PDFreactor accepts a Java XML InputSource as mode of input. The InputSource allows you to specify the input using an input stream, a reader, or a systemId. This enables you to load input directly as string, using a file path or from an URL.