new-words

annotate new-words.py @ 68:846240941452

find changesets by author, revision, files, or words in the commit message
added -C key: compress to lines; fixed bug with #90-line
author Igor Chubin <igor@chub.in>
date Sun Sep 23 16:07:29 2012 +0300 (2012-09-23)
parents 5a003076eb11
children
rev   line source
igor@37 1 #!/usr/bin/env python
igor@38 2 # -*- coding: utf-8 -*-
igor@37 3
igor@40 4 from __future__ import with_statement
igor@38 5 import codecs
igor@49 6 import difflib
igor@38 7 import logging
igor@38 8 import os
igor@37 9 import optparse
igor@38 10 import re
igor@38 11 import subprocess
igor@38 12 import sys
igor@38 13 import Stemmer
igor@54 14 import tempfile
igor@42 15 try:
igor@42 16 import psyco
igor@42 17 psyco.full()
igor@42 18 except:
igor@42 19 pass
igor@38 20
igor@38 21 config = {
igor@38 22 'config_directory': os.environ['HOME'] + '/.new-words',
igor@38 23 'language': 'en',
igor@38 24 }
igor@38 25
igor@38 26 logging.basicConfig(filename='/tmp/new-words-py.log', level=logging.DEBUG)
igor@38 27
igor@38 28 class Normalizator:
igor@38 29 def __init__(self, language, linked_words={}):
igor@38 30 stemmer_algorithm = {
igor@38 31 'de' : 'german',
igor@63 32 'fr' : 'french',
igor@38 33 'en' : 'english',
igor@51 34 'es' : 'spanish',
igor@38 35 'ru' : 'russian',
igor@51 36 'it' : 'italian',
igor@38 37 'uk' : 'ukrainian',
igor@38 38 }
igor@65 39 try:
igor@65 40 self.stemmer = Stemmer.Stemmer(stemmer_algorithm[language])
igor@65 41 except:
igor@65 42 self.stemmer = None
igor@38 43 self.linked_words = linked_words
igor@38 44
igor@38 45 def normalize(self, word):
igor@38 46 word_chain = []
igor@38 47 while word in self.linked_words and not word in word_chain:
igor@38 48 word_chain.append(word)
igor@38 49 word = self.linked_words[word]
igor@65 50 if self.stemmer:
igor@65 51 return self.stemmer.stemWord(word.lower())
igor@65 52 else:
igor@65 53 return word.lower()
igor@37 54
igor@47 55 def best_word_from_group(self, wordpairs_group):
igor@47 56 """Returns the word that is the most relevant to the wordpairs_group.
igor@47 57
igor@47 58 At the moment: returns the word with minimal length"""
igor@49 59
igor@49 60 def f(x, y):
igor@49 61 return difflib.SequenceMatcher(
igor@49 62 None,
igor@49 63 #(x[-2:] == 'en' and x[:-2].lower() or x.lower()),
igor@49 64 x.lower(),
igor@49 65 y.lower()).ratio()
igor@47 66
igor@47 67 minimal_length = min(len(pair[1]) for pair in wordpairs_group)
igor@49 68 best_match = list(x[1] for x in sorted(
igor@47 69 (x for x in wordpairs_group if len(x[1]) == minimal_length),
igor@47 70 key=lambda x:x[0],
igor@47 71 reverse=True))[0]
igor@47 72
igor@51 73 return best_match
igor@51 74
igor@49 75 suggestions = self.dictionary_suggestions(best_match)
igor@49 76 if len(suggestions) == 1:
igor@49 77 return best_match
igor@49 78
igor@49 79 verb = False
igor@49 80 corrected_best_match = best_match
igor@49 81 if best_match[-2:] == 'et':
igor@49 82 word = best_match[:-1]+"n"
igor@49 83 sugg = self.dictionary_suggestions(word)
igor@49 84 if len(sugg) == 1:
igor@49 85 return word
igor@49 86 suggestions += sugg
igor@49 87 corrected_best_match = word
igor@49 88 corrected_best_match = best_match[:-2]
igor@49 89 verb = True
igor@49 90
igor@49 91 if best_match[-1] == 't':
igor@49 92 word = best_match[:-1]+"en"
igor@49 93 sugg = self.dictionary_suggestions(word)
igor@49 94 if len(sugg) == 1:
igor@49 95 return word
igor@49 96 suggestions += sugg
igor@49 97 corrected_best_match = best_match[:-1]
igor@49 98 verb = True
igor@49 99
igor@49 100 if corrected_best_match[0].lower() == corrected_best_match[0]:
igor@49 101 suggestions = [ x for x in suggestions
igor@49 102 if x[0].lower() == x[0] ]
igor@49 103
igor@49 104 if suggestions == []:
igor@49 105 return best_match+"_"
igor@49 106 return best_match+" "+(" ".join(
igor@49 107 sorted(
igor@49 108 suggestions,
igor@49 109 key = lambda x: f(x, corrected_best_match),
igor@49 110 reverse = True
igor@49 111 )
igor@49 112 )
igor@49 113 )
igor@49 114
igor@49 115 def dictionary_suggestions(self, word):
igor@49 116 return [
igor@49 117 x.decode('utf-8').rstrip('\n')
igor@49 118 for x
igor@49 119 in subprocess.Popen(
igor@49 120 ["de-variants", word],
igor@49 121 stdout=subprocess.PIPE
igor@49 122 ).stdout.readlines() ]
igor@49 123
igor@49 124
igor@37 125 parser = optparse.OptionParser()
igor@37 126
igor@37 127 parser.add_option(
igor@37 128 "-a", "--no-marks",
igor@55 129 help="don't add marks (and don't save marks added by user) [NOT IMPLEMENTED YET]",
igor@37 130 action="store_true",
igor@37 131 dest="no_marks")
igor@37 132
igor@37 133 parser.add_option(
igor@37 134 "-c", "--compressed",
igor@37 135 help="show compressed wordlist: one word per group",
igor@37 136 action="store_true",
igor@37 137 dest="compressed")
igor@37 138
igor@37 139 parser.add_option(
igor@68 140 "-C", "--compressed-to-line",
igor@68 141 help="show compressed wordlist: all words of the group in a line",
igor@68 142 action="store_true",
igor@68 143 dest="compressed_to_line")
igor@68 144
igor@68 145 parser.add_option(
igor@37 146 "-k", "--known-words",
igor@37 147 help="put higher words that are similar to the known words (only for English)",
igor@37 148 action="store_true",
igor@37 149 dest="compressed")
igor@37 150
igor@37 151 parser.add_option(
igor@37 152 "-l", "--language",
igor@37 153 help="specify language of text",
igor@37 154 action="store",
igor@37 155 dest="language")
igor@37 156
igor@37 157 parser.add_option(
igor@54 158 "-f", "--allowed-words",
igor@54 159 help="file with list of allowed words (words that will be shown in the output)",
igor@54 160 action="store",
igor@54 161 dest="allowed_words")
igor@54 162
igor@54 163 parser.add_option(
igor@55 164 "-G", "--words-grouping",
igor@55 165 help="turn off word grouping",
igor@55 166 action="store_true",
igor@55 167 dest="no_words_grouping")
igor@55 168
igor@55 169 parser.add_option(
igor@54 170 "-X", "--function",
igor@38 171 help="filter through subsystem [INTERNAL]",
igor@38 172 action="store",
igor@38 173 dest="function")
igor@38 174
igor@38 175 parser.add_option(
igor@37 176 "-m", "--merge-tag",
igor@55 177 help="merge words tagged with specified tag into the main vocabulary [NOT IMPLEMENTED YET]",
igor@37 178 action="store",
igor@37 179 dest="merge_tag")
igor@37 180
igor@37 181 parser.add_option(
igor@37 182 "-M", "--merge-tagged",
igor@55 183 help="merge words tagged with ANY tag into the main vocabulary [NOT IMPLEMENTED YET]",
igor@37 184 action="store_true",
igor@37 185 dest="merge_tagged")
igor@37 186
igor@37 187 parser.add_option(
igor@37 188 "-n", "--non-interactive",
igor@37 189 help="non-interactive mode (don't run vi)",
igor@37 190 action="store_true",
igor@37 191 dest="non_interactive")
igor@37 192
igor@37 193 parser.add_option(
igor@37 194 "-N", "--no-filter",
igor@37 195 help="switch off known words filtering",
igor@37 196 action="store_true",
igor@37 197 dest="no_filter")
igor@37 198
igor@37 199 parser.add_option(
igor@37 200 "-p", "--pages",
igor@37 201 help="work with specified pages only (pages = start-stop/total )",
igor@37 202 action="store",
igor@37 203 dest="pages")
igor@37 204
igor@37 205 parser.add_option(
igor@48 206 "-d", "--delete-tag",
igor@48 207 help="delete subvocabulary of specified tag",
igor@37 208 action="store",
igor@48 209 dest="delete_tag")
igor@37 210
igor@37 211 parser.add_option(
igor@55 212 "-r", "--show-range",
igor@55 213 help="show only words specified number of words",
igor@55 214 action="store",
igor@55 215 dest="show_range")
igor@55 216
igor@55 217 parser.add_option(
igor@54 218 "-R", "--show-range-percentage",
igor@54 219 help="show only words that cover specified percentage of the text, skip the rest",
igor@54 220 action="store",
igor@54 221 dest="show_range_percentage")
igor@54 222
igor@54 223 parser.add_option(
igor@37 224 "-s", "--text-stats",
igor@37 225 help="show the text statistics (percentage of known words and so on) and exit",
igor@37 226 action="store_true",
igor@37 227 dest="text_stats")
igor@37 228
igor@37 229 parser.add_option(
igor@37 230 "-S", "--voc-stats",
igor@55 231 help="show your vocabulary statistics (number of words and word groups) [NOT IMPLEMENTED YET]",
igor@37 232 action="store_true",
igor@37 233 dest="voc_stats")
igor@37 234
igor@37 235 parser.add_option(
igor@37 236 "-t", "--tag",
igor@37 237 help="tag known words with tag",
igor@37 238 action="store",
igor@37 239 dest="tag")
igor@37 240
igor@37 241 parser.add_option(
igor@37 242 "-T", "--show-tags",
igor@37 243 help="tag known words with tag",
igor@37 244 action="store_true",
igor@37 245 dest="show_tags")
igor@37 246
igor@37 247 parser.add_option(
igor@63 248 "-v", "--vocabulary-filename",
igor@63 249 help="use specified file as a vocabulary",
igor@63 250 action="store",
igor@63 251 dest="vocabulary_filename")
igor@63 252
igor@63 253 parser.add_option(
igor@65 254 "-w", "--web",
igor@65 255 help="Web browser version",
igor@65 256 action="store_true",
igor@65 257 dest="web")
igor@65 258
igor@65 259 parser.add_option(
igor@37 260 "-2", "--two-words",
igor@37 261 help="find 2 words' sequences",
igor@37 262 action="store_true",
igor@37 263 dest="two_words")
igor@37 264
igor@37 265 parser.add_option(
igor@37 266 "-3", "--three-words",
igor@37 267 help="find 3 words' sequences",
igor@37 268 action="store_true",
igor@37 269 dest="three_words")
igor@37 270
igor@38 271 def readlines_from_file(filename):
igor@38 272 res = []
igor@38 273 with codecs.open(filename, "r", "utf-8") as f:
igor@38 274 for line in f.readlines():
igor@38 275 res += [line]
igor@38 276 return res
igor@38 277
igor@54 278 def readlines_from_url(url):
igor@54 279 return [x.decode('utf-8') for x in
igor@54 280 subprocess.Popen(
igor@54 281 "lynx -dump '{url}' | perl -p -e 's@http://[a-zA-Z&_.:/0-9%?=,#+()\[\]~-]*@@'".format(url=url),
igor@54 282 shell = True,
igor@54 283 stdout = subprocess.PIPE,
igor@54 284 stderr = subprocess.STDOUT
igor@54 285 ).communicate()[0].split('\n')
igor@54 286 ]
igor@54 287
igor@38 288 def readlines_from_stdin():
igor@38 289 return codecs.getreader("utf-8")(sys.stdin).readlines()
igor@38 290
igor@38 291 def words_from_line(line):
igor@38 292 line = line.rstrip('\n')
igor@38 293 #return re.split('(?:\s|[*\r,.:#@()+=<>$;"?!|\[\]^%&~{}«»–])+', line)
igor@38 294 #return re.split('[^a-zA-ZäöëüßÄËÖÜß]+', line)
igor@44 295 return re.compile("(?!['_])(?:\W)+", flags=re.UNICODE).split(line)
igor@38 296
igor@44 297 def get_words(lines, group_by=[1]):
igor@38 298 """
igor@38 299 Returns hash of words in a file
igor@38 300 word => number
igor@38 301 """
igor@38 302 result = {}
igor@44 303 (a, b, c) = ("", "", "")
igor@38 304 for line in lines:
igor@38 305 words = words_from_line(line)
igor@38 306 for word in words:
igor@41 307 if re.match('[0-9]*$', word):
igor@41 308 continue
igor@38 309 result.setdefault(word, 0)
igor@38 310 result[word] += 1
igor@44 311 if 2 in group_by and a != "" and b != "":
igor@44 312 w = "%s_%s" % (a,b)
igor@44 313 result.setdefault(w, 0)
igor@44 314 result[w] += 1
igor@44 315 if 3 in group_by and not "" in [a,b,c]:
igor@44 316 w = "%s_%s_%s" % (a,b,c)
igor@44 317 result.setdefault(w, 0)
igor@44 318 result[w] += 1
igor@44 319 (a,b,c) = (b, c, word)
igor@44 320
igor@44 321 logging.debug(result)
igor@38 322 return result
igor@38 323
igor@54 324 def voc_filename():
igor@63 325 if 'vocabulary_filename' in config:
igor@63 326 return config['vocabulary_filename']
igor@54 327 return "%s/%s.txt"%(config['config_directory'], config['language'])
igor@54 328
igor@38 329 def load_vocabulary():
igor@54 330 return get_words(readlines_from_file(voc_filename()))
igor@38 331
igor@38 332 def notes_filenames():
igor@38 333 return ["%s/notes-%s.txt"%(config['config_directory'], config['language'])]
igor@38 334
igor@38 335 def load_notes(files):
igor@38 336 notes = {}
igor@38 337 for filename in files:
igor@39 338 with codecs.open(filename, "r", "utf-8") as f:
igor@38 339 for line in f.readlines():
igor@38 340 (word, note) = re.split('\s+', line.rstrip('\n'), maxsplit=1)
igor@38 341 notes.setdefault(word, {})
igor@38 342 notes[word][filename] = note
igor@38 343 return notes
igor@38 344
igor@39 345 def add_notes(lines, notes):
igor@39 346 notes_filename = notes_filenames()[0]
igor@39 347 result = []
igor@39 348 for line in lines:
igor@39 349 if line.startswith('#'):
igor@39 350 result += [line]
igor@39 351 else:
igor@39 352 match_object = re.search('^\s*\S+\s*(\S+)', line)
igor@39 353 if match_object:
igor@39 354 word = match_object.group(1)
igor@39 355 if word in notes:
igor@39 356 if notes_filename in notes[word]:
igor@39 357 line = line.rstrip('\n')
igor@39 358 line = "%-30s %s\n" % (line, notes[word][notes_filename])
igor@39 359 result += [line]
igor@39 360 else:
igor@39 361 result += [line]
igor@39 362 else:
igor@39 363 result += [line]
igor@39 364 return result
igor@39 365
igor@39 366 def remove_notes(lines, notes_group):
igor@39 367 notes_filename = notes_filenames()[0]
igor@39 368 notes = {}
igor@39 369 for k in notes_group.keys():
igor@39 370 if notes_filename in notes_group[k]:
igor@39 371 notes[k] = notes_group[k][notes_filename]
igor@39 372
igor@39 373 result = []
igor@39 374 for line in lines:
igor@39 375 line = line.rstrip('\n')
igor@39 376 match_object = re.match('(\s+)(\S+)(\s+)(\S+)(\s+)(.*)', line)
igor@39 377 if match_object:
igor@39 378 result.append("".join([
igor@39 379 match_object.group(1),
igor@39 380 match_object.group(2),
igor@39 381 match_object.group(3),
igor@39 382 match_object.group(4),
igor@39 383 "\n"
igor@39 384 ]))
igor@39 385 notes[match_object.group(4)] = match_object.group(6)
igor@39 386 else:
igor@39 387 result.append(line+"\n")
igor@39 388
igor@39 389 save_notes(notes_filename, notes)
igor@39 390 return result
igor@39 391
igor@39 392 def save_notes(filename, notes):
igor@39 393 lines = []
igor@39 394 saved_words = []
igor@39 395 with codecs.open(filename, "r", "utf-8") as f:
igor@39 396 for line in f.readlines():
igor@39 397 (word, note) = re.split('\s+', line.rstrip('\n'), maxsplit=1)
igor@39 398 if word in notes:
igor@39 399 line = "%-29s %s\n" % (word, notes[word])
igor@39 400 saved_words.append(word)
igor@39 401 lines.append(line)
igor@39 402 for word in [x for x in notes.keys() if not x in saved_words]:
igor@39 403 line = "%-29s %s\n" % (word, notes[word])
igor@39 404 lines.append(line)
igor@39 405
igor@39 406 with codecs.open(filename, "w", "utf-8") as f:
igor@39 407 for line in lines:
igor@39 408 f.write(line)
igor@39 409
igor@39 410
igor@38 411 def substract_dictionary(dict1, dict2):
igor@38 412 """
igor@38 413 returns dict1 - dict2
igor@38 414 """
igor@38 415 result = {}
igor@38 416 for (k,v) in dict1.items():
igor@38 417 if not k in dict2:
igor@38 418 result[k] = v
igor@38 419 return result
igor@38 420
igor@38 421 def dump_words(words, filename):
igor@38 422 with codecs.open(filename, "w+", "utf-8") as f:
igor@38 423 for word in words.keys():
igor@38 424 f.write(("%s\n"%word)*words[word])
igor@38 425
igor@38 426 def error_message(text):
igor@38 427 print text
igor@38 428
igor@40 429 def find_wordgroups_weights(word_pairs, normalizator):
igor@38 430 weight = {}
igor@40 431 for (num, word) in word_pairs:
igor@38 432 normalized = normalizator.normalize(word)
igor@38 433 weight.setdefault(normalized, 0)
igor@40 434 weight[normalized] += num
igor@38 435 return weight
igor@38 436
igor@38 437 def find_linked_words(notes):
igor@38 438 linked_words = {}
igor@38 439 for word in notes.keys():
igor@38 440 for note in notes[word].values():
igor@38 441 if "@" in note:
igor@38 442 result = re.search(r'\@(\S*)', note)
igor@38 443 if result:
igor@38 444 main_word = result.group(1)
igor@38 445 if main_word:
igor@38 446 linked_words[word] = main_word
igor@38 447 return linked_words
igor@38 448
igor@40 449 def compare_word_pairs(pair1, pair2, wgw, normalizator, linked_words):
igor@40 450 (num1, word1) = pair1
igor@40 451 (num2, word2) = pair2
igor@38 452
igor@38 453 normalized_word1 = normalizator.normalize(word1)
igor@38 454 normalized_word2 = normalizator.normalize(word2)
igor@38 455
igor@38 456 cmp_res = cmp(wgw[normalized_word1], wgw[normalized_word2])
igor@38 457 if cmp_res != 0:
igor@38 458 return cmp_res
igor@38 459 else:
igor@38 460 cmp_res = cmp(normalized_word1, normalized_word2)
igor@38 461 if cmp_res != 0:
igor@38 462 return cmp_res
igor@38 463 else:
igor@38 464 return cmp(int(num1), int(num2))
igor@38 465
igor@47 466
igor@48 467 def print_words_sorted(
igor@48 468 word_pairs,
igor@48 469 stats,
igor@48 470 normalizator,
igor@48 471 print_stats=True,
igor@48 472 stats_only=False,
igor@48 473 compressed_wordlist=False,
igor@68 474 compressed_to_line = False,
igor@48 475 show_range=0,
igor@48 476 show_range_percentage=0,
igor@48 477 ):
igor@54 478 result = []
igor@40 479 if stats_only:
igor@54 480 #codecs.getwriter("utf-8")(sys.stdout).write(
igor@54 481 result.append(
igor@43 482 " ".join([
igor@43 483 "%-10s" % x for x in [
igor@43 484 "LANG",
igor@43 485 "KNOWN%",
igor@43 486 "UNKNOWN%",
igor@43 487 "KNOWN",
igor@43 488 "TOTAL",
igor@43 489 "WPS",
igor@43 490 "UWPS*10"
igor@43 491 ]]) + "\n")
igor@54 492 result.append(
igor@43 493 " ".join([
igor@43 494 "%(language)-10s",
igor@43 495 "%(percentage)-10.2f",
igor@43 496 "%(percentage_unknown)-10.2f",
igor@43 497 "%(total_known)-11d"
igor@43 498 "%(total)-11d"
igor@43 499 "%(wps)-11d"
igor@43 500 "%(uwps)-11d"
igor@43 501 ]) % stats + "\n")
igor@54 502 return "".join(result)
igor@38 503
igor@40 504 if print_stats:
igor@54 505 result.append(
igor@43 506 "# %(language)s, %(percentage)-7.2f, <%(total_known)s/%(total)s>, <%(groups)s/%(words)s>\n" % stats)
igor@38 507
igor@40 508 known = int(stats['total_known'])
igor@40 509 total = int(stats['total'])
igor@68 510 level_lines = range(int(float(stats['percentage']))/5*5+5,90,5)+range(90,102)
igor@68 511 if 100.0*known/total >= level_lines[0]:
igor@68 512 current_level = level_lines[0]
igor@68 513 while 100.0*known/total > level_lines[0]:
igor@68 514 current_level = level_lines[0]
igor@68 515 level_lines = level_lines[1:]
igor@47 516 old_normalized_word = None
igor@47 517 words_of_this_group = []
igor@48 518 printed_words = 0
igor@40 519 for word_pair in word_pairs:
igor@47 520
igor@47 521 normalized_word = normalizator.normalize(word_pair[1])
igor@47 522 if old_normalized_word and old_normalized_word != normalized_word:
igor@47 523 if compressed_wordlist:
igor@49 524 compressed_word_pair = (
igor@49 525 sum(x[0] for x in words_of_this_group),
igor@49 526 normalizator.best_word_from_group(words_of_this_group)
igor@49 527 )
igor@68 528 if compressed_to_line:
igor@68 529 result.append("%10s %s %s\n" % (compressed_word_pair + (" ".join(y for x,y in words_of_this_group if y not in compressed_word_pair),)))
igor@68 530 else:
igor@68 531 result.append("%10s %s\n" % compressed_word_pair)
igor@48 532 printed_words += 1
igor@47 533 words_of_this_group = []
igor@47 534
igor@47 535 old_normalized_word = normalized_word
igor@47 536 words_of_this_group.append(word_pair)
igor@47 537
igor@47 538 if not compressed_wordlist:
igor@54 539 result.append("%10s %s\n" % word_pair)
igor@48 540 printed_words += 1
igor@47 541
igor@47 542
igor@40 543 known += word_pair[0]
igor@40 544 if 100.0*known/total >= level_lines[0]:
igor@40 545 current_level = level_lines[0]
igor@40 546 while 100.0*known/total > level_lines[0]:
igor@40 547 current_level = level_lines[0]
igor@40 548 level_lines = level_lines[1:]
igor@54 549 result.append("# %s\n" % current_level)
igor@38 550
igor@48 551 if show_range >0 and printed_words >= show_range:
igor@48 552 break
igor@48 553 if show_range_percentage >0 and 100.0*known/total >= show_range_percentage:
igor@48 554 break
igor@48 555
igor@54 556 return result
igor@39 557
igor@53 558 def parse_parts_description(parts_description):
igor@53 559 """
igor@53 560 Returns triad (start, stop, step)
igor@53 561 basing on parts_description string.
igor@53 562 from-to/step
igor@53 563 from+delta/step
igor@53 564 """
igor@53 565
igor@53 566 try:
igor@53 567 (a, step) = parts_description.split("/", 1)
igor@53 568 step = int(step)
igor@53 569 start = 0
igor@53 570 stop = 0
igor@53 571 if '-' in a:
igor@53 572 (start, stop) = a.split("-", 1)
igor@53 573 start = int(start)
igor@53 574 stop = int(stop)
igor@53 575 elif '+' in a:
igor@53 576 (start, stop) = a.split("+", 1)
igor@53 577 start = int(start)
igor@53 578 stop = int(stop)
igor@53 579 else:
igor@53 580 start = int(a)
igor@53 581 stop = start + 1
igor@53 582 return (start, stop, step)
igor@53 583
igor@53 584 except:
igor@54 585 raise ValueError("Parts description must be in format: num[[+-]num]/num; this [%s] is incorrect" % parts_description)
igor@53 586
igor@53 587
igor@53 588 def take_part(lines, part_description = None):
igor@55 589 if part_description == None or part_description == '':
igor@53 590 return lines
igor@53 591 (start, stop, step) = parse_parts_description(part_description)
igor@53 592 n = len(lines)
igor@53 593 part_size = (1.0*n) / step
igor@53 594 result = []
igor@53 595 for i in range(n):
igor@54 596 if i >= start * part_size and i <= stop * part_size:
igor@54 597 result += [lines[i]]
igor@53 598 return result
igor@53 599
igor@65 600 def web_editor(output):
igor@65 601 from twisted.internet import reactor
igor@65 602 from twisted.web.server import Site
igor@65 603 from twisted.web.static import File
igor@65 604 from twisted.web.resource import Resource
igor@65 605 import json
igor@65 606
igor@65 607 word_list = []
igor@65 608
igor@65 609 for o in output:
igor@65 610 a = re.split('\s+', o.strip(), 2)
igor@65 611 a = a + ['']*(3-len(a))
igor@65 612 word_list.append({'number':a[0], 'word':a[1], 'comment':a[2]})
igor@65 613
igor@65 614 print "Loaded ", len(word_list)
igor@65 615
igor@65 616 new_words_html = "/home/igor/hg/new-words/web"
igor@65 617
igor@65 618 class JSONPage(Resource):
igor@65 619 isLeaf = True
igor@65 620 def render_GET(self, request):
igor@65 621 return json.dumps({"word_list": word_list})
igor@65 622
igor@65 623 class SaveJSON(Resource):
igor@65 624 isLeaf = True
igor@65 625 def render_POST(self, request):
igor@65 626 print json.loads(request.args["selected_words"][0])
igor@65 627 return json.dumps({"status": "ok"})
igor@65 628
igor@65 629 json_page = JSONPage()
igor@65 630 save_json = SaveJSON()
igor@65 631
igor@65 632 resource = File(new_words_html)
igor@65 633 resource.putChild("json", json_page)
igor@65 634 resource.putChild("save", save_json)
igor@65 635
igor@65 636 factory = Site(resource)
igor@65 637 reactor.listenTCP(8880, factory)
igor@65 638 reactor.run()
igor@65 639
igor@65 640
igor@40 641 def filter_get_words_group_words_add_stat(args):
igor@40 642 vocabulary = load_vocabulary()
igor@40 643 notes = load_notes(notes_filenames())
igor@54 644
igor@65 645 input_lines = []
igor@54 646 if len(args) > 0:
igor@65 647 for arg in args:
igor@65 648 if 'http://' in arg:
igor@65 649 input_lines += readlines_from_url(arg)
igor@65 650 else:
igor@65 651 input_lines += readlines_from_file(arg)
igor@54 652 else:
igor@65 653 input_lines += readlines_from_stdin()
igor@54 654
igor@54 655 if len(input_lines) == 0:
igor@54 656 print >> sys.stderr, "Nothing to do, standard input is empty, exiting."
igor@54 657 sys.exit(1)
igor@54 658
igor@54 659 lines = take_part(input_lines, config.get('pages', ''))
igor@54 660
igor@54 661 (_, original_text_tempfile) = tempfile.mkstemp(prefix='new-word')
igor@54 662 with codecs.open(original_text_tempfile, "w", "utf-8") as f:
igor@54 663 f.write("".join(lines))
igor@54 664
igor@44 665 group_by = [1]
igor@48 666
igor@54 667 if 'two_words' in config:
igor@44 668 group_by.append(2)
igor@54 669 if 'three_words' in config:
igor@44 670 group_by.append(3)
igor@44 671 words = get_words(lines, group_by)
igor@43 672 stats_only = False
igor@54 673 if 'text_stats' in config:
igor@43 674 stats_only = True
igor@40 675
igor@47 676 compressed_wordlist = False
igor@68 677 if 'compressed' in config or 'compressed_to_line' in config:
igor@47 678 compressed_wordlist = True
igor@47 679
igor@68 680 compressed_to_line = 'compressed_to_line' in config
igor@68 681
igor@55 682 if 'show_range' in config:
igor@55 683 show_range = int(config['show_range'])
igor@48 684 else:
igor@48 685 show_range = 0
igor@54 686
igor@54 687 if 'show_range_percentage' in config:
igor@54 688 show_range_percentage = int(config['show_range_percentage'])
igor@48 689 else:
igor@48 690 show_range_percentage = 0
igor@48 691
igor@44 692
igor@40 693 stats = {}
igor@40 694 stats['total'] = sum(words[x] for x in words.keys())
igor@54 695 if not 'no_filter' in config:
igor@45 696 words = substract_dictionary(words, vocabulary)
igor@40 697
igor@40 698 stats['total_unknown'] = sum(words[x] for x in words.keys())
igor@40 699 stats['total_known'] = stats['total'] - stats['total_unknown']
igor@43 700 stats['percentage'] = 100.0*stats['total_known']/stats['total']
igor@43 701 stats['percentage_unknown'] = 100.0-100.0*stats['total_known']/stats['total']
igor@40 702 stats['groups'] = 0
igor@40 703 stats['words'] = len(words)
igor@43 704 stats['sentences'] = 0 #FIXME
igor@43 705 stats['wps'] = 0 #FIXME
igor@43 706 stats['uwps'] = 0 #FIXME
igor@40 707 stats['language'] = config['language']
igor@40 708
igor@40 709 linked_words = find_linked_words(notes)
igor@40 710 normalizator = Normalizator(config['language'], linked_words)
igor@40 711
igor@50 712 # filter words by allowed_words_filter
igor@54 713 if 'allowed_words' in config:
igor@54 714 allowed_words_filename = config['allowed_words']
igor@50 715 normalized_allowed_words = [
igor@50 716 normalizator.normalize(w.rstrip('\n'))
igor@50 717 for w in readlines_from_file(allowed_words_filename)
igor@50 718 ]
igor@50 719
igor@50 720 result = {}
igor@50 721 for w, wn in words.iteritems():
igor@50 722 if normalizator.normalize(w) in normalized_allowed_words:
igor@50 723 result[w] = wn
igor@50 724 words = result
igor@50 725
igor@44 726 words_with_freq = []
igor@40 727 for k in sorted(words.keys(), key=lambda k: words[k], reverse=True):
igor@44 728 words_with_freq.append((words[k], k))
igor@40 729
igor@44 730 wgw = find_wordgroups_weights(words_with_freq, normalizator)
igor@55 731 if not 'no_words_grouping' in config or not config['no_words_grouping']:
igor@45 732 words_with_freq = sorted(
igor@44 733 words_with_freq,
igor@40 734 cmp=lambda x,y:compare_word_pairs(x,y, wgw, normalizator, linked_words),
igor@40 735 reverse=True)
igor@40 736
igor@54 737 output = print_words_sorted(
igor@47 738 words_with_freq,
igor@47 739 stats,
igor@47 740 normalizator,
igor@47 741 stats_only=stats_only,
igor@48 742 compressed_wordlist=compressed_wordlist,
igor@68 743 compressed_to_line=compressed_to_line,
igor@48 744 show_range=show_range,
igor@48 745 show_range_percentage=show_range_percentage,
igor@47 746 )
igor@40 747
igor@54 748
igor@54 749 if ('non_interactive' in config or 'text_stats' in config):
igor@54 750 codecs.getwriter("utf-8")(sys.stdout).write("".join(output))
igor@65 751 elif config.get('web', False):
igor@65 752 web_editor(output)
igor@54 753 else:
igor@54 754 (_, temp1) = tempfile.mkstemp(prefix='new-word')
igor@54 755 (_, temp2) = tempfile.mkstemp(prefix='new-word')
igor@54 756
igor@54 757 with codecs.open(temp1, "w", "utf-8") as f:
igor@54 758 f.write("".join(output))
igor@54 759 with codecs.open(temp2, "w", "utf-8") as f:
igor@54 760 f.write("".join(add_notes(output, notes)))
igor@54 761
igor@54 762 os.putenv('ORIGINAL_TEXT', original_text_tempfile)
igor@54 763 os.system((
igor@54 764 "vim"
igor@54 765 " -c 'setlocal spell spelllang={language}'"
igor@54 766 " -c 'set keywordprg={language}'"
igor@54 767 " -c 'set iskeyword=@,48-57,/,.,-,_,+,,,#,$,%,~,=,48-255'"
igor@54 768 " {filename}"
igor@54 769 " < /dev/tty > /dev/tty"
igor@54 770 ).format(language=config['language'], filename=temp2))
igor@54 771
igor@54 772 lines = remove_notes(readlines_from_file(temp2), notes)
igor@54 773
igor@54 774 # compare lines_before and lines_after and return deleted words
igor@54 775 lines_before = output
igor@54 776 lines_after = lines
igor@54 777 deleted_words = []
igor@54 778
igor@60 779 lines_after_set = set(lines_after)
igor@54 780 for line in lines_before:
igor@60 781 if line not in lines_after_set:
igor@54 782 line = line.strip()
igor@54 783 if ' ' in line:
igor@54 784 word = re.split('\s+', line, 1)[1]
igor@54 785 if ' ' in word:
igor@54 786 word = re.split('\s+', word, 1)[0]
igor@54 787 deleted_words.append(word)
igor@54 788
igor@54 789 with codecs.open(voc_filename(), "a", "utf-8") as f:
igor@54 790 f.write("\n".join(deleted_words + ['']))
igor@54 791
igor@54 792 os.unlink(temp1)
igor@54 793 os.unlink(temp2)
igor@54 794
igor@54 795 os.unlink(original_text_tempfile)
igor@54 796
igor@37 797 (options, args) = parser.parse_args()
igor@38 798 if options.language:
igor@38 799 config['language'] = options.language
igor@37 800
igor@54 801 if options.pages:
igor@54 802 config['pages'] = options.pages
igor@54 803 else:
igor@54 804 config['pages'] = ""
igor@54 805
igor@54 806 if options.allowed_words:
igor@54 807 config['allowed_words'] = options.allowed_words
igor@54 808
igor@55 809 if options.show_range:
igor@55 810 config['show_range'] = options.show_range
igor@55 811
igor@54 812 if options.show_range_percentage:
igor@54 813 config['show_range_percentage'] = options.show_range_percentage
igor@54 814
igor@54 815 if options.non_interactive:
igor@54 816 config['non_interactive'] = True
igor@54 817
igor@54 818 if options.text_stats:
igor@54 819 config['text_stats'] = True
igor@54 820
igor@54 821 if options.compressed:
igor@54 822 config['compressed'] = True
igor@54 823
igor@68 824 if options.compressed_to_line:
igor@68 825 config['compressed_to_line'] = True
igor@68 826
igor@54 827 if options.no_filter:
igor@54 828 config['no_filter'] = True
igor@54 829
igor@54 830 if options.two_words:
igor@54 831 config['two_words'] = True
igor@54 832
igor@54 833 if options.three_words:
igor@54 834 config['three_words'] = True
igor@54 835
igor@55 836 if options.no_words_grouping:
igor@55 837 config['no_words_grouping'] = True
igor@37 838
igor@65 839 if options.web:
igor@65 840 config['web'] = True
igor@65 841
igor@55 842 filter_get_words_group_words_add_stat(args)
igor@55 843
igor@55 844 #if options.function:
igor@55 845 # function_names = {
igor@55 846 # 'get_words_group_words_add_stat': ,
igor@55 847 # }
igor@55 848 # if options.function in function_names:
igor@55 849 # function_names[options.function](args)
igor@55 850 # else:
igor@55 851 # error_message("Unkown function %s.\nAvailable functions:\n%s" % (
igor@55 852 # options.function, "".join([" "+x for x in sorted(function_names.keys())])))
igor@55 853 # sys.exit(1)
igor@55 854 #
igor@37 855
igor@37 856
igor@37 857
igor@38 858 #os.system("vim")
igor@37 859