1//
  2// Copyright (c) 2019 Ted Unangst <tedu@tedunangst.com>
  3//
  4// Permission to use, copy, modify, and distribute this software for any
  5// purpose with or without fee is hereby granted, provided that the above
  6// copyright notice and this permission notice appear in all copies.
  7//
  8// THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
  9// WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
 10// MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
 11// ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 12// WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
 13// ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 14// OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 15
 16package main
 17
 18import (
 19	"regexp"
 20
 21	"github.com/mattn/go-runewidth"
 22)
 23
 24// these lists are mostly (?) accurate
 25var bigboldshitz = "𝐀𝐁𝐂𝐃𝐄𝐅𝐆𝐇𝐈𝐉𝐊𝐋𝐌𝐍𝐎𝐏𝐐𝐑𝐒𝐓𝐔𝐕𝐖𝐗𝐘𝐙"
 26var lilboldshitz = "𝐚𝐛𝐜𝐝𝐞𝐟𝐠𝐡𝐢𝐣𝐤𝐥𝐦𝐧𝐨𝐩𝐪𝐫𝐬𝐭𝐮𝐯𝐰𝐱𝐲𝐳"
 27var moeboldshitz = "𝗔𝗕𝗖𝗗𝗘𝗙𝗚𝗛𝗜𝗝𝗞𝗟𝗠𝗡𝗢𝗣𝗤𝗥𝗦𝗧𝗨𝗩𝗪𝗫𝗬𝗭"
 28var morboldshitz = "𝗮𝗯𝗰𝗱𝗲𝗳𝗴𝗵𝗶𝗷𝗸𝗹𝗺𝗻𝗼𝗽𝗾𝗿𝘀𝘁𝘂𝘃𝘄𝘅𝘆𝘇"
 29var biggothshitz = "𝕬𝕭𝕮𝕯𝕰𝕱𝕲𝕳𝕴𝕵𝕶𝕷𝕸𝕹𝕺𝕻𝕼𝕽𝕾𝕿𝖀𝖁𝖂𝖃𝖄𝖅"
 30var lilgothshitz = "𝖆𝖇𝖈𝖉𝖊𝖋𝖌𝖍𝖎𝖏𝖐𝖑𝖒𝖓𝖔𝖕𝖖𝖗𝖘𝖙𝖚𝖛𝖜𝖝𝖞𝖟"
 31var moegothshitz = "𝔄𝔅𝕮𝔇𝔈𝔉𝔊𝕳ℑ𝔍𝔎𝔏𝔐𝔑𝔒𝔓𝔔ℜ𝔖𝔗𝔘𝔙𝔚𝔛𝔜𝖅"
 32var morgothshitz = "𝔞𝔟𝔠𝔡𝔢𝔣𝔤𝔥𝔦𝔧𝔨𝔩𝔪𝔫𝔬𝔭𝔮𝔯𝔰𝔱𝔲𝔳𝔴𝔵𝔶𝔷"
 33var bigitalshitz = "𝑨𝑩𝑪𝑫𝑬𝑭𝑮𝑯𝑰𝑱𝑲𝑳𝑴𝑵𝑶𝑷𝑸𝑹𝑺𝑻𝑼𝑽𝑾𝑿𝒀𝒁"
 34var lilitalshitz = "𝒂𝒃𝒄𝒅𝒆𝒇𝒈𝒉𝒊𝒋𝒌𝒍𝒎𝒏𝒐𝒑𝒒𝒓𝒔𝒕𝒖𝒗𝒘𝒙𝒚𝒛"
 35var moeitalshitz = "𝘼𝘽𝘾𝘿𝙀𝙁𝙂𝙃𝙄𝙅𝙆𝙇𝙈𝙉𝙊𝙋𝙌𝙍𝙎𝙏𝙐𝙑𝙒𝙓𝙔𝙕"
 36var moritalshitz = "𝙖𝙗𝙘𝙙𝙚𝙛𝙜𝙝𝙞𝙟𝙠𝙡𝙢𝙣𝙤𝙥𝙦𝙧𝙨𝙩𝙪𝙫𝙬𝙭𝙮𝙯"
 37var bigbangshitz = "𝔸𝔹ℂ𝔻𝔼𝔽𝔾ℍ𝕀𝕁𝕂𝕃𝕄ℕ𝕆ℙℚℝ𝕊𝕋𝕌𝕍𝕎𝕏𝕐ℤ"
 38var lilbangshitz = "𝕒𝕓𝕔𝕕𝕖𝕗𝕘𝕙𝕚𝕛𝕜𝕝𝕞𝕟𝕠𝕡𝕢𝕣𝕤𝕥𝕦𝕧𝕨𝕩𝕪𝕫"
 39var bigwideshitz = "ＡＢＣＤＥＦＧＨＩＪＫＬＭＮＯＰＱＲＳＴＵＶＷＸＹＺ"
 40var lilwideshitz = "ａｂｃｄｅｆｇｈｉｊｋｌｍｎｏｐｑｒｓｔｕｖｗｘｙｚ"
 41var bigblokshitz = "🅰🅱🅲🅳🅴🅵🅶🅷🅸🅹🅺🅻🅼🅽🅾🅿🆀🆁🆂🆃🆄🆅🆆🆇🆈🆉"
 42var evenmoeshitz = "𝐴𝐵𝐶𝐷𝐸𝐹𝐺𝐻𝐼𝐽𝐾𝐿𝑀𝑁𝑂𝑃𝑄𝑅𝑆𝑇𝑈𝑉𝑊𝑋𝑌𝑍"
 43var evenmorshitz = "𝒶𝒷𝒸𝒹𝑒𝒻𝓰𝒽𝒾𝒿𝓀𝓁𝓂𝓃𝓸𝓅𝓆𝓇𝓈𝓉𝓊𝓋𝓌𝓍𝓎𝓏"
 44
 45var re_alltheshitz = regexp.MustCompile(`([` +
 46	bigboldshitz + lilboldshitz +
 47	moeboldshitz + morboldshitz +
 48	biggothshitz + lilgothshitz +
 49	moegothshitz + morgothshitz +
 50	bigitalshitz + lilitalshitz +
 51	moeitalshitz + moritalshitz +
 52	bigbangshitz + lilbangshitz +
 53	bigwideshitz + lilwideshitz +
 54	evenmoeshitz + evenmorshitz +
 55	bigblokshitz +
 56	"][ '\ufe0f]?){3,}")
 57
 58var allUppers = []string{bigboldshitz, moeboldshitz, biggothshitz, bigwideshitz, moegothshitz, bigitalshitz, moeitalshitz, bigbangshitz, bigblokshitz, evenmoeshitz}
 59var allLowers = []string{lilboldshitz, morboldshitz, lilgothshitz, lilwideshitz, morgothshitz, lilitalshitz, moritalshitz, lilbangshitz, evenmorshitz}
 60
 61var skinTones = "\U0001F3FB\U0001F3FC\U0001F3FD\U0001F3FE\U0001F3FF"
 62var re_moredumb = regexp.MustCompile("[\U0001f44f\U0001f6a8\U000026a0][" + skinTones + "\ufe0f]*")
 63
 64// this may not be especially fast
 65func unpucker(s string) string {
 66	fixer := func(r string) string {
 67		x := make([]byte, len(r))
 68		xi := 0
 69	loop1:
 70		for _, c := range r {
 71			xi++
 72			if c == ' ' || c == '\'' {
 73				x[xi] = byte(c)
 74				continue
 75			}
 76			for _, set := range allUppers {
 77				i := 0
 78				for _, rr := range set {
 79					if rr == c {
 80						x[xi] = byte('A' + i)
 81						continue loop1
 82					}
 83					i++
 84				}
 85			}
 86			for _, set := range allLowers {
 87				i := 0
 88				for _, rr := range set {
 89					if rr == c {
 90						x[xi] = byte('a' + i)
 91						continue loop1
 92					}
 93					i++
 94				}
 95			}
 96			x[xi] = '.'
 97		}
 98		return string(x)
 99	}
100	s = re_alltheshitz.ReplaceAllStringFunc(s, fixer)
101
102	s = re_moredumb.ReplaceAllString(s, ".")
103
104	zw := false
105	for _, c := range s {
106		if runewidth.RuneWidth(c) == 0 {
107			zw = true
108			break
109		}
110	}
111	if zw {
112		x := make([]byte, 0, len(s))
113		zw = false
114		for _, c := range s {
115			if runewidth.RuneWidth(c) == 0 {
116				if zw {
117					continue
118				}
119				zw = true
120			} else {
121				zw = false
122			}
123			q := string(c)
124			x = append(x, []byte(q)...)
125		}
126		return string(x)
127	}
128	return s
129}