++ed by:

11 PAUSE users
1 non-PAUSE user.

Randy J Ray


Image::Size - read the dimensions of an image in several popular formats


    use Image::Size;
    # Get the size of globe.gif
    ($globe_x, $globe_y) = imgsize("globe.gif");
    # Assume X=60 and Y=40 for remaining examples

    use Image::Size 'html_imgsize';
    # Get the size as "HEIGHT=X WIDTH=Y" for HTML generation
    $size = html_imgsize("globe.gif");
    # $size == "HEIGHT=60 WIDTH=40"

    use Image::Size 'attr_imgsize';
    # Get the size as a list passable to routines in CGI.pm
    @attrs = attr_imgsize("globe.gif");
    # @attrs == ('-HEIGHT', 60, '-WIDTH', 40)

    use Image::Size;
    # Get the size of an in-memory buffer
    ($buf_x, $buf_y) = imgsize($buf);


The Image::Size library is based upon the wwwis script written by Alex Knowles (alex@ed.ac.uk), a tool to examine HTML and add HEIGHT and WIDTH parameters to image tags. The sizes are cached internally based on file name, so multiple calls on the same file name (such as images used in bulleted lists, for example) do not result in repeated computations.

Image::Size provides three interfaces for possible import:


Returns a three-item list of the X and Y dimensions (height and width, in that order) and image type of stream. Errors are noted by a -1 value for the first element, and an error string for the second. The third element can be (and usually is) ignored, but is useful when sizing data whose type is unknown.


Returns the height and width (X and Y) of stream pre-formatted as a single string "HEIGHT=X WIDTH=Y" suitable for addition into generated HTML IMG tags.


Returns the height and width of stream as part of a 4-element list useful for routines that use hash tables for the manipulation of named parameters, such as the Tk or CGI libraries. A typical return value looks like ("-HEIGHT", X, "-WIDTH", Y).

By default, only imgsize() is imported. Any one or combination of the three may be imported, or all three may be with the tag :all.

Input Types

The sort of data passed as stream can be one of three forms:


If an ordinary scalar (string) is passed, it is assumed to be a file name (either absolute or relative to the current working directory of the process) and is searched for and opened (if found) as the source of data. Possible error messages (see DIAGNOSTICS below) may include file-access problems.

scalar reference

If the passed-in stream is a scalar reference, it is interpreted as pointing to an in-memory buffer containing the image data.

        # Assume that &read_data gets data somewhere (WWW, etc.)
        $img = &read_data;
        ($x, $y, $id) = imgsize(\$img);
        # $x and $y are dimensions, $id is the type of the image
IO::File object reference

The third option is to pass in an object of the IO::File class that has already been instantiated on the target image file. The file pointer will necessarily move, but will be restored to its original position before subroutine end.

        # $fh was passed in, is IO::File reference:
        ($x, $y, $id) = imgsize($fh);
        # Same as calling with filename, but more abstract.

Recognizd Formats

Image::Size understands and sizes data in the following formats:

  • GIF

  • JPG

  • XBM

  • XPM

  • PPM family (PPM/PGM/PBM)

  • PNG

When using the imgsize interface, there is a third, unused value returned if the programmer wishes to save and examine it. This value is the three- letter identity of the data type. This is useful when operating on open file handles or in-memory data, where the type is as unknown as the size. The two support routines ignore this third return value, so those wishing to use it must use the base imgsize routine.


The base routine, imgsize, returns a -1 as the first value in its list when an error has occured. The second return element contains a descriptive error message.

The second and third forms blindly format the returned data of imgsize, and as such may return corrupted data in the event of an error.


This will reliably work on perl 5.002 or newer. Perl versions prior to 5.003 do not have the IO::File module by default, which this module requires. You will have to retrieve and install it, or upgrade to 5.003, in which it is included as part of the core.

Caching of size data can only be done on inputs that are file names. Open file handles and scalar references cannot be reliably transformed into a unique key for the table of cache data. Buffers could be cached using the MD5 module, and perhaps in the future I will make that an option. I do not, however, wish to lengthen the dependancy list by another item at this time.


http://www.tardis.ed.ac.uk/~ark/wwwis/ for a description of wwwis and how to obtain it.


Perl module interface by Randy J. Ray (rjray@uswest.com), original image-sizing code by Alex Knowles (alex@ed.ac.uk) and Andrew Tong (werdna@ugcs.caltech.edu), used with their joint permission.

Some bug fixes submitted by Bernd Leibing (bernd.leibing@rz.uni-ulm.de). PPM/PGM/PBM sizing code contributed by Carsten Dominik (dominik@strw.LeidenUniv.nl). Tom Metro (tmetro@vl.com) re-wrote the JPG and PNG code, and also provided a PNG image for the test suite. Dan Klein (dvk@lonewolf.com) contributed a re-write of the GIF code.