Rev 291 | Rev 293 | Go to most recent revision | Only display areas with differences | Ignore whitespace | Details | Blame | Last modification | View Log | RSS feed
| Rev 291 | Rev 292 | ||
|---|---|---|---|
| - | 1 | #!/usr/bin/env python3
|
|
| - | 2 | ||
| 1 | '''
|
3 | '''
|
| 2 | Created on 2014-10-20
|
4 | Created on 2014-10-20
|
| 3 | 5 | ||
| 4 | @author: Thomas 'PointedEars' Lahn <mail@PointedEars.de>
|
6 | @author: Thomas 'PointedEars' Lahn <mail@PointedEars.de>
|
| 5 | '''
|
7 | '''
|
| 6 | from sys import argv, stderr |
8 | from sys import argv, stderr |
| 7 | from re import findall, DOTALL, IGNORECASE, match, sub, compile, \ |
9 | from re import findall, DOTALL, IGNORECASE, match, sub, compile, \ |
| 8 | split |
10 | split |
| 9 | from os import chdir |
11 | from os import chdir, stat |
| 10 | from os.path import dirname, realpath |
12 | from os.path import dirname, realpath, basename |
| 11 | from collections import OrderedDict |
13 | from collections import OrderedDict |
| 12 | from functools import cmp_to_key |
14 | from functools import cmp_to_key |
| 13 | from copy import deepcopy |
15 | from copy import deepcopy |
| 14 | from collections.abc import MutableSequence |
16 | from collections.abc import MutableSequence |
| - | 17 | from pickle import dump, load |
|
| 15 | 18 | ||
| - | 19 | debug_level = 2 |
|
| 16 | dictionary = {} |
20 | dictionary = {} |
| 17 | 21 | ||
| 18 | prepositions = { |
22 | prepositions = { |
| 19 | "fi'": 'on', |
23 | "fi'": 'on', |
| 20 | "na'": 'at|to', |
24 | "na'": 'at|to', |
| 21 | "t'": 'of' |
25 | "t'": 'of' |
| 22 | }
|
26 | }
|
| 23 | 27 | ||
| - | 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 | ||
| 24 | class MutableString2(MutableSequence): |
39 | class MutableString2(MutableSequence): |
| 25 | def __init__(self, value=None): |
40 | def __init__(self, value=None): |
| 26 | self._values = [str(value)] if value is not None else [] |
41 | self._values = [str(value)] if value is not None else [] |
| 27 | 42 | ||
| 28 | def __add__(self, value): |
43 | def __add__(self, value): |
| 29 | return ''.join([self, value]) |
44 | return ''.join([self, value]) |
| 30 | 45 | ||
| 31 | def __delitem__(self): |
46 | def __delitem__(self): |
| 32 | raise NotImplementedError |
47 | raise NotImplementedError |
| 33 | 48 | ||
| 34 | def __getitem__(self, index): |
49 | def __getitem__(self, index): |
| 35 | return str(self)[index] |
50 | return str(self)[index] |
| 36 | 51 | ||
| 37 | def __len__(self): |
52 | def __len__(self): |
| 38 | return len(str(self)) |
53 | return len(str(self)) |
| 39 | 54 | ||
| 40 | def __repr__(self): |
55 | def __repr__(self): |
| 41 | return ''.join(self._values) |
56 | return ''.join(self._values) |
| 42 | 57 | ||
| 43 | def __setitem__(self, index, value): |
58 | def __setitem__(self, index, value): |
| 44 | raise NotImplementedError |
59 | raise NotImplementedError |
| 45 | 60 | ||
| 46 | def __str__(self): |
61 | def __str__(self): |
| 47 | return self.__repr__() |
62 | return self.__repr__() |
| 48 | 63 | ||
| 49 | def extend(self, values): |
64 | def extend(self, values): |
| 50 | self._values.append(values) |
65 | self._values.append(values) |
| 51 | 66 | ||
| 52 | def insert(self, index, value): |
67 | def insert(self, index, value): |
| 53 | raise NotImplementedError |
68 | raise NotImplementedError |
| 54 | 69 | ||
| - | 70 | def cli_help(): |
|
| - | 71 | print('Usage: {0} TEXT...'.format(basename(argv[0]))) |
|
| - | 72 | ||
| 55 | def load_dictionary(dictionary, dictionary_file): |
73 | def load_dictionary(dictionary, dictionary_file): |
| 56 | print('Loading dictionary {0} ...'.format(dictionary_file), end='', file=stderr) |
74 | dmsg('Loading dictionary '.format(dictionary_file), end='', min_level=1) |
| 57 | 75 | ||
| 58 | chdir(dirname(realpath(__file__))) |
76 | chdir(dirname(realpath(__file__))) |
| 59 | with open(dictionary_file) as f: |
- | |
| 60 | keys = "ipa|en|lit|pos|com|tag|ex" |
- | |
| 61 | indent = None |
- | |
| 62 | value = None |
- | |
| 63 | 77 | ||
| - | 78 | pickle_file = basename(dictionary_file) + '.pickle' |
|
| - | 79 | ||
| 64 | for line in f: |
80 | try:
|
| 65 | m = match(r'^\s*vuh:\s*(?P<phrase>.+)', line) |
81 | pickle_mtime = stat(pickle_file).st_mtime |
| - | 82 | except FileNotFoundError:
|
|
| 66 | if m is not None: |
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) |
|
| 67 | phrase = m.group("phrase") |
87 | with open(dictionary_file) as f: |
| 68 | dictionary[phrase] = {} |
88 | keys = "ipa|en|lit|pos|com|tag|ex" |
| 69 | indent = None |
89 | indent = None |
| 70 | else:
|
90 | value = None |
| - | 91 | ||
| 71 | m = match( |
92 | for line in f: |
| 72 | r'(?P<indent>\s*)(?P<key>{0}):\s*(?P<value>.+)'.format(keys), |
93 | m = match(r'^\s*vuh:\s*(?P<phrase>.+)', line) |
| 73 | line)
|
- | |
| 74 | if m is not None: |
94 | if m is not None: |
| 75 | indent = m.group("indent") |
95 | phrase = m.group("phrase") |
| 76 | key = m.group("key") |
96 | dictionary[phrase] = {} |
| 77 | value = m.group("value") |
97 | indent = None |
| 78 | value = dictionary[phrase][key] = MutableString2(value) |
- | |
| 79 | elif indent is not None: |
98 | else:
|
| 80 | m = match(r'(?P<indent>\s+)(?P<continuation>\S.*)', line) |
99 | m = match(r'(?P<indent>\s*)(?P<key>{0}):\s*(?P<value>.+)'.format(keys), line) |
| 81 | if m is not None: |
100 | if m is not None: |
| - | 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: |
|
| 82 | if len(m.group("indent")) == len(indent) + 2: |
108 | if len(m.group("indent")) == len(indent) + 2: |
| 83 | dictionary[phrase][key] += (" " + m.group("continuation")) |
109 | dictionary[phrase][key] += (" " + m.group("continuation")) |
| - | 110 | ||
| - | 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 |
|
| 84 | 120 | ||
| 85 | print(' done ({0} entries).'.format(len(dictionary)), file=stderr) |
121 | dmsg(' done ({0} entries).'.format(len(dictionary)), min_level=1) |
| 86 | 122 | ||
| 87 | def clean_dictionary(dictionary): |
123 | def clean_dictionary(dictionary): |
| 88 | braces_re = compile(r'^\s*\{(.+)\}\s*$') |
124 | braces_re = compile(r'^\s*\{(.+)\}\s*$') |
| 89 | semicolon_re = compile(r'\s*;\s*') |
125 | semicolon_re = compile(r'\s*;\s*') |
| 90 | 126 | ||
| 91 | for orig_phrase, data in list(dictionary.items()): |
127 | for orig_phrase, data in list(dictionary.items()): |
| 92 | if orig_phrase.find(";") > -1: |
128 | if orig_phrase.find(";") > -1: |
| 93 | synonyms = map( |
129 | synonyms = map( |
| 94 | lambda x: sub(braces_re, r'\1', orig_phrase), |
130 | lambda x: sub(braces_re, r'\1', orig_phrase), |
| 95 | split(semicolon_re, orig_phrase)) |
131 | split(semicolon_re, orig_phrase)) |
| 96 | 132 | ||
| 97 | for synonym in synonyms: |
133 | for synonym in synonyms: |
| 98 | dictionary[synonym] = deepcopy(data) |
134 | dictionary[synonym] = deepcopy(data) |
| 99 | 135 | ||
| 100 | del dictionary[orig_phrase] |
136 | del dictionary[orig_phrase] |
| 101 | else:
|
137 | else:
|
| 102 | m = match(braces_re, orig_phrase) |
138 | m = match(braces_re, orig_phrase) |
| 103 | if m is not None: |
139 | if m is not None: |
| 104 | dictionary[m.group(1)] = deepcopy(dictionary[orig_phrase]) |
140 | dictionary[m.group(1)] = deepcopy(dictionary[orig_phrase]) |
| 105 | del dictionary[orig_phrase] |
141 | del dictionary[orig_phrase] |
| 106 | 142 | ||
| 107 | def sort_dict_alnum_english_key(phrase): |
143 | def sort_dict_alnum_english_key(phrase): |
| 108 | return sub(r'\{(.+)\}', r'\1', phrase[0]).lower() |
144 | return sub(r'\{(.+)\}', r'\1', phrase[0]).lower() |
| 109 | 145 | ||
| 110 | def get_sort_dict_alnum_vulcan_key(): |
146 | def get_sort_dict_alnum_vulcan_key(): |
| 111 | letters = list(map(str.lower, [ |
147 | letters = list(map(str.lower, [ |
| 112 | " ", 'S', 'T', 'P', 'K', 'R', 'L', 'A', 'Sh', 'O', 'U', 'D', |
148 | " ", 'S', 'T', 'P', 'K', 'R', 'L', 'A', 'Sh', 'O', 'U', 'D', |
| 113 | 'V', 'Kh', 'E', 'H', 'G', 'Ch', 'I', 'N', 'Zh', 'M', 'Y', 'F', 'Z', |
149 | 'V', 'Kh', 'E', 'H', 'G', 'Ch', 'I', 'N', 'Zh', 'M', 'Y', 'F', 'Z', |
| 114 | 'Th', 'W', 'B', "'", '-'])) |
150 | 'Th', 'W', 'B', "'", '-'])) |
| 115 | letter_values = dict(map(lambda x: (x[1], x[0]), enumerate(letters))) |
151 | letter_values = dict(map(lambda x: (x[1], x[0]), enumerate(letters))) |
| 116 | letters_re = compile(r'(?:{0})'.format('|'.join(sorted(letters, key=lambda char:-len(char))))) |
152 | letters_re = compile(r'(?:{0})'.format('|'.join(sorted(letters, key=lambda char:-len(char))))) |
| 117 | 153 | ||
| 118 | def sort_dict_alnum_vulcan (a, b): |
154 | def sort_dict_alnum_vulcan (a, b): |
| 119 | # split into Vulcan letters
|
155 | # split into Vulcan letters
|
| 120 | a = findall(letters_re, sort_dict_alnum_english_key(a)) |
156 | a = findall(letters_re, sort_dict_alnum_english_key(a)) |
| 121 | b = findall(letters_re, sort_dict_alnum_english_key(b)) |
157 | b = findall(letters_re, sort_dict_alnum_english_key(b)) |
| 122 | 158 | ||
| 123 | if len(a) < len(b): |
159 | if len(a) < len(b): |
| 124 | for index, char in enumerate(a): |
160 | for index, char in enumerate(a): |
| 125 | diff = letter_values[char] - letter_values[b[index]] |
161 | diff = letter_values[char] - letter_values[b[index]] |
| 126 | if diff != 0: |
162 | if diff != 0: |
| 127 | return diff
|
163 | return diff
|
| 128 | return -1 |
164 | return -1 |
| 129 | 165 | ||
| 130 | # len(b) <= len(a)
|
166 | # len(b) <= len(a)
|
| 131 | for index, char in enumerate(b): |
167 | for index, char in enumerate(b): |
| 132 | diff = letter_values[a[index]] - letter_values[char] |
168 | diff = letter_values[a[index]] - letter_values[char] |
| 133 | if diff != 0: |
169 | if diff != 0: |
| 134 | return diff
|
170 | return diff
|
| 135 | 171 | ||
| 136 | return 1 if len(b) < len(a) else 0 |
172 | return 1 if len(b) < len(a) else 0 |
| 137 | 173 | ||
| 138 | return cmp_to_key(sort_dict_alnum_vulcan) |
174 | return cmp_to_key(sort_dict_alnum_vulcan) |
| 139 | 175 | ||
| 140 | def translate (word, recursion=False): |
176 | def translate (phrase): |
| 141 | translation = dictionary.get(word.lower(), None) |
177 | translation = dictionary.get(phrase.lower(), None) |
| 142 | if translation is not None: |
178 | if translation is not None: |
| 143 | translation = translation["en"] |
- | |
| 144 | if match('[A-Z]', word): |
- | |
| 145 | return sub('[a-z]', lambda ch: ch.group(0).upper(), str(translation), count=1) |
- | |
| 146 | return translation
|
179 | return translation
|
| 147 | 180 | ||
| 148 | if not recursion: |
- | |
| 149 | # prepositions attached?
|
- | |
| 150 | for prep, prep_transl in prepositions.items(): |
- | |
| 151 | if (match(prep, word)): |
- | |
| 152 | real_word = word.replace(r'^' + prep, '') |
- | |
| 153 | real_word_transl = translate(real_word, recursion=True) |
- | |
| 154 | if real_word_transl is not None: |
- | |
| 155 | return prep_transl + ' ' + real_word_transl |
- | |
| 156 | - | ||
| 157 | if recursion:
|
- | |
| 158 | return None |
181 | return None |
| 159 | else:
|
- | |
| 160 | # Not in dictionary: proper name or missing for other reasons
|
- | |
| 161 | return '{{{0}}}'.format(word) |
- | |
| 162 | 182 | ||
| 163 | if __name__ == '__main__': |
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) |
|
| - | 188 | ||
| - | 189 | text = argv[1] |
|
| - | 190 | ||
| 164 | load_dictionary(dictionary, 'vuh-gol-en.dict.zdb.txt') |
191 | load_dictionary(dictionary, 'vuh-gol-en.dict.zdb.txt') |
| 165 | clean_dictionary(dictionary) |
192 | clean_dictionary(dictionary) |
| 166 | 193 | ||
| 167 | # try:
|
194 | # try:
|
| 168 | # for phrase, data in OrderedDict(sorted(
|
195 | # for phrase, data in OrderedDict(sorted(
|
| 169 | # dictionary.items(),
|
196 | # dictionary.items(),
|
| 170 | # key=get_sort_dict_alnum_vulcan_key()
|
197 | # key=get_sort_dict_alnum_vulcan_key()
|
| 171 | # )).items():
|
198 | # )).items():
|
| 172 | # print(phrase, "=", data)
|
199 | # print(phrase, "=", data)
|
| 173 | # except BrokenPipeError:
|
200 | # except BrokenPipeError:
|
| 174 | # pass
|
201 | # pass
|
| 175 | 202 | ||
| 176 | text = argv[1] |
- | |
| 177 | sentences = findall(r'(?!\s+)(?:.+?\.{1,3}|.+$)', text, DOTALL) |
203 | sentences = findall(r'(?!\s+)(?:.+?\.{1,3}|.+$)', text, DOTALL) |
| - | 204 | dmsg("sentences:", sentences, min_level=2) |
|
| 178 | for sentence in sentences: |
205 | for sentence in sentences: |
| - | 206 | dmsg("sentence:", sentence, min_level=2) |
|
| - | 207 | ||
| - | 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) |
|
| - | 212 | ||
| - | 213 | words = findall(r'[^\s.]+', clause) |
|
| - | 214 | dmsg("words:", words, min_level=2) |
|
| - | 215 | ||
| 179 | print(sentence) |
216 | offset = 0 |
| - | 217 | while offset < len(words): |
|
| - | 218 | translation = None |
|
| - | 219 | ||
| - | 220 | for i in reversed(range(offset + 1, len(words) + 1)): |
|
| - | 221 | phrase = ' '.join(words[offset:i]) |
|
| 180 | 222 | ||
| 181 | words = findall(r"(?!\s+)[a-z'-]{2,}", sentence, IGNORECASE) |
223 | dmsg("phrase:", phrase, min_level=2) |
| 182 | print(words) |
- | |
| 183 | 224 | ||
| 184 | translated_words = list(map(translate, words)) |
225 | translation = translate(phrase) |
| 185 | print(translated_words) |
- | |
| 186 | 226 | ||
| 187 | for index, word in enumerate(words): |
227 | if translation is not None: |
| 188 | sentence = sentence.replace(word, str(translated_words[index])) |
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 |
|
| 189 | print(sentence) |
232 | break
|
| 190 | 233 | ||
| 191 | # replace punctuation
|
234 | if translation is None: |
| 192 | for symbol, replacement in ({" - ": ", "}).items(): |
235 | dmsg("phrase-translation:", translation, min_level=2) |
| 193 | sentence = sentence.replace(symbol, replacement) |
236 | offset += 1 |
| 194 | 237 | ||
| 195 | print(sentence) |
238 | dmsg("words-translation:", words, min_level=2) |
| 196 | 239 | ||