer</TT> had count been sufficiently large. 826 * @stable ICU 3.0 827 */ 828 U_STABLE int32_t U_EXPORT2 829 u_snprintf_u(UChar *buffer, 830 int32_t count, 831 const UChar *patternSpecification, 832 ... ); 833 834 /** 835 * Write formatted data to a Unicode string. 836 * This is identical to <TT>u_sprintf_u</TT>, except that it will 837 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 838 * 839 * @param buffer The Unicode string to which to write. 840 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will 841 * interpret the variable arguments received and format the data. 842 * @param ap The argument list to use. 843 * @return The number of Unicode characters written to <TT>f</TT>. 844 * @see u_sprintf_u 845 * @stable ICU 3.0 846 */ 847 U_STABLE int32_t U_EXPORT2 848 u_vsprintf_u(UChar *buffer, 849 const UChar *patternSpecification, 850 va_list ap); 851 852 /** 853 * Write formatted data to a Unicode string. 854 * This is identical to <TT>u_snprintf_u</TT>, except that it will 855 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 856 * When the number of code units required to store the data exceeds 857 * <TT>count</TT>, then <TT>count</TT> code units of data are stored in 858 * <TT>buffer</TT> and a negative value is returned. When the number of code 859 * units required to store the data equals <TT>count</TT>, the string is not 860 * null terminated and <TT>count</TT> is returned. 861 * 862 * @param buffer The Unicode string to which to write. 863 * @param count The number of code units to read. 864 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will 865 * interpret the variable arguments received and format the data. 866 * @param ap The argument list to use. 867 * @return The number of Unicode characters that would have been written to 868 * <TT>f</TT> had count been sufficiently large. 869 * @see u_sprintf_u 870 * @stable ICU 3.0 871 */ 872 U_STABLE int32_t U_EXPORT2 873 u_vsnprintf_u(UChar *buffer, 874 int32_t count, 875 const UChar *patternSpecification, 876 va_list ap); 877 878 /* Input string functions */ 879 880 /** 881 * Read formatted data from a Unicode string. 882 * 883 * @param buffer The Unicode string from which to read. 884 * @param patternSpecification A pattern specifying how <TT>u_sscanf</TT> will 885 * interpret the variable arguments received and parse the data. 886 * @return The number of items successfully converted and assigned, or EOF 887 * if an error occurred. 888 * @stable ICU 3.0 889 */ 890 U_STABLE int32_t U_EXPORT2 891 u_sscanf(const UChar *buffer, 892 const char *patternSpecification, 893 ... ); 894 895 /** 896 * Read formatted data from a Unicode string. 897 * This is identical to <TT>u_sscanf</TT>, except that it will 898 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 899 * 900 * @param buffer The Unicode string from which to read. 901 * @param patternSpecification A pattern specifying how <TT>u_sscanf</TT> will 902 * interpret the variable arguments received and parse the data. 903 * @param ap The argument list to use. 904 * @return The number of items successfully converted and assigned, or EOF 905 * if an error occurred. 906 * @see u_sscanf 907 * @stable ICU 3.0 908 */ 909 U_STABLE int32_t U_EXPORT2 910 u_vsscanf(const UChar *buffer, 911 const char *patternSpecification, 912 va_list ap); 913 914 /** 915 * Read formatted data from a Unicode string. 916 * 917 * @param buffer The Unicode string from which to read. 918 * @param patternSpecification A pattern specifying how <TT>u_sscanf</TT> will 919 * interpret the variable arguments received and parse the data. 920 * @return The number of items successfully converted and assigned, or EOF 921 * if an error occurred. 922 * @stable ICU 3.0 923 */ 924 U_STABLE int32_t U_EXPORT2 925 u_sscanf_u(const UChar *buffer, 926 const UChar *patternSpecification, 927 ... ); 928 929 /** 930 * Read formatted data from a Unicode string. 931 * This is identical to <TT>u_sscanf_u</TT>, except that it will 932 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>. 933 * 934 * @param buffer The Unicode string from which to read. 935 * @param patternSpecification A pattern specifying how <TT>u_sscanf</TT> will 936 * interpret the variable arguments received and parse the data. 937 * @param ap The argument list to use. 938 * @return The number of items successfully converted and assigned, or EOF 939 * if an error occurred. 940 * @see u_sscanf_u 941 * @stable ICU 3.0 942 */ 943 U_STABLE int32_t U_EXPORT2 944 u_vsscanf_u(const UChar *buffer, 945 const UChar *patternSpecification, 946 va_list ap); 947 948 #endif 949 #endif 950 951 952