<p:declare-step xmlns:p="http://www.w3.org/ns/xproc" xmlns:px="http://www.daisy.org/ns/pipeline/xproc" xmlns:pxi="http://www.daisy.org/ns/pipeline/xproc/internal" version="1.0" type="px:file-peek" exclude-inline-prefixes="#all"> <p:option name="href" required="true"> <p:documentation xmlns="http://www.w3.org/1999/xhtml"> URI to the file you want to read bytes from. May also be an entry inside a ZIP file. </p:documentation> </p:option> <p:option name="offset" required="true"> <p:documentation xmlns="http://www.w3.org/1999/xhtml"> Number of bytes to skip into the file. </p:documentation> </p:option> <p:option name="length" required="true"> <p:documentation xmlns="http://www.w3.org/1999/xhtml"> Number of bytes to include in the result. </p:documentation> </p:option> <p:option name="use-base64" select="'false'"> <p:documentation xmlns="http://www.w3.org/1999/xhtml"> By default, the output will be hex-ecoded, which is normally easier to manipulate. If you want base64-encoded output then you can set this to true. </p:documentation> </p:option> <p:output port="result"/> <p:declare-step type="pxi:file-peek"> <p:option name="href" required="true"/> <p:option name="offset" required="true"/> <p:option name="length" required="true"/> <p:output port="result"/> </p:declare-step> <pxi:file-peek> <p:with-option name="href" select="$href"/> <p:with-option name="offset" select="$offset"/> <p:with-option name="length" select="$length"/> </pxi:file-peek> <p:add-attribute match="/*" attribute-name="offset"> <p:with-option name="attribute-value" select="$offset"/> </p:add-attribute> <p:add-attribute match="/*" attribute-name="length"> <p:with-option name="attribute-value" select="$length"/> </p:add-attribute> <p:choose> <p:when test="$use-base64 = 'true'"> <p:identity/> </p:when> <p:otherwise> <p:xslt> <p:with-param name="offset" select="$offset"/> <p:with-param name="length" select="$length"/> <p:input port="stylesheet"> <p:document href="../xslt/base64-to-hex.xsl"/> </p:input> </p:xslt> </p:otherwise> </p:choose> </p:declare-step>