2 ******************************************************************************
4 * Copyright (C) 1998-2014, International Business Machines
5 * Corporation and others. All Rights Reserved.
7 ******************************************************************************
11 * Modification History:
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 ******************************************************************************
28 #include "unicode/utypes.h"
29 #include "unicode/ucnv.h"
30 #include "unicode/utrans.h"
31 #include "unicode/localpointer.h"
32 #include "unicode/unum.h"
34 #if !UCONFIG_NO_CONVERSION
38 The following is a small list as to what is currently wrong/suggestions for
41 * Make sure that * in the scanf format specification works for all formats.
42 * Each UFILE takes up at least 2KB.
43 Look into adding setvbuf() for configurable buffers.
44 * This library does buffering. The OS should do this for us already. Check on
45 this, and remove it from this library, if this is the case. Double buffering
46 wastes a lot of time and space.
47 * Test stdin and stdout with the u_f* functions
48 * Testing should be done for reading and writing multi-byte encodings,
49 and make sure that a character that is contained across buffer boundries
50 works even for incomplete characters.
51 * Make sure that the last character is flushed when the file/string is closed.
52 * snprintf should follow the C99 standard for the return value, which is
53 return the number of characters (excluding the trailing '\0')
54 which would have been written to the destination string regardless
55 of available space. This is like pre-flighting.
56 * Everything that uses %s should do what operator>> does for UnicodeString.
57 It should convert one byte at a time, and once a character is
58 converted then check to see if it's whitespace or in the scanset.
59 If it's whitespace or in the scanset, put all the bytes back (do nothing
61 * If bad string data is encountered, make sure that the function fails
62 without memory leaks and the unconvertable characters are valid
63 substitution or are escaped characters.
64 * u_fungetc() can't unget a character when it's at the beginning of the
65 internal conversion buffer. For example, read the buffer size # of
66 characters, and then ungetc to get the previous character that was
67 at the end of the last buffer.
68 * u_fflush() and u_fclose should return an int32_t like C99 functions.
69 0 is returned if the operation was successful and EOF otherwise.
70 * u_fsettransliterator does not support U_READ side of transliteration.
71 * The format specifier should limit the size of a format or honor it in
72 order to prevent buffer overruns. (e.g. %256.256d).
73 * u_fread and u_fwrite don't exist. They're needed for reading and writing
74 data structures without any conversion.
75 * u_file_read and u_file_write are used for writing strings. u_fgets and
76 u_fputs or u_fread and u_fwrite should be used to do this.
77 * The width parameter for all scanf formats, including scanset, needs
78 better testing. This prevents buffer overflows.
79 * Figure out what is suppose to happen when a codepage is changed midstream.
80 Maybe a flush or a rewind are good enough.
81 * Make sure that a UFile opened with "rw" can be used after using
82 u_fflush with a u_frewind.
83 * scanf(%i) should detect what type of number to use.
84 * Add more testing of the alternate format, %#
85 * Look at newline handling of fputs/puts
86 * Think more about codeunit/codepoint error handling/support in %S,%s,%C,%c,%[]
87 * Complete the file documentation with proper doxygen formatting.
88 See http://oss.software.ibm.com/pipermail/icu/2003-July/005647.html
93 * \brief C API: Unicode stdio-like API
95 * <h2>Unicode stdio-like C API</h2>
97 * <p>This API provides an stdio-like API wrapper around ICU's other
98 * formatting and parsing APIs. It is meant to ease the transition of adding
99 * Unicode support to a preexisting applications using stdio. The following
100 * is a small list of noticable differences between stdio and ICU I/O's
101 * ustdio implementation.</p>
104 * <li>Locale specific formatting and parsing is only done with file IO.</li>
105 * <li>u_fstropen can be used to simulate file IO with strings.
106 * This is similar to the iostream API, and it allows locale specific
107 * formatting and parsing to be used.</li>
108 * <li>This API provides uniform formatting and parsing behavior between
109 * platforms (unlike the standard stdio implementations found on various
111 * <li>This API is better suited for text data handling than binary data
112 * handling when compared to the typical stdio implementation.</li>
113 * <li>You can specify a Transliterator while using the file IO.</li>
114 * <li>You can specify a file's codepage separately from the default
115 * system codepage.</li>
118 * <h2>Formatting and Parsing Specification</h2>
120 * General printf format:<br>
121 * %[format modifier][width][.precision][type modifier][format]
123 * General scanf format:<br>
124 * %[*][format modifier][width][type modifier][format]
126 <table cellspacing="3">
127 <tr><td>format</td><td>default<br>printf<br>type</td><td>default<br>scanf<br>type</td><td>description</td></tr>
128 <tr><td>%E</td><td>double</td><td>float</td><td>Scientific with an uppercase exponent</td></tr>
129 <tr><td>%e</td><td>double</td><td>float</td><td>Scientific with a lowercase exponent</td></tr>
130 <tr><td>%G</td><td>double</td><td>float</td><td>Use %E or %f for best format</td></tr>
131 <tr><td>%g</td><td>double</td><td>float</td><td>Use %e or %f for best format</td></tr>
132 <tr><td>%f</td><td>double</td><td>float</td><td>Simple floating point without the exponent</td></tr>
133 <tr><td>%X</td><td>int32_t</td><td>int32_t</td><td>ustdio special uppercase hex radix formatting</td></tr>
134 <tr><td>%x</td><td>int32_t</td><td>int32_t</td><td>ustdio special lowercase hex radix formatting</td></tr>
135 <tr><td>%d</td><td>int32_t</td><td>int32_t</td><td>Decimal format</td></tr>
136 <tr><td>%i</td><td>int32_t</td><td>int32_t</td><td>Same as %d</td></tr>
137 <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>
138 <tr><td>%o</td><td>int32_t</td><td>int32_t</td><td>ustdio special octal radix formatting</td></tr>
139 <tr><td>%u</td><td>uint32_t</td><td>uint32_t</td><td>Decimal format</td></tr>
140 <tr><td>%p</td><td>void *</td><td>void *</td><td>Prints the pointer value</td></tr>
141 <tr><td>%s</td><td>char *</td><td>char *</td><td>Use default converter or specified converter from fopen</td></tr>
142 <tr><td>%c</td><td>char</td><td>char</td><td>Use default converter or specified converter from fopen<br>
143 When width is specified for scanf, this acts like a non-NULL-terminated char * string.<br>
144 By default, only one char is written.</td></tr>
145 <tr><td>%S</td><td>UChar *</td><td>UChar *</td><td>Null terminated UTF-16 string</td></tr>
146 <tr><td>%C</td><td>UChar</td><td>UChar</td><td>16-bit Unicode code unit<br>
147 When width is specified for scanf, this acts like a non-NULL-terminated UChar * string<br>
148 By default, only one codepoint is written.</td></tr>
149 <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>
150 <tr><td>%%</td><td> </td><td> </td><td>Show a percent sign</td></tr>
155 <tr><td>modifier</td><td>formats</td><td>type</td><td>comments</td></tr>
156 <tr><td>%h</td><td>%d, %i, %o, %x</td><td>int16_t</td><td>short format</td></tr>
157 <tr><td>%h</td><td>%u</td><td>uint16_t</td><td>short format</td></tr>
158 <tr><td>%h</td><td>c</td><td>char</td><td><b>(Unimplemented)</b> Use invariant converter</td></tr>
159 <tr><td>%h</td><td>s</td><td>char *</td><td><b>(Unimplemented)</b> Use invariant converter</td></tr>
160 <tr><td>%h</td><td>C</td><td>char</td><td><b>(Unimplemented)</b> 8-bit Unicode code unit</td></tr>
161 <tr><td>%h</td><td>S</td><td>char *</td><td><b>(Unimplemented)</b> Null terminated UTF-8 string</td></tr>
162 <tr><td>%l</td><td>%d, %i, %o, %x</td><td>int32_t</td><td>long format (no effect)</td></tr>
163 <tr><td>%l</td><td>%u</td><td>uint32_t</td><td>long format (no effect)</td></tr>
164 <tr><td>%l</td><td>c</td><td>N/A</td><td><b>(Unimplemented)</b> Reserved for future implementation</td></tr>
165 <tr><td>%l</td><td>s</td><td>N/A</td><td><b>(Unimplemented)</b> Reserved for future implementation</td></tr>
166 <tr><td>%l</td><td>C</td><td>UChar32</td><td><b>(Unimplemented)</b> 32-bit Unicode code unit</td></tr>
167 <tr><td>%l</td><td>S</td><td>UChar32 *</td><td><b>(Unimplemented)</b> Null terminated UTF-32 string</td></tr>
168 <tr><td>%ll</td><td>%d, %i, %o, %x</td><td>int64_t</td><td>long long format</td></tr>
169 <tr><td>%ll</td><td>%u</td><td>uint64_t</td><td><b>(Unimplemented)</b> long long format</td></tr>
170 <tr><td>%-</td><td><i>all</i></td><td>N/A</td><td>Left justify</td></tr>
171 <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>
172 <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>
173 <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
174 decimal point for floats.</td></tr>
175 <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
176 some large number.</td></tr>
177 <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
178 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>
182 %* int32_t Next argument after this one specifies the width
185 %* N/A This field is scanned, but not stored
187 <p>If you are using this C API instead of the ustream.h API for C++,
188 you can use one of the following u_fprintf examples to display a UnicodeString.</p>
191 UFILE *out = u_finit(stdout, NULL, NULL);
192 UnicodeString string1("string 1");
193 UnicodeString string2("string 2");
194 u_fprintf(out, "%S\n", string1.getTerminatedBuffer());
195 u_fprintf(out, "%.*S\n", string2.length(), string2.getBuffer());
203 * When an end of file is encountered, this value can be returned.
209 /** Forward declaration of a Unicode-aware file @stable 3.0 */
210 typedef struct UFILE UFILE
;
213 * Enum for which direction of stream a transliterator applies to.
214 * @see u_fsettransliterator
220 U_READWRITE
=3 /* == (U_READ | U_WRITE) */
225 * A UFILE is a wrapper around a FILE* that is locale and codepage aware.
226 * That is, data written to a UFILE will be formatted using the conventions
227 * specified by that UFILE's Locale; this data will be in the character set
228 * specified by that UFILE's codepage.
229 * @param filename The name of the file to open.
230 * @param perm The read/write permission for the UFILE; one of "r", "w", "rw"
231 * @param locale The locale whose conventions will be used to format
232 * and parse output. If this parameter is NULL, the default locale will
234 * @param codepage The codepage in which data will be written to and
235 * read from the file. If this paramter is NULL the system default codepage
237 * @return A new UFILE, or NULL if an error occurred.
240 U_STABLE UFILE
* U_EXPORT2
241 u_fopen(const char *filename
,
244 const char *codepage
);
246 #ifndef U_HIDE_DRAFT_API
248 * Open a UFILE with a UChar* filename
249 * A UFILE is a wrapper around a FILE* that is locale and codepage aware.
250 * That is, data written to a UFILE will be formatted using the conventions
251 * specified by that UFILE's Locale; this data will be in the character set
252 * specified by that UFILE's codepage.
253 * @param filename The name of the file to open.
254 * @param perm The read/write permission for the UFILE; one of "r", "w", "rw"
255 * @param locale The locale whose conventions will be used to format
256 * and parse output. If this parameter is NULL, the default locale will
258 * @param codepage The codepage in which data will be written to and
259 * read from the file. If this paramter is NULL the system default codepage
261 * @return A new UFILE, or NULL if an error occurred.
264 U_DRAFT UFILE
* U_EXPORT2
265 u_fopen_u(const UChar
*filename
,
268 const char *codepage
);
269 #endif /* U_HIDE_DRAFT_API */
272 * Open a UFILE on top of an existing FILE* stream. The FILE* stream
273 * ownership remains with the caller. To have the UFILE take over
274 * ownership and responsibility for the FILE* stream, use the
276 * @param f The FILE* to which this UFILE will attach and use.
277 * @param locale The locale whose conventions will be used to format
278 * and parse output. If this parameter is NULL, the default locale will
280 * @param codepage The codepage in which data will be written to and
281 * read from the file. If this paramter is NULL, data will be written and
282 * read using the default codepage for <TT>locale</TT>, unless <TT>locale</TT>
283 * is NULL, in which case the system default codepage will be used.
284 * @return A new UFILE, or NULL if an error occurred.
287 U_STABLE UFILE
* U_EXPORT2
290 const char *codepage
);
293 * Open a UFILE on top of an existing FILE* stream. The FILE* stream
294 * ownership is transferred to the new UFILE. It will be closed when the
296 * @param f The FILE* which this UFILE will take ownership of.
297 * @param locale The locale whose conventions will be used to format
298 * and parse output. If this parameter is NULL, the default locale will
300 * @param codepage The codepage in which data will be written to and
301 * read from the file. If this paramter is NULL, data will be written and
302 * read using the default codepage for <TT>locale</TT>, unless <TT>locale</TT>
303 * is NULL, in which case the system default codepage will be used.
304 * @return A new UFILE, or NULL if an error occurred. If an error occurs
305 * the ownership of the FILE* stream remains with the caller.
308 U_STABLE UFILE
* U_EXPORT2
311 const char *codepage
);
314 * Create a UFILE that can be used for localized formatting or parsing.
315 * The u_sprintf and u_sscanf functions do not read or write numbers for a
316 * specific locale. The ustdio.h file functions can be used on this UFILE.
317 * The string is usable once u_fclose or u_fflush has been called on the
319 * @param stringBuf The string used for reading or writing.
320 * @param capacity The number of code units available for use in stringBuf
321 * @param locale The locale whose conventions will be used to format
322 * and parse output. If this parameter is NULL, the default locale will
324 * @return A new UFILE, or NULL if an error occurred.
327 U_STABLE UFILE
* U_EXPORT2
328 u_fstropen(UChar
*stringBuf
,
333 * Close a UFILE. Implies u_fflush first.
334 * @param file The UFILE to close.
338 U_STABLE
void U_EXPORT2
339 u_fclose(UFILE
*file
);
341 #if U_SHOW_CPLUSPLUS_API
346 * \class LocalUFILEPointer
347 * "Smart pointer" class, closes a UFILE via u_fclose().
348 * For most methods see the LocalPointerBase base class.
350 * @see LocalPointerBase
354 U_DEFINE_LOCAL_OPEN_POINTER(LocalUFILEPointer
, UFILE
, u_fclose
);
361 * Tests if the UFILE is at the end of the file stream.
362 * @param f The UFILE from which to read.
363 * @return Returns TRUE after the first read operation that attempts to
364 * read past the end of the file. It returns FALSE if the current position is
368 U_STABLE UBool U_EXPORT2
372 * Flush output of a UFILE. Implies a flush of
373 * converter/transliterator state. (That is, a logical break is
374 * made in the output stream - for example if a different type of
375 * output is desired.) The underlying OS level file is also flushed.
376 * Note that for a stateful encoding, the converter may write additional
377 * bytes to return the stream to default state.
378 * @param file The UFILE to flush.
381 U_STABLE
void U_EXPORT2
382 u_fflush(UFILE
*file
);
385 * Rewind the file pointer to the beginning of the file.
386 * @param file The UFILE to rewind.
390 u_frewind(UFILE
*file
);
393 * Get the FILE* associated with a UFILE.
395 * @return A FILE*, owned by the UFILE. (The FILE <EM>must not</EM> be modified or closed)
398 U_STABLE
FILE* U_EXPORT2
399 u_fgetfile(UFILE
*f
);
401 #if !UCONFIG_NO_FORMATTING
404 * Get the locale whose conventions are used to format and parse output.
405 * This is the same locale passed in the preceding call to<TT>u_fsetlocale</TT>
406 * or <TT>u_fopen</TT>.
407 * @param file The UFILE to set.
408 * @return The locale whose conventions are used to format and parse output.
411 U_STABLE
const char* U_EXPORT2
412 u_fgetlocale(UFILE
*file
);
415 * Set the locale whose conventions will be used to format and parse output.
416 * @param locale The locale whose conventions will be used to format
418 * @param file The UFILE to query.
419 * @return NULL if successful, otherwise a negative number.
422 U_STABLE
int32_t U_EXPORT2
423 u_fsetlocale(UFILE
*file
,
429 * Get the codepage in which data is written to and read from the UFILE.
430 * This is the same codepage passed in the preceding call to
431 * <TT>u_fsetcodepage</TT> or <TT>u_fopen</TT>.
432 * @param file The UFILE to query.
433 * @return The codepage in which data is written to and read from the UFILE,
434 * or NULL if an error occurred.
437 U_STABLE
const char* U_EXPORT2
438 u_fgetcodepage(UFILE
*file
);
441 * Set the codepage in which data will be written to and read from the UFILE.
442 * All Unicode data written to the UFILE will be converted to this codepage
443 * before it is written to the underlying FILE*. It it generally a bad idea to
444 * mix codepages within a file. This should only be called right
445 * after opening the <TT>UFile</TT>, or after calling <TT>u_frewind</TT>.
446 * @param codepage The codepage in which data will be written to
447 * and read from the file. For example <TT>"latin-1"</TT> or <TT>"ibm-943"</TT>.
448 * A value of NULL means the default codepage for the UFILE's current
449 * locale will be used.
450 * @param file The UFILE to set.
451 * @return 0 if successful, otherwise a negative number.
455 U_STABLE
int32_t U_EXPORT2
456 u_fsetcodepage(const char *codepage
,
461 * Returns an alias to the converter being used for this file.
462 * @param f The UFILE to get the value from
463 * @return alias to the converter (The converter <EM>must not</EM> be modified or closed)
466 U_STABLE UConverter
* U_EXPORT2
u_fgetConverter(UFILE
*f
);
468 #if !UCONFIG_NO_FORMATTING
470 * Returns an alias to the number formatter being used for this file.
471 * @param f The UFILE to get the value from
472 * @return alias to the number formatter (The formatter <EM>must not</EM> be modified or closed)
475 U_STABLE
const UNumberFormat
* U_EXPORT2
u_fgetNumberFormat(UFILE
*f
);
477 /* Output functions */
480 * Write formatted data to <TT>stdout</TT>.
481 * @param patternSpecification A pattern specifying how <TT>u_printf</TT> will
482 * interpret the variable arguments received and format the data.
483 * @return The number of Unicode characters written to <TT>stdout</TT>
486 U_STABLE
int32_t U_EXPORT2
487 u_printf(const char *patternSpecification
,
491 * Write formatted data to a UFILE.
492 * @param f The UFILE to which to write.
493 * @param patternSpecification A pattern specifying how <TT>u_fprintf</TT> will
494 * interpret the variable arguments received and format the data.
495 * @return The number of Unicode characters written to <TT>f</TT>.
498 U_STABLE
int32_t U_EXPORT2
500 const char *patternSpecification
,
504 * Write formatted data to a UFILE.
505 * This is identical to <TT>u_fprintf</TT>, except that it will
506 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
507 * @param f The UFILE to which to write.
508 * @param patternSpecification A pattern specifying how <TT>u_fprintf</TT> will
509 * interpret the variable arguments received and format the data.
510 * @param ap The argument list to use.
511 * @return The number of Unicode characters written to <TT>f</TT>.
515 U_STABLE
int32_t U_EXPORT2
517 const char *patternSpecification
,
521 * Write formatted data to <TT>stdout</TT>.
522 * @param patternSpecification A pattern specifying how <TT>u_printf_u</TT> will
523 * interpret the variable arguments received and format the data.
524 * @return The number of Unicode characters written to <TT>stdout</TT>
527 U_STABLE
int32_t U_EXPORT2
528 u_printf_u(const UChar
*patternSpecification
,
532 * Get a UFILE for <TT>stdout</TT>.
533 * @return UFILE that writes to <TT>stdout</TT>
536 U_STABLE UFILE
* U_EXPORT2
540 * Write formatted data to a UFILE.
541 * @param f The UFILE to which to write.
542 * @param patternSpecification A pattern specifying how <TT>u_fprintf</TT> will
543 * interpret the variable arguments received and format the data.
544 * @return The number of Unicode characters written to <TT>f</TT>.
547 U_STABLE
int32_t U_EXPORT2
548 u_fprintf_u(UFILE
*f
,
549 const UChar
*patternSpecification
,
553 * Write formatted data to a UFILE.
554 * This is identical to <TT>u_fprintf_u</TT>, except that it will
555 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
556 * @param f The UFILE to which to write.
557 * @param patternSpecification A pattern specifying how <TT>u_fprintf</TT> will
558 * interpret the variable arguments received and format the data.
559 * @param ap The argument list to use.
560 * @return The number of Unicode characters written to <TT>f</TT>.
564 U_STABLE
int32_t U_EXPORT2
565 u_vfprintf_u(UFILE
*f
,
566 const UChar
*patternSpecification
,
570 * Write a Unicode to a UFILE. The null (U+0000) terminated UChar*
571 * <TT>s</TT> will be written to <TT>f</TT>, excluding the NULL terminator.
572 * A newline will be added to <TT>f</TT>.
573 * @param s The UChar* to write.
574 * @param f The UFILE to which to write.
575 * @return A non-negative number if successful, EOF otherwise.
579 U_STABLE
int32_t U_EXPORT2
580 u_fputs(const UChar
*s
,
584 * Write a UChar to a UFILE.
585 * @param uc The UChar to write.
586 * @param f The UFILE to which to write.
587 * @return The character written if successful, EOF otherwise.
590 U_STABLE UChar32 U_EXPORT2
595 * Write Unicode to a UFILE.
596 * The ustring passed in will be converted to the UFILE's underlying
597 * codepage before it is written.
598 * @param ustring A pointer to the Unicode data to write.
599 * @param count The number of Unicode characters to write
600 * @param f The UFILE to which to write.
601 * @return The number of Unicode characters written.
605 U_STABLE
int32_t U_EXPORT2
606 u_file_write(const UChar
*ustring
,
611 /* Input functions */
612 #if !UCONFIG_NO_FORMATTING
615 * Read formatted data from a UFILE.
616 * @param f The UFILE from which to read.
617 * @param patternSpecification A pattern specifying how <TT>u_fscanf</TT> will
618 * interpret the variable arguments received and parse the data.
619 * @return The number of items successfully converted and assigned, or EOF
620 * if an error occurred.
623 U_STABLE
int32_t U_EXPORT2
625 const char *patternSpecification
,
629 * Read formatted data from a UFILE.
630 * This is identical to <TT>u_fscanf</TT>, except that it will
631 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
632 * @param f The UFILE from which to read.
633 * @param patternSpecification A pattern specifying how <TT>u_fscanf</TT> will
634 * interpret the variable arguments received and parse the data.
635 * @param ap The argument list to use.
636 * @return The number of items successfully converted and assigned, or EOF
637 * if an error occurred.
641 U_STABLE
int32_t U_EXPORT2
643 const char *patternSpecification
,
647 * Read formatted data from a UFILE.
648 * @param f The UFILE from which to read.
649 * @param patternSpecification A pattern specifying how <TT>u_fscanf</TT> will
650 * interpret the variable arguments received and parse the data.
651 * @return The number of items successfully converted and assigned, or EOF
652 * if an error occurred.
655 U_STABLE
int32_t U_EXPORT2
657 const UChar
*patternSpecification
,
661 * Read formatted data from a UFILE.
662 * This is identical to <TT>u_fscanf_u</TT>, except that it will
663 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
664 * @param f The UFILE from which to read.
665 * @param patternSpecification A pattern specifying how <TT>u_fscanf</TT> will
666 * interpret the variable arguments received and parse the data.
667 * @param ap The argument list to use.
668 * @return The number of items successfully converted and assigned, or EOF
669 * if an error occurred.
673 U_STABLE
int32_t U_EXPORT2
674 u_vfscanf_u(UFILE
*f
,
675 const UChar
*patternSpecification
,
680 * Read one line of text into a UChar* string from a UFILE. The newline
681 * at the end of the line is read into the string. The string is always
683 * @param f The UFILE from which to read.
684 * @param n The maximum number of characters - 1 to read.
685 * @param s The UChar* to receive the read data. Characters will be
686 * stored successively in <TT>s</TT> until a newline or EOF is
687 * reached. A null character (U+0000) will be appended to <TT>s</TT>.
688 * @return A pointer to <TT>s</TT>, or NULL if no characters were available.
691 U_STABLE UChar
* U_EXPORT2
697 * Read a UChar from a UFILE. It is recommended that <TT>u_fgetcx</TT>
698 * used instead for proper parsing functions, but sometimes reading
699 * code units is needed instead of codepoints.
701 * @param f The UFILE from which to read.
702 * @return The UChar value read, or U+FFFF if no character was available.
705 U_STABLE UChar U_EXPORT2
709 * Read a UChar32 from a UFILE.
711 * @param f The UFILE from which to read.
712 * @return The UChar32 value read, or U_EOF if no character was
713 * available, or U+FFFFFFFF if an ill-formed character was
718 U_STABLE UChar32 U_EXPORT2
722 * Unget a UChar from a UFILE.
723 * If this function is not the first to operate on <TT>f</TT> after a call
724 * to <TT>u_fgetc</TT>, the results are undefined.
725 * If this function is passed a character that was not recieved from the
726 * previous <TT>u_fgetc</TT> or <TT>u_fgetcx</TT> call, the results are undefined.
727 * @param c The UChar to put back on the stream.
728 * @param f The UFILE to receive <TT>c</TT>.
729 * @return The UChar32 value put back if successful, U_EOF otherwise.
732 U_STABLE UChar32 U_EXPORT2
737 * Read Unicode from a UFILE.
738 * Bytes will be converted from the UFILE's underlying codepage, with
739 * subsequent conversion to Unicode. The data will not be NULL terminated.
740 * @param chars A pointer to receive the Unicode data.
741 * @param count The number of Unicode characters to read.
742 * @param f The UFILE from which to read.
743 * @return The number of Unicode characters read.
746 U_STABLE
int32_t U_EXPORT2
747 u_file_read(UChar
*chars
,
751 #if !UCONFIG_NO_TRANSLITERATION
754 * Set a transliterator on the UFILE. The transliterator will be owned by the
756 * @param file The UFILE to set transliteration on
757 * @param adopt The UTransliterator to set. Can be NULL, which will
758 * mean that no transliteration is used.
759 * @param direction either U_READ, U_WRITE, or U_READWRITE - sets
760 * which direction the transliterator is to be applied to. If
761 * U_READWRITE, the "Read" transliteration will be in the inverse
763 * @param status ICU error code.
764 * @return The previously set transliterator, owned by the
765 * caller. If U_READWRITE is specified, only the WRITE transliterator
766 * is returned. In most cases, the caller should call utrans_close()
767 * on the result of this function.
770 U_STABLE UTransliterator
* U_EXPORT2
771 u_fsettransliterator(UFILE
*file
, UFileDirection direction
,
772 UTransliterator
*adopt
, UErrorCode
*status
);
777 /* Output string functions */
778 #if !UCONFIG_NO_FORMATTING
782 * Write formatted data to a Unicode string.
784 * @param buffer The Unicode String to which to write.
785 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will
786 * interpret the variable arguments received and format the data.
787 * @return The number of Unicode code units written to <TT>buffer</TT>. This
788 * does not include the terminating null character.
791 U_STABLE
int32_t U_EXPORT2
792 u_sprintf(UChar
*buffer
,
793 const char *patternSpecification
,
797 * Write formatted data to a Unicode string. When the number of code units
798 * required to store the data exceeds <TT>count</TT>, then <TT>count</TT> code
799 * units of data are stored in <TT>buffer</TT> and a negative value is
800 * returned. When the number of code units required to store the data equals
801 * <TT>count</TT>, the string is not null terminated and <TT>count</TT> is
804 * @param buffer The Unicode String to which to write.
805 * @param count The number of code units to read.
806 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will
807 * interpret the variable arguments received and format the data.
808 * @return The number of Unicode characters that would have been written to
809 * <TT>buffer</TT> had count been sufficiently large. This does not include
810 * the terminating null character.
813 U_STABLE
int32_t U_EXPORT2
814 u_snprintf(UChar
*buffer
,
816 const char *patternSpecification
,
820 * Write formatted data to a Unicode string.
821 * This is identical to <TT>u_sprintf</TT>, except that it will
822 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
824 * @param buffer The Unicode string to which to write.
825 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will
826 * interpret the variable arguments received and format the data.
827 * @param ap The argument list to use.
828 * @return The number of Unicode characters written to <TT>buffer</TT>.
832 U_STABLE
int32_t U_EXPORT2
833 u_vsprintf(UChar
*buffer
,
834 const char *patternSpecification
,
838 * Write formatted data to a Unicode string.
839 * This is identical to <TT>u_snprintf</TT>, except that it will
840 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.<br><br>
841 * When the number of code units required to store the data exceeds
842 * <TT>count</TT>, then <TT>count</TT> code units of data are stored in
843 * <TT>buffer</TT> and a negative value is returned. When the number of code
844 * units required to store the data equals <TT>count</TT>, the string is not
845 * null terminated and <TT>count</TT> is returned.
847 * @param buffer The Unicode string to which to write.
848 * @param count The number of code units to read.
849 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will
850 * interpret the variable arguments received and format the data.
851 * @param ap The argument list to use.
852 * @return The number of Unicode characters that would have been written to
853 * <TT>buffer</TT> had count been sufficiently large.
857 U_STABLE
int32_t U_EXPORT2
858 u_vsnprintf(UChar
*buffer
,
860 const char *patternSpecification
,
864 * Write formatted data to a Unicode string.
866 * @param buffer The Unicode string to which to write.
867 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will
868 * interpret the variable arguments received and format the data.
869 * @return The number of Unicode characters written to <TT>buffer</TT>.
872 U_STABLE
int32_t U_EXPORT2
873 u_sprintf_u(UChar
*buffer
,
874 const UChar
*patternSpecification
,
878 * Write formatted data to a Unicode string. When the number of code units
879 * required to store the data exceeds <TT>count</TT>, then <TT>count</TT> code
880 * units of data are stored in <TT>buffer</TT> and a negative value is
881 * returned. When the number of code units required to store the data equals
882 * <TT>count</TT>, the string is not null terminated and <TT>count</TT> is
885 * @param buffer The Unicode string to which to write.
886 * @param count The number of code units to read.
887 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will
888 * interpret the variable arguments received and format the data.
889 * @return The number of Unicode characters that would have been written to
890 * <TT>buffer</TT> had count been sufficiently large.
893 U_STABLE
int32_t U_EXPORT2
894 u_snprintf_u(UChar
*buffer
,
896 const UChar
*patternSpecification
,
900 * Write formatted data to a Unicode string.
901 * This is identical to <TT>u_sprintf_u</TT>, except that it will
902 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
904 * @param buffer The Unicode string to which to write.
905 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will
906 * interpret the variable arguments received and format the data.
907 * @param ap The argument list to use.
908 * @return The number of Unicode characters written to <TT>f</TT>.
912 U_STABLE
int32_t U_EXPORT2
913 u_vsprintf_u(UChar
*buffer
,
914 const UChar
*patternSpecification
,
918 * Write formatted data to a Unicode string.
919 * This is identical to <TT>u_snprintf_u</TT>, except that it will
920 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
921 * When the number of code units required to store the data exceeds
922 * <TT>count</TT>, then <TT>count</TT> code units of data are stored in
923 * <TT>buffer</TT> and a negative value is returned. When the number of code
924 * units required to store the data equals <TT>count</TT>, the string is not
925 * null terminated and <TT>count</TT> is returned.
927 * @param buffer The Unicode string to which to write.
928 * @param count The number of code units to read.
929 * @param patternSpecification A pattern specifying how <TT>u_sprintf</TT> will
930 * interpret the variable arguments received and format the data.
931 * @param ap The argument list to use.
932 * @return The number of Unicode characters that would have been written to
933 * <TT>f</TT> had count been sufficiently large.
937 U_STABLE
int32_t U_EXPORT2
938 u_vsnprintf_u(UChar
*buffer
,
940 const UChar
*patternSpecification
,
943 /* Input string functions */
946 * Read formatted data from a Unicode string.
948 * @param buffer The Unicode string from which to read.
949 * @param patternSpecification A pattern specifying how <TT>u_sscanf</TT> will
950 * interpret the variable arguments received and parse the data.
951 * @return The number of items successfully converted and assigned, or EOF
952 * if an error occurred.
955 U_STABLE
int32_t U_EXPORT2
956 u_sscanf(const UChar
*buffer
,
957 const char *patternSpecification
,
961 * Read formatted data from a Unicode string.
962 * This is identical to <TT>u_sscanf</TT>, except that it will
963 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
965 * @param buffer The Unicode string from which to read.
966 * @param patternSpecification A pattern specifying how <TT>u_sscanf</TT> will
967 * interpret the variable arguments received and parse the data.
968 * @param ap The argument list to use.
969 * @return The number of items successfully converted and assigned, or EOF
970 * if an error occurred.
974 U_STABLE
int32_t U_EXPORT2
975 u_vsscanf(const UChar
*buffer
,
976 const char *patternSpecification
,
980 * Read formatted data from a Unicode string.
982 * @param buffer The Unicode string from which to read.
983 * @param patternSpecification A pattern specifying how <TT>u_sscanf</TT> will
984 * interpret the variable arguments received and parse the data.
985 * @return The number of items successfully converted and assigned, or EOF
986 * if an error occurred.
989 U_STABLE
int32_t U_EXPORT2
990 u_sscanf_u(const UChar
*buffer
,
991 const UChar
*patternSpecification
,
995 * Read formatted data from a Unicode string.
996 * This is identical to <TT>u_sscanf_u</TT>, except that it will
997 * <EM>not</EM> call <TT>va_start</TT> and <TT>va_end</TT>.
999 * @param buffer The Unicode string from which to read.
1000 * @param patternSpecification A pattern specifying how <TT>u_sscanf</TT> will
1001 * interpret the variable arguments received and parse the data.
1002 * @param ap The argument list to use.
1003 * @return The number of items successfully converted and assigned, or EOF
1004 * if an error occurred.
1008 U_STABLE
int32_t U_EXPORT2
1009 u_vsscanf_u(const UChar
*buffer
,
1010 const UChar
*patternSpecification
,