Rev 291 | Rev 293 | Go to most recent revision | Details | Compare with Previous | Last modification | View Log | RSS feed
| Rev | Author | Line No. | Line |
|---|---|---|---|
| 292 | PointedEar | 1 | #!/usr/bin/env python3 |
| 2 | |||
| 291 | PointedEar | 3 | ''' |
| 4 | Created on 2014-10-20 |
||
| 5 | |||
| 6 | @author: Thomas 'PointedEars' Lahn <mail@PointedEars.de> |
||
| 7 | ''' |
||
| 8 | from sys import argv, stderr |
||
| 9 | from re import findall, DOTALL, IGNORECASE, match, sub, compile, \ |
||
| 10 | split |
||
| 292 | PointedEar | 11 | from os import chdir, stat |
| 12 | from os.path import dirname, realpath, basename |
||
| 291 | PointedEar | 13 | from collections import OrderedDict |
| 14 | from functools import cmp_to_key |
||
| 15 | from copy import deepcopy |
||
| 16 | from collections.abc import MutableSequence |
||
| 292 | PointedEar | 17 | from pickle import dump, load |
| 291 | PointedEar | 18 | |
| 292 | PointedEar | 19 | debug_level = 2 |
| 291 | PointedEar | 20 | dictionary = {} |
| 21 | |||
| 22 | prepositions = { |
||
| 23 | "fi'": 'on', |
||
| 24 | "na'": 'at|to', |
||
| 25 | "t'": 'of' |
||
| 26 | } |
||
| 27 | |||
| 292 | PointedEar | 28 | def dmsg(*args, **kwargs): |
| 29 | if not hasattr(kwargs, 'min_level') or kwargs['min_level'] is None: |
||
| 30 | kwargs['min_level'] = 1 |
||
| 31 | |||
| 32 | if not hasattr(kwargs, 'file'): |
||
| 33 | kwargs['file'] = stderr |
||
| 34 | |||
| 35 | if debug_level >= kwargs['min_level']: |
||
| 36 | del kwargs['min_level'] |
||
| 37 | print(*args, **kwargs) |
||
| 38 | |||
| 291 | PointedEar | 39 | class MutableString2(MutableSequence): |
| 40 | def __init__(self, value=None): |
||
| 41 | self._values = [str(value)] if value is not None else [] |
||
| 42 | |||
| 43 | def __add__(self, value): |
||
| 44 | return ''.join([self, value]) |
||
| 45 | |||
| 46 | def __delitem__(self): |
||
| 47 | raise NotImplementedError |
||
| 48 | |||
| 49 | def __getitem__(self, index): |
||
| 50 | return str(self)[index] |
||
| 51 | |||
| 52 | def __len__(self): |
||
| 53 | return len(str(self)) |
||
| 54 | |||
| 55 | def __repr__(self): |
||
| 56 | return ''.join(self._values) |
||
| 57 | |||
| 58 | def __setitem__(self, index, value): |
||
| 59 | raise NotImplementedError |
||
| 60 | |||
| 61 | def __str__(self): |
||
| 62 | return self.__repr__() |
||
| 63 | |||
| 64 | def extend(self, values): |
||
| 65 | self._values.append(values) |
||
| 66 | |||
| 67 | def insert(self, index, value): |
||
| 68 | raise NotImplementedError |
||
| 69 | |||
| 292 | PointedEar | 70 | def cli_help(): |
| 71 | print('Usage: {0} TEXT...'.format(basename(argv[0]))) |
||
| 72 | |||
| 291 | PointedEar | 73 | def load_dictionary(dictionary, dictionary_file): |
| 292 | PointedEar | 74 | dmsg('Loading dictionary '.format(dictionary_file), end='', min_level=1) |
| 291 | PointedEar | 75 | |
| 76 | chdir(dirname(realpath(__file__))) |
||
| 77 | |||
| 292 | PointedEar | 78 | pickle_file = basename(dictionary_file) + '.pickle' |
| 79 | |||
| 80 | try: |
||
| 81 | pickle_mtime = stat(pickle_file).st_mtime |
||
| 82 | except FileNotFoundError: |
||
| 83 | pickle_mtime = None |
||
| 84 | |||
| 85 | if pickle_mtime is None or stat(dictionary_file).st_mtime > pickle_mtime: |
||
| 86 | dmsg('from {0} ...'.format(dictionary_file), end='', min_level=1) |
||
| 87 | with open(dictionary_file) as f: |
||
| 88 | keys = "ipa|en|lit|pos|com|tag|ex" |
||
| 89 | indent = None |
||
| 90 | value = None |
||
| 91 | |||
| 92 | for line in f: |
||
| 93 | m = match(r'^\s*vuh:\s*(?P<phrase>.+)', line) |
||
| 291 | PointedEar | 94 | if m is not None: |
| 292 | PointedEar | 95 | phrase = m.group("phrase") |
| 96 | dictionary[phrase] = {} |
||
| 97 | indent = None |
||
| 98 | else: |
||
| 99 | m = match(r'(?P<indent>\s*)(?P<key>{0}):\s*(?P<value>.+)'.format(keys), line) |
||
| 291 | PointedEar | 100 | if m is not None: |
| 292 | PointedEar | 101 | indent = m.group("indent") |
| 102 | key = m.group("key") |
||
| 103 | value = m.group("value") |
||
| 104 | value = dictionary[phrase][key] = MutableString2(value) |
||
| 105 | elif indent is not None: |
||
| 106 | m = match(r'(?P<indent>\s+)(?P<continuation>\S.*)', line) |
||
| 107 | if m is not None: |
||
| 108 | if len(m.group("indent")) == len(indent) + 2: |
||
| 109 | dictionary[phrase][key] += (" " + m.group("continuation")) |
||
| 291 | PointedEar | 110 | |
| 292 | PointedEar | 111 | dmsg('\nSaving pickle {0} ...'.format(pickle_file), end='', min_level=1) |
| 112 | # TODO: Pickle should only contain strings to be small |
||
| 113 | with open(pickle_file, mode='wb') as f: dump(dictionary, f) |
||
| 114 | dmsg(' done.', min_level=1) |
||
| 115 | else: |
||
| 116 | dmsg('from {0} ...'.format(pickle_file), end='', min_level=1) |
||
| 117 | with open(pickle_file, mode='rb') as f: pickle = load(f) |
||
| 118 | for key, value in pickle.items(): |
||
| 119 | dictionary[key] = value |
||
| 291 | PointedEar | 120 | |
| 292 | PointedEar | 121 | dmsg(' done ({0} entries).'.format(len(dictionary)), min_level=1) |
| 122 | |||
| 291 | PointedEar | 123 | def clean_dictionary(dictionary): |
| 124 | braces_re = compile(r'^\s*\{(.+)\}\s*$') |
||
| 125 | semicolon_re = compile(r'\s*;\s*') |
||
| 126 | |||
| 127 | for orig_phrase, data in list(dictionary.items()): |
||
| 128 | if orig_phrase.find(";") > -1: |
||
| 129 | synonyms = map( |
||
| 130 | lambda x: sub(braces_re, r'\1', orig_phrase), |
||
| 131 | split(semicolon_re, orig_phrase)) |
||
| 132 | |||
| 133 | for synonym in synonyms: |
||
| 134 | dictionary[synonym] = deepcopy(data) |
||
| 135 | |||
| 136 | del dictionary[orig_phrase] |
||
| 137 | else: |
||
| 138 | m = match(braces_re, orig_phrase) |
||
| 139 | if m is not None: |
||
| 140 | dictionary[m.group(1)] = deepcopy(dictionary[orig_phrase]) |
||
| 141 | del dictionary[orig_phrase] |
||
| 142 | |||
| 143 | def sort_dict_alnum_english_key(phrase): |
||
| 144 | return sub(r'\{(.+)\}', r'\1', phrase[0]).lower() |
||
| 145 | |||
| 146 | def get_sort_dict_alnum_vulcan_key(): |
||
| 147 | letters = list(map(str.lower, [ |
||
| 148 | " ", 'S', 'T', 'P', 'K', 'R', 'L', 'A', 'Sh', 'O', 'U', 'D', |
||
| 149 | 'V', 'Kh', 'E', 'H', 'G', 'Ch', 'I', 'N', 'Zh', 'M', 'Y', 'F', 'Z', |
||
| 150 | 'Th', 'W', 'B', "'", '-'])) |
||
| 151 | letter_values = dict(map(lambda x: (x[1], x[0]), enumerate(letters))) |
||
| 152 | letters_re = compile(r'(?:{0})'.format('|'.join(sorted(letters, key=lambda char:-len(char))))) |
||
| 153 | |||
| 154 | def sort_dict_alnum_vulcan (a, b): |
||
| 155 | # split into Vulcan letters |
||
| 156 | a = findall(letters_re, sort_dict_alnum_english_key(a)) |
||
| 157 | b = findall(letters_re, sort_dict_alnum_english_key(b)) |
||
| 158 | |||
| 159 | if len(a) < len(b): |
||
| 160 | for index, char in enumerate(a): |
||
| 161 | diff = letter_values[char] - letter_values[b[index]] |
||
| 162 | if diff != 0: |
||
| 163 | return diff |
||
| 164 | return -1 |
||
| 165 | |||
| 166 | # len(b) <= len(a) |
||
| 167 | for index, char in enumerate(b): |
||
| 168 | diff = letter_values[a[index]] - letter_values[char] |
||
| 169 | if diff != 0: |
||
| 170 | return diff |
||
| 171 | |||
| 172 | return 1 if len(b) < len(a) else 0 |
||
| 173 | |||
| 174 | return cmp_to_key(sort_dict_alnum_vulcan) |
||
| 175 | |||
| 292 | PointedEar | 176 | def translate (phrase): |
| 177 | translation = dictionary.get(phrase.lower(), None) |
||
| 291 | PointedEar | 178 | if translation is not None: |
| 179 | return translation |
||
| 180 | |||
| 292 | PointedEar | 181 | return None |
| 291 | PointedEar | 182 | |
| 292 | PointedEar | 183 | if __name__ == '__main__': |
| 184 | if len(argv) < 2: |
||
| 185 | print('Nothing to translate.', end='\n\n', file=stderr) |
||
| 186 | cli_help() |
||
| 187 | exit(1) |
||
| 291 | PointedEar | 188 | |
| 292 | PointedEar | 189 | text = argv[1] |
| 190 | |||
| 291 | PointedEar | 191 | load_dictionary(dictionary, 'vuh-gol-en.dict.zdb.txt') |
| 192 | clean_dictionary(dictionary) |
||
| 193 | |||
| 194 | # try: |
||
| 195 | # for phrase, data in OrderedDict(sorted( |
||
| 196 | # dictionary.items(), |
||
| 197 | # key=get_sort_dict_alnum_vulcan_key() |
||
| 198 | # )).items(): |
||
| 199 | # print(phrase, "=", data) |
||
| 200 | # except BrokenPipeError: |
||
| 201 | # pass |
||
| 202 | |||
| 203 | sentences = findall(r'(?!\s+)(?:.+?\.{1,3}|.+$)', text, DOTALL) |
||
| 292 | PointedEar | 204 | dmsg("sentences:", sentences, min_level=2) |
| 291 | PointedEar | 205 | for sentence in sentences: |
| 292 | PointedEar | 206 | dmsg("sentence:", sentence, min_level=2) |
| 291 | PointedEar | 207 | |
| 292 | PointedEar | 208 | clauses = split(r'\s+[-–—]\s+', sentence) |
| 209 | dmsg("clauses:", clauses, min_level=2) |
||
| 210 | for clause in clauses: |
||
| 211 | dmsg("clause:", clause, min_level=2) |
||
| 291 | PointedEar | 212 | |
| 292 | PointedEar | 213 | words = findall(r'[^\s.]+', clause) |
| 214 | dmsg("words:", words, min_level=2) |
||
| 291 | PointedEar | 215 | |
| 292 | PointedEar | 216 | offset = 0 |
| 217 | while offset < len(words): |
||
| 218 | translation = None |
||
| 291 | PointedEar | 219 | |
| 292 | PointedEar | 220 | for i in reversed(range(offset + 1, len(words) + 1)): |
| 221 | phrase = ' '.join(words[offset:i]) |
||
| 291 | PointedEar | 222 | |
| 292 | PointedEar | 223 | dmsg("phrase:", phrase, min_level=2) |
| 224 | |||
| 225 | translation = translate(phrase) |
||
| 226 | |||
| 227 | if translation is not None: |
||
| 228 | dmsg("phrase-translation:", translation, min_level=2) |
||
| 229 | dmsg("words[{0}:{1}] = [\"{2}\"]".format(offset, i, translation), min_level=2) |
||
| 230 | words[offset:i] = [translation] |
||
| 231 | offset += i - 1 |
||
| 232 | break |
||
| 233 | |||
| 234 | if translation is None: |
||
| 235 | dmsg("phrase-translation:", translation, min_level=2) |
||
| 236 | offset += 1 |
||
| 237 | |||
| 238 | dmsg("words-translation:", words, min_level=2) |