/[pcre]/code/trunk/doc/html/pcre16.html

Diff of /code/trunk/doc/html/pcre16.html

Parent Directory | Revision Log | View Patch Patch

-revision 869 by ph10,
Sat Jan 14 11:16:23 2012 UTC
+revision 903 by ph10,
Sat Jan 21 16:37:17 2012 UTC
 Line 160 
 man page, in case the conversion went wr
  <br><a name="SEC5" href="#TOC1">PCRE 16-BIT API 16-BIT-ONLY FUNCTION</a><br>
  <P>
  <b>int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *<i>output</i>,</b>
- <b>PCRE_SPTR16 <i>input</i>, int <i>length</i>, int *<i>byte_order</i>, </b>
+ <b>PCRE_SPTR16 <i>input</i>, int <i>length</i>, int *<i>byte_order</i>,</b>
  <b>int <i>keep_boms</i>);</b>
  </P>
  <br><a name="SEC6" href="#TOC1">THE PCRE 16-BIT LIBRARY</a><br>
 Line 177 
 to the 16-bit library. This page describ
 -bit library.
  </P>
  <P>
  WARNING: A single application can be linked with both libraries, but you must
  take care when processing any particular pattern to use functions from just one
  library. For example, if you want to study a pattern that was compiled with
  <b>pcre16_compile()</b>, you must do so with <b>pcre16_study()</b>, not
  <b>pcre_study()</b>, and you must free the study data with
 Line 186 
 library. For example, if you want to stu
  </P>
  <br><a name="SEC7" href="#TOC1">THE HEADER FILE</a><br>
  <P>
  There is only one header file, <b>pcre.h</b>. It contains prototypes for all the
  functions in both libraries, as well as definitions of flags, structures, error
  codes, etc.
  </P>
  <br><a name="SEC8" href="#TOC1">THE LIBRARY NAME</a><br>
  <P>
  In Unix-like systems, the 16-bit library is called <b>libpcre16</b>, and can
  normally be accesss by adding <b>-lpcre16</b> to the command for linking an
  application that uses PCRE.
  </P>
  <br><a name="SEC9" href="#TOC1">STRING TYPES</a><br>
  <P>
  In the 8-bit library, strings are passed to PCRE library functions as vectors
  of bytes with the C type "char *". In the 16-bit library, strings are passed as
  vectors of unsigned 16-bit quantities. The macro PCRE_UCHAR16 specifies an
  appropriate data type, and PCRE_SPTR16 is defined as "const PCRE_UCHAR16 *". In
  very many environments, "short int" is a 16-bit data type. When PCRE is built,
  it defines PCRE_UCHAR16 as "short int", but checks that it really is a 16-bit
  data type. If it is not, the build fails with an error message telling the
  maintainer to modify the definition appropriately.
  </P>
  <br><a name="SEC10" href="#TOC1">STRUCTURE TYPES</a><br>
  <P>
  The types of the opaque structures that are used for compiled 16-bit patterns
  and JIT stacks are <b>pcre16</b> and <b>pcre16_jit_stack</b> respectively. The
  type of the user-accessible structure that is returned by <b>pcre16_study()</b>
  is <b>pcre16_extra</b>, and the type of the structure that is used for passing
  data to a callout function is <b>pcre16_callout_block</b>. These structures
  contain the same fields, with the same names, as their 8-bit counterparts. The
  only difference is that pointers to character strings are 16-bit instead of
 -bit types.
  </P>
  <br><a name="SEC11" href="#TOC1">16-BIT FUNCTIONS</a><br>
  <P>
  For every function in the 8-bit library there is a corresponding function in
  the 16-bit library with a name that starts with <b>pcre16_</b> instead of
  <b>pcre_</b>. The prototypes are listed above. In addition, there is one extra
  function, <b>pcre16_utf16_to_host_byte_order()</b>. This is a utility function
  that converts a UTF-16 character string to host byte order if necessary. The
  other 16-bit functions expect the strings they are passed to be in host byte
  order.
  </P>
  <P>
  The <i>input</i> and <i>output</i> arguments of
  <b>pcre16_utf16_to_host_byte_order()</b> may point to the same address, that is,
  conversion in place is supported. The output buffer must be at least as long as
  the input.
  </P>
  <P>
 Line 239 
 The length argument specifies the
  input string; a negative value specifies a zero-terminated string.
  </P>
  <P>
  If <i>byte_order</i> is NULL, it is assumed that the string starts off in host
  byte order. This may be changed by byte-order marks (BOMs) anywhere in the
  string (commonly as the first character).
  </P>
  <P>
  If <i>byte_order</i> is not NULL, a non-zero value of the integer to which it
  points means that the input starts off in host byte order, otherwise the
  opposite order is assumed. Again, BOMs in the string can change this. The final
  byte order is passed back at the end of processing.
  </P>
  <P>
  If <i>keep_boms</i> is not zero, byte-order mark characters (0xfeff) are copied
  into the output string. Otherwise they are discarded.
  </P>
  <P>
 Line 259 
 buffer, including the zero terminator if
  </P>
  <br><a name="SEC12" href="#TOC1">SUBJECT STRING OFFSETS</a><br>
  <P>
  The offsets within subject strings that are returned by the matching functions
  are in 16-bit units rather than bytes.
  </P>
  <br><a name="SEC13" href="#TOC1">NAMED SUBPATTERNS</a><br>
  <P>
  The name-to-number translation table that is maintained for named subpatterns
  uses 16-bit characters. The <b>pcre16_get_stringtable_entries()</b> function
  returns the length of each entry in the table as the number of 16-bit data
  units.
  </P>
  <br><a name="SEC14" href="#TOC1">OPTION NAMES</a><br>
 Line 276 
 which correspond to PCRE_UTF8 and PCRE_N
  fact, these new options define the same bits in the options word.
  </P>
  <P>
  For the <b>pcre16_config()</b> function there is an option PCRE_CONFIG_UTF16
  that returns 1 if UTF-16 support is configured, otherwise 0. If this option is
  given to <b>pcre_config()</b>, or if the PCRE_CONFIG_UTF8 option is given to
  <b>pcre16_config()</b>, the result is the PCRE_ERROR_BADOPTION error.
  </P>
  <br><a name="SEC15" href="#TOC1">CHARACTER CODES</a><br>
  <P>
  In 16-bit mode, when PCRE_UTF16 is not set, character values are treated in the
  same way as in 8-bit, non UTF-8 mode, except, of course, that they can range
  from 0 to 0xffff instead of 0 to 0xff. Character types for characters less than
 xff can therefore be influenced by the locale in the same way as before.
  Characters greater than 0xff have only one case, and no "type" (such as letter
  or digit).
  </P>
  <P>
  In UTF-16 mode, the character code is Unicode, in the range 0 to 0x10ffff, with
  the exception of values in the range 0xd800 to 0xdfff because those are
  "surrogate" values that are used in pairs to encode values greater than 0xffff.
  </P>
  <P>
  A UTF-16 string can indicate its endianness by special code knows as a
  byte-order mark (BOM). The PCRE functions do not handle this, expecting strings
  to be in host byte order. A utility function called
  <b>pcre16_utf16_to_host_byte_order()</b> is provided to help with this (see
 Line 304 
 above).
  </P>
  <br><a name="SEC16" href="#TOC1">ERROR NAMES</a><br>
  <P>
  The errors PCRE_ERROR_BADUTF16_OFFSET and PCRE_ERROR_SHORTUTF16 correspond to
  their 8-bit counterparts. The error PCRE_ERROR_BADMODE is given when a compiled
  pattern is passed to a function that processes patterns in the other
  mode, for example, if a pattern compiled with <b>pcre_compile()</b> is passed to
  <b>pcre16_exec()</b>.
  </P>
  <P>
  There are new error codes whose names begin with PCRE_UTF16_ERR for invalid
  UTF-16 strings, corresponding to the PCRE_UTF8_ERR codes for UTF-8 strings that
  are described in the section entitled
  <a href="pcreapi.html#badutf8reasons">"Reason codes for invalid UTF-8 strings"</a>
  in the main
  <a href="pcreapi.html"><b>pcreapi</b></a>
  page. The UTF-16 errors are:
  <pre>
 Line 327 
 page. The UTF-16 errors are:
  </P>
  <br><a name="SEC17" href="#TOC1">ERROR TEXTS</a><br>
  <P>
  If there is an error while compiling a pattern, the error text that is passed
  back by <b>pcre16_compile()</b> or <b>pcre16_compile2()</b> is still an 8-bit
  character string, zero-terminated.
  </P>
  <br><a name="SEC18" href="#TOC1">CALLOUTS</a><br>
 Line 338 
 a callout function point to 16-bit vecto
  </P>
  <br><a name="SEC19" href="#TOC1">TESTING</a><br>
  <P>
  The <b>pcretest</b> program continues to operate with 8-bit input and output
  files, but it can be used for testing the 16-bit library. If it is run with the
  command line option <b>-16</b>, patterns and subject strings are converted from
 -bit to 16-bit before being passed to PCRE, and the 16-bit library functions
  are used instead of the 8-bit ones. Returned 16-bit strings are converted to
 -bit for output. If the 8-bit library was not compiled, <b>pcretest</b>
  defaults to 16-bit and the <b>-16</b> option is ignored.
  </P>
  <P>
  When PCRE is being built, the <b>RunTest</b> script that is called by "make
  check" uses the <b>pcretest</b> <b>-C</b> option to discover which of the 8-bit
  and 16-bit libraries has been built, and runs the tests appropriately.
  </P>
  <br><a name="SEC20" href="#TOC1">NOT SUPPORTED IN 16-BIT MODE</a><br>
  <P>
  Not all the features of the 8-bit library are available with the 16-bit
  library. The C++ and POSIX wrapper functions support only the 8-bit library,
  and the <b>pcregrep</b> program is at present 8-bit only.
  </P>
  <br><a name="SEC21" href="#TOC1">AUTHOR</a><br>

 Legend:



Removed from v.869
 


changed lines


 
Added in v.903
 Legend:



Removed from v.869
 


changed lines


 
Added in v.903
-Removed from v.869
+Added in v.903

webmaster@exim.org	ViewVC Help
Powered by ViewVC 1.1.12