autodoc.git/ traditional_manual/ chapter_9.html

Branch: Tag:

2003-10-28

2003-10-28 11:46:02 by Martin Stjernholm <mast@lysator.liu.se>

Made the backend debug checks somewhat more accurate.

Rev: src/modules/files/file.c:1.306

916:    as many bytes as you have asked for, but will merely return as    much as the system read function will return. This mainly useful    with stream devices which can return exactly one row or packet at -  a time.</p> +  a time. If <code>not_all</code> is used in blocking mode, <code>read()</code> will +  only block if there's no data at all available.</p>   <p> If something goes wrong and <code>not_all</code> is set, zero will be    returned. If something goes wrong and <code>not_all</code> is zero or left    out, then either zero or a string shorter than <code>len</code> is returned.
1544:   <dt class='head--doc'>Description</dt>   <dd class='body--doc'><p>Write data to a file or a stream.</p>   <p> Writes <code>data</code> and returns the number of bytes that were -  actually written.</p> - <p> If more than one argument is given, <code>sprintf()</code> will be -  used to format them.</p> - <p> If <code>data</code> is an array, it will be concatenated, and then written.</p> - <p> 0 is returned in nonblocking mode if it was not possible to write -  anything without blocking.</p> - <p> -1 is returned if something went wrong and no bytes were written.</p> +  actually written. It can be less than the size of the given data if</p> + <ul> + <li><p>some data was written successfully and then something went +  wrong, or</p> + </li><li><p>nonblocking mode is used and not all data could be written +  without blocking.</p> + </li></ul><p>-1 is returned if something went wrong and no bytes were written. +  If only some data was written due to an error and that error +  persists, then a later call to <code>write()</code> will fail and return -1.</p> + <p> If everything went fine, a call to <code>errno()</code> directly afterwards +  will return zero.</p> + <p> If <code>data</code> is an array of strings, they will be written sequence.</p> + <p> If more than one argument is given, <code>sprintf()</code> will be used to +  format them using <code>format</code>. If <code>format</code> is an array, the strings +  in it are concatenated and the result is used as format string.</p>   </dd>   <dt class='head--doc'>Note</dt>   <dd class='body--doc'><p>Writing of wide strings is not supported. You have to encode the
1572:      <dt class='head--doc'>Description</dt>   <dd class='body--doc'><p>Write out-of-band data to a stream.</p> - <p> Writes out-of-band data to a stream and returns how many bytes that were -  actually written.</p> + <p> Writes out-of-band data to a stream and returns how many bytes +  that were actually written. It can be less than the size of the +  given data if some data was written successfully and then +  something went wrong.</p> + <p> -1 is returned if something went wrong and no bytes were written. +  If only some data was written due to an error and that error +  persists, then a later call to <code>write_oob()</code> will fail and return -1.</p> + <p> If everything went fine, a call to <code>errno()</code> directly afterwards +  will return zero.</p>   <p> If more than one argument is given, <code>sprintf()</code> will be    used to format them.</p> - <p> -1 is returned if something went wrong and no bytes were written.</p> +    </dd>   <dt class='head--doc'>Note</dt>   <dd class='body--doc'><p>Out-of-band data was not be supported on Pike 0.5 and earlier,