NAME App::pandoc::preprocess - Preprocess Pandoc before Processing Pandoc VERSION version 0.5.0 ppp - pandoc pre-process USAGE cat chapters/input-*.pandoc | ppp | pandoc -o output.pdf --smart [more pandoc options...] Additionally see `etc/input.txt` for concrete examples. PREREQUISITES * dot/neato (neato is new!) * rdfdot * ditaa * Image::Magick (for downscaling of large images) BACKGROUND * much simpler design than version 1: pipeable & chainable, reading line-by-line * parallelized work on image file creation How it works 1. while-loop will iterate line by line and is using the flip-flop-operator: * as soon, as a ditaa/rdfdot/dot-block starts, globals ($fileno, $outfile, etc) are set, so all other routines can see them * when actually *inside* the block, the block's contents are printed to the newly generated file (image-X.(ditaa/rdfdot/dot)) 2. once the flip-flop-operator hits the end of the ditaa/rdfdot/dot-block, a child will be spawned to take over the actual ditaa/rdfdot/dot-process to create the png-file and the globals are reset 3. all other lines which are not part of a ditaa/rdfdot/dot-block will simply be piped through to stdout 4. at the end of the program, all children are waited for 5. in the meantime, the new pandoc contents are printed to stdout 6. all child-processes will remain quiert as far as stdout is concerned and write to their individual log-files Todo * Captions * Checks whether ditaa... are available * check whether ditaa has file.encoding set * bundle ditaa with this 'make CPAN happy -- we only have a main in bin/ppp' AUTHOR DBR COPYRIGHT AND LICENSE This software is Copyright (c) 2014 by DBR. This is free software, licensed under: DO WHAT THE FUCK YOU WANT TO PUBLIC LICENSE, Version 2, December 2004