man3/wcsnrtombs.3

   1 .\" Copyright (c) Bruno Haible <haible@clisp.cons.org>
   2 .\"
   3 .\" This is free documentation; you can redistribute it and/or
   4 .\" modify it under the terms of the GNU General Public License as
   5 .\" published by the Free Software Foundation; either version 2 of
   6 .\" the License, or (at your option) any later version.
   7 .\"
   8 .\" References consulted:
   9 .\"   GNU glibc-2 source code and manual
  10 .\"   Dinkumware C library reference http://www.dinkumware.com/
  11 .\"   OpenGroup's Single Unix specification http://www.UNIX-systems.org/online.html
  12 .\"
  13 .TH WCSNRTOMBS 3  1999-07-25 "GNU" "Linux Programmer's Manual"
  14 .SH NAME
  15 wcsnrtombs \- convert a wide-character string to a multibyte string
  16 .SH SYNOPSIS
  17 .nf
  18 .B #include <wchar.h>
  19 .sp
  20 .BI "size_t wcsnrtombs(char *" dest ", const wchar_t **" src ", size_t " nwc ,
  21 .BI "                  size_t " len ", mbstate_t *" ps );
  22 .fi
  23 .SH DESCRIPTION
  24 The
  25 .BR wcsnrtombs ()
  26 function is like the
  27 .BR wcsrtombs ()
  28 function,
  29 except that the number of wide characters to be converted,
  30 starting at \fI*src\fP, is limited to \fInwc\fP.
  31 .PP
  32 If \fIdest\fP is not a NULL pointer,
  33 the
  34 .BR wcsnrtombs ()
  35 function converts
  36 at most \fInwc\fP wide characters from
  37 the wide-character string \fI*src\fP to a multibyte string starting at
  38 \fIdest\fP.
  39 At most \fIlen\fP bytes are written to \fIdest\fP.
  40 The shift state
  41 \fI*ps\fP is updated.
  42 The conversion is effectively performed by repeatedly
  43 calling wcrtomb(\fIdest\fP,\fI*src\fP,\fIps\fP),
  44 as long as this call succeeds,
  45 and then incrementing \fIdest\fP by the
  46 number of bytes written and \fI*src\fP
  47 by one.
  48 The conversion can stop for three reasons:
  49 .PP
  50 1. A wide character has been encountered that can not be represented as a
  51 multibyte sequence (according to the current locale).
  52 In this case \fI*src\fP
  53 is left pointing to the invalid wide character, (size_t)(\-1) is returned,
  54 and \fIerrno\fP is set to \fBEILSEQ\fP.
  55 .PP
  56 2. \fInwc\fP wide characters have been
  57 converted without encountering a L'\\0',
  58 or the length limit forces a stop.
  59 In this case \fI*src\fP is left pointing
  60 to the next wide character to be converted, and the number of bytes written
  61 to \fIdest\fP is returned.
  62 .PP
  63 3. The wide-character string has been completely converted, including the
  64 terminating L'\\0' (which has the side effect of bringing back \fI*ps\fP
  65 to the initial state).
  66 In this case \fI*src\fP is set to NULL, and the number
  67 of bytes written to \fIdest\fP, excluding the terminating '\\0' byte, is
  68 returned.
  69 .PP
  70 If \fIdest\fP is NULL, \fIlen\fP is ignored,
  71 and the conversion proceeds as above,
  72 except that the converted bytes are not written out to memory, and that
  73 no destination length limit exists.
  74 .PP
  75 In both of the above cases,
  76 if \fIps\fP is a NULL pointer, a static anonymous
  77 state only known to the wcsnrtombs function is used instead.
  78 .PP
  79 The programmer must ensure that there is room for at least \fIlen\fP bytes
  80 at \fIdest\fP.
  81 .SH "RETURN VALUE"
  82 The
  83 .BR wcsnrtombs ()
  84 function returns
  85 the number of bytes that make up the
  86 converted part of multibyte sequence,
  87 not including the terminating null byte.
  88 If a wide character was encountered which
  89 could not be converted, (size_t)(\-1)
  90 is returned, and \fIerrno\fP set to \fBEILSEQ\fP.
  91 .SH "CONFORMING TO"
  92 This function is a GNU extension.
  93 .SH NOTES
  94 The behavior of
  95 .BR wcsnrtombs ()
  96 depends on the LC_CTYPE category of the
  97 current locale.
  98 .PP
  99 Passing NULL as \fIps\fP is not multi-thread safe.
 100 .SH "SEE ALSO"
 101 .BR iconv (3),
 102 .BR wcsrtombs (3)