# encoding: utf-8 module Slideshow ## fix:/todo: move generation code out of command into its own class ## not residing/depending on cli class Gen ## todo: rename command to build include LogUtils::Logging include ManifestHelper ### fix: remove opts, use config (wrapped!!) def initialize( opts, config, headers ) @opts = opts @config = config @headers = headers ## todo: check if we need to use expand_path - Dir.pwd always absolute (check ~/user etc.) @usrdir = File.expand_path( Dir.pwd ) # save original (current) working directory end attr_reader :usrdir # original working dir (user called slideshow from) attr_reader :srcdir, :outdir, :pakdir # NB: "initalized" in create_slideshow attr_reader :opts, :config, :headers attr_reader :session # give helpers/plugins a session-like hash attr_reader :markup_type # :textile, :markdown, :mediawiki, :rest # uses configured markup processor (textile,markdown,rest,mediawiki) to generate html def text_to_html( content ) content = case @markup_type when :markdown markdown_to_html( content ) when :textile textile_to_html( content ) when :mediawiki mediawiki_to_html( content ) when :rest rest_to_html( content ) end content end def guard_text( text ) # todo/fix 2: note for Textile we need to differentiate between blocks and inline # thus, to avoid runs - use guard_block (add a leading newline to avoid getting include in block that goes before) # todo/fix: remove wrap_markup; replace w/ guard_text # why: text might be css, js, not just html ## todo: add print depreciation warning wrap_markup( text ) end def guard_block( text ) if markup_type == :textile # saveguard with notextile wrapper etc./no further processing needed # note: add leading newlines to avoid block-runons "\n\n\n#{text}\n\n" elsif markup_type == :markdown # wrap in newlines to avoid runons "\n\n#{text}\n\n" elsif markup_type == :mediawiki "\n\n\n#{text}\n\n" else text end end def guard_inline( text ) wrap_markup( text ) end def wrap_markup( text ) if markup_type == :textile # saveguard with notextile wrapper etc./no further processing needed "\n#{text}\n" else text end end # move into a filter?? def post_processing_slides( content ) # 1) add slide breaks if config.slide? # only allow !SLIDE directives fo slide breaks? # do nothing (no extra automagic slide breaks wanted) else if (@markup_type == :markdown && Markdown.lib == 'pandoc-ruby') || @markup_type == :rest content = add_slide_directive_before_div_h1( content ) else if config.header_level == 2 content = add_slide_directive_before_h2( content ) else # assume level 1 content = add_slide_directive_before_h1( content ) end end end dump_content_to_file_debug_html( content ) # 2) use generic slide break processing instruction to # split content into slides slide_counter = 0 slides = [] slide_buf = "" content.each_line do |line| if line.include?( '/m, '').gsub( /[\n\r\t ]/, '').length > 0 logger.debug "add slide with missing leading slide directive >#{slide_buf}< with slide_counter == 0" slides << slide_buf slide_buf = "" else logger.debug "skipping slide_buf >#{slide_buf}< with slide_counter == 0" end end slide_counter += 1 end slide_buf << line end if slide_counter > 0 slides << slide_buf # add slide to slide stack slide_buf = "" # reset slide source buffer end slides2 = [] slides.each do |source| slides2 << Slide.new( source, config ) end puts "#{slides2.size} slides found:" slides2.each_with_index do |slide,i| print " [#{i+1}] " if slide.header.present? print slide.header else # remove html comments print "-- no header -- | #{slide.content.gsub(//m, '').gsub(/\n/,'$')[0..40]}" end puts end # make content2 and slide2 available to erb template # -- todo: cleanup variable names and use attr_readers for content and slide ### fix: use class SlideDeck or Deck?? for slides array? content2 = "" slides2.each do |slide| content2 << slide.to_classic_html end @content = content2 @slides = slides2 # strutured content end def create_slideshow( fn ) manifest_path_or_name = opts.manifest # add .txt file extension if missing (for convenience) if manifest_path_or_name.downcase.ends_with?( '.txt' ) == false manifest_path_or_name << '.txt' end logger.debug "manifest=#{manifest_path_or_name}" # check if file exists (if yes use custom template package!) - allows you to override builtin package with same name if File.exists?( manifest_path_or_name ) manifestsrc = manifest_path_or_name else # check for builtin manifests manifests = installed_template_manifests matches = manifests.select { |m| m[0] == manifest_path_or_name } if matches.empty? puts "*** error: unknown template manifest '#{manifest_path_or_name}'" # todo: list installed manifests exit 2 end manifestsrc = matches[0][1] end ### todo: use File.expand_path( xx, relative_to ) always with second arg ## do NOT default to cwd (because cwd will change!) # Reference src with absolute path, because this can be used with different pwd manifestsrc = File.expand_path( manifestsrc, usrdir ) # expand output path in current dir and make sure output path exists @outdir = File.expand_path( opts.output_path, usrdir ) logger.debug "setting outdir to >#{outdir}<" FileUtils.makedirs( outdir ) unless File.directory? outdir dirname = File.dirname( fn ) basename = File.basename( fn, '.*' ) extname = File.extname( fn ) logger.debug "dirname=#{dirname}, basename=#{basename}, extname=#{extname}" # change working dir to sourcefile dir # todo: add a -c option to commandline? to let you set cwd? @srcdir = File.expand_path( dirname, usrdir ) logger.debug "setting srcdir to >#{srcdir}<" unless usrdir == srcdir logger.debug "changing cwd to src - new >#{srcdir}<, old >#{Dir.pwd}<" Dir.chdir srcdir end puts "Preparing slideshow '#{basename}'..." if config.known_textile_extnames.include?( extname ) @markup_type = :textile elsif config.known_rest_extnames.include?( extname ) @markup_type = :rest elsif config.known_mediawiki_extnames.include?( extname ) @markup_type = :mediawiki else # default/fallback use markdown @markup_type = :markdown end # shared variables for templates (binding) @content_for = {} # reset content_for hash @name = basename @extname = extname @session = {} # reset session hash for plugins/helpers inname = "#{basename}#{extname}" logger.debug "inname=#{inname}" content = File.read( inname ) # run text filters config.text_filters.each do |filter| mn = filter.tr( '-', '_' ).to_sym # construct method name (mn) content = send( mn, content ) # call filter e.g. include_helper_hack( content ) end if config.takahashi? content = takahashi_slide_breaks( content ) end # convert light-weight markup to hypertext content = text_to_html( content ) # post-processing # make content2 and slide2 available to erb template # -- todo: cleanup variable names and use attr_readers for content and slide if @markup_type == :markdown && config.markdown_post_processing?( Markdown.lib ) == false puts " Skipping post-processing (passing content through as is)..." @content = content # content all-in-one - make it available in erb templates else # sets @content (all-in-one string) and @slides (structured content; split into slides) post_processing_slides( content ) end #### pak merge # nb: change cwd to template pak root @pakdir = File.dirname( manifestsrc ) # template pak root - make availabe too in erb via binding logger.debug " setting pakdir to >#{pakdir}<" # todo/fix: change current work dir (cwd) in pakman gem itself # for now lets do it here logger.debug "changing cwd to pak - new >#{pakdir}<, old >#{Dir.pwd}<" Dir.chdir( pakdir ) pakpath = outdir logger.debug( "manifestsrc >#{manifestsrc}<, pakpath >#{pakpath}<" ) Pakman::Templater.new.merge_pak( manifestsrc, pakpath, binding, basename ) logger.debug "restoring cwd to src - new >#{srcdir}<, old >#{Dir.pwd}<" Dir.chdir( srcdir ) ## pop/restore org (original) working folder/dir unless usrdir == srcdir logger.debug "restoring cwd to usr - new >#{usrdir}<, old >#{Dir.pwd}<" Dir.chdir( usrdir ) end puts "Done." end # method create_slideshow end # class Gen end # class Slideshow