libvips/doc/using-cpp.xml

<?xml version="1.0"?>
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN"
               "http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd" [
]>
<!-- vim: set ts=2 sw=2 expandtab: --> 
<refentry id="using-from-cpp">
  <refmeta>
    <refentrytitle>VIPS from C++</refentrytitle>
    <manvolnum>3</manvolnum>
    <refmiscinfo>VIPS Library</refmiscinfo>
  </refmeta>

  <refnamediv>
    <refname>Using VIPS</refname>
    <refpurpose>How to use the VIPS library from C++</refpurpose>
  </refnamediv>

  <refsect3 id="using-cpp">
    <title>Introduction</title>
    <para>
      VIPS comes with a convenient C++ API. It is a very thin wrapper over the 
      C API and adds automatic reference counting, exceptions, operator 
      overloads, and automatic constant expansion. You can drop down to the C 
      API at any point, so all the C API docs also work for C++. 
    </para>

<programlisting language="cpp">
/* compile with:
 *      g++ -g -Wall example.cc `pkg-config vips-cpp --cflags --libs`
 */

#include &lt;vips/vips8&gt;

using namespace vips;

int
main (int argc, char **argv)
{ 
  if (VIPS_INIT (argv[0])) 
    vips_error_exit (NULL);
  
  if (argc != 3)
    vips_error_exit ("usage: %s input-file output-file", argv[0]);

  VImage in = VImage::new_from_file (argv[1],
    VImage::option ()-&gt;set ("access", VIPS_ACCESS_SEQUENTIAL));
  
  double avg = in.avg ();
  
  printf ("avg = %g\n", avg);
  printf ("width = %d\n", in.width ());
  
  in = VImage::new_from_file (argv[1],
    VImage::option ()-&gt;set ("access", VIPS_ACCESS_SEQUENTIAL));
  
  VImage out = in.embed (10, 10, 1000, 1000,
    VImage::option ()-&gt;
      set ("extend", "background")-&gt;
      set ("background", 128));
  
  out.write_to_file (argv[2]);
  
  vips_shutdown ();
  
  return (0);
}
</programlisting>

    <para>
      Everything before <code>VImage in = VImage::..</code> is exactly
      as the C API. vips_error_exit() just prints the arguments plus the 
      libvips error log and exits with an error code. 
    </para>

    <para>
      <code>VImage in = VImage::..</code> is the C++ equivalent of 
      vips_image_new_from_file(). It works 
      in the same way, the differences being:

      <itemizedlist>
        <listitem>
          <para>
            <code>VImage</code> lifetime is managed automatically, like a smart 
            pointer. You don't need to call g_object_unref(). 
          </para>
        </listitem>

        <listitem>
          <para>
            Instead of using varargs and a %NULL-terminated option list, this 
            function takes an optional <code>VOption</code> pointer. This 
            gives a list of name / value pairs for optional arguments to the 
            function.
          </para>

          <para>
            In this case we request unbuffered IO for the image, meaning, we 
            expect to do a single top-to-bottom scan of the image and do not 
            need it to be decompressed entirely. You can use the C enum name, 
            as is done in this case, or use a string and have the string 
            looked up. See below.
          </para>

          <para>
            The function will delete the <code>VOption</code> pointer for
            us when it's finished with it.
          </para>
        </listitem>

        <listitem>
          <para>
            Instead of returning %NULL on error, this constructor will 
            raise a <code>VError</code> exception. 
          </para>
        </listitem>
      </itemizedlist>

      There are a series of similar constructors which parallel the other 
      constructors in the C API, see VImage::new_from_memory(),
      VImage::new_from_buffer(), and VImage::new_matrix(). 
    </para>

    <para>
      The convenience function `VImage::new_from_image()` makes a constant 
      image from an existing image. The image it returns will have the same 
      size, interpretation, resolution and format as the image you call it on, 
      but with every pixel having the constant value you specify. For example:

<programlisting language="cpp">
      new_image = image.new_from_image(12);
</programlisting>

      Now `new_image` has the same size as `image`, but has one band, and every 
      pixel has the value 12. You can pass a `std::vector&lt;double&gt;` as the 
      argument to make a constant image with a different number of bands. 
    </para>

    <para>
      There's also 
      VImage::new_memory() and VImage::new_temp_file(), which when written to 
      with VImage::write() will create whole images on memory or on disc. 
    </para>

    <para>
      The next line finds the average pixel value, it's the equivalent of the 
      vips_avg() function. The differences from the C API are:

      <itemizedlist>
        <listitem>
          <para>
            VImage::avg() is a member function: the <code>this</code> 
            parameter is the first (the only, in this case) input image. 
          </para>

          <para>
            The function returns the first output parameter, in this case the 
            average pixel value. Other return values are via pointer arguments, 
            as in the C API. 
          </para>

          <para>
            Like VImage::new_from_file(), function raises the 
            <code>VError</code> exception on error.
          </para>

          <para>
            Like VImage::new_from_file(), extra arguments are passed 
            via an optional <code>VOption</code> parameter. There are none
            in this case, so the function brackets can be left empty.
          </para>
        </listitem>
      </itemizedlist>

      All other operations follow the same pattern, for example the C API call 
      vips_add():

<programlisting language="cpp">
int vips_add( VipsImage *left, VipsImage *right, VipsImage **out, ... );
</programlisting>

      appears in C++ as:

<programlisting language="cpp">
VImage VImage::add( VImage right, VOption *options = 0 );
</programlisting>
    </para>

    <para>
      The next line uses VImage::width() to get the image width in pixels. 
      There are similar functions paralleling vips_image_get_format() and 
      friends. Use VImage::set() to set metadata fields, VImage::get_int() and 
      c. to fetch metadata. 
    </para>

    <para>
      Next we reload the image. The VImage::avg() will have scanned the image 
      and reached the end of the file, we need to scan again for the next 
      operation. If we'd selected random access mode (the default) in the 
      original VImage::new_from_file(), we would not need to reload.
    </para>

    <para>
      The next line runs vips_embed() with two optional parameters. The first 
      sets the value to an enum (here we use a string to set the value, it'll 
      be looked up in the list of possible enum values, or you can use the 
      symbols from the C API), the 
      second sets the value to an <code>int</code>. The 
      <code>"background"</code> 
      parameter is actually a #VipsArrayDouble: if you pass an 
      <code>int</code> instead, it will be automatically converted to a 
      one-element array for you. You can pass a 
      <code>std::vector&lt;double&gt;</code> too: the utility function 
      VImage::to_vectorv() is a convenient way to make one. 
    </para>

    <para>
      Finally, VImage::write_to_file() will write the new image to the 
      filesystem. You can add a #VOption as a final parameter and set options 
      for the writer if you wish. Again, the operation will throw a #VError 
      exception on error. The other writers from the C API are also present: 
      you can write to a memory array, to a formatted image in memory, or to 
      another image. 
    </para>

    <para>
      The API docs have a <link linkend="function-list">handy table of all vips 
      operations</link>, if you want to find out how to do something, try 
      searching that. 
    </para>
  </refsect3>

  <refsect3 id="cpp-expansion">
    <title>Automatic constant expansion</title>

    <para>
      The C++ API will automatically turn constants into images in some cases. 
      For example, you can join two images together bandwise (the 
      bandwise join of two RGB images would be a six-band image) with:

<programlisting language="cpp">
VImage rgb = ...; 
VImage six_band = rgb.bandjoin( rgb );
</programlisting>

      You can also bandjoin a constant, for example: 

<programlisting language="cpp">
VImage rgb_with_alpha = rgb.bandjoin( 255 );
</programlisting>

      Will add an extra band to an image, with every element in the new band 
      having the value 255. This is quite a general feature. You can use a 
      constant in most places where you can use an image and it will be 
      converted. For example:

<programlisting language="cpp">
VImage a = (a &lt; 128).ifthenelse( 128, a ); 
</programlisting>

      Will set every band element of <code>a</code> less than 128 to 128. 
    </para>

    <para>
      The C++ API includes the usual range of arithmetic operator overloads. 
      You can mix constants, vectors and images freely.
    </para>

    <para>
      The API overloads <code>[]</code> to be vips_extract_band(). You can 
      write:

<programlisting language="cpp">
VImage xyz = VImage::xyz( 256, 256 ) - VImage::to_vectorv( 2, 128.0, 128.0 ); 
VImage mask = (xyz[0].pow( 2 ) + xyz[1].pow( 2 )).pow( 0.5 ) &lt; 100;
</programlisting>

      to make a circular mask, for example.
    </para>

    <para>
      The API overloads <code>()</code> to be vips_getpoint(). You can 
      write:

<programlisting language="cpp">
VImage xyz = VImage::xyz( 256, 256 ) - VImage::to_vectorv( 2, 128.0, 128.0 ); 
// this will have the value [0, 0]
std::vector&lt;double&gt; point = xyz(128, 128);
</programlisting>
    </para>

  </refsect3>

  <refsect3 id="cpp-enum">
    <title>Enum expansion</title>

    <para>
      VIPS operations which implement several functions with a controlling 
      enum, such as vips_math(), are expanded to a set of member functions 
      named after the enum. For example, the C function:

<programlisting language="cpp">
int vips_math( VipsImage *in, VipsImage **out, VipsOperationMath math, ... );
</programlisting>

      where #VipsOperationMath has the member #VIPS_OPERATION_MATH_SIN, has a 
      C convenience function vips_sin():

<programlisting language="cpp">
int vips_sin( VipsImage *in, VipsImage **out, ... );
</programlisting>

      and a C++ member function VImage::sin():

<programlisting language="cpp">
VImage VImage::sin( VOption *options = 0 );
</programlisting>

    </para>
  </refsect3>

  <refsect3 id="cpp-metadata">
    <title>Image metadata</title>

    <para>
      VIPS images can have a lot of metadata attached to them, giving things
      like ICC profiles, EXIF data, and so on. You can use the command-line
      program <code>vipsheader</code> with the <code>-a</code> flag to list
      all the fields.
    </para>

    <para>
      You can read metadata items with the member functions
      <code>get_int()</code>, <code>get_double()</code>, 
      <code>get_string()</code> and <code>get_blob()</code>. Use
      <code>get_typeof()</code> to call vips_image_get_typeof() and read the
      type of an item. This will return 0 for undefined fields.

<programlisting language="cpp">
const char *VImage::get_string( const char *field ) throw( VError );
</programlisting>
    </para>

    <para>
      You can use the <code>set()</code> family of overloaded members to set
      metadata, for example:

<programlisting language="cpp">
void VImage::set( const char *field, const char *value );
</programlisting>
    </para>

    <para>
      You can use these functions to manipulate exif metadata, for example:

<programlisting language="cpp">
VImage im = VImage::new_from_file( "x.jpg" )
int orientation = atoi( im.get_string( "exif-ifd0-Orientation" ) );
im.set( "exif-ifd0-Orientation", "2" );
im.write_to_file( "y.jpg" );
</programlisting>
    </para>
  </refsect3>

  <refsect3 id="cpp-extend">
    <title>Extending the C++ interface</title>

    <para>
      The C++ interface comes in two parts. First, <code>VImage8.h</code>
      defines a simple layer over #GObject for automatic reference counting,
      then a generic way to call any vips8 operation with VImage::call(),
      then a few convenience functions, then a set of overloads.
    </para>

    <para>
      The member function for each operation, for example VImage::add(), is 
      generated by a small Python program called <code>gen-operators.py</code>,
      and its companion, <code>gen-operators-h.py</code> to generate the 
      headers. If you write a new VIPS operator, you'll need to rerun these 
      programs to make the new member function. 
    </para>

    <para>
      You can write the wrapper yourself, of course, they are very simple. 
      The one for VImage::add() looks like this:

<programlisting language="cpp">
VImage VImage::add(VImage right, VOption *options)
{
  VImage out;

  call("add" ,
    (options ? options : VImage::option()) ->
      set("out", &amp;out) ->
      set("left", *this) ->
      set("right", right));

  return out;
}
</programlisting>

      Where VImage::call() is the generic call-a-vips8-operation function.
    </para>
  </refsect3>
</refentry>