CVE-2016-1238 mitigation

[imager.git] / lib / Imager / ImageTypes.pod

diff --git a/lib/Imager/ImageTypes.pod b/lib/Imager/ImageTypes.pod
index 44190d148604fc2f5ad87c01b3f1f19c0ea11e2c..c383ea00a5e87693964e9a754c48450b1b5ea471 100644 (file)
--- a/lib/Imager/ImageTypes.pod
+++ b/lib/Imager/ImageTypes.pod
@@ -1,6 +1,6 @@
 =head1 NAME
 
 =head1 NAME
 

-Imager::ImageTypes - Internal image representation information
+Imager::ImageTypes - image models for Imager

 
 =head1 SYNOPSIS
 
 
 =head1 SYNOPSIS
 


@@ -52,9 +52,10 @@ Imager::ImageTypes - Internal image representation information
     print "\n";
   
   } else {
     print "\n";
   
   } else {

-  
+    # palette info
+    my $count = $img->colorcount;  

     @colors = $img->getcolors();
     @colors = $img->getcolors();

-    print "Palette size: ".@colors."\n";
+    print "Palette size: $count\n";

     my $mx = @colors > 4 ? 4 : 0+@colors;
     print "First $mx entries:\n";
     for (@colors[0..$mx-1]) {
     my $mx = @colors > 4 ? 4 : 0+@colors;
     print "First $mx entries:\n";
     for (@colors[0..$mx-1]) {


@@ -73,20 +74,53 @@ Imager::ImageTypes - Internal image representation information
     print "No tags in image\n";
   }
   
     print "No tags in image\n";
   }
   

-

 =head1 DESCRIPTION
 
 =head1 DESCRIPTION
 

-Imager supports various internal image representations of images.  The
-two major classes are direct mode and paletted mode.  In paletted mode
-an image has a numbered list of colors and the color of each pixel is
-determined by an index into the table.  In direct mode there is no
-color palette and each pixel has a seperate value for red green and
-blue for RGB images.  To complicate matters it's possible to have
-other color spaces than RGB, for example, gray, gray and alpha, or
-red, green, blue and alpha.
+Imager supports two basic models of image:
+
+=over
+
+=item *
+
+direct color - all samples are stored for every pixel.  eg. for an
+8-bit/sample RGB image, 24 bits are stored for each pixel.
+
+=item *
+
+paletted - an index into a table of colors is stored for each pixel.
+
+=back
+
+Direct color or paletted images can have 1 to 4 samples per color
+stored.  Imager treats these as follows:
+
+=over
+
+=item *
+
+1 sample per color - gray scale image.
+
+=item *
+
+2 samples per color - gray scale image with alpha channel, allowing
+transparency.
+
+=item *
+
+3 samples per color - RGB image.
+
+=item *

 
 

-In addition it's possible to have direct type images with 8 bits/channel
-16 bits/channel or double/channel (64 bits on many systems).
+4 samples per color - RGB image with alpha channel, allowing
+transparency.
+
+=back
+
+Direct color images can have sample sizes of 8-bits per sample,
+16-bits per sample or a double precision floating point number per
+sample (64-bits on many systems).
+
+Paletted images are always 8-bits/sample.

 
 To query an existing image about it's parameters see the C<bits()>,
 C<type()>, C<getwidth()>, C<getheight()>, C<getchannels()> and
 
 To query an existing image about it's parameters see the C<bits()>,
 C<type()>, C<getwidth()>, C<getheight()>, C<getchannels()> and


@@ -95,11 +129,16 @@ C<virtual()> methods.
 The coordinate system in Imager has the origin in the upper left
 corner, see L<Imager::Draw> for details.
 
 The coordinate system in Imager has the origin in the upper left
 corner, see L<Imager::Draw> for details.
 

+The alpha channel when one is present is considered unassociated -
+ie the color data has not been scaled by the alpha channel.  Note
+that not all code follows this (recent) rule, but will over time.
+

 =head2 Creating Imager Objects
 
 =over
 
 =head2 Creating Imager Objects
 
 =over
 

-=item new
+=item new()
+X<new(), Imager methods>

 
   $img = Imager->new();
   $img->read(file=>"alligator.ppm") or die $img->errstr;
 
   $img = Imager->new();
   $img->read(file=>"alligator.ppm") or die $img->errstr;


@@ -112,11 +151,115 @@ method on later.
   $img = Imager->new(%opts); # create direct mode RGBA image
   $img = Imager->new(%opts, channels=>4); # create direct mode RGBA image
 
   $img = Imager->new(%opts); # create direct mode RGBA image
   $img = Imager->new(%opts, channels=>4); # create direct mode RGBA image
 

+You can also read a file from new():
+
+  $img = Imager->new(file => "someimage.png");
+
+The parameters for new are:
+
+=over
+
+=item *
+
+C<xsize>, C<ysize> - Defines the width and height in pixels of the
+image.  These must be positive.
+
+If not supplied then only placeholder object is created, which can be
+supplied to the C<read()> or C<img_set()> methods.
+
+=item *
+
+C<channels> - The number of channels for the image.  Default 3.  Valid
+values are from 1 to 4.
+
+=item *
+
+C<model> - this overrides the value, if any, supplied for C<channels>.
+This can be one of C<gray>, C<graya>, C<rgb> or C<rgba>.
+
+=item *
+
+C<bits> - The storage type for samples in the image.  Default: 8.
+Valid values are:
+
+=over
+
+=item *
+
+C<8> - One byte per sample.  256 discrete values.
+
+=item *
+
+C<16> - 16-bits per sample, 65536 discrete values.
+
+=item *
+
+C<double> - one C double per sample.
+
+=back
+
+Note: you can use any Imager function on any sample size image.
+
+Paletted images always use 8 bits/sample.
+
+=item *
+
+C<type> - either C<'direct'> or C<'paletted'>.  Default: C<'direct'>.
+
+Direct images store color values for each pixel.  
+
+Paletted images keep a table of up to 256 colors called the palette,
+each pixel is represented as an index into that table.
+
+In most cases when working with Imager you will want to use the
+C<direct> image type.
+
+If you draw on a C<paletted> image with a color not in the image's
+palette then Imager will transparently convert it to a C<direct>
+image.
+
+=item *
+
+C<maxcolors> - the maximum number of colors in a paletted image.
+Default: 256.  This must be in the range 1 through 256.
+
+=item *
+
+C<file>, C<fh>, C<fd>, C<callback>, C<readcb>, or C<io> - specify a
+file name, filehandle, file descriptor or callback to read image data
+from.  See L<Imager::Files> for details.  The typical use is:
+
+  my $im = Imager->new(file => $filename);
+
+=item *
+
+C<filetype> - treated as the file format parameter, as for C<type>
+with the read() method, eg:
+
+  my $im = Imager->new(file => $filename, filetype => "gif");
+
+In most cases Imager will detect the file's format itself.
+
+=back
+
+In the simplest case just supply the width and height of the image:
+
+  # 8 bit/sample, RGB image
+  my $img = Imager->new(xsize => $width, ysize => $height);
+
+or if you want an alpha channel:
+
+  # 8 bits/sample, RGBA image
+  my $img = Imager->new(xsize => $width, ysize => $height, channels=>4);
+
+Note that it I<is> possible for image creation to fail, for example if
+channels is out of range, or if the image would take too much memory.
+

 To create paletted images, set the 'type' parameter to 'paletted':
 
   $img = Imager->new(xsize=>200, ysize=>200, type=>'paletted');
 
 To create paletted images, set the 'type' parameter to 'paletted':
 
   $img = Imager->new(xsize=>200, ysize=>200, type=>'paletted');
 

-which creates an image with a maxiumum of 256 colors, which you can
+which creates an image with a maximum of 256 colors, which you can

 change by supplying the C<maxcolors> parameter.
 
 For improved color precision you can use the bits parameter to specify
 change by supplying the C<maxcolors> parameter.
 
 For improved color precision you can use the bits parameter to specify


@@ -136,84 +279,203 @@ Note that as of this writing all functions should work on images with
 more than 8-bits/channel, but many will only work at only
 8-bit/channel precision.
 
 more than 8-bits/channel, but many will only work at only
 8-bit/channel precision.
 

-Currently only 8-bit, 16-bit, and double per channel image types are
-available, this may change later.
+If you want an empty Imager object to call the read() method on, just
+call new() with no parameters:
+
+  my $img = Imager->new;
+  $img->read(file=>$filename)
+    or die $img->errstr;

 
 

-=item img_set
+Though it's much easier now to just call new() with a C<file>
+parameter:

 
 

-If you have an existing image, use img_set() to change it's dimensions
-- this will destroy any existing image data:
+  my $img = Imager->new(file => $filename)
+    or die Imager->errstr;
+
+If none of C<xsize>, C<ysize>, C<file>, C<fh>, C<fd>, C<callback>,
+C<readcb>, C<data>, C<io> is supplied, and other parameters I<are> supplied
+C<< Imager->new >> will return failure rather than returning an empty
+image object.
+
+=item img_set()
+X<img_set>
+
+img_set destroys the image data in the object and creates a new one
+with the given dimensions and channels.  For a way to convert image
+data between formats see the C<convert()> method.

 
   $img->img_set(xsize=>500, ysize=>500, channels=>4);
 
 
   $img->img_set(xsize=>500, ysize=>500, channels=>4);
 

+This takes exactly the same parameters as the new() method, excluding
+those for reading from files.
+

 =back
 
 =back
 

-=head2 Getting Information About an Imager Object
+=head2 Image Attribute functions
+
+These return basic attributes of an image object.

 
 =over
 
 
 =over
 

-=item getwidth
+=item getwidth()

 
   print "Image width: ", $img->getwidth(), "\n";
 
 The C<getwidth()> method returns the width of the image.  This value
 
   print "Image width: ", $img->getwidth(), "\n";
 
 The C<getwidth()> method returns the width of the image.  This value

-comes either from C<new()> with xsize,ysize parameters or from reading
-data from a file with C<read()>.  If called on an image that has no
-valid data in it like C<Imager-&gt;new()> returns, the return value of
-C<getwidth()> is undef.
+comes either from C<new()> with C<xsize>, C<ysize> parameters or from
+reading data from a file with C<read()>.  If called on an image that
+has no valid data in it like C<Imager-E<gt>new()> returns, the return
+value of C<getwidth()> is undef.

 
 

-=item getheight
+=item getheight()

 
   print "Image height: ", $img->getheight(), "\n";
 
 
   print "Image height: ", $img->getheight(), "\n";
 

-Same details apply as for L<getwidth>.
+Same details apply as for L</getwidth()>.

 
 

-=item getchannels
+=item getchannels()
+X<getchannels() method>X<methods, getchannels()>

 
   print "Image has ",$img->getchannels(), " channels\n";
 
 
   print "Image has ",$img->getchannels(), " channels\n";
 

-To get the number of channels in an image C<getchannels()> is used.
+Returns the number of channels in an image.

 
 

+Note: previously the number of channels in an image mapped directly to
+the color model of the image, ie a 4 channel image was always RGBA.
+This may change in a future release of Imager.

 
 

-=item getcolorcount
+Returns an empty list if the image object is not initialized.

 
 

-It is possible to have Imager find the number of colors in an image by
-with the C<getcolorcount()> method. It requires memory proportionally
-to the number of colors in the image so it is possible to have it stop
-sooner if you only need to know if there are more than a certain
-number of colors in the image.  If there are more colors than asked
-for the function return undef.  Examples:
+=item colorchannels()
+X<colorchannels() method>X<methods, colorchannels()>

 
 

-  if (!defined($img->getcolorcount(maxcolors=>512)) {
-    print "Less than 512 colors in image\n";
-  }
+Returns the number of color channels.
+
+Currently this is always 1 or 3, but may be 0 for some rare images in
+a future version of Imager.
+
+Returns an empty list if the image object is not initialized.
+
+=item colormodel()
+X<colormodel method>X<methods, colormodel>
+
+Returns the color model of the image, including whether there is an
+alpha channel.
+
+By default this is returned as a string, one of C<unknown>, C<gray>,
+C<graya>, C<rgb> or C<rgba>.
+
+If you call C<colormodel()> with a true numeric parameter:
+
+  my $model = $img->colormodel(numeric => 1);

 
 

+then the color model is returned as a number, mapped as follows:

 
 

-=item bits
+  Numeric  String
+  -------  ------
+      0    unknown
+      1    gray
+      2    graya
+      3    rgb
+      4    rgba
+
+=item alphachannel()
+X<alphachannel() method>X<methods, alphachannel()>
+
+Returns the channel index of the alpha channel of the image.
+
+This is 1 for grayscale images with alpha, 3 for RGB images with alpha
+and will return C<undef> for all other images.
+
+Returns an empty list if the image object is not initialized.
+
+=item bits()

 
 The bits() method retrieves the number of bits used to represent each
 channel in a pixel, 8 for a normal image, 16 for 16-bit image and
 'double' for a double/channel image.
 
 
 The bits() method retrieves the number of bits used to represent each
 channel in a pixel, 8 for a normal image, 16 for 16-bit image and
 'double' for a double/channel image.
 

-=item type
+  if ($img->bits eq 8) {
+    # fast but limited to 8-bits/sample
+  }
+  else {
+    # slower but more precise
+  }
+
+Returns an empty list if the image object is not initialized.
+
+=item type()
+
+The type() method returns either 'direct' for direct color images or
+'paletted' for paletted images.
+
+  if ($img->type eq 'paletted') {
+    # print the palette
+    for my $color ($img->getcolors) {
+      print join(",", $color->rgba), "\n";
+    }
+  }
+
+Returns an empty list if the image object is not initialized.
+
+=item virtual()
+
+The virtual() method returns non-zero if the image contains no actual
+pixels, for example masked images.
+
+=for stopwords SDL
+
+This may also be used for non-native Imager images in the future, for
+example, for an Imager object that draws on an SDL surface.
+
+=item is_bilevel()
+
+Tests if the image will be written as a monochrome or bi-level image
+for formats that support that image organization.
+
+In scalar context, returns true if the image is bi-level.
+
+In list context returns a list:
+
+  ($is_bilevel, $zero_is_white) = $img->is_bilevel;
+
+An image is considered bi-level, if all of the following are true:
+
+=over
+
+=item *
+
+the image is a paletted image
+
+=item *

 
 

-The type() method returns either 'direct' for truecolor images or
-'paletted' for paletted images.  
+the image has 1 or 3 channels

 
 

-=item virtual
+=item *

 
 

-The virtual() method returns non-zero
-if the image contains no actual pixels, for example masked images.
+the image has only 2 colors in the palette

 
 

+=item *
+
+those 2 colors are black and white, in either order.

 
 =back
 
 
 =back
 

+If a real bi-level organization image is ever added to Imager, this
+function will return true for that too.
+
+Returns an empty list if the image object is not initialized.
+
+=back

 
 =head2 Direct Type Images
 
 
 =head2 Direct Type Images
 

+Direct images store the color value directly for each pixel in the
+image.
+

 =over
 
 =over
 

-=item getmask
+=item getmask()

 
   @rgbanames = qw( red green blue alpha );
   my $mask = $img->getmask();
 
   @rgbanames = qw( red green blue alpha );
   my $mask = $img->getmask();


@@ -222,12 +484,15 @@ if the image contains no actual pixels, for example masked images.
     print $rgbanames[$_],"\n" if $mask & 1<<$_;
   }
 
     print $rgbanames[$_],"\n" if $mask & 1<<$_;
   }
 

+=for stopwords th
+

 C<getmask()> is used to fetch the current channel mask.  The mask
 determines what channels are currently modifiable in the image.  The
 C<getmask()> is used to fetch the current channel mask.  The mask
 determines what channels are currently modifiable in the image.  The

-channel mask is an integer value, if the i-th lsb is set the i-th
-channel is modifiable.
+channel mask is an integer value, if the C<i-th> least significant bit
+is set the C<i-th> channel is modifiable.  eg. a channel mask of 0x5
+means only channels 0 and 2 are writable.

 
 

-=item setmask
+=item setmask()

 
   $mask = $img->getmask();
   $img->setmask(mask=>8);     # modify alpha only
 
   $mask = $img->getmask();
   $img->setmask(mask=>8);     # modify alpha only


@@ -237,12 +502,15 @@ channel is modifiable.
   $img->setmask(mask=>$mask); # restore previous mask
 
 C<setmask()> is used to set the channel mask of the image.  See
   $img->setmask(mask=>$mask); # restore previous mask
 
 C<setmask()> is used to set the channel mask of the image.  See

-L<getmask> for details.
+L</getmask()> for details.

 
 =back
 
 =head2 Palette Type Images
 
 
 =back
 
 =head2 Palette Type Images
 

+Paletted images keep an array of up to 256 colors, and each pixel is
+stored as an index into that array.
+

 In general you can work with paletted images in the same way as RGB
 images, except that if you attempt to draw to a paletted image with a
 color that is not in the image's palette, the image will be converted
 In general you can work with paletted images in the same way as RGB
 images, except that if you attempt to draw to a paletted image with a
 color that is not in the image's palette, the image will be converted


@@ -254,7 +522,7 @@ C<getcolors()> and C<findcolor()>:
 
 =over
 
 
 =over
 

-=item addcolors
+=item addcolors()

 
 You can add colors to a paletted image with the addcolors() method:
 
 
 You can add colors to a paletted image with the addcolors() method:
 


@@ -265,14 +533,31 @@ You can add colors to a paletted image with the addcolors() method:
 The return value is the index of the first color added, or undef if
 adding the colors would overflow the palette.
 
 The return value is the index of the first color added, or undef if
 adding the colors would overflow the palette.
 

-=item setcolors
+The only parameter is C<colors> which must be a reference to an array
+of Imager::Color objects.
+
+=item setcolors()

 
   $img->setcolors(start=>$start, colors=>\@colors);
 
 Once you have colors in the palette you can overwrite them with the
 
   $img->setcolors(start=>$start, colors=>\@colors);
 
 Once you have colors in the palette you can overwrite them with the

-C<setcolors()> method:  C<sercolors()> returns true on success.
+C<setcolors()> method:  C<setcolors()> returns true on success.

 
 

-=item getcolors
+Parameters:
+
+=over
+
+=item *
+
+start - the first index to be set.  Default: 0
+
+=item *
+
+colors - reference to an array of Imager::Color objects.
+
+=back
+
+=item getcolors()

 
 To retrieve existing colors from the palette use the getcolors() method:
 
 
 To retrieve existing colors from the palette use the getcolors() method:
 


@@ -283,7 +568,7 @@ To retrieve existing colors from the palette use the getcolors() method:
   # get a range of colors
   my @colors = $img->getcolors(start=>$index, count=>$count);
 
   # get a range of colors
   my @colors = $img->getcolors(start=>$index, count=>$count);
 

-=item findcolor
+=item findcolor()

 
 To quickly find a color in the palette use findcolor():
 
 
 To quickly find a color in the palette use findcolor():
 


@@ -291,8 +576,97 @@ To quickly find a color in the palette use findcolor():
 
 which returns undef on failure, or the index of the color.
 
 
 which returns undef on failure, or the index of the color.
 

-You can get the current palette size with $img->colorcount, and the
-maximum size of the palette with $img->maxcolors.
+Parameter:
+
+=over
+
+=item *
+
+color - an Imager::Color object.
+
+=back
+
+=item colorcount()
+
+Returns the number of colors in the image's palette:
+
+  my $count = $img->colorcount;
+
+=item maxcolors()
+
+Returns the maximum size of the image's palette.
+
+  my $maxcount = $img->maxcolors;
+
+=back
+
+=head2 Color Distribution
+
+=over
+
+=item getcolorcount()
+
+Calculates the number of colors in an image.
+
+The amount of memory used by this is proportional to the number of
+colors present in the image, so to avoid using too much memory you can
+supply a maxcolors() parameter to limit the memory used.
+
+Note: getcolorcount() treats the image as an 8-bit per sample image.
+
+=over
+
+=item *
+
+X<maxcolors!getcolorcount>C<maxcolors> - the maximum number of colors to
+return.  Default: unlimited.
+
+=back
+
+  if (defined($img->getcolorcount(maxcolors=>512)) {
+    print "Less than 512 colors in image\n";
+  }
+
+=item getcolorusagehash()
+
+Calculates a histogram of colors used by the image.
+
+=over
+
+=item *
+
+X<maxcolors!getcolorusagehash>C<maxcolors> - the maximum number of colors
+to return.  Default: unlimited.
+
+=back
+
+Returns a reference to a hash where the keys are the raw color as
+bytes, and the values are the counts for that color.
+
+The alpha channel of the image is ignored.  If the image is gray scale
+then the hash keys will each be a single character.
+
+  my $colors = $img->getcolorusagehash;
+  my $blue_count = $colors->{pack("CCC", 0, 0, 255)} || 0;
+  print "#0000FF used $blue_count times\n";
+
+=item getcolorusage()
+
+Calculates color usage counts and returns just the counts.
+
+=over
+
+=item *
+
+X<maxcolors!getcolorusage>C<maxcolors> - the maximum number of colors
+to return.  Default: unlimited.
+
+=back
+
+Returns a list of the color frequencies in ascending order.
+
+  my @counts = $img->getcolorusage;
+  print "The most common color is used $counts[0] times\n";

 
 =back
 
 
 =back
 


@@ -303,24 +677,101 @@ the palette, the image will be internally converted to a normal image.
 
 =over
 
 
 =over
 

-=item to_paletted
+=item to_paletted()

 
 You can create a new paletted image from an existing image using the
 to_paletted() method:
 
  $palimg = $img->to_paletted(\%opts)
 
 
 You can create a new paletted image from an existing image using the
 to_paletted() method:
 
  $palimg = $img->to_paletted(\%opts)
 

-where %opts contains the options specified under L<Quantization options>.
+where %opts contains the options specified under L</Quantization options>.

 
 

-=item to_rgb8
+  # convert to a paletted image using the web palette
+  # use the closest color to each pixel
+  my $webimg = $img->to_paletted({ make_colors => 'webmap' });
+
+  # convert to a paletted image using a fairly optimal palette
+  # use an error diffusion dither to try to reduce the average error
+  my $optimag = $img->to_paletted({ make_colors => 'mediancut',
+                                    translate => 'errdiff' });
+
+=item to_rgb8()

 
 You can convert a paletted image (or any image) to an 8-bit/channel
 RGB image with:
 
   $rgbimg = $img->to_rgb8;
 
 
 You can convert a paletted image (or any image) to an 8-bit/channel
 RGB image with:
 
   $rgbimg = $img->to_rgb8;
 

+No parameters.
+
+=item to_rgb16()
+
+Convert a paletted image (or any image) to a 16-bit/channel RGB image.
+
+  $rgbimg = $img->to_rgb16;
+
+No parameters.
+
+=item to_rgb_double()
+
+Convert a paletted image (or any image) to an double/channel direct
+color image.
+
+  $rgbimg = $img->to_rgb_double;

 
 

-=head2 Masked Images
+No parameters.
+
+=item masked()
+
+Creates a masked image.  A masked image lets you create an image proxy
+object that protects parts of the underlying target image.
+
+In the discussion below there are 3 image objects involved:
+
+=over
+
+=item *
+
+the masked image - the return value of the masked() method.  Any
+writes to this image are written to the target image, assuming the
+mask image allows it.
+
+=item *
+
+the mask image - the image that protects writes to the target image.
+Supplied as the C<mask> parameter to the masked() method.
+
+=item *
+
+the target image - the image you called the masked() method on.  Any
+writes to the masked image end up on this image.
+
+=back
+
+Parameters:
+
+=over
+
+=item *
+
+mask - the mask image.  If not supplied then all pixels in the target
+image are writable.  On each write to the masked image, only pixels
+that have non-zero in channel 0 of the mask image will be written to
+the original image.  Default: none, if not supplied then no masking is
+done, but the other parameters are still honored.
+
+=item *
+
+left, top - the offset of writes to the target image.  eg. if you
+attempt to set pixel (x,y) in the masked image, then pixel (x+left,
+y+top) will be written to in the original image.
+
+=item *
+
+bottom, right - the bottom right of the area in the target available
+from the masked image.
+
+=back

 
 Masked images let you control which pixels are modified in an
 underlying image.  Where the first channel is completely black in the
 
 Masked images let you control which pixels are modified in an
 underlying image.  Where the first channel is completely black in the


@@ -328,22 +779,37 @@ mask image, writes to the underlying image are ignored.
 
 For example, given a base image called $img:
 
 
 For example, given a base image called $img:
 

-  my $mask = Imager->new(xsize=>$img->getwidth, ysize=>getheight,
+  my $mask = Imager->new(xsize=>$img->getwidth, ysize=>$img->getheight,

                          channels=>1);
   # ... draw something on the mask
   my $maskedimg = $img->masked(mask=>$mask);
 
                          channels=>1);
   # ... draw something on the mask
   my $maskedimg = $img->masked(mask=>$mask);
 

-You can specifiy the region of the underlying image that is masked
+  # now draw on $maskedimg and it will only draw on areas of $img 
+  # where $mask is non-zero in channel 0.
+
+You can specify the region of the underlying image that is masked

 using the left, top, right and bottom options.
 
 If you just want a subset of the image, without masking, just specify
 using the left, top, right and bottom options.
 
 If you just want a subset of the image, without masking, just specify

-the region without specifying a mask.
+the region without specifying a mask.  For example:
+
+  # just work with a 100x100 region of $img
+  my $maskedimg = $img->masked(left => 100, top=>100,
+                               right=>200, bottom=>200);

 
 

+=item make_palette()

 
 

+This doesn't perform an image conversion, but it can be used to
+construct a common palette for use in several images:

 
 

+  my @colors = Imager->make_palette(\%opts, @images);

 
 

+You must supply at least one image, even if the C<make_colors>
+parameter produces a fixed palette.

 
 

+On failure returns no colors and you can check C<< Imager->errstr >>.

 
 

+=back

 
 =head2 Tags
 
 
 =head2 Tags
 


@@ -352,22 +818,50 @@ stored as pixels of the image.
 
 At the perl level each tag has a name or code and a value, which is an
 integer or an arbitrary string.  An image can contain more than one
 
 At the perl level each tag has a name or code and a value, which is an
 integer or an arbitrary string.  An image can contain more than one

-tag with the same name or code.
+tag with the same name or code, but having more than one tag with the
+same name is discouraged.

 
 You can retrieve tags from an image using the tags() method, you can
 get all of the tags in an image, as a list of array references, with
 
 You can retrieve tags from an image using the tags() method, you can
 get all of the tags in an image, as a list of array references, with

-the code or name of the tag followed by the value of the tag:
+the code or name of the tag followed by the value of the tag.

 
 

+Imager's support for fairly limited, for access to pretty much all
+image metadata you may want to try L<Image::ExifTool>.
+
+=over
+
+=item tags()
+
+Retrieve tags from the image.
+
+With no parameters, retrieves a list array references, each containing
+a name and value: all tags in the image:
+
+  # get a list of ( [ name1 => value1 ], [ name2 => value2 ] ... )

   my @alltags = $img->tags;
   my @alltags = $img->tags;

+  print $_->[0], ":", $_->[1], "\n" for @all_tags;
+
+  # or put it in a hash, but this will lose duplicates
+  my %alltags = map @$_, $img->tags;
+
+in scalar context this returns the number of tags:
+
+  my $num_tags = $img->tags;
+
+or you can get all tags values for the given name:
+
+  my @namedtags = $img->tags(name => $name);

 
 

-or you can get all tags that have a given name:
+in scalar context this returns the first tag of that name:

 
 

-  my @namedtags = $img->tags(name=>$name);
+  my $firstnamed = $img->tags(name => $name);

 
 or a given code:
 
   my @tags = $img->tags(code=>$code);
 
 
 or a given code:
 
   my @tags = $img->tags(code=>$code);
 

+=item addtag()
+

 You can add tags using the addtag() method, either by name:
 
   my $index = $img->addtag(name=>$name, value=>$value);
 You can add tags using the addtag() method, either by name:
 
   my $index = $img->addtag(name=>$name, value=>$value);


@@ -376,6 +870,8 @@ or by code:
 
   my $index = $img->addtag(code=>$code, value=>$value);
 
 
   my $index = $img->addtag(code=>$code, value=>$value);
 

+=item deltag()
+

 You can remove tags with the deltag() method, either by index:
 
   $img->deltag(index=>$index);
 You can remove tags with the deltag() method, either by index:
 
   $img->deltag(index=>$index);


@@ -390,6 +886,12 @@ or by code:
 
 In each case deltag() returns the number of tags deleted.
 
 
 In each case deltag() returns the number of tags deleted.
 

+=item settag()
+
+settag() replaces any existing tags with a new tag.  This is
+equivalent to calling deltag() then addtag().
+
+=back

 
 =head2 Common Tags
 
 
 =head2 Common Tags
 


@@ -404,256 +906,425 @@ some standard information.
 
 =over
 
 
 =over
 

-=item i_xres
+=item *

 
 

-=item i_yres
+X<i_xres tag>X<i_yres tag>X<tags, i_xres>X<tags, i_yres>C<i_xres>, C<i_yres>
+- The spatial resolution of the image in pixels per inch.  If the
+image format uses a different scale, eg. pixels per meter, then this
+value is converted.  A floating point number stored as a string.

 
 

-The spatial resolution of the image in pixels per inch.  If the image
-format uses a different scale, eg. pixels per meter, then this value
-is converted.  A floating point number stored as a string.
+  # our image was generated as a 300 dpi image
+  $img->settag(name => 'i_xres', value => 300);
+  $img->settag(name => 'i_yres', value => 300);

 
 

-=item i_aspect_only
+  # 100 pixel/cm for a TIFF image
+  $img->settag(name => 'tiff_resolutionunit', value => 3); # RESUNIT_CENTIMETER
+  # convert to pixels per inch, Imager will convert it back
+  $img->settag(name => 'i_xres', value => 100 * 2.54);
+  $img->settag(name => 'i_yres', value => 100 * 2.54);

 
 

-If this is non-zero then the values in i_xres and i_yres are treated
-as a ratio only.  If the image format does not support aspect ratios
-then this is scaled so the smaller value is 72dpi.
+=item *

 
 

-=item i_incomplete
+X<i_aspect_only tag>X<tags, i_aspect_only>C<i_aspect_only> - If this is
+non-zero then the values in i_xres and i_yres are treated as a ratio
+only.  If the image format does not support aspect ratios then this is
+scaled so the smaller value is 72 DPI.

 
 

-If this tag is present then the whole image could not be read.  This
-isn't implemented for all images yet.
+=item *

 
 

-=back
+X<i_incomplete tag>X<tags, i_incomplete>C<i_incomplete> - If this tag is
+present then the whole image could not be read.  This isn't
+implemented for all images yet, and may not be.
+
+=item *
+
+X<i_lines_read tag>X<tags, i_lines_read>C<i_lines_read> - If
+C<i_incomplete> is set then this tag may be set to the number of
+scan lines successfully read from the file.  This can be used to decide
+whether an image is worth processing.
+
+=item *
+
+X<i_format tag>X<tags, i_format>i_format - The file format this file
+was read from.

 
 

+=item *

 
 

+X<i_background>X<tags, i_background>i_background - used when writing
+an image with an alpha channel to a file format that doesn't support
+alpha channels.  The C<write> method will convert a normal color
+specification like "#FF0000" into a color object for you, but if you
+set this as a tag you will need to format it like
+C<color(>I<red>C<,>I<green>C<,>I<blue>C<)>, eg color(255,0,0).

 
 

+=item *

 
 

+X<i_comment>C<i_comment> - used when reading or writing several image
+formats.  If the format has only one text field it will be read into
+the C<i_comment> tag or written to the file.
+
+=back

 
 =head2 Quantization options
 
 
 =head2 Quantization options
 

-These options can be specified when calling write_multi() for gif
-files, when writing a single image with the gifquant option set to
-'gen', or for direct calls to i_writegif_gen and i_writegif_callback.
+These options can be specified when calling
+L<Imager::ImageTypes/to_paletted()>, write_multi() for GIF files, when
+writing a single image with the C<gifquant> option set to C<gen>, or for
+direct calls to i_writegif_gen() and i_writegif_callback().

 
 =over
 
 
 =over
 

-=item colors
+=item *

 
 

-A arrayref of colors that are fixed.  Note that some color generators
-will ignore this.
+C<colors> - An arrayref of colors that are fixed.  Note that some
+color generators will ignore this.  If this is supplied it will be
+filled with the color table generated for the image.

 
 

-=item transp
+=item *

 
 

-The type of transparency processing to perform for images with an
-alpha channel where the output format does not have a proper alpha
-channel (eg. gif).  This can be any of:
+C<transp> - The type of transparency processing to perform for images
+with an alpha channel where the output format does not have a proper
+alpha channel (eg. GIF).  This can be any of:

 
 =over
 
 
 =over
 

-=item none
+=item *

 
 

-No transparency processing is done. (default)
+C<none> - No transparency processing is done. (default)

 
 

-=item threshold
+=item *

 
 

-Pixels more transparent that tr_threshold are rendered as transparent.
+C<threshold> - pixels more transparent than C<tr_threshold> are
+rendered as transparent.

 
 

-=item errdiff
+=item *

 
 

-An error diffusion dither is done on the alpha channel.  Note that
-this is independent of the translation performed on the colour
-channels, so some combinations may cause undesired artifacts.
+C<errdiff> - An error diffusion dither is done on the alpha channel.
+Note that this is independent of the translation performed on the
+color channels, so some combinations may cause undesired artifacts.

 
 

-=item ordered
+=item *

 
 

-The ordered dither specified by tr_orddith is performed on the alpha
-channel.
+C<ordered> - the ordered dither specified by tr_orddith is performed
+on the alpha channel.

 
 =back
 
 This will only be used if the image has an alpha channel, and if there
 
 =back
 
 This will only be used if the image has an alpha channel, and if there

-is space in the palette for a transparency colour.
+is space in the palette for a transparency color.

 
 

-=item tr_threshold
+=item *

 
 

-The highest alpha value at which a pixel will be made transparent when
-transp is 'threshold'. (0-255, default 127)
+C<tr_threshold> - the highest alpha value at which a pixel will be
+made transparent when C<transp> is 'threshold'. (0-255, default 127)

 
 

-=item tr_errdiff
+=item *

 
 

-The type of error diffusion to perform on the alpha channel when
-transp is 'errdiff'.  This can be any defined error diffusion type
-except for custom (see errdiff below).
+C<tr_errdiff> - The type of error diffusion to perform on the alpha
+channel when C<transp> is C<errdiff>.  This can be any defined error
+diffusion type except for custom (see C<errdiff> below).

 
 

-=item tr_orddith
+=item *

 
 

-The type of ordered dither to perform on the alpha channel when transp
-is 'ordered'.  Possible values are:
+C<tr_orddith> - The type of ordered dither to perform on the alpha
+channel when C<transp> is 'ordered'.  Possible values are:

 
 =over
 
 
 =over
 

-=item random
-
-A semi-random map is used.  The map is the same each time.
+=item *

 
 

-=item dot8
+C<random> - A semi-random map is used.  The map is the same each time.

 
 

-8x8 dot dither.
+=item *

 
 

-=item dot4
+C<dot8> - 8x8 dot dither.

 
 

-4x4 dot dither
+=item *

 
 

-=item hline
+C<dot4> - 4x4 dot dither

 
 

-horizontal line dither.
+=item *

 
 

-=item vline
+C<hline> - horizontal line dither.

 
 

-vertical line dither.
+=item *

 
 

-=item "/line"
+C<vline> - vertical line dither.

 
 

-=item slashline
+=item *

 
 

-diagonal line dither
+C</line>, C<slashline> - diagonal line dither

 
 

-=item '\line'
+=item *

 
 

-=item backline
+C<\line>, C<backline> - diagonal line dither

 
 

-diagonal line dither
+=item *

 
 

-=item tiny
+C<tiny> - dot matrix dither (currently the default).  This is probably
+the best for displays (like web pages).

 
 

-dot matrix dither (currently the default).  This is probably the best
-for displays (like web pages).
+=item *

 
 

-=item custom
-
-A custom dither matrix is used - see tr_map
+C<custom> - A custom dither matrix is used - see C<tr_map>.

 
 =back
 
 
 =back
 

-=item tr_map
+=item *

 
 

-When tr_orddith is custom this defines an 8 x 8 matrix of integers
-representing the transparency threshold for pixels corresponding to
-each position.  This should be a 64 element array where the first 8
-entries correspond to the first row of the matrix.  Values should be
-betweern 0 and 255.
+C<tr_map> - When tr_orddith is custom this defines an 8 x 8 matrix of
+integers representing the transparency threshold for pixels
+corresponding to each position.  This should be a 64 element array
+where the first 8 entries correspond to the first row of the matrix.
+Values should be between 0 and 255.

 
 

-=item make_colors
+=item *

 
 

-Defines how the quantization engine will build the palette(s).
-Currently this is ignored if 'translate' is 'giflib', but that may
-change.  Possible values are:
+C<make_colors> - Defines how the quantization engine will build the
+palette(s).  Currently this is ignored if C<translate> is C<giflib>,
+but that may change.  Possible values are:

 
 =over
 
 
 =over
 

-=item none
+=item *
+
+C<none> - only colors supplied in 'colors' are used.
+
+=item *
+
+C<webmap> - the web color map is used (need URL here.)
+
+=item *
+
+C<addi> - The original code for generating the color map (Addi's code) is
+used.
+
+=item *

 
 

-Only colors supplied in 'colors' are used.
+C<mediancut> - Uses a median-cut algorithm, faster than C<addi>, but not
+as good a result.

 
 

-=item webmap
+=item *

 
 

-The web color map is used (need url here.)
+C<mono>, C<monochrome> - a fixed black and white palette, suitable for
+producing bi-level images (eg. facsimile)

 
 

-=item addi
+=item *

 
 

-The original code for generating the color map (Addi's code) is used.
+C<gray>, C<gray4>, C<gray16> - make fixed gray palette with 256, 4 or
+16 entries respectively.

 
 =back
 
 Other methods may be added in the future.
 
 
 =back
 
 Other methods may be added in the future.
 

-=item colors
+=item *

 
 

-A arrayref containing Imager::Color objects, which represents the
-starting set of colors to use in translating the images.  webmap will
-ignore this.  The final colors used are copied back into this array
-(which is expanded if necessary.)
+C<colors> - an arrayref containing Imager::Color objects, which
+represents the starting set of colors to use in translating the
+images.  C<webmap> will ignore this.  On return the final colors used
+are copied back into this array (which is expanded if necessary.)

 
 

-=item max_colors
+=item *

 
 

-The maximum number of colors to use in the image.
+C<max_colors> - the maximum number of colors to use in the image.

 
 

-=item translate
+=item *

 
 

-The method used to translate the RGB values in the source image into
-the colors selected by make_colors.  Note that make_colors is ignored
-whene translate is 'giflib'.
+C<translate> - The method used to translate the RGB values in the
+source image into the colors selected by make_colors.  Note that
+make_colors is ignored when C<translate> is C<giflib>.

 
 Possible values are:
 
 =over
 
 
 Possible values are:
 
 =over
 

-=item giflib
+=item *

 
 

-The giflib native quantization function is used.
+C<giflib> - this is a historical equivalent for C<closest> that also
+forces C<make_colors> to C<mediancut>.

 
 

-=item closest
+=item *

 
 

-The closest color available is used.
+C<closest> - the closest color available is used.

 
 

-=item perturb
+=item *

 
 

-The pixel color is modified by perturb, and the closest color is chosen.
+C<perturb> - the pixel color is modified by C<perturb>, and the
+closest color is chosen.

 
 

-=item errdiff
+=item *

 
 

-An error diffusion dither is performed.
+C<errdiff> - an error diffusion dither is performed.  If the supplied
+(or generated) palette contains only grays the source colors are
+converted to gray before error diffusion is performed.

 
 =back
 
 
 =back
 

-It's possible other transate values will be added.
+It's possible other C<translate> values will be added.

 
 

-=item errdiff
+=item *

 
 

-The type of error diffusion dither to perform.  These values (except
-for custom) can also be used in tr_errdif.
+C<errdiff> - The type of error diffusion dither to perform.  These
+values (except for custom) can also be used in tr_errdif.
+
+=for stopwords Floyd-Steinberg Jarvis Judice Ninke Stucki

 
 =over
 
 
 =over
 

-=item floyd
+=item *
+
+C<floyd> - Floyd-Steinberg dither
+
+=item *
+
+C<jarvis> - Jarvis, Judice and Ninke dither
+
+=item *

 
 

-Floyd-Steinberg dither
+C<stucki> - Stucki dither

 
 

-=item jarvis
+=item *

 
 

-Jarvis, Judice and Ninke dither
+C<custom> - custom.  If you use this you must also set C<errdiff_width>,
+C<errdiff_height> and C<errdiff_map>.

 
 

-=item stucki
+=back
+
+=item *

 
 

-Stucki dither
+C<errdiff_width>, C<errdiff_height>, C<errdiff_orig>, C<errdiff_map> -
+When C<translate> is C<errdiff> and C<errdiff> is C<custom> these
+define a custom error diffusion map.  C<errdiff_width> and
+C<errdiff_height> define the size of the map in the arrayref in
+C<errdiff_map>.  C<errdiff_orig> is an integer which indicates the
+current pixel position in the top row of the map.

 
 

-=item custom
+=item *

 
 

-Custom.  If you use this you must also set errdiff_width,
-errdiff_height and errdiff_map.
+C<perturb> - When translate is C<perturb> this is the magnitude of the
+random bias applied to each channel of the pixel before it is looked
+up in the color table.

 
 =back
 
 
 =back
 

-=item errdiff_width
+=head1 INITIALIZATION

 
 

-=item errdiff_height
+This documents the Imager initialization function, which you will
+almost never need to call.

 
 

-=item errdiff_orig
+=over

 
 

-=item errdiff_map
+=item init()

 
 

-When translate is 'errdiff' and errdiff is 'custom' these define a
-custom error diffusion map.  errdiff_width and errdiff_height define
-the size of the map in the arrayref in errdiff_map.  errdiff_orig is
-an integer which indicates the current pixel position in the top row
-of the map.
+This is a function, not a method.

 
 

-=item perturb
+This function is a mess, it can take the following named parameters:
+
+=over

 
 

-When translate is 'perturb' this is the magnitude of the random bias
-applied to each channel of the pixel before it is looked up in the
-color table.
+=item *
+
+C<log> - name of a log file to log Imager's actions to.  Not all
+actions are logged, but the debugging memory allocator does log
+allocations here.  Ignored if Imager has been built without logging
+support.  Preferably use the open_log() method instead.
+
+=item *
+
+C<loglevel> - the maximum level of message to log.  Default: 1.
+
+=item *
+
+C<warn_obsolete> - if this is non-zero then Imager will warn when you
+attempt to use obsoleted parameters or functionality.  This currently
+only includes the old GIF output options instead of tags.
+
+=item *
+
+C<t1log> - if non-zero then T1lib will be configured to produce a log
+file.  This will fail if there are any existing T1lib font objects.

 
 =back
 
 
 =back
 

-=cut
\ No newline at end of file
+Example:
+
+  Imager::init(log => 'trace.log', loglevel => 9);
+
+=back
+
+=head1 LOGGING METHODS
+
+Imager can open an internal log to send debugging information to.
+This log is extensively used in Imager's tests, but you're unlikely to
+use it otherwise.
+
+If Imager has been built with logging disabled, the methods fail
+quietly.
+
+=over
+
+=item open_log()
+
+Open the Imager debugging log file.
+
+=over
+
+=item *
+
+C<log> - the file name to log to.  If this is undef logging information
+is sent to the standard error stream.
+
+=item *
+
+C<loglevel> the level of logging to produce.  Default: 1.
+
+=back
+
+Returns a true value if the log file was opened successfully.
+
+  # send debug output to test.log
+  Imager->open_log(log => "test.log");
+
+  # send debug output to stderr
+  Imager->open_log();
+
+=item close_log()
+
+Close the Imager debugging log file and disable debug logging.
+
+No parameters.
+
+  Imager->close_log();
+
+=item log()
+
+ Imager->log($message)
+ Imager->log($message, $level)
+
+This method does not use named parameters.
+
+The default for C<$level> is 1.
+
+Send a message to the debug log.
+
+  Imager->log("My code got here!");
+
+=item is_logging()
+
+Returns a true value if logging is enabled.
+
+=back
+
+=head1 AUTHOR
+
+Tony Cook <tonyc@cpan.org>, Arnar M. Hrafnkelsson
+
+=head1 SEE ALSO
+
+Imager(3), Imager::Files(3), Imager::Draw(3),
+Imager::Color(3), Imager::Fill(3), Imager::Font(3),
+Imager::Transformations(3), Imager::Engines(3), Imager::Filters(3),
+Imager::Expr(3), Imager::Matrix2d(3), Imager::Fountain(3)
+
+=cut