count-word-occurrences

#!/usr/bin/env bash

basename="$(basename "$0")"

# Check for required arguments, explain usage if not present.
if [ $# -lt 1 ]; then
    >&2 echo "usage: $basename <filename> [minimum]"
    exit 1
fi

# Optionally filter out words below a minimum number of occurrences.
if [ $# -gt 1 ]; then
    min_count="$2"
else
    min_count=1
fi

# Can we access the given filename?
filename="$1"
if [ ! -f "$filename" ]; then
    >&2 echo "fatal: file '$filename' does not exist"
    exit 2
fi
if [ ! -r "$filename" ]; then
    >&2 echo "fatal: cannot read file '$filename'"
    exit 3
fi

# Strip link URLs from Markdown documents.
textfile="/tmp/${basename}-${filename}.txt"
sed 's/\!\{0,1\}\[[^]]*\]([^)]*)//g' "$filename" > "$textfile" || {
    >&2 echo "fatal: cannot write to file '$textfile'"
    exit 4
}

words=$(cat /usr/share/dict/words | tr '[:upper:]' '[:lower:]' | uniq)

for word in $words; do
    occurrences="$(grep -ciw "$word" "$textfile")"
    if [ "$occurrences" -ge "$min_count" ]; then
        echo "$word: $occurrences"
    fi
done