rmaicle · April 24, 2017 13:58
diff --git a/convert_asciidoc_html_to_markdown b/convert_asciidoc_html_to_markdown
 #!/bin/bash
 # Convert asciidoc to markdown
 # Argument(s):
 #   - filename - assumed to have an extension of .adoc
 #
 # TODO: Lists

 filename="$1.md"
 filename="${filename//-/_}"

 asciidoc -b html5 -o tmp.html "$1".txt

 cp -f tmp.html original.html

 # Store copy of HTML with deleted HTML header part for debugging
 sed -i '/<head>/,/<\/head>/d' original.html

 # HTML replacements
 # =================

 # Delete HTML header and closing elements
 sed -i '/DOCTYPE/,/<\/head>/d' tmp.html
 sed -i 's/<body class="article">//g' tmp.html
 sed -i '/<\/body>/,/<\/head>/d' tmp.html

 cp -f tmp.html no_headers.html

 # Definition lists
 sed -i 's/<dt class="hdlist1">/†/' tmp.html
 sed -i 's/<\/dt>/‡/' tmp.html
 sed -ir 's/<dd>/»/' tmp.html
 sed -ir 's/<\/dd>/«/' tmp.html

 # Code blocks
 # ===========

 # Put <pre> on its own line
 sed -i 's/^<pre>\(.*\)/<pre>\n\n~~~\n\1/' tmp.html
 # Put </pre> on its own line
 sed -i '/<\/pre>$/,/^<\/div><\/div>$/ N; {s/<\/pre>/\n~~~\n\n<\/pre>/}' tmp.html

 # Delete HTML block for code blocks
 sed -i '/^<div class="listingblock">/,/^<pre>/d' tmp.html
 sed -i '/^<div class="literalblock">/,/^<pre>/d' tmp.html
 sed -i '/^<\/pre>/,/^<\/div><\/div>/d' tmp.html

 # Pandoc strips code span HTML elements
 # so we pre-process them here, delimited using ͼ and ͽ
 # <span class="monospaced">
 # </span>
 sed -ir '/<span class="monospaced">/,/<\/span>/ s/<span class="monospaced">\(.*\)<\/span>/ͼ\1ͽ/g' tmp.html

 cp -f tmp.html pre_pandoc.html

 pandoc -f html -t markdown_github+blank_before_header+all_symbols_escapable+blank_before_blockquote+definition_lists+fenced_code_blocks+footnotes+pipe_tables+yaml_metadata_block tmp.html -o "$filename"

 cp -f "$filename" no_fixes.md

 # Fixes
 # =====

 # Put texts between † and ‡ on its own line
 #sed -i 's/\(†.*‡\)/\n\1\n/' "$filename"
 sed -i 's/‡[[:space:]]†/‡\n†/' "$filename"
 # Delete extra spaces before and after the delimiters
 #sed -i 's/†[[:space:]]/†/' "$filename"
 #sed -i 's/[[:space:]]‡/‡/' "$filename"
 # Re-combine lines
 sed -ir '/†/{N;N;s/†\n\(.*\)\n‡/†\1‡/}' "$filename"
 sed -ir '/^‡$/d' "$filename"

 cp -f "$filename" delimited.md

 # Delete spaces between » and «.
 # These delimiters must be on its own line.
 sed -i 's/^[[:space:]]»$/»/' "$filename"
 sed -i 's/^[[:space:]]«$/«/' "$filename"
 sed -i 's/^«[[:space:]]/«/' "$filename"
 # Delete emtpy line before « and append an empty line after
 sed -i '/^$/{N;s/^\n«$/«\n/}' "$filename"

 # Compress two empty lines to a single line
 sed -i '/^$/N;/^\n$/D' "$filename"

 # Paragraphs within » and « after the first one must be
 # 'terminated' or formatted as another dd element
 #sed -i '/»/,/«/ N; {s/^\n\(.*\)/\n: \1/ }' "$filename"
 # -- the above delimits too many paragraphs which should not be

 # yaml
 # ====

 sed -i '/``` content/ s//~~~/' "$filename"
 sed -i '/```/ s//~~~/' "$filename"
 sed -i '/~~~~ content/ s//~~~/' "$filename"
 sed -i '/~~~~/ s//~~~/' "$filename"

 # Code spans
 # ==========

 # Wrap code span between temporary delimiter characters
 # Must consider successive code spans.
 #sed -i 's/<span class="monospaced">/†/g' "$filename"
 #sed -i 's/<\/span>/‡/g' "$filename"
 #sed -ir  's/<span class="monospaced">\(.*\)<\/span>/†\1‡/g' "$filename"
 #sed -ir '/<span class="monospaced">/,/<\/span>/ s/<span class="monospaced">\(.*\)<\/span>/†\1‡/g' "$filename"

 # Delete successive delimiters
 sed -i 's/†[[:space:]]†/†/' "$filename"
 sed -i 's/‡[[:space:]]‡/‡/' "$filename"
 # Delete consecutive delimiters
 sed -i 's/††/†/' "$filename"
 sed -i 's/‡‡/‡/' "$filename"

 # Delete spaces
 # =============

 # Delete space before a definition list term
 #sed -i 's/^ -'/-/ "$filename"

 # Delete trailing whitespace at end of each line
 #sed -in '/^---$/,/^---$/ !{ /\s*$/ s///g }' "$filename"

 # Convert < and > characters within code spans
 # ============================================

 # Convert code span between custom delimeters containing &gt; and &lt;
 # to verbatim equivalent
 sed -i '/†/,/‡/ {s/&lt;/\</g}' "$filename"
 sed -i '/†/,/‡/ {s/&gt;/\>/g}' "$filename"
 # Convert text containing &gt; and &lt; to escaped equivalent
 sed -i '/†/,/‡/ !{/&lt;/ s//\\</g}' "$filename"
 sed -i '/†/,/‡/ !{/&gt;/ s//\\>/g}' "$filename"

 # Convert temporary dt delimiters
 # ======================================

 sed -i 's/†/\`/g' "$filename"
 sed -i 's/‡/\`/g' "$filename"

 # Convert temporary code span delimiters
 # ======================================

 sed -ir 's/ͼ/`/g' "$filename"
 sed -ir 's/ͽ/`/g' "$filename"

 # Delete temporary dd delimiters » «
 # ==================================

 sed -i '/^»$/ N; s/^»\n\(.*\)$/: \1/' "$filename"
 sed -i '/^«$/N;/\n/D' "$filename"






 title=`cat $filename | sed q`
 title=${title%%(*}

 #prefix="/mnt/work/projects/_github/rmaicle/rmaicle.github.io.2017/_source/documentations/"
 directory=`pwd`
 directory=${directory#*documentations/}
 directory="/doc/"$directory"/"$filename
 directory=${directory%%.md}

 echo "$directory"

 > x.md
 echo "---" >> x.md
 echo "title: $title" >> x.md
 echo "layout: documentation" >> x.md
 echo "categories: [documentation]" >> x.md
 echo "tags: [git]" >> x.md
 echo "draft: true" >> x.md
 echo "published: true" >> x.md
 echo "permalink: $directory" >> x.md
 echo "group: git" >> x.md
 echo "---" >> x.md
 echo "" >> x.md

 cat x.md "$filename" > tmp.md
 mv tmp.md "$filename"

 # Cleanup
 #rm -f ./tmp.html
 rm -f x.md
 rm -f *.mdn
 rm -f *.mdr
	#!/bin/bash
	# Convert asciidoc to markdown
	# Argument(s):
	# - filename - assumed to have an extension of .adoc
	#
	# TODO: Lists

	filename="$1.md"
	filename="${filename//-/_}"

	asciidoc -b html5 -o tmp.html "$1".txt

	cp -f tmp.html original.html

	# Store copy of HTML with deleted HTML header part for debugging
	sed -i '/<head>/,/<\/head>/d' original.html

	# HTML replacements
	# =================

	# Delete HTML header and closing elements
	sed -i '/DOCTYPE/,/<\/head>/d' tmp.html
	sed -i 's/<body class="article">//g' tmp.html
	sed -i '/<\/body>/,/<\/head>/d' tmp.html

	cp -f tmp.html no_headers.html

	# Definition lists
	sed -i 's/<dt class="hdlist1">/†/' tmp.html
	sed -i 's/<\/dt>/‡/' tmp.html
	sed -ir 's/<dd>/»/' tmp.html
	sed -ir 's/<\/dd>/«/' tmp.html

	# Code blocks
	# ===========

	# Put <pre> on its own line
	sed -i 's/^<pre>\(.*\)/<pre>\n\n~~~\n\1/' tmp.html
	# Put </pre> on its own line
	sed -i '/<\/pre>$/,/^<\/div><\/div>$/ N; {s/<\/pre>/\n~~~\n\n<\/pre>/}' tmp.html

	# Delete HTML block for code blocks
	sed -i '/^<div class="listingblock">/,/^<pre>/d' tmp.html
	sed -i '/^<div class="literalblock">/,/^<pre>/d' tmp.html
	sed -i '/^<\/pre>/,/^<\/div><\/div>/d' tmp.html

	# Pandoc strips code span HTML elements
	# so we pre-process them here, delimited using ͼ and ͽ
	# <span class="monospaced">
	# </span>
	sed -ir '/<span class="monospaced">/,/<\/span>/ s/<span class="monospaced">\(.*\)<\/span>/ͼ\1ͽ/g' tmp.html

	cp -f tmp.html pre_pandoc.html

	pandoc -f html -t markdown_github+blank_before_header+all_symbols_escapable+blank_before_blockquote+definition_lists+fenced_code_blocks+footnotes+pipe_tables+yaml_metadata_block tmp.html -o "$filename"

	cp -f "$filename" no_fixes.md

	# Fixes
	# =====

	# Put texts between † and ‡ on its own line
	#sed -i 's/\(†.*‡\)/\n\1\n/' "$filename"
	sed -i 's/‡[[:space:]]†/‡\n†/' "$filename"
	# Delete extra spaces before and after the delimiters
	#sed -i 's/†[[:space:]]/†/' "$filename"
	#sed -i 's/[[:space:]]‡/‡/' "$filename"
	# Re-combine lines
	sed -ir '/†/{N;N;s/†\n\(.*\)\n‡/†\1‡/}' "$filename"
	sed -ir '/^‡$/d' "$filename"

	cp -f "$filename" delimited.md

	# Delete spaces between » and «.
	# These delimiters must be on its own line.
	sed -i 's/^[[:space:]]»$/»/' "$filename"
	sed -i 's/^[[:space:]]«$/«/' "$filename"
	sed -i 's/^«[[:space:]]/«/' "$filename"
	# Delete emtpy line before « and append an empty line after
	sed -i '/^$/{N;s/^\n«$/«\n/}' "$filename"

	# Compress two empty lines to a single line
	sed -i '/^$/N;/^\n$/D' "$filename"

	# Paragraphs within » and « after the first one must be
	# 'terminated' or formatted as another dd element
	#sed -i '/»/,/«/ N; {s/^\n\(.*\)/\n: \1/ }' "$filename"
	# -- the above delimits too many paragraphs which should not be

	# yaml
	# ====

	sed -i '/``` content/ s//~~~/' "$filename"
	sed -i '/```/ s//~~~/' "$filename"
	sed -i '/~~~~ content/ s//~~~/' "$filename"
	sed -i '/~~~~/ s//~~~/' "$filename"

	# Code spans
	# ==========

	# Wrap code span between temporary delimiter characters
	# Must consider successive code spans.
	#sed -i 's/<span class="monospaced">/†/g' "$filename"
	#sed -i 's/<\/span>/‡/g' "$filename"
	#sed -ir 's/<span class="monospaced">\(.*\)<\/span>/†\1‡/g' "$filename"
	#sed -ir '/<span class="monospaced">/,/<\/span>/ s/<span class="monospaced">\(.*\)<\/span>/†\1‡/g' "$filename"

	# Delete successive delimiters
	sed -i 's/†[[:space:]]†/†/' "$filename"
	sed -i 's/‡[[:space:]]‡/‡/' "$filename"
	# Delete consecutive delimiters
	sed -i 's/††/†/' "$filename"
	sed -i 's/‡‡/‡/' "$filename"

	# Delete spaces
	# =============

	# Delete space before a definition list term
	#sed -i 's/^ -'/-/ "$filename"

	# Delete trailing whitespace at end of each line
	#sed -in '/^---$/,/^---$/ !{ /\s*$/ s///g }' "$filename"

	# Convert < and > characters within code spans
	# ============================================

	# Convert code span between custom delimeters containing > and <
	# to verbatim equivalent
	sed -i '/†/,/‡/ {s/</\</g}' "$filename"
	sed -i '/†/,/‡/ {s/>/\>/g}' "$filename"
	# Convert text containing > and < to escaped equivalent
	sed -i '/†/,/‡/ !{/</ s//\\</g}' "$filename"
	sed -i '/†/,/‡/ !{/>/ s//\\>/g}' "$filename"

	# Convert temporary dt delimiters
	# ======================================

	sed -i 's/†/\`/g' "$filename"
	sed -i 's/‡/\`/g' "$filename"

	# Convert temporary code span delimiters
	# ======================================

	sed -ir 's/ͼ/`/g' "$filename"
	sed -ir 's/ͽ/`/g' "$filename"

	# Delete temporary dd delimiters » «
	# ==================================

	sed -i '/^»$/ N; s/^»\n\(.*\)$/: \1/' "$filename"
	sed -i '/^«$/N;/\n/D' "$filename"






	title=`cat $filename \| sed q`
	title=${title%%(*}

	#prefix="/mnt/work/projects/_github/rmaicle/rmaicle.github.io.2017/_source/documentations/"
	directory=`pwd`
	directory=${directory#*documentations/}
	directory="/doc/"$directory"/"$filename
	directory=${directory%%.md}

	echo "$directory"

	> x.md
	echo "---" >> x.md
	echo "title: $title" >> x.md
	echo "layout: documentation" >> x.md
	echo "categories: [documentation]" >> x.md
	echo "tags: [git]" >> x.md
	echo "draft: true" >> x.md
	echo "published: true" >> x.md
	echo "permalink: $directory" >> x.md
	echo "group: git" >> x.md
	echo "---" >> x.md
	echo "" >> x.md

	cat x.md "$filename" > tmp.md
	mv tmp.md "$filename"

	# Cleanup
	#rm -f ./tmp.html
	rm -f x.md
	rm -f *.mdn
	rm -f *.mdr