#!/bin/bash
set -euo pipefail
IFS=$'\n\t'
OUTPUT_DIR=out
mkdir -p html
mkdir -p "$OUTPUT_DIR"
MAIN_STORY_OUTPUT_FILE="$OUTPUT_DIR/read-the-story.html"
HTML_FILE=ickabog.html
LC=${LC:-""}
if [[ "$LC" != "" ]]; then
LC="/$LC"
fi
if [[ "$LC" == "" || "$LC" == "en-US" ]]; then
MAIN_STORY_URL="https://web.archive.org/web/20200713135650/https://www.theickabog.com/$LC/read-the-story/"
else
# Non-english editions are still accessible on the original website.
MAIN_STORY_URL="https://www.theickabog.com$LC/read-the-story/"
fi
echo "[+] Fetching $MAIN_STORY_URL"
wget --quiet --timeout=10 "$MAIN_STORY_URL" --output-document "$MAIN_STORY_OUTPUT_FILE"
LANG=$(cat "$MAIN_STORY_OUTPUT_FILE"| pup 'html attr{lang}')
echo "[+] Language set to $LANG"
MAIN_TITLE=$(cat "$MAIN_STORY_OUTPUT_FILE" | pup 'ul.chapters__list a json{}' | jq -r '[.[] | {url: .href, chapter: .children[0].children[0].children[0].children[0].text, title: .children[0].children[0].children[0].children[1].text}] | sort_by(.chapter) | .[]|[.chapter, .title, .url] | @tsv' | grep $' 2\t' | while IFS=$'\t' read -r chapter title url; do echo "$title"; done)
echo "[+] Title set to $MAIN_TITLE"
echo "$MAIN_TITLE" > "$HTML_FILE"
# args = "$url" "$chapter" "$title"
function download_chapter() {
[[ $2 =~ 1$ ]] && MAIN_TITLE=$3
# We have a direct link, so lets use that
if [[ $1 =~ ^http ]]; then
URL="$1"
# For english, we need to make sure we are using the archive link
elif [[ "$LANG" == "en-US" || "$LANG" == "en-GB" ]]; then
URL="https://web.archive.org$1"
# For others cases, let us make it a absolute URL
else
URL="https://www.theickabog.com$1"
fi
echo $URL
[ -s "html/$2.html" ] || wget --timeout=10 --quiet "$URL" -O "html/$2.html"
echo "$3
" >> "$HTML_FILE"
cat "html/$2.html" | pup -p --charset UTF-8 'article div.row:nth-child(2) div.entry-content' >> "$HTML_FILE"
}
cat "$MAIN_STORY_OUTPUT_FILE" |
pup -p --charset UTF-8 'ul.chapters__list a json{}' |
jq -r '[.[] | {url: .href, chapter: .children[0].children[0].children[0].children[0].text, title: .children[0].children[0].children[0].children[1].text}] | sort_by(.chapter | match("[0-9]+$")) | .[]|[.chapter, .title, .url] | @tsv' |
while IFS=$'\t' read -r chapter title url; do download_chapter "$url" "$chapter" "$title"; done
echo "" >> "$HTML_FILE"
cat <<__METADATA__ > metadata.xml
J.K Rowling
__METADATA__
pandoc --from=html \
--output="$OUTPUT_DIR/ickabog.epub" \
--epub-metadata=metadata.xml \
--epub-cover-image=cover.jpg \
--epub-chapter-level=1 \
"$HTML_FILE"
echo "[+] Generated $OUTPUT_DIR/ickabog.epub"
if command -v kindlegen > /dev/null; then
kindlegen "$OUTPUT_DIR/ickabog.epub" > /dev/null 2>&1
echo "[+] Generated MOBI using kindlegen: $OUTPUT_DIR/ickabog.mobi"
elif command -v ebook-convert > /dev/null; then
ebook-convert "$OUTPUT_DIR/ickabog.epub" \
"$OUTPUT_DIR/ickabog.mobi" \
--metadata title="$MAIN_TITLE" \
> /dev/null 2>&1
echo "[+] Generated MOBI using ebook-convert: $OUTPUT_DIR/ickabog.mobi"
else
echo "[-] Could not generate MOBI, install kindlegen or calibre"
fi
if command -v xelatex >/dev/null; then
pandoc --from=html \
--pdf-engine=xelatex \
--metadata title="$MAIN_TITLE" \
--metadata author="J.K Rowling" \
--toc \
--output="$OUTPUT_DIR/ickabog-no-cover.pdf" \
-V lang="$LANG" \
-V geometry=margin=1.5cm \
"$HTML_FILE"
else
echo "[-] Missing xelatex, please install xelatex before trying again"
exit
fi
if command -v qpdf > /dev/null; then
qpdf --empty --pages cover.pdf "$OUTPUT_DIR/ickabog-no-cover.pdf" -- "$OUTPUT_DIR/ickabog.pdf"
else
echo "[-] qpdf not installed, PDF won't have cover"
mv "$OUTPUT_DIR/ickabog-no-cover.pdf" "$OUTPUT_DIR/ickabog.pdf"
fi
echo "[+] Generated PDF using xelatex: $OUTPUT_DIR/ickabog.pdf"
# Run only if context is available
if command -v context>/dev/null; then
pandoc --from=html --to=pdf \
-V fontsize=18pt \
--output="$OUTPUT_DIR/ickabog-large-no-cover.pdf" \
--metadata title="$MAIN_TITLE" \
--metadata author="J.K Rowling" \
--pdf-engine=context \
-V margin-left=0cm \
-V margin-right=0cm \
-V margin-top=0cm \
-V margin-bottom=0cm \
-V geometry=margin=0cm \
-V lang="$LANG" \
"$HTML_FILE"
if command -v qpdf > /dev/null; then
qpdf --empty --pages cover.pdf "$OUTPUT_DIR/ickabog-large-no-cover.pdf" -- "$OUTPUT_DIR/ickabog-large.pdf"
else
mv "$OUTPUT_DIR/ickabog-no-cover.pdf" "$OUTPUT_DIR/ickabog-large.pdf"
fi
echo "[+] Generated PDF using context: $OUTPUT_DIR/ickabog-large.pdf"
else
echo "[-] Can't find ConTeXt, not generating large-text PDF"
fi