Added a section on how to compute pixel offsets for

2004-01-06  Federico Mena Quintero  <federico@ximian.com>

* gdk-pixbuf/tmpl/gdk-pixbuf.sgml: Added a section on how to
compute pixel offsets for gdk_pixbuf_get_pixels().
This commit is contained in:
Federico Mena Quintero 2004-01-07 03:16:10 +00:00 committed by Federico Mena Quintero
parent 65fa5b0f98
commit 9d6ed04dff
2 changed files with 126 additions and 0 deletions

View File

@ -15,6 +15,11 @@ Wed Jan 7 01:26:07 2004 Matthias Clasen <maclas@gmx.de>
* gdk-pixbuf/gdk-pixbuf-sections.txt: Add gdk_pixbuf_get_file_info.
2004-01-06 Federico Mena Quintero <federico@ximian.com>
* gdk-pixbuf/tmpl/gdk-pixbuf.sgml: Added a section on how to
compute pixel offsets for gdk_pixbuf_get_pixels().
Mon Dec 29 01:40:20 2003 Matthias Clasen <maclas@gmx.de>
* gtk/gtk-sections.txt: Add gtk_accel_map_{un,}lock_path.

View File

@ -11,6 +11,81 @@ Information that describes an image.
information that describes an image in memory.
</para>
<section id="image-data">
<title>Image Data</title>
<para>
Image data in a pixbuf is stored in memory in uncompressed,
packed format. Rows in the image are stored top to bottom, and
in each row pixels are stored from left to right. There may be
padding at the end of a row. The "rowstride" value of a pixbuf,
as returned by gdk_pixbuf_get_rowstride(), indicates the number
of bytes between rows.
</para>
<example id="put-pixel">
<title>put_pixel() example</title>
<para>
The following code illustrates a simple put_pixel() function
for RGB pixbufs with 8 bits per channel with an alpha channel.
It is not included in the gdk-pixbuf library for performance
reasons; rather than making several function calls for each
pixel, your own code can take shortcuts.
</para>
<programlisting>
static void
put_pixel (GdkPixbuf *pixbuf, int x, int y, guchar red, guchar green, guchar blue, guchar alpha)
{
int width, height, rowstride, n_channels;
guchar *pixels, *p;
n_channels = gdk_pixbuf_get_n_channels (pixbuf);
g_assert (gdk_pixbuf_get_colorspace (pixbuf) == GDK_COLORSPACE_RGB);
g_assert (gdk_pixbuf_get_bits_per_sample (pixbuf) == 8);
g_assert (gdk_pixbuf_get_has_alpha (pixbuf));
g_assert (n_channels == 4);
width = gdk_pixbuf_get_width (pixbuf);
height = gdk_pixbuf_get_height (pixbuf);
g_assert (x &gt;= 0 && x < width);
g_assert (y &gt;= 0 && y < height);
rowstride = gdk_pixbuf_get_rowstride (pixbuf);
pixels = gdk_pixbuf_get_pixels (pixbuf);
p = pixels + y * rowstride + x * n_channels;
p[0] = red;
p[1] = green;
p[2] = blue;
p[3] = alpha;
}
</programlisting>
<para>
This function will not work for pixbufs with images that are
other than 8 bits per sample or channel, but it will work for
most of the pixbufs that GTK+ uses.
</para>
</example>
<note>
<para>
If you are doing memcpy() of raw pixbuf data, note that the
last row in the pixbuf may not be as wide as the full
rowstride, but rather just as wide as the pixel data needs to
be. That is, it is unsafe to do <literal>memcpy (dest,
pixels, rowstride * height)</literal> to copy a whole pixbuf.
Use gdk_pixbuf_copy() instead, or compute the width in bytes
of the last row as <literal>width * ((n_channels *
bits_per_sample + 7) / 8)</literal>.
</para>
</note>
</section>
<!-- ##### SECTION See_Also ##### -->
<para>
</para>
@ -76,6 +151,46 @@ In the future it will do full alpha compositing.
</para>
<!-- ##### ARG GdkPixbuf:bits-per-sample ##### -->
<para>
</para>
<!-- ##### ARG GdkPixbuf:colorspace ##### -->
<para>
</para>
<!-- ##### ARG GdkPixbuf:has-alpha ##### -->
<para>
</para>
<!-- ##### ARG GdkPixbuf:height ##### -->
<para>
</para>
<!-- ##### ARG GdkPixbuf:n-channels ##### -->
<para>
</para>
<!-- ##### ARG GdkPixbuf:pixels ##### -->
<para>
</para>
<!-- ##### ARG GdkPixbuf:rowstride ##### -->
<para>
</para>
<!-- ##### ARG GdkPixbuf:width ##### -->
<para>
</para>
<!-- ##### FUNCTION gdk_pixbuf_get_colorspace ##### -->
<para>
@ -163,3 +278,9 @@ End:
@Returns:
<!--
Local variables:
mode: sgml
sgml-parent-document: ("../gdk-pixbuf.sgml" "book" "refsect2")
End:
-->