Multiple scripts that are useful but don't deserve their own repository.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
multiple_scripts/books_search.sh

206 lines
7.6 KiB

#!/usr/bin/env bash
##############################################################################
#
# books_search.sh
# By Steven Saus
# (c) 2020; licensed under the MIT license
#
# Find and open an ebook from your Calibre library without, ahem, opening your
# Calibre library. Feeds in format, author, and title to fzf / rofi for you
# to choose, then uses epy/epr/conversion for TUI/CLI and xdg-open for GUI.
#
# Originally inspired by a script by [Miroslav Vidovic](https://github.com/miroslavvidovic/rofi-scripts)
# though rewritten, expanded, and changed quite a freaking lot.
#
# Requires Calibre books be stored in the default structure of
# /AUTHOR/TITLE (NUBMER)/BOOKFILE
#
# For example, an example is
# /home/steven/documents/Calibre Library/Douglas Adams/So Long, and Thanks for All the Fish (362)/So Long, and Thanks for All the Fish - Douglas Adams.epub
#
# Creates a flat file "database" on first run in $HOME/.cache/book_search_cache
# which can be regenerated (cronjob, etc) by running this script with the -e
# switch.
#
# You can also regenerate it immediately before a run by using the -r switch
#
# Use -m to utilize ebook metadata instead of the file structure mentioned above.
#
# add in subject from exiftool (may need to add cpan install Activity::Zip (I think?) for tags
#
#
# Why wasn't I using this???
# calibredb list -f authors,formats,cover,tags,series,series_index,title --separator §
# id is first row, so can run multiple times to attach id to each of these. Or maybe feed into
# memory dynamically for fzf? That's probably better, innit?
# calibredb list -f authors,formats,cover,tags,series,series_index,title --separator §
# # need check for multi-format
# # preview window for cover and/or summary?
#With Preview
#SelectedFile=$(cat "$CacheFile" | fzf --no-hscroll -m --height 80% --border --ansi --no-bold --preview="$SCRIPTDIR/quite-intriguing-preview {}" | sed 's/ (/./g' | sed 's/)//g' | sed 's/:man:/:man -Pcat:/g' | awk -F ':' '{print $2 " " $1}'
#
#calibredb list -f title,authors | fzf +x -e -i --header-lines=2 --preview="notify-send {}"
#
# use both so can use whatever backend, even without calibre at all.
##############################################################################
# Books directory
BOOKS_DIR="/home/steven/documents/Calibre Library/"
EXIFTOOL=$(which exiftool)
FD_FIND=$(which fdfind)
EPY=$(which epy) #https://github.com/wustho/epy
EPR=$(which epr) #https://github.com/wustho/epr
declare -a FILES
REGEN="false"
CliOnly="true"
CacheFile="$HOME/.cache/book_search_cache"
CALIBRE_STRUCTURE="true"
# So that I don't have to worry about the structure of the path too hard
SLASHES=$(echo "$BOOKS_DIR" | grep -o '/' | wc -l)
((SLASHES++))
gen_list (){
# I guess this could be done with a do-while loop and process substitution
# instead, but I'm on call and it's 0318 and I don't want to tear it apart.
if [ -f "$FD_FIND" ];then
mapfile FILES < <(fdfind -a -e epub -e pdf -e mobi -e azw3 . "$BOOKS_DIR")
else
mapfile FILES < <(find -H "$BOOKS_DIR" -type f -iname "*.pdf" -or -iname "*.epub" -or -iname "*.mobi" -or -iname "*.azw3")
fi
echo "" > $CacheFile
if [ "$CALIBRE_STRUCTURE" == "true" ];then
for ((i = 0; i < ${#FILES[@]}; i++));do
FILES[$i]=$(echo "${FILES[$i]}" | tr -d '\n')
shortfile=$(basename "${FILES[$i]}")
EXT=${shortfile##*.}
PATH0=$(dirname "${FILES[$i]}")
PATH1=$(echo "$PATH0" | cut -d '/' -f ${SLASHES}- )
#echo "$PATH1 @"
AUTHOR=$(echo "$PATH1" | awk -F '/' '{print $1}')
TITLE=$(echo "$PATH1" | awk -F '/' '{print $2}' | awk -F '(' '{print $1}')
printf "%s | %s | %s | %s \n" "${EXT}" "${AUTHOR}" "${TITLE}" "${FILES[$i]}" >> $CacheFile
done
else
for ((i = 0; i < ${#FILES[@]}; i++));do
metadata=""
FILES[$i]=$(echo "${FILES[$i]}" | tr -d '\n')
metadata=$(exiftool "${FILES[$i]}")
AUTHOR=$(echo "$metadata" | grep -e "^Creator" | grep -v "Creator Id" | awk -F ':' '{print $2}'| xargs)
TITLE=$(echo "$metadata" | grep -e "^Title" | grep -v "Title Id" | awk -F ':' '{print $2}'| xargs)
EXT=$(echo "$metadata" | grep -e "^File Type Extension" | awk -F ':' '{print $2}' | xargs)
printf "%s | %s | %s | %s \n" "${EXT}" "${AUTHOR}" "${TITLE}" "${FILES[$i]}" >> $CacheFile
done
fi
}
main() {
if [ "$REGEN" != "false" ];then
gen_list
fi
if [ ! -f $CacheFile ];then
gen_list
fi
if [ "$CliOnly" == "true" ];then
SelectedBook=$(cat $CacheFile | fzf --no-hscroll -m --height 60% --border --ansi --no-bold --header "Which Book?" )
else
#use ROFI, not zenity
SelectedBook=$(cat $CacheFile | rofi -i -dmenu -p "Which Book?" -theme DarkBlue)
fi
#extra xargs to strip newlines and whitespace
book=$( echo "$SelectedBook" | awk -F '|' '{print $4}' | xargs)
type=$( echo "$SelectedBook" | awk -F '|' '{print $1}' | xargs)
if [ -n "$book" ]; then
if [ "$CliOnly" == "false" ];then
xdg-open "${book}"
else
if [ ! -f "$EPY" ] && [ ! -f "$EPR" ];then
# there is no sense doing extra processing if xdg-open is just
# going to handle it.
xdg-open \""${book}"\"
else
case "$type" in
pdf)
tmpfile=$(mktemp)
## Pick the one you like!
pdftotext -nopgbrk -layout "${book}" "$tmpfile"; bat "$tmpfile"; rm "$tmpfile"
#pdftohtml -c -i -s -zoom 1 "${book}" "$tmpfile"; www-browser "$tmpfile"; rm "$tmpfile"
exit 0
;;
mobi|azw3)
tmpdir=$(mktemp -d)
# You hopefully have calibre?
ebook-convert "${book}" "$tmpdir/out.epub"
book=$(echo "$tmpdir/out.epub")
;;
*) echo "Calling reader"
;;
esac
if [ -f "$EPY" ];then
executeme=$(printf "%s \"%s\"" "$EPY" "$book")
eval "$executeme"
elif [ -f "$EPR" ];then
executeme=$(printf "%s \"%s\"" "$EPR" "$book")
eval "$executeme"
fi
fi
fi
else
echo "Book file not found!" 1>&2
exit 1
fi
}
##############################################################################
# Show the Help
##############################################################################
display_help(){
echo "###################################################################"
echo "# booksearch.sh [-h|-r]"
echo "# -h show help "
echo "# -r regenerate booklist (and run) "
echo "# -e regenerate booklist (and exit) "
echo "# -g use GUI (rofi) "
echo "# -m Use ebook metadata, not file structure"
echo "# FAIR WARNING: PDF METADATA IS OFTEN BORKED BEYOND RECOGNITION!"
echo "###################################################################"
}
# Read in variables
while [ $# -gt 0 ]; do
option="$1"
case $option in
-m) CALIBRE_STRUCTURE="false"
shift ;;
-h) display_help
exit
shift ;;
-r) REGEN="true"
shift ;;
-e) gen_list
exit
shift ;;
-g) CliOnly="false"
shift ;;
esac
done
main