There are basically two options: either add image dimensions to the DITA source files, or adjust the resolution of the images themselves. The latter can be done by hand, via batch processing, or with a system-wide service.

Specifying Dimension Attributes in DITA Markup

Recent toolkit versions (as of 1.4.3) honor the dimension-related attributes in the <image> element, so specifying the height or width of the image is often enough: <image href="../path/to/file.png" placement="break" width="15cm"/>.

This first method affects both HTML and PDF output. However, if you’re using an older toolkit version, if modifying your DITA source files isn’t an option, or if you only want to change the scale in PDFs (but not in HTML), you may be better off changing the metadata of the image files instead.

Adjusting Image Resolution

If you’re using RenderX XEP to produce PDFs, setting the image resolution to 120 dpi will often do the trick. The XEP User Guide includes this useful bit of wisdom:

Bitmap graphic[s] that have no built-in resolution or dimension data, default to a resolution of 120 dpi (5 dots of a 600-dpi printer) as prescribed by the CSS2 Spec.

So that’s what we need to know: XEP assumes a resolution of 120 dpi.

As the User Guide explains:

This is always the case for GIF images, but may also occur with other image types. The XSL Recommendation suggests using 0.28 mm as a pixel size in such cases, which corresponds to 90 dpi resolution. A smaller pixel size gives better print results because the proportion between pixel size and page width is similar to that of a computer screen. With lower resolutions, often the large GIF/JPEG images fit onto a screen but not into the printable area on the page.

You can use your favorite pixel graphics program to change the image resolution to 120 dpi. You don’t want to resample the image, only change the dpi setting (or “pixel density”) in the image metadata.

This is easy to do by hand for a single image, but it quickly gets tedious for large numbers of files.

Processing with ImageMagick

As the name suggests, the ImageMagick software suite provides a powerful arsenal of tools and wizardry that can be used to perform any number of tricks on individual images, or entire directories.

If you’re running Mac OS X, you can use MacPorts to install ImageMagick:

sudo port install ImageMagick  

The versatile mogrify command included with ImageMagick can be used to edit images in place, overwriting the original files. The units and density options will do just what we need:

mogrify -units PixelsPerInch -density 120 test.png  

This sets the pixel density of test.png to 120 dpi. To process all PNG files in a folder, use:

mogrify -units PixelsPerInch -density 120 *.png  

These commands can be adjusted to process all image types in a folder at once, or chain other effects (such as drop shadows) into a single command.

Creating a Mac OS X Service with Automator

If you’re a little leery of the command line and prefer something a bit more user-friendly, you can use Automator to create a system-wide service. Then changing the resolution for a selection of files is as easy as selecting a command from the Services menu.

To create the service with Mac OS 10.6 Snow Leopard:

1. Open /Applications/Automator.app, select the Service template and click Choose.
2. At the top of the workflow area on the right side of the window, change the first list option to image files.
3. From the Automator library, add the Run Shell Script action. Leave the Shell set to /bin/bash, set Pass input to as arguments, and replace echo in the default script with /opt/local/bin/mogrify -units PixelsPerInch -density 120.

When you’re done, the Automator dialog should look like this:

Save the workflow as a service, and you can run it on files & folders in the Finder by Control-clicking and choosing your new command from the Services submenu in the shortcut menu.

If you’d rather just download a copy of my service, you’ll find it here. Unpack the archive and copy the contents to ~/Library/Services.

• one covers removing page numbers from cross references, and
• another explains how to modify PDF2 cover pages to include a company logo.

Nice job Dan!

This test build provides insight into development progress toward the next toolkit release, which is scheduled to support the upcoming DITA 1.2 specification. For details, see the wiki page titled “Plans for DITA Open Toolkit version 1.5.”

Among the changes in Milestone 10 are a number of updates to the PDF2 plugin. The print=no attribute setting is now honored, and a new basic settings file has been added to the plugin to make customization easier. For the full list of changes, see “Plans for the DITA Open Toolkit PDF2 Plug-in.”