[sewtha,#5][xl]: markdown version of david mackay's sustainable energ…

…y without the hot air.

In /sewtha/ directory. Working with the website.

* epub source
* extract.sh that does all the extraction and cleaning (quite a bit of debugging to make sure html is well-behaved for next/mdx)
* images for the text in site/public/img/sewtha (could be nicer to keep these with the text and copy over in build ...)

sewtha/src/extract.sh

@@ -0,0 +1,64 @@
+#!/bin/zsh
+
+# Main epub version linked on https://www.withouthotair.com/epubVersions.html
+# curl http://www.inference.eng.cam.ac.uk/sustainable/book/translate/SustainableEnergy-withoutthehotair-DavidJCMacKay.epub > sewtha.epub
+
+# let's extract it with the images
+mkdir -p img
+pandoc without-the-hot-air.epub -t gfm -o main.md --wrap=none --extract-media=img
+
+cp main.md tidy.md
+
+## let's try and tidy it up ...
+# remove stuff like <span id="titlepage.xhtml"></span>
+sed -i '' 's/^<span.*><\/span>$//g' tidy.md
+# <span class="figurenumber">Figure 1.2.</span> Are "our" fossil fuels running out? Total crude oil production from the North Sea, and oil price in 2006 dollars per barrel. [<span class="darkred">\[10\]</span>](#chap01.xhtml#ch01n10)
+
+# remove stuff like <div class="imgcap">
+# ~~in fact, let's leave for now~~
+sed -i '' 's/^<div.*//g' tidy.md
+sed -i '' 's/^<\/div>//g' tidy.md
+
+# strip trailng white space
+sed -i '' 's/ *$//g' tidy.md
+
+# let's delete all 2+ blank lines
+# see https://stackoverflow.com/questions/12598916/how-to-use-sed-to-remove-only-double-empty-lines
+sed -i '' 'N;/^\n$/d;P;D' tidy.md
+
+# correct quotes to normal quotes
+sed -i '' 's/”/"/g' tidy.md
+sed -i '' 's/“/"/g' tidy.md
+
+# fix headings e.g. ...
+# # <span class="smallfont">*Part I*</span>
+sed -i '' 's/^# <span class="smallfont">\*\(.*\)\*<\/span>/# \1/g' tidy.md
+
+# delete the nav
+# sed '/PATTERN-1/,/PATTERN-2/{//!d}'
+
+## Footnotes
+# [<span class="darkred">\[13\]</span>](#chap01.xhtml#ch01n13)
+sed -i '' 's/ \[<span class="darkred">\\\[[0-9]*[^#]*(#chap[0-9]*.xhtml#ch\([0-9]*\)n\([0-9]*\))/[^\1-\2]/g' tidy.md
+# And fn is ... [<span class="mark">\[1\]</span>](#chap01.xhtml#ret01)*
+sed -i '' 's/\[<span class="mark">\\\[[0-9*\\\][^)]*#chap\([0-9]*\).xhtml#ret\([0-9]*\))/[^\1-\2]: /g' tidy.md
+# TODO: remove special span class red case ...
+
+## fix up references section e.g.
+# -  <span class="smallcaps"> {OECD</span> Nuclear Energy Agency}. (2006).
+sed -i '' -e 's/<span class="smallcaps">\([^<]*\)<\/span>/\1/g' tidy.md
+
+## final hacks ...
+# replace unmatched closing </div> usually looking like
+# </div>\n\n</th>
+# TODo: I cannot find an easy way to do multiline replace so giving up for now!
+# sed -i '' -e 's/<\/div>\n\n<\/th>/<\/th>/' tidy.md
+
+# let's fix Images which are in img/Images
+mv img/Images/* img/
+rm -Rf img/Images
+sed -i '' 's/img\/Images/\/img\/sewtha/g' tidy.md
+
+cp tidy.md ../README.md
+mkdir -p ../../site/public/img/sewtha
+cp -a img/* ../../site/public/img/sewtha/