2ascii

#!/bin/sed -f
# -*- coding: utf-8 -*-
#
# NAME
#
#   2ascii - convert UTF-8 punctuation to "equivalent" ASCII
#
# SYNOPSIS
#
#   2ascii [-i] [<file>...]
#
# DESCRIPTION
#
#   2ascii is a sed(1) script to convert UTF-8 punctuation to ASCII.
#   You can sanitize the file further by piping it through iconv:
#
#     iconv -c -f utf-8 -t ascii//TRANSLIT [<file>...]
#
# OPTIONS
#
#   -i  Edit files in place
#

# Fix (long) dashes.
s/\([[:alnum:]]\)[–‒—―]\([[:alnum:]]\)/\1 - \2/g

# Punctuation.
s/[¤]/*/g
s/[§]/*/g
s/[©]/(C)/g
s/[«]/"/g
s/[­]/-/g
s/[®]/(R)/g
s/[°]/deg. /g
s/[´]/'/g
s/[»]/"/g
s/[÷]/\//g
s/[ǀ]/|/g
s/[ǃ]/!/g
s/[ʹ]/'/g
s/[ʺ]/"/g
s/[ʼ]/'/g
s/[˄]/^/g
s/[ˆ]/^/g
s/[ˈ]/'/g
s/[ˋ]/`/g
s/[ˍ]/_/g
s/[˜]/~/g
s/[։]/:/g
s/[׀]/|/g
s/[׃]/:/g
s/[٪]/%/g
s/[٭]/*/g
s/[‐]/-/g
s/[‑]/-/g
s/[‒]/--/g
s/[–]/--/g
s/[—]/--/g
s/[―]/--/g
s/[‖]/||/g
s/[‗]/_/g
s/[‘]/'/g
s/[’]/'/g
s/[‚]/,/g
s/[‛]/'/g
s/[“]/"/g
s/[”]/"/g
s/[„]/"/g
s/[‟]/"/g
s/[…]/.../g
s/[′]/'/g
s/[″]/"/g
s/[‴]/''/g
s/[‵]/`/g
s/[‶]/"/g
s/[‷]/''/g
s/[‸]/^/g
s/[‹]/</g
s/[›]/>/g
s/[‽]/?/g
s/[⁄]/\//g
s/[⁎]/*/g
s/[⁒]/%/g
s/[⁓]/~/g
s/[№]/No./g
s/[℗]/ (P)/g
s/[℠]/ (SM)/g
s/[™]/ (TM)/g
s/[−]/-/g
s/[∕]/\//g
s/[∖]/\\/g
s/[∗]/*/g
s/[∣]/|/g
s/[∶]/:/g
s/[∼]/~/g
s/[≤]/<=/g
s/[≥]/>=/g
s/[≦]/<=/g
s/[≧]/>=/g
s/[⌃]/^/g
s/[〈]/</g
s/[〉]/>/g
s/[◊]/*/g
s/[♯]/#/g
s/[✱]/*/g
s/[❘]/|/g
s/[❢]/!/g
s/[⟦]/[/g
s/[⟨]/</g
s/[⟩]/>/g
s/[⦃]/{/g
s/[⦄]/}/g
s/[〃]/"/g
s/[〈]/</g
s/[〉]/>/g
s/[〛]/]/g
s/[〜]/~/g
s/[〝]/"/g
s/[〞]/"/g

# Convert n-space separated sentences into two space separated
# sentences.  This is still buggy as it fails to differentiate between
# periods used to end sentences, and periods used to shorten words like
# in "Mrs.".
s/\([^[:space:]]\{2\}[.!?][`'"]\?\)[[:space:]]\+/\1  /g

# Convert DOS style line ending to *nix style.
s/\r$//
s/\r/\n/g