phraseology_generator.py 16.5 KB
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
# -*- coding: utf-8 -*-

from dictionary.models import sort_arguments, sort_positions, sortatributes
from settings import MORFEUSZ2
from copy import deepcopy

def lexicalisation(argument, subj, base, negativity, reference=None):
    b = argument.type
    if b == 'fixed':
        return (get_words(sortatributes(argument)[-1]), [], 1)
    attributes = sortatributes(argument)
    lexicalisation_type = attributes[0].values.all()[0].argument.type
    lexicalisation_parameters = sortatributes(attributes[0].values.all()[0].argument)
    if lexicalisation_type == 'xp': # xp(...)[np/prepnp], ...
       lexicalisation_type = lexicalisation_parameters[0].values.all()[0].argument.type
       lexicalisation_parameters = sortatributes(lexicalisation_parameters[0].values.all()[0].argument)
    if lexicalisation_type == 'np': # np(case), number, nouns, atr
        nps = get_nps(get_case(lexicalisation_parameters[0], subj, negativity), get_number(attributes[1], subj), get_words(attributes[2]), attributes[3])
        if subj:
            return (nps, get_verb(base, get_number(attributes[1], subj), subj), -1)
        else:
            return (nps, [], 1)
    elif lexicalisation_type == 'prepnp': #prepnp(prep, case), number, nouns, atr
        prepnps = get_prepnps(get_preposition(lexicalisation_parameters[0]), get_case(lexicalisation_parameters[1], subj, negativity), get_number(attributes[1], subj), get_words(attributes[2]), attributes[3])
        return (prepnps, [], 1)
    elif lexicalisation_type == 'adjp': # adjp(case), number, gender, degree, adjectives, atr
        adjps = get_adjps(get_case(lexicalisation_parameters[0], subj, negativity, reference), get_number(attributes[1], subj, reference), get_gender(attributes[2], reference), get_degree(attributes[3]), get_words(attributes[4]), attributes[5])
        if reference is None:
            if lexicalisation_parameters[0].values.all()[0].parameter.type.name == u'agr':
                if get_number(attributes[1], subj, reference) == u'pl':
                    return (adjps, get_plural(base), 0)
                else:
                    return (adjps, [], 0)
            else:
                return (adjps, [], -1)
        elif subj:
            return (adjps, get_verb(base, get_number(attributes[1], subj), subj), -1)
        else:
            return (adjps, [], 1)
    elif lexicalisation_type == 'prepadjp': #prepadjp(prep, case), number, gender, degree, adjectives, atr
        prepadjps = get_prepadjps(get_preposition(lexicalisation_parameters[0]), get_case(lexicalisation_parameters[1], subj, False, reference), get_number(attributes[1], subj, reference), get_gender(attributes[2], reference), get_degree(attributes[3]), get_words(attributes[4]), attributes[5])
        return (prepadjps, [], 1)
    elif lexicalisation_type == 'infp':
        infps = get_infps(get_aspect(lexicalisation_parameters[0]), get_words(attributes[2]), attributes[4])
        return (infps, [], 1)
    elif lexicalisation_type == 'advp': #advp(type), degree, adverb, atr
        advps = get_advps(get_degree(attributes[1]), get_words(attributes[2]), attributes[3])
        return (advps, [], -1)
    elif lexicalisation_type == 'nump': # nump(case), num, noun, atr
        numps = get_numps(get_case(lexicalisation_parameters[0], subj, negativity, reference), get_words(attributes[1]), get_words(attributes[2]), attributes[3])
        return (numps, get_verb(base, 'pl', subj), -1)
    elif lexicalisation_type == 'prepnump': # prepnump(prep,case), num, noun, atr
        prepnumps = get_prepnumps(get_preposition(lexicalisation_parameters[0]), get_case(lexicalisation_parameters[1], subj, False, reference), get_words(attributes[1]), get_words(attributes[2]), attributes[3])
        return (prepnumps, [], 1)
    elif lexicalisation_type == 'qub': # qub, form, atr
        qubs = get_qubs(get_words(attributes[1]), attributes[2])
        return (qubs, [], -1)
    else:
        return ([], [], 0)
    return ([], [], 0)

def is_subj(categories):
    for cat in categories:
        if cat.category == u'subj':
            return True
    return False

def get_preposition(attribute):
    return attribute.values.all()[0].parameter.type.name

def get_numerals(attribute):
    return get_words(attribute)

def get_words(attribute):
    words = [word.text[1:-1] for word in attribute.values.all()]
    return words

def get_aspect(attribute):
    return attribute.values.all()[0].parameter.type.name

def get_case(attribute, is_subj, negativity, reference=None):
    case = attribute.values.all()[0].parameter.type.name
    if case == u'str':
        if is_subj:
            case = [u'nom']
        elif negativity:
            case = [u'gen']
        else:
            case = [u'acc']
    elif case == u'part':
        case = [u'gen', u'acc']
    elif case == u'agr' and reference is not None:
        _, tag = reference
        base = tag.split(':')[0]
        if base == u'siebie':
            case = tag.split(':')[1].split('.')
        else:
            case = tag.split(':')[2].split('.')
    elif case == u'agr' and reference is None:
        case = [u'nom']
    else:
        case = [case]
    return case

def get_number(attribute, is_subj, reference=None):
    number = attribute.values.all()[0].parameter.type.name
    if number == u'_':
        if is_subj:
            number = u'sg'
    elif number == u'agr' and reference is not None:
        _, tag = reference
        base = tag.split(':')[0]
        if base == u'siebie':
            number = u'_'
        else:
            number = tag.split(':')[1]
    elif number == u'agr' and reference is None:
        number = u'sg'
    return number

def get_gender(attribute, reference=None):
    gender = attribute.values.all()[0].parameter.type.name
    if gender == u'_':
        gender = u'n'
    elif gender == u'm':
        gender = u'm1'
    elif gender == u'agr' and reference is not None:
        _, tag = reference
        base = tag.split(':')[0]
        if base == u'siebie':
            gender = u'_'
        else:
            gender = tag.split(':')[3]
    elif gender == u'agr' and reference is None:
        gender = 'm1'
    return gender

def get_degree(attribute):
    degree = attribute.values.all()[0].parameter.type.name
    if degree == u'_':
        degree = u'pos'
    return degree

def in_tag(what, tag):
    if u':' + what + u':' in tag or u':' + what + u'.' in tag or  u'.' + what + u'.' in tag or  u'.' + what + u':' in tag or tag.endswith(u'.' + what) or tag.endswith(u':' + what):
        return True
    else:
        return False

def get_nps(cases, number, nouns, atr):
    result = []
    for noun in nouns:
        options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(noun)]
        options_temp = []
        for case in cases:
            if case != u'_':
                filtered = []
                for option in options:
                    (orth, tag) = option
                    if in_tag(case, tag):
                        filtered.append(option)
                options_temp += filtered
            else:
                options_temp += filtered
        options = options_temp
        if number != u'_':
            filtered = []
            for option in options:
                (orth, tag) = option
                if in_tag(number, tag):
                    filtered.append(option)
            options = filtered
        result += options
    return dependents(atr, result)

def get_prepnps(prep, cases, number, nouns, _atr):
    nps = get_nps(cases, number, nouns, _atr)
    return [prep + ' ' + np for np in nps]

def get_infps(aspect, verbs, atr):
    result = []
    for verb in verbs:
        options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(verb)]
        filtered = []
        for option in options:
            (orth, tag) = option
            if u'inf:' in tag:
                filtered.append(option)
        options = filtered
        filtered = []
        if aspect != u'_':
            for option in options:
                (orth, tag) = option
                if in_tag(aspect, tag):
                    filtered.append(option)
            options = filtered
        result += options
    return dependents(atr, result)

def get_adjps(cases, number, gender, degree, adjectives, atr):
    result = []
    for adjective in adjectives:
        options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(adjective)]
        filtered = []
        for option in options:
            (orth, tag) = option
            if u'adj:' in tag:
                filtered.append(option)
        options = filtered
        options_temp = []
        for case in cases:
            if case != u'_':
                filtered = []
                for option in options:
                    (orth, tag) = option
                    if in_tag(case, tag):
                        filtered.append(option)
                options_temp += filtered
            else:
                options_temp += options
        options = options_temp
        if number != u'_':
            filtered = []
            for option in options:
                (orth, tag) = option
                if in_tag(number, tag):
                    filtered.append(option)
            options = filtered
        if gender != u'_':
            filtered = []
            for option in options:
                (orth, tag) = option
                if in_tag(gender, tag):
                    filtered.append(option)
            options = filtered
        if degree != u'_':
            filtered = []
            for option in options:
                (orth, tag) = option
                if in_tag(degree, tag):
                    filtered.append(option)
            options = filtered
        result += options
    return dependents(atr, result)

def get_prepadjps(prep, case, number, gender, degree, adjectives, _atr):
     adjps = get_adjps(case, number, gender, degree, adjectives, _atr)
     return [prep + ' ' + adjp for adjp in adjps]

def get_advps(degree, adverbs, atr):
    result = []
    for adverb in adverbs:
        options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(adverb)]
        filtered = []
        for option in options:
            (orth, tag) = option
            if tag.startswith(u'adv'):
                filtered.append(option)
        options = filtered
        if ':' in tag and degree != u'_':
            filtered = []
            for option in options:
                (orth, tag) = option
                if in_tag(degree, tag):
                    filtered.append(option)
            options = filtered
        result += options
    return dependents(atr, result)

def get_qubs(qubs, atr):
    result = []
    for qub in qubs:
        options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(qub)]
        filtered = []
        for option in options:
            (orth, tag) = option
            if tag.startswith(u'qub'):
                filtered.append(option)
        options = filtered
        result += options
    return dependents(atr, result)

def get_numps(cases, numerals, nouns, atr):
    results = []
    nums = []
    for numeral in numerals:
        options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(numeral)]
        filtered = []
        for option in options:
            (orth, tag) = option
            if tag.startswith(u'num:'):
                filtered.append(option)
        options = filtered
        options_temp = []
        for case in cases:
            if case != u'_':
                filtered = []
                for option in options:
                    (orth, tag) = option
                    if in_tag(case, tag):
                        filtered.append(option)
                options_temp += filtered
            else:
                options_temp += options
        nums = options_temp
        if len(nums) == 0:
            continue
        for (num_orth, num_tag) in nums:
            rec = num_tag.split(':')[4]
            for noun in nouns:
                options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(noun) if 'subst:' in tag]
                filtered = []
                for option in options:
                    (orth, tag) = option
                    if in_tag(u'pl', tag):
                        filtered.append(option)
                options = filtered
                if rec == 'rec':
                    c = ['gen']
                elif rec == 'congr':
                    c = cases
                else:
                    c = list(set(['gen'] + cases))
                options_temp = []
                for case in c:
                    if case != u'_':
                        filtered = []
                        for option in options:
                            (orth, tag) = option
                            if in_tag(case, tag):
                                filtered.append(option)
                        options_temp += filtered
                    else:
                        options_temp += options
                options = options_temp
                for (orth, tag) in options:
                    gender = tag.split(':')[3]
                    if in_tag(gender, num_tag):
                        results.append(num_orth + ' ' + orth)

    return results #ignoring ambiguos atr for numps

def get_prepnumps(prep, cases, numerals, nouns, atr):
    numps = get_numps(cases, numerals, nouns, atr)
    return [prep + ' ' + nump for nump in numps]

 
def get_verb(inf, number, is_subj):
    if not is_subj:
        return None
    else:
        options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(inf)]
        filtered = []
        for option in options:
            (orth, tag) = option
            if tag.startswith(u'fin') and in_tag(number, tag) and in_tag(u'ter', tag):
                filtered.append(option)
        options = filtered
        return [orth for orth, _ in options]

def get_plural(noun):
    options = [(orth, tag) for (orth, base, tag, _, _) in MORFEUSZ2.generate(noun)]
    filtered = []
    for option in options:
        (orth, tag) = option
        if in_tag(u'pl', tag) and in_tag(u'nom', tag):
            filtered.append(option)
        options = filtered
    return [orth for orth, _ in options]
    
    
def dependents(atr, options):
    if atr.selection_mode.name == u'ratr' or atr.selection_mode.name == u'ratr1':
        result = []
        for option in options:
            result += phrase(option, atr.values.all())
        return result
    else:
       return [orth for orth, _ in options]

def phrase(head, dependents):
    modifiers = {'pre': [], 'post': []}
    for dependent in dependents:
        values = []
        type = None
        for argument in dependent.position.arguments.all():
            if argument.type == u'fixed':
                type = argument.type
            elif argument.type == u'lex':
                type = sortatributes(argument)[0].values.all()[0].argument.type
                value, _, _ = lexicalisation(argument, False, '', False, head)
                values += value
        if type == u'adjp':
            modifiers['pre'].append(values)
        else:
            modifiers['post'].append(values)
    pre = []
    for permutation in permutations(modifiers['pre']):
        pre += cartesian(permutation)
    pre = [' '.join(words) for words in pre]
    pre = list(set(pre))
    post = []
    for permutation in permutations(modifiers['post']):
        post += cartesian(permutation)
    post = [' '.join(words) for words in post]
    post = list(set(post))
    orth, _ = head
    result = []
    if len(pre) == 0 and len(post) == 0:
        result.append(orth)
    for prefix in pre:
        for suffix in post:
            if prefix == '' and suffix == '':
                if len(pre) == 1 and len(post) == 1:
                    result.append(orth)
            elif prefix == '':
                result.append(orth + ' ' + suffix)
            elif suffix == '':
                result.append(prefix + ' ' + orth)
            else:
                result.append(prefix + ' ' + orth + ' ' + suffix)
    return result
        

def cartesian(llist):
    if len(llist) == 0:
        result = [[]]
    else:
        result = []
        tail = cartesian(llist[1:])
        for element in llist[0]:
            tailcopy = deepcopy(tail)
            for cart in tailcopy:
                cart.insert(0, element)
                result.append(cart)
        result += tail
    return result

def permutations(llist):
    if len(llist) == 0:
        result = [[]]
    else:
        result = []
        perms = permutations(llist[1:])
        for perm in perms:
            for i in range(0, len(perm) + 1):
                permcopy = deepcopy(perm)
                permcopy.insert(i, llist[0])
                result.append(permcopy)
    return result