say

#!/usr/local/bin/python3
# -*- coding: utf-8 -*-

# Phoneme guide:
# http://developer.apple.com/library/mac/#documentation/UserExperience/Conceptual/SpeechSynthesisProgrammingGuide/Phonemes/Phonemes.html
#
# Regular Expressions guide:
# http://docs.python.org/library/re.html

import sys, re
from subprocess import call

array1 = (
    (r'U\.S\.', r' U S '),
    (r'@', r' at-sign '),
    (r'\s\|\s', r' pipe '),
    (r'\s\.\.\s', r' dot dot '),
    (r'\s_\s', r' underscore '),
    (r'\(\)', r'empty-parentheses'),
    (r'\{\}', r'empty-curly-braces'),
    (r'\[\]', r'empty-square-brackets'),
    (r'<(.{1,20}?)>', r' over \1'),
    # (r'\((.{1,20}?)\)', r' of \1'),
    (r'\s\?\s', r' question-mark '),
    (r'\s=>\s', r' fat-arrow '),
    (r'\s->\s', r' arrow '),
    (r'\s<\s', r'less-than'),
    (r'\s>\s', r'greater-than'),
    (r'\s>=\s', r'greater-than-or-equal-to'),
    (r'\s<=\s', r'less-than-or-equal-to'),
    (r'\^', r'(carat)'),
    (r'´', r"'"),
    (r'\.\.\.', r'[[slnc 500]]'), # Pause on ellipsis
    (r'·', "point"),
    (r"$", r'. . .'),
    (r'(\d+)\s?[-–—]\s?(\d+)%', r'\1 to \2 percent'),
    #(r'\s"(\S)', r' quote, \1'),
    #(r'(\S)"\s', r'\1, end quote, '),
    (r'“(.{1,50}?)”', r', \1, '),
    #(r'“', r'quote,'),
    #(r'”', r', end quote, '),
    (r'\.$', r'.\n\n'),
    (r'(\d)\.(\d)', r'\1 point \2'),
    (r'(\D)\.(\w)', r'\1 dot \2'),
    (r'(\w)\.(\w)', r'\1 dot \2'),
    (r'(\D)\.(\d+)', r'\1. Reference \2.'),
    (r'Mt\.', r'mount'),
    (r'e\.g\.', r'for example'),
    (r'U\.S\.', r'United States'),
    (r'\n+', r'[[slnc 200]]'), # Pause on newlines
    (r'(\w)-(\w)', r'\1 \2'),
    (r'([^:]*)::([^:]*)', r'\1 double-colon \2'),
    (r'/', r' slash '),
    (r'\[\d+\]', r''), # Remove Wikipedia footnote references [xx]
    (r'(\w+?)- (\w+?)', r'\1\2') # Close hyphenated words
)

array2 = (
    # General
#	('live', 'lIHv'),

    # Occult
#	('Aleister', '1AElEHstr'),
#	('Anubis', '1AEn2AXbIXs'),
#	('Aquino', 'AXkw1IYnOW'),
#	('Crowley', 'kr1OWlIY'),
#	('Enochian', 'AXn1OWkIYAXn'),
#	('grimoire', 'grIHmw1AOr'),
#	('Gurdjieff', 'g1UXrJAXf'),
#	('initiates', 'IHn1IXSIYAXts'),
#	('Ipsissimus', '2IHps1IHsIHmUXs'),
#	('LaVey', 'lAXv1EY'),
#	('LaVeyan', 'lUX +vEYIHn'),
#	('lycanthropy', 'lAYk1AEnTrAXp2IY'),
#	('mage', 'm1EYJ'),
#	('magister', 'm1AEJihstAXr'),
#	('magistry', 'm1AEJihstrIY'),
#	('magus', 'm1EYgAXs'),
#	('Nemo', 'n1IYmOW'),
#	('Satanism', 's1EYtn2IHzm'),
#	('templi', 't1EHmplIY'),
#	('vampirism', 'v1AEmpAXr2IHzAXm'),
#	('Walpurgisnacht', 'wAOlp1AXrgIHsnAOkt'),
#	('Xem', 'k1EHm'),
#	('Xeper', 'k1EHfAXr'),
#	('gods', 'gAAdz'),
#	(r'I\s*(°|degree)', 'fUXrst dUXgr1IY'),
#	(r'II\s*(°|degree)', 's1EHk2AXndUXgr1IY'),
#	(r'III\s*(°|degree)', 'T1IXrdUXgr1IY'),
#	(r'IV\s*(°|degree)', 'f1OWrT dUXgr1IY'),
#	(r'V\s*(°|degree)', 'f1IHfT dUXgr1IY'),
#	(r'VI\s*(°|degree)', 's1IXkT dUXgr1IY'),
#	(r'VII\s*(°|degree)', 's1EHvAXnT dUXgr1IY'),
#	(r'VIII\s*(°|degree)', '1EYtT dUXgr1IY'),
#	(r'IX\s*(°|degree)', 'n1AYnT dUXgr1IY'),
#	(r'X\s*(°|degree)', 't1EHnT dUXgr1IY'),

    # Philosophy
#	('Nietzsche', 'n1IYCIY'),
#	('praxis', 'prAEksIHs'),

    # Politics
#	('anarcho', '2AEn1AArkOW'),
#	('memetic', 'mIYmEHt1IHk'),
#   ('memeplexes', 'mIYm1plEHksAXz'),
#	('memeplex', 'mIYm1plEHks'),
#	('meme', 'mIYm'),

    # Physiology
#	('androgenic', 'AEndrOWJ1EHnIXk'),
#	('endogenous', 'EHnd1AOJEHnIXs'),
#	('exogenous', 'IXks1AOJEHnIXs'),
#	('periodization', 'pIXrIYUXd2UXz1EYSUXn'),
#	('periodized', 'pIXrIYUXdAYzd'),
#	('periodize', 'pIXrIYUXdAYz'),
#	('progestin', 'prOWJ1EHstIXn'),
#	('reps', '~rEHps'),

    # Personal Improvement
#	('Csikszentmihalyi', 'C2IXksEHntm1IYhAY'),

    # History
#	('Elizabeth I', 'EHl1IHzAXb2EHT DAX f1UXrst'),

    # Miscellaneous
#	('kazillion', 'kUXz1IHlIYUXn'),
#	('snazzy', 'sn1AEzIY'),

    # Men's Rights
#	('hypergamy', 'hAYp1IXrgAXmIY'),
#	('hypergamous', 'hAYp1IXrgAXmUXs'),

  # Technology
    ('destructuring', 'de-structuring'),
    ('tuples', 'tooples'),
    ('tuple', 'toople'),
    ('Rustaceans', 'rustaishuns'),
    ('Rustacean', 'rustaishun'),
    ('iteration', 'iteraishun'),
    ('enums', 'eenums'),
    ('enum', 'eenum'),
    ('conciseness', 'concise-ness'),
    ('enumerated', 'enumerayted'),
    ('iterator', 'itarator'),
    ('iterators', 'itarators'),
    ('dyn', 'dine'),
    ('deref', 'dee-ref'),
    ('println', 'print-line'),
    ('rustc', 'rust-c'),
    ('serde', 'ser-day'),
    ('TOML', 'Tomal'),
    ('an SQLite', 'a SQLite'),
    ('SQLite store', 'sequel light store'),
    ('SQLite', 'sequel light'),
    ('async', 'a-sink'),
    ('uniquing', 'unique-ing'),
    ('pecs', 'peks'),
    ('cbor', 'see-bore'),
    ('dcbor', 'd-see-bore')
)

originalText = ''
for f in sys.stdin:
    originalText += f

# Split the text into paragraphs
paragraphs = re.split(r'\s*\n\s*\n\s*', originalText)

# Remove leading and trailing whitespace and join lines for each paragraph
processedParagraphs = [' '.join(p.strip() for p in paragraph.split('\n')) for paragraph in paragraphs]

# Join the paragraphs back together with blank lines in between
originalText = '\n\n'.join(processedParagraphs)

modifiedText = originalText

for t in array1:
    pattern = t[0]
    replacement = t[1]
    modifiedText = re.sub(pattern, replacement, modifiedText, flags = re.IGNORECASE | re.MULTILINE);

for t in array2:
    pattern = t[0]
    #replacement = ' [[inpt PHON]]' + t[1] + '[[inpt TEXT]] '
    replacement = t[1]
    pattern = r'\b' + pattern + r'\b'
    modifiedText = re.sub(pattern, replacement, modifiedText, flags = re.IGNORECASE | re.MULTILINE);

call(["say", modifiedText])

# print(modifiedText)