(v13) Input file seekability and direct access

This page applies to Harlequin v13.1r0 and later; and to Harlequin Core but not Harlequin MultiRIP

The method that Harlequin Core uses to supply data to the RIP for PostScript files changed for Harlequin 12. Older versions of Harlequin Core used to read PostScript files in the skin, and supply a buffer of data at a time through the %console% device using the SwLePs() call. This method of supplying data has the disadvantage that the input stream is not seekable, so some operations (for example, supporting DCS comments) do not work. Files supplied on the command line, or via hotfolder, are now set up as the input stream directly by the %config% device.

PDF files previously used a small stub job that opened the file and accessed it indirectly. With this change and the switch to using HqnInputTypes , all file formats are now treated consistently.

This change improves consistency with Harlequin MultiRIP, it makes DCS and other operations requiring seekable input work. It will also change the apparent input filename (accessible through various statusdict parameters and operators) to be the real filename, rather than %console%. Customers may find differences in Genoa CET and other system test files because of this. Files supplied on the command line or via hotfolder have their filesystem devices automatically mounted, including UNC paths on Windows.

The previous method of using SwLePs( ) to supply data buffers is still present and supported in the SDK for those who have already integrated its use.