mdwc

#! /usr/bin/env bash

# Copyright (c) 2016, 2018 Phillip Alday
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.

# argument parsing adapted from http://stackoverflow.com/a/29754866/2022326

getopt --test > /dev/null
if [[ $? != 4 ]]; then
    echo "I'm sorry, 'getopt --test' failed in this environment."
    echo "Exiting (your shell doesn't support needed functionality)."
    exit 1
fi

SHORT=cmlLwtbhf:k:
LONG=bytes,chars,lines,max-line-length,words,files0-from:,title,body,help,field:,key:

# -temporarily store output to be able to check for errors
# -activate advanced mode getopt quoting e.g. via “--options”
# -pass arguments only via   -- "$@"   to separate them correctly
PARSED=$(getopt --options $SHORT --longoptions $LONG --name "$0" -- "$@")
if [[ $? != 0 ]]; then
    # e.g. $? == 1
    #  then getopt has complained about wrong arguments to stdout
    exit 2
fi

# use eval with "$PARSED" to properly handle the quoting
eval set -- "$PARSED"

WCOPTS=""
BODY=false
TITLE=false
FIELDS=""

# now enjoy the options in order and nicely split until we see --
while true; do
    case "$1" in
        -c|--bytes)
            WCOPTS="${WCOPTS} -c"
            shift
            ;;
        -m|--chars)
            WCOPTS="${WCOPTS} -m"
            shift
            ;;
        -l|--lines)
            WCOPTS="${WCOPTS} -l"
            shift
            ;;
        -L|--max-line-length)
            WCOPTS="${WCOPTS} -L"
            shift
            ;;
        -w|--words)
            WCOPTS="${WCOPTS} -w"
            shift
            ;;
        --files0-from)
            WCOPTS="${WCOPTS} --files0-from=$2"
            shift 2
            ;;
        --title|-t)
            TITLE=true
            shift
            ;;
        --field|--key|-f|-k)
            FIELDS="${FIELDS},$2"
            shift 2
            ;;
        --body|-b)
            BODY=true
            shift
            ;;
        --help|-h)
            echo "$0: wc wrapper for pandoc MarkDown files with YAML block."
            echo
            echo "arguments:"
            echo -e "  --title, -t   \t run wc on title line in initial YAML block"
            echo -e "                \t (note: will not work on multiline titles)"
            echo
            echo -e "  --body,  -b   \t run wc on body (everything after initial YAML block)"
            echo
            echo -e "  --field, -f   \t run wc on specified top-level YAML field"
            echo -e "                \t Note: this will extract multiline fields and can be"
            echo -e "                \t used to extract multliline titles. This method is "
            echo -e "                \t slower, which may be come apparant when applied to "
            echo -e "                \t large (numbers of) files."
            echo
            echo -e "  --key,   -k   \t synonym for --field. One of these may be removed"
            echo -e "                \t in the future"
            echo
            echo "wc arguments supported: (see man wc for more details)"
            echo "  --bytes, -c"
            echo "  --chars, -m"
            echo "  --words, -w"
            echo "  --lines, -l"
            echo "  --max-line-length, -L"
            echo "  --files0-from"
            echo
            echo "Default is wc on just the body (useful for situations with length limits such as abstracts)" | fold -s
            exit
            ;;
        --) # end of options
            shift
            break
            ;;
        *)
            echo "Unrecognized option: $1"
            echo "(How did you get here? Please consider reporting a bug)"
            exit 3
            ;;
    esac
done

# handle non-option arguments

# set default behavior to report body stats
if [[ "$TITLE" = false && "$BODY" = false && "$FIELDS" = "" ]]; then
    BODY=true
fi

# if there are no arguments, then something is being piped in, so we need to handle that
# the temporary file is not terribly efficient, but it's easy and we expect this to be a corner case
if [[ $# == 0 ]]; then
    TMP=$(mktemp)
    tee $TMP 1>/dev/null
    eval set -- "$TMP"
fi

while [[ $# != 0 ]]; do
    # don't display the gibberish temporary filename
    if [ "$1" != "$TMP" ]; then
        FNAME=$1
    else
        FNAME=""
    fi

    if [ "$FIELDS" ]; then

        for key in $(echo ${FIELDS} | tr ", " "\n\n")
        do
            value=$(sed -Ee '/^---\w*$/,/^(---|...)$/!d' $1 | egrep "^${key}" | awk -e'{$1="";print $0}')
            # trim whitespace
            value=$(echo $value| xargs echo -n)

            if [[ "$value" = "|" ]]; then
                # extract the relevant chunk then
                # remove first empty line and last line, which is the next key
                value=$(sed -Ee '/^---\w*$/,/^(---|...)$/!d' $1 | sed -Ee "/^${key}/,/^[a-z]/!d" | sed -e '1d;$d')
            fi

            echo $value | wc $WCOPTS | awk -e '{print $0, "'${FNAME}'", "('${key}')"}'
        done
    fi

    if [ "$TITLE" = true ]; then
        sed -Ee '/^---\w*$/,/^(---|...)$/!d' $1 | egrep '^title:' | awk -e'{$1="";print $0}'| wc $WCOPTS | awk -e '{print $0, "'${FNAME}'", "(title)"}'
    fi
    if [ "$BODY" = true ]; then
        sed -Ee '/^---\w*$/,/^(---|...)\w*$/d' $1 | wc $WCOPTS | awk -e '{print $0, "'${FNAME}'", "(body)"}'
    fi
    shift
done

# clean up the tempfile if necessary
if [ -e "$TMP" ]; then
    rm "$TMP"
fi