syntagrus.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "electric-making",
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "from pathlib import Path\n",
    "from collections import namedtuple\n",
    "from itertools import chain\n",
    "from typing import List, Tuple, Dict, Union, Optional\n",
    "\n",
    "import bs4\n",
    "from bs4 import BeautifulSoup as Soup\n",
    "import pandas as pd\n",
    "from pprint import pprint\n",
    "\n",
    "from tqdm.auto import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "national-partnership",
   "metadata": {},
   "outputs": [],
   "source": [
    "Feats = Dict[str, str]\n",
    "WordDict = Dict[str, Union[str, Feats]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "continuous-armstrong",
   "metadata": {},
   "outputs": [],
   "source": [
    "SYNTAGRUS_DIR = './syntagrus/syntagrus/syntagrus2020'\n",
    "PARSER = 'lxml-xml'\n",
    "\n",
    "syntagrus_dir = Path(SYNTAGRUS_DIR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "conscious-membership",
   "metadata": {},
   "outputs": [],
   "source": [
    "files = list(syntagrus_dir.glob('**/*.tgt'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "choice-roommate",
   "metadata": {},
   "outputs": [],
   "source": [
    "desired_cols = [\n",
    "    'sent_num_file', 'left_context', 'text', 'right_context',\n",
    "        'full_analysis', 'sent_pos_chain',\n",
    "    'has_many_dos', 'has_many_ios',     # whether sentence has many objs with same deprel\n",
    "    'has_nodetype_in_our_forms', \n",
    "    'verb_inf', 'verb',\n",
    "    'verb_feats',                       # full verb features string\n",
    "    'verb_has_neg',                     # binary: 1 - there is a \"ne\", 0 - there isn't\n",
    "    'DO_group', 'IO_group',             # full NPs/PPs\n",
    "        'DO_had_appos', 'IO_had_appos',\n",
    "        'DO_has_relcl', 'IO_has_relcl',\n",
    "            'DO_has_wh_relcl', 'IO_has_wh_relcl',\n",
    "            'DO_has_participle', 'IO_has_participle',\n",
    "    'DO_depth', 'IO_depth',\n",
    "    'DO_length', 'IO_length', \n",
    "    'DO_pos_feat_chain', 'IO_pos_feat_chain',\n",
    "    'DO',                               # word forms\n",
    "    'DO_pos', 'DO_lemma',    \n",
    "        'DO_animacy', 'DO_case',\n",
    "        'DO_number',\n",
    "    'IO',\n",
    "    'IO_pos', 'IO_head', 'IO_head_pos', 'IO_lemma',\n",
    "        'IO_animacy', 'IO_case',\n",
    "        'IO_number',\n",
    "    'order',                            # in terms of V/IO/DO\n",
    "    ]\n",
    "Res = namedtuple('res', desired_cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "swedish-brazil",
   "metadata": {},
   "outputs": [],
   "source": [
    "PRON_STR = (\n",
    "    'я мы ты вы он она оно они '\n",
    "    'себя '\n",
    "    'свой мой наш твой ваш его её их '\n",
    "    'кто что где откуда зачем когда сколько какой чей который каковой каков '\n",
    "    'то тот этот столько такой таков сей оный '\n",
    "    'всяк всякий каждый сам самый любой иной другой весь все '\n",
    "    'никто ничто некого нечего нигде ниоткуда неоткуда незачем никогда нисколько никакой ничей '\n",
    "    'некто нечто некий некоторый несколько '\n",
    ")\n",
    "\n",
    "PRONS = set(PRON_STR.split())\n",
    "\n",
    "\n",
    "PRON_ONLY_MODIFIER_STR = 'кое- -либо -нибудь -то было'\n",
    "PRON_ONLY_MODIFIERS = set(PRON_ONLY_MODIFIER_STR.split())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "every-scholarship",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "кто бы то ни было True\n",
      " кое-кто True\n",
      " кому-нибудь True\n",
      " что-то True\n",
      " что-либо True\n",
      " кем бы то ни было True\n",
      " былое False\n",
      " коек False\n",
      " либонуть False\n"
     ]
    }
   ],
   "source": [
    "lemmas = 'кто бы то ни было, кое-кто, кому-нибудь, что-то, что-либо, кем бы то ни было, былое, коек, либонуть'\n",
    "\n",
    "for lemma in lemmas.split(','):\n",
    "    has_modif = any(re.search(rf'\\b{modif}\\b', lemma) for modif in PRON_ONLY_MODIFIERS)\n",
    "    print(f'{lemma} {has_modif}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "linear-extreme",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'я мы ты вы он она оно они себя свой мой наш твой ваш его её их кто что где откуда зачем когда сколько какой чей который каковой каков то тот этот столько такой таков сей оный всяк всякий каждый сам самый любой иной другой весь все никто ничто некого нечего нигде ниоткуда неоткуда незачем никогда нисколько никакой ничей некто нечто некий некоторый несколько '"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "PRON_STR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "sophisticated-kennedy",
   "metadata": {},
   "outputs": [],
   "source": [
    "common_features = {\n",
    "    'is_composite': 'сл',\n",
    "}\n",
    "\n",
    "nominal_features = {\n",
    "    'gender': 'муж жен сред',\n",
    "    'number': 'ед мн',\n",
    "    'case': 'им род парт дат вин твор пр местн зв',\n",
    "    'animacy': 'неод од',\n",
    "}\n",
    "\n",
    "adjectival_features = {\n",
    "    'brevity': 'кр',\n",
    "}\n",
    "\n",
    "comparative_features = {\n",
    "    'comp_type': 'смяг',\n",
    "    'comp': 'срав прев',\n",
    "}\n",
    "\n",
    "verb_format = {\n",
    "    'representation': 'инф прич деепр',\n",
    "    'mood': 'изъяв пов',\n",
    "    'aspect': 'несов сов',\n",
    "    'tense': 'непрош прош наст',\n",
    "    'person': '1-л 2-л 3-л',\n",
    "    'voice': 'страд',\n",
    "    **adjectival_features,\n",
    "    **nominal_features,\n",
    "}\n",
    "\n",
    "noun_format = {\n",
    "    **nominal_features,\n",
    "}\n",
    "\n",
    "adjective_format = {\n",
    "    **comparative_features,\n",
    "    **adjectival_features,\n",
    "    **nominal_features\n",
    "}\n",
    "\n",
    "adverb_format = {\n",
    "    **comparative_features,\n",
    "}\n",
    "\n",
    "numeral_format = {\n",
    "    **nominal_features\n",
    "}\n",
    "\n",
    "def get_pos_to_feats_format():\n",
    "    pos_to_feats_format = {\n",
    "        'V': dict(verb_format), 'S': dict(noun_format), 'A': dict(adjective_format),\n",
    "        'ADV': dict(adverb_format), 'NUM': dict(numeral_format), 'COM': {},\n",
    "    }\n",
    "    \n",
    "    for pos in pos_to_feats_format.keys():\n",
    "        pos_to_feats_format[pos].update(common_features)\n",
    "\n",
    "    \n",
    "    for pos, pos_format in pos_to_feats_format.items():\n",
    "        for param, values in pos_format.items():\n",
    "            pos_format[param] = values.split()\n",
    "            \n",
    "    return pos_to_feats_format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "christian-checklist",
   "metadata": {},
   "outputs": [],
   "source": [
    "def is_sentence_with_accdo_io(\n",
    "    tag: bs4.element.Tag,\n",
    "    first_object_possible_roles: List[str] = ['1-компл'],\n",
    "    second_object_possible_roles: List[str] = ['2-компл'],\n",
    ") -> bool:\n",
    "    is_sentence = tag.name == 'S'\n",
    "    if not is_sentence:\n",
    "        return False\n",
    "    \n",
    "    sent_str = str(tag)\n",
    "    \n",
    "    first_object_rel_str = (\n",
    "        f'{first_object_possible_roles[0]}' if len(first_object_possible_roles) == 1\n",
    "        else '(?:{})'.format('|'.join(second_object_possible_roles))\n",
    "    )\n",
    "    \n",
    "    first_object_pat = re.compile(\n",
    "        rf'^<W DOM=\".*\" FEAT=\"(?:\\w+ )*?ВИН(?: \\w+)*?\".*LINK=\"{first_object_rel_str}\">.*?$',\n",
    "        re.MULTILINE\n",
    "    )\n",
    "    first_object = re.search(\n",
    "        first_object_pat,\n",
    "        sent_str)\n",
    "    if not first_object:\n",
    "        return False\n",
    "    \n",
    "    second_object_rel_str = (\n",
    "        f'{second_object_possible_roles[0]}' if len(second_object_possible_roles) == 1\n",
    "        else '(?:{})'.format('|'.join(second_object_possible_roles))\n",
    "    )\n",
    "    second_object = re.search(rf'LINK=\"{second_object_rel_str}\"', sent_str)\n",
    "    if not second_object:\n",
    "        return False\n",
    "    else:\n",
    "        return True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "protecting-torture",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_soup(filename: Path) -> bs4.BeautifulSoup:\n",
    "    with open(filename, 'r', encoding='utf-8') as f:\n",
    "        text = f.read()\n",
    "        soup = Soup(text, PARSER)\n",
    "    \n",
    "    return soup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "positive-basis",
   "metadata": {},
   "outputs": [],
   "source": [
    "def find_sentences_do_io(\n",
    "    body: bs4.element.Tag,\n",
    "    is_sentence_with_do_io_kwargs = {}\n",
    ") -> List[bs4.element.Tag]:\n",
    "    \n",
    "    sents = body.find_all(\n",
    "        lambda tag: is_sentence_with_accdo_io(tag, **is_sentence_with_do_io_kwargs),\n",
    "        recursive=False\n",
    "    )\n",
    "    \n",
    "    return sents"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "sufficient-release",
   "metadata": {},
   "outputs": [],
   "source": [
    "file = syntagrus_dir / Path('2019/baden-baden.tgt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "printable-investor",
   "metadata": {},
   "outputs": [],
   "source": [
    "file_soup = get_soup(file).find('body')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "adjusted-semester",
   "metadata": {},
   "outputs": [],
   "source": [
    "sents = file_soup.find_all('S')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "vertical-template",
   "metadata": {},
   "outputs": [],
   "source": [
    "ss = find_sentences_do_io(file_soup)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "adjusted-christopher",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<S ID=\"9\">\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"1\" KSNAME=\"ОН\" LEMMA=\"ОН\" LINK=\"предик\">Он</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ ПРОШ ЕД МУЖ\" ID=\"2\" LEMMA=\"ЧУВСТВОВАТЬ\">чувствовал</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"3\" LEMMA=\"СЕБЯ\" LINK=\"1-компл\">себя</W>\n",
       " <W DOM=\"2\" FEAT=\"ADV\" ID=\"4\" LEMMA=\"ЗДЕСЬ\" LINK=\"обст\">здесь</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ТВОР ОД\" ID=\"5\" KSNAME=\"РУССКИЙ2\" LEMMA=\"РУССКИЙ\" LINK=\"2-компл\">русским</W>, \n",
       " <W DOM=\"5\" FEAT=\"V СОВ СТРАД ПРИЧ ПРОШ ЕД МУЖ ТВОР\" ID=\"6\" LEMMA=\"ТЕРЯТЬ\" LINK=\"опред\">потерянным</W>\n",
       " <W DOM=\"6\" FEAT=\"CONJ\" ID=\"7\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"7\" FEAT=\"V СОВ СТРАД ПРИЧ ПРОШ ЕД МУЖ ТВОР\" ID=\"8\" KSNAME=\"ОТРЫВАТЬ1\" LEMMA=\"ОТРЫВАТЬ\" LINK=\"соч-союзн\">оторванным</W>\n",
       " <W DOM=\"8\" FEAT=\"PR\" ID=\"9\" LEMMA=\"ОТ\" LINK=\"2-компл\">от</W>\n",
       " <W DOM=\"9\" FEAT=\"S МН МУЖ РОД НЕОД\" ID=\"10\" LEMMA=\"КОРЕНЬ\" LINK=\"предл\">корней</W>…\n",
       " </S>,\n",
       " <S ID=\"10\">\n",
       " <W DOM=\"2\" FEAT=\"A ЕД СРЕД ИМ\" ID=\"1\" KSNAME=\"ЭТОТ\" LEMMA=\"ЭТОТ\" LINK=\"опред\">Это</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД СРЕД ИМ НЕОД\" ID=\"2\" LEMMA=\"ЧУВСТВО\" LINK=\"предик\">чувство</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД СРЕД РОД НЕОД\" ID=\"3\" LEMMA=\"РОДСТВО\" LINK=\"1-компл\">родства</W>\n",
       " <W DOM=\"3\" FEAT=\"PR\" ID=\"4\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"1-компл\">с</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД МУЖ ТВОР ОД\" ID=\"5\" KSNAME=\"НЕГО1\" LEMMA=\"ОН\" LINK=\"предл\">ним</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"6\" KSNAME=\"ДАВАТЬ1\" LEMMA=\"ДАВАТЬ\">дает</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД ЖЕН ДАТ ОД\" ID=\"7\" LEMMA=\"Я\" LINK=\"2-компл\">мне</W>\n",
       " <W DOM=\"6\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"8\" LEMMA=\"СИЛА\" LINK=\"1-компл\">силы</W>\n",
       " <W DOM=\"8\" FEAT=\"V НЕСОВ ИНФ\" ID=\"9\" LEMMA=\"ЖИТЬ\" LINK=\"1-компл\">жить</W>\n",
       " <W DOM=\"9\" FEAT=\"ADV\" ID=\"10\" LEMMA=\"ЗДЕСЬ\" LINK=\"1-компл\">здесь</W>.\n",
       " </S>,\n",
       " <S ID=\"14\">\n",
       " <W DOM=\"2\" FEAT=\"ADV\" ID=\"1\" LEMMA=\"НАКОНЕЦ\" LINK=\"вводн\">Наконец</W>, \n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"2\" LEMMA=\"ПРЕКЛОНЯТЬ\">преклоняет</W>\n",
       " <W DOM=\"4\" FEAT=\"A МН ВИН НЕОД\" ID=\"3\" LEMMA=\"СВОЙ\" LINK=\"опред\">свои</W>\n",
       " <W DOM=\"2\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"4\" LEMMA=\"ВЕТВЬ\" LINK=\"1-компл\">ветви</W>\n",
       " <W DOM=\"2\" FEAT=\"PR\" ID=\"5\" LEMMA=\"К\" LINK=\"обст\">к</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД ЖЕН ДАТ НЕОД\" ID=\"6\" LEMMA=\"ВОДА\" LINK=\"предл\">воде</W>\n",
       " <W DOM=\"9\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"7\" LEMMA=\"РОЗОВЫЙ\" LINK=\"опред\">розовая</W>\n",
       " <W DOM=\"9\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"8\" LEMMA=\"МАХРОВЫЙ\" LINK=\"опред\">махровая</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД ЖЕН ИМ НЕОД\" ID=\"9\" LEMMA=\"САКУРА\" LINK=\"предик\">сакура</W>, \n",
       " <W DOM=\"2\" FEAT=\"CONJ\" ID=\"10\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сент-соч\">и</W>\n",
       " <W DOM=\"12\" FEAT=\"A МН ИМ\" ID=\"11\" KSNAME=\"ВЕСЬ\" LEMMA=\"ВЕСЬ\" LINK=\"опред\">все</W>\n",
       " <W DOM=\"14\" FEAT=\"S МН СРЕД ИМ НЕОД\" ID=\"12\" KSNAME=\"ДЕРЕВО1\" LEMMA=\"ДЕРЕВО\" LINK=\"предик\">деревья</W>\n",
       " <W DOM=\"12\" FEAT=\"ADV\" ID=\"13\" KSNAME=\"ВОКРУГ2\" LEMMA=\"ВОКРУГ\" LINK=\"атриб\">вокруг</W>\n",
       " <W DOM=\"10\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ МН 3-Л\" ID=\"14\" LEMMA=\"ПРЕВРАЩАТЬСЯ\" LINK=\"соч-союзн\">превращаются</W>\n",
       " <W DOM=\"14\" FEAT=\"PR\" ID=\"15\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"18\" FEAT=\"A МН ВИН НЕОД\" ID=\"16\" LEMMA=\"ОГРОМНЫЙ\" LINK=\"опред\">огромные</W>\n",
       " <W DOM=\"18\" FEAT=\"A МН ВИН НЕОД\" ID=\"17\" LEMMA=\"БЕЛОСНЕЖНЫЙ\" LINK=\"опред\">белоснежные</W>\n",
       " <W DOM=\"15\" FEAT=\"S МН МУЖ ВИН НЕОД\" ID=\"18\" LEMMA=\"КОРАЛЛ\" LINK=\"предл\">кораллы</W>…\n",
       " </S>,\n",
       " <S ID=\"19\">\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"1\" LEMMA=\"ПОЯВЛЯТЬСЯ\">Появляется</W>\n",
       " <W DOM=\"3\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"2\" LEMMA=\"РАЗНООБРАЗНЫЙ\" LINK=\"опред\">разнообразная</W>\n",
       " <W DOM=\"1\" FEAT=\"S ЕД ЖЕН ИМ НЕОД\" ID=\"3\" LEMMA=\"ПУБЛИКА\" LINK=\"предик\">публика</W>, \n",
       " <W DOM=\"3\" FEAT=\"V НЕСОВ ПРИЧ НЕПРОШ ЕД ЖЕН ИМ\" ID=\"4\" LEMMA=\"ЗАПОЛНЯТЬ\" LINK=\"опред\">заполняющая</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД МУЖ ТВОР ОД\" ID=\"5\" LEMMA=\"СЕБЯ\" LINK=\"2-компл\">собой</W>\n",
       " <W DOM=\"4\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"6\" LEMMA=\"АЛЛЕЯ\" LINK=\"1-компл\">аллеи</W>\n",
       " <W DOM=\"6\" FEAT=\"S МН МУЖ РОД НЕОД\" ID=\"7\" LEMMA=\"ПАРК\" LINK=\"атриб\">парков</W>.\n",
       " </S>,\n",
       " <S ID=\"38\">\n",
       " <W DOM=\"2\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"1\" LEMMA=\"Я\" LINK=\"предик\">Я</W>\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"2\" LEMMA=\"ДОПОЛНЯТЬ\">дополнила</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД ЖЕН ВИН ОД\" ID=\"3\" LEMMA=\"ОНА\" LINK=\"1-компл\">ее</W>\n",
       " <W DOM=\"5\" FEAT=\"NUM ТВОР\" ID=\"4\" KSNAME=\"НЕСКОЛЬКО1\" LEMMA=\"НЕСКОЛЬКО\" LINK=\"количест\">несколькими</W>\n",
       " <W DOM=\"2\" FEAT=\"S МН ЖЕН ТВОР НЕОД\" ID=\"5\" LEMMA=\"КАРТИНА\" LINK=\"2-компл\">картинами</W>: \n",
       " \"<W DOM=\"7\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"6\" LEMMA=\"БЕЛЫЙ\" LINK=\"опред\">Белая</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"7\" LEMMA=\"ЛОШАДЬ\" LINK=\"разъяснит\">лошадь</W>\", \n",
       " \"<W DOM=\"7\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"8\" KSNAME=\"ЛЕВ1\" LEMMA=\"ЛЕВ\" LINK=\"сочин\">Лев</W>\n",
       " <W DOM=\"8\" FEAT=\"CONJ\" ID=\"9\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"9\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"10\" LEMMA=\"ВОЛК\" LINK=\"соч-союзн\">волк</W>\" \n",
       " <W DOM=\"8\" FEAT=\"CONJ\" ID=\"11\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W> \n",
       " \"<W DOM=\"11\" FEAT=\"S ЕД СРЕД ИМ НЕОД\" ID=\"12\" LEMMA=\"ПОХИЩЕНИЕ\" LINK=\"соч-союзн\">Похищение</W>\n",
       " <W DOM=\"12\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"13\" LEMMA=\"ЕВРОПА\" LINK=\"1-компл\">Европы</W>\" \n",
       " <W DOM=\"7\" FEAT=\"S ЕД МУЖ РОД ОД\" ID=\"14\" KSNAME=\"СЕРОВ\" LEMMA=\"СЕРОВ\" LINK=\"атриб\">Серова</W> - \n",
       " <W DOM=\"16\" FEAT=\"ADV\" ID=\"15\" LEMMA=\"ПОЧЕМУ-ТО\" LINK=\"обст\">почему-то</W>\n",
       " <W DOM=\"2\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД СРЕД\" ID=\"16\" LEMMA=\"ЗАХОТЕТЬСЯ\" LINK=\"разъяснит\">захотелось</W>\n",
       " <W DOM=\"16\" FEAT=\"V СОВ ИНФ\" ID=\"17\" LEMMA=\"ДЕЛАТЬ\" LINK=\"2-компл\">сделать</W>\n",
       " <W DOM=\"17\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"18\" LEMMA=\"КОПИЯ\" LINK=\"1-компл\">копии</W>.\n",
       " </S>,\n",
       " <S ID=\"52\">\n",
       " <W DOM=\"8\" FEAT=\"ADV\" ID=\"1\" LEMMA=\"ОДНАЖДЫ\" LINK=\"обст\">Однажды</W>, \n",
       " <W DOM=\"8\" FEAT=\"V СОВ ДЕЕПР ПРОШ\" ID=\"2\" KSNAME=\"ВЫВОДИТЬ4\" LEMMA=\"ВЫВОДИТЬ\" LINK=\"обст\">выведя</W>\n",
       " <W DOM=\"2\" FEAT=\"PR\" ID=\"3\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"обст\">на</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"4\" LEMMA=\"ПРОГУЛКА\" LINK=\"предл\">прогулку</W>\n",
       " <W DOM=\"6\" FEAT=\"A ЕД МУЖ ВИН ОД\" ID=\"5\" LEMMA=\"НАШ\" LINK=\"опред\">нашего</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"6\" LEMMA=\"ПЕС\" LINK=\"1-компл\">пса</W>, \n",
       " <W DOM=\"8\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"7\" LEMMA=\"Я\" LINK=\"предик\">я</W>\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"8\" LEMMA=\"ЗНАКОМИТЬСЯ\">познакомилась</W>\n",
       " <W DOM=\"8\" FEAT=\"PR\" ID=\"9\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"1-компл\">с</W>\n",
       " <W DOM=\"12\" FEAT=\"A ЕД ЖЕН ТВОР\" ID=\"10\" KSNAME=\"ОДИН2\" LEMMA=\"ОДИН\" LINK=\"опред\">одной</W>\n",
       " <W DOM=\"12\" FEAT=\"A ЕД ЖЕН ТВОР\" ID=\"11\" LEMMA=\"МИЛЫЙ\" LINK=\"опред\">милой</W>\n",
       " <W DOM=\"9\" FEAT=\"S ЕД ЖЕН ТВОР ОД\" ID=\"12\" LEMMA=\"ЖЕНЩИНА\" LINK=\"предл\">женщиной</W>\n",
       " <W DOM=\"14\" FEAT=\"A ЕД МУЖ РОД\" ID=\"13\" LEMMA=\"ВОСТОЧНЫЙ\" LINK=\"опред\">восточного</W>\n",
       " <W DOM=\"12\" FEAT=\"S ЕД МУЖ РОД НЕОД\" ID=\"14\" KSNAME=\"ВИД2\" LEMMA=\"ВИД\" LINK=\"атриб\">вида</W>, \n",
       " <W DOM=\"12\" FEAT=\"V СОВ СТРАД ПРИЧ ПРОШ ЕД ЖЕН ТВОР\" ID=\"15\" KSNAME=\"ОБВЕШИВАТЬ1\" LEMMA=\"ОБВЕШИВАТЬ\" LINK=\"опред\">обвешенной</W>\n",
       " <W DOM=\"15\" FEAT=\"S ЕД СРЕД ТВОР НЕОД\" ID=\"16\" LEMMA=\"МНОЖЕСТВО\" LINK=\"2-компл\">множеством</W>\n",
       " <W DOM=\"18\" FEAT=\"A МН РОД\" ID=\"17\" LEMMA=\"ПЕСТРЫЙ\" LINK=\"опред\">пестрых</W>\n",
       " <W DOM=\"16\" FEAT=\"S МН ЖЕН РОД НЕОД\" ID=\"18\" LEMMA=\"БУСЫ\" LINK=\"квазиагент\">бус</W>.\n",
       " </S>,\n",
       " <S ID=\"58\">\n",
       " <W DOM=\"2\" FEAT=\"ADV\" ID=\"1\" LEMMA=\"ПОСТОЯННО\" LINK=\"обст\">Постоянно</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"2\" KSNAME=\"ЕСТЬ1\" LEMMA=\"ЕСТЬ\">ест</W>\n",
       " <W DOM=\"4\" FEAT=\"A ЕД ЖЕН ВИН\" ID=\"3\" LEMMA=\"СВЕЖИЙ\" LINK=\"опред\">свежую</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"4\" LEMMA=\"ТРАВА\" LINK=\"1-компл\">траву</W>\n",
       " <W DOM=\"2\" FEAT=\"CONJ\" ID=\"5\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"5\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"6\" LEMMA=\"ПИТЬ\" LINK=\"соч-союзн\">пьет</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"7\" LEMMA=\"ВОДА\" LINK=\"1-компл\">воду</W>\n",
       " <W DOM=\"6\" FEAT=\"PR\" ID=\"8\" LEMMA=\"ИЗ\" LINK=\"2-компл\">из</W>\n",
       " <W DOM=\"8\" FEAT=\"S МН МУЖ РОД НЕОД\" ID=\"9\" LEMMA=\"РУЧЕЙ\" LINK=\"предл\">ручьев</W>\n",
       " <W DOM=\"9\" FEAT=\"CONJ\" ID=\"10\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"10\" FEAT=\"S МН ЖЕН РОД НЕОД\" ID=\"11\" LEMMA=\"РЕКА\" LINK=\"соч-союзн\">рек</W>, \n",
       " <W DOM=\"16\" FEAT=\"S МН РОД\" ID=\"12\" LEMMA=\"КОТОРЫЙ\" LINK=\"предик\">которых</W>, \n",
       " <W DOM=\"16\" FEAT=\"PR\" ID=\"13\" LEMMA=\"К\" LINK=\"вводн\">к</W>\n",
       " <W DOM=\"13\" FEAT=\"S ЕД СРЕД ДАТ НЕОД\" ID=\"14\" LEMMA=\"СЧАСТЬЕ\" LINK=\"предл\">счастью</W>, \n",
       " <W DOM=\"16\" FEAT=\"ADV\" ID=\"15\" KSNAME=\"ТУТ1\" LEMMA=\"ТУТ\" LINK=\"обст\">тут</W>\n",
       " <W DOM=\"9\" FEAT=\"ADV\" ID=\"16\" KSNAME=\"МНОГО1\" LEMMA=\"МНОГО\" LINK=\"релят\">много</W>.\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"59\">\n",
       " <W DOM=\"3\" FEAT=\"ADV\" ID=\"1\" LEMMA=\"ОДНАЖДЫ\" LINK=\"огранич\">Однажды</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"2\" KSNAME=\"ОН\" LEMMA=\"ОН\" LINK=\"1-компл\">его</W>\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД МУЖ\" ID=\"3\" LEMMA=\"ВИДЕТЬ\">увидел</W>\n",
       " <W DOM=\"3\" FEAT=\"PR\" ID=\"4\" KSNAME=\"В3\" LEMMA=\"В\" LINK=\"обст\">в</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД МУЖ МЕСТН НЕОД\" ID=\"5\" KSNAME=\"ЛЕС1\" LEMMA=\"ЛЕС\" LINK=\"предл\">лесу</W>\n",
       " <W DOM=\"7\" FEAT=\"A ЕД МУЖ ИМ\" ID=\"6\" LEMMA=\"НАШ\" LINK=\"опред\">наш</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"7\" LEMMA=\"СОСЕД\" LINK=\"предик\">сосед</W> - \n",
       " <W DOM=\"9\" FEAT=\"A ЕД МУЖ ИМ\" ID=\"8\" KSNAME=\"МОЛОДОЙ1\" LEMMA=\"МОЛОДОЙ\" LINK=\"опред\">молодой</W>\n",
       " <W DOM=\"7\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"9\" LEMMA=\"ПАРЕНЬ\" LINK=\"разъяснит\">парень</W> - \n",
       " <W DOM=\"3\" FEAT=\"CONJ\" ID=\"10\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"12\" FEAT=\"PART\" ID=\"11\" KSNAME=\"НЕ1\" LEMMA=\"НЕ\" LINK=\"огранич\">не</W>\n",
       " <W DOM=\"10\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД МУЖ\" ID=\"12\" KSNAME=\"ПОВЕРИТЬ\" LEMMA=\"ПОВЕРИТЬ\" LINK=\"соч-союзн\">поверил</W>\n",
       " <W DOM=\"14\" FEAT=\"A МН ДАТ\" ID=\"13\" LEMMA=\"СВОЙ\" LINK=\"опред\">своим</W>\n",
       " <W DOM=\"12\" FEAT=\"S МН МУЖ ДАТ НЕОД\" ID=\"14\" LEMMA=\"ГЛАЗ\" LINK=\"2-компл\">глазам</W>: \n",
       " <W DOM=\"12\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"15\" KSNAME=\"БЕЖАТЬ1\" LEMMA=\"БЕЖАТЬ\" LINK=\"разъяснит\">бежит</W>\n",
       " <W DOM=\"15\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"16\" LEMMA=\"КОСУЛЯ\" LINK=\"предик\">косуля</W>, \n",
       " <W DOM=\"15\" FEAT=\"CONJ\" ID=\"17\" KSNAME=\"А1\" LEMMA=\"А\" LINK=\"сент-соч\">а</W>\n",
       " <W DOM=\"20\" FEAT=\"PR\" ID=\"18\" KSNAME=\"ЗА2\" LEMMA=\"ЗА\" LINK=\"обст\">за</W>\n",
       " <W DOM=\"18\" FEAT=\"S ЕД ЖЕН ТВОР ОД\" ID=\"19\" LEMMA=\"ОНА\" LINK=\"предл\">ней</W>\n",
       " <W DOM=\"17\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"20\" KSNAME=\"БЕЖАТЬ1\" LEMMA=\"БЕЖАТЬ\" LINK=\"соч-союзн\" NODETYPE=\"FANTOM\"/>\n",
       " <W DOM=\"20\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"21\" LEMMA=\"ВОЛК\" LINK=\"предик\">волк</W>…\n",
       " <LF LFARG=\"5\" LFFUNC=\"_LOC\" LFVAL=\"4\"/>\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"62\">\n",
       " <W DOM=\"3\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"1\" LEMMA=\"Я\" LINK=\"предик\">Я</W>\n",
       " <W DOM=\"3\" FEAT=\"ADV\" ID=\"2\" LEMMA=\"ВСЕГДА\" LINK=\"огранич\">всегда</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 1-Л\" ID=\"3\" LEMMA=\"ЧУВСТВОВАТЬ\">чувствую</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"4\" LEMMA=\"СЕБЯ\" LINK=\"1-компл\">себя</W>\n",
       " <W DOM=\"3\" FEAT=\"PR\" ID=\"5\" KSNAME=\"В3\" LEMMA=\"В\" LINK=\"обст\">в</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД МУЖ МЕСТН НЕОД\" ID=\"6\" KSNAME=\"ЛЕС1\" LEMMA=\"ЛЕС\" LINK=\"предл\">лесу</W>\n",
       " <W DOM=\"3\" FEAT=\"CONJ\" ID=\"7\" KSNAME=\"КАК1\" LEMMA=\"КАК\" LINK=\"2-компл\">как</W>\n",
       " <W DOM=\"7\" FEAT=\"ADV\" ID=\"8\" LEMMA=\"ДОМА\" LINK=\"сравн-союзн\">дома</W>, \n",
       " <W DOM=\"3\" FEAT=\"CONJ\" ID=\"9\" LEMMA=\"НО\" LINK=\"сочин\">но</W>\n",
       " <W DOM=\"19\" FEAT=\"ADV\" ID=\"10\" LEMMA=\"ОДНАЖДЫ\" LINK=\"обст\">однажды</W>, \n",
       " <W DOM=\"19\" FEAT=\"V СОВ ДЕЕПР ПРОШ\" ID=\"11\" LEMMA=\"ВИДЕТЬ\" LINK=\"обст\">увидев</W>\n",
       " <W DOM=\"11\" FEAT=\"PR\" ID=\"12\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"обст\">в</W>\n",
       " <W DOM=\"14\" FEAT=\"NUM\" ID=\"13\" KSNAME=\"ЧИСЛ-БОЛЬШ\" LEMMA=\"20\" LINK=\"количест\">20</W>\n",
       " <W DOM=\"12\" FEAT=\"S МН МУЖ ПР НЕОД\" ID=\"14\" LEMMA=\"ШАГ\" LINK=\"предл\">шагах</W>\n",
       " <W DOM=\"16\" FEAT=\"A ЕД МУЖ ВИН ОД\" ID=\"15\" LEMMA=\"ОГРОМНЫЙ\" LINK=\"опред\">огромного</W>\n",
       " <W DOM=\"11\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"16\" KSNAME=\"КАБАН1\" LEMMA=\"КАБАН\" LINK=\"1-компл\">кабана</W>, \n",
       " <W DOM=\"16\" FEAT=\"V НЕСОВ ПРИЧ НЕПРОШ ЕД МУЖ ВИН ОД\" ID=\"17\" LEMMA=\"ПЕРЕСЕКАТЬ\" LINK=\"опред\">пересекающего</W>\n",
       " <W DOM=\"17\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"18\" LEMMA=\"ТРОПА\" LINK=\"1-компл\">тропу</W>, \n",
       " <W DOM=\"9\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"19\" LEMMA=\"ИСПУГАТЬСЯ\" LINK=\"соч-союзн\">испугалась</W>.\n",
       " <LF LFARG=\"6\" LFFUNC=\"_LOC\" LFVAL=\"5\"/>\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"67\">\n",
       " <W DOM=\"6\" FEAT=\"PR\" ID=\"1\" KSNAME=\"В3\" LEMMA=\"В\" LINK=\"обст\">В</W>\n",
       " <W DOM=\"3\" FEAT=\"A ЕД МУЖ ПР\" ID=\"2\" LEMMA=\"ОГРОМНЫЙ\" LINK=\"опред\">огромном</W>\n",
       " <W DOM=\"1\" FEAT=\"S ЕД МУЖ МЕСТН НЕОД\" ID=\"3\" KSNAME=\"ТАЗ1\" LEMMA=\"ТАЗ\" LINK=\"предл\">тазу</W>\n",
       " <W DOM=\"5\" FEAT=\"A ЕД СРЕД ИМ\" ID=\"4\" KSNAME=\"ВЕСЬ\" LEMMA=\"ВЕСЬ\" LINK=\"опред\">все</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД СРЕД ИМ НЕОД\" ID=\"5\" KSNAME=\"ЭТО\" LEMMA=\"ЭТО\" LINK=\"предик\">это</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ СТРАД ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"6\" KSNAME=\"ПЕРЕМЕШИВАТЬ1\" LEMMA=\"ПЕРЕМЕШИВАТЬ\">перемешивается</W>, \n",
       " <W DOM=\"6\" FEAT=\"V НЕСОВ СТРАД ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"7\" LEMMA=\"РАСКЛАДЫВАТЬ\" LINK=\"сочин\">раскладывается</W>\n",
       " <W DOM=\"7\" FEAT=\"PR\" ID=\"8\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"обст\">на</W>\n",
       " <W DOM=\"8\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"9\" LEMMA=\"ПОРЦИЯ\" LINK=\"предл\">порции</W>\n",
       " <W DOM=\"7\" FEAT=\"CONJ\" ID=\"10\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"10\" FEAT=\"V НЕСОВ СТРАД ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"11\" LEMMA=\"ЗАСОВЫВАТЬ\" LINK=\"соч-союзн\">засовывается</W>\n",
       " <W DOM=\"11\" FEAT=\"PR\" ID=\"12\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"12\" FEAT=\"S ЕД МУЖ ВИН НЕОД\" ID=\"13\" LEMMA=\"ХОЛОДИЛЬНИК\" LINK=\"предл\">холодильник</W>, \n",
       " <W DOM=\"17\" FEAT=\"PR\" ID=\"14\" KSNAME=\"ПОСЛЕ1\" LEMMA=\"ПОСЛЕ\" LINK=\"обст\">после</W>\n",
       " <W DOM=\"14\" FEAT=\"S ЕД СРЕД РОД НЕОД\" ID=\"15\" KSNAME=\"ЧТО2\" LEMMA=\"ЧТО\" LINK=\"предл\">чего</W>\n",
       " <W DOM=\"17\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"16\" KSNAME=\"ФИКТ-ЛИЧИМЯ\" LEMMA=\"ЮМА\" LINK=\"предик\">Юма</W>\n",
       " <W DOM=\"6\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"17\" LEMMA=\"ВСТАВАТЬ\" LINK=\"изъясн\">встает</W>\n",
       " <W DOM=\"17\" FEAT=\"PR\" ID=\"18\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"18\" FEAT=\"S ЕД МУЖ ВИН НЕОД\" ID=\"19\" KSNAME=\"ТАЗ1\" LEMMA=\"ТАЗ\" LINK=\"предл\">таз</W>\n",
       " <W DOM=\"17\" FEAT=\"CONJ\" ID=\"20\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"22\" FEAT=\"ADV\" ID=\"21\" LEMMA=\"ТЩАТЕЛЬНО\" LINK=\"обст\">тщательно</W>\n",
       " <W DOM=\"20\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"22\" LEMMA=\"ВЫЛИЗЫВАТЬ\" LINK=\"соч-союзн\">вылизывает</W>\n",
       " <W DOM=\"22\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"23\" KSNAME=\"ОН\" LEMMA=\"ОН\" LINK=\"1-компл\">его</W>\n",
       " <W DOM=\"22\" FEAT=\"ADV\" ID=\"24\" KSNAME=\"ИЗНУТРИ2\" LEMMA=\"ИЗНУТРИ\" LINK=\"обст\">изнутри</W>.\n",
       " <LF LFARG=\"3\" LFFUNC=\"_LOC\" LFVAL=\"1\"/>\n",
       " <LF LFARG=\"19\" LFFUNC=\"_LOC-AD\" LFVAL=\"18\"/>\n",
       " </S>,\n",
       " <S ID=\"71\">\n",
       " <W DOM=\"2\" FEAT=\"S ЕД ЖЕН ВИН ОД\" ID=\"1\" LEMMA=\"СОБАКА\" LINK=\"1-компл\">Собаку</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"2\" LEMMA=\"ЛЮБИТЬ\">любит</W>\n",
       " <W DOM=\"2\" FEAT=\"ADV\" ID=\"3\" LEMMA=\"БЕЗМЕРНО\" LINK=\"обст\">безмерно</W>, \n",
       " <W DOM=\"2\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"4\" LEMMA=\"НАЗЫВАТЬ\" LINK=\"сочин\">называет</W>\n",
       " <W DOM=\"6\" FEAT=\"A ЕД МУЖ ТВОР\" ID=\"5\" LEMMA=\"ЛУЧШИЙ\" LINK=\"опред\">лучшим</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД МУЖ ТВОР ОД\" ID=\"6\" LEMMA=\"ДРУГ\" LINK=\"2-компл\">другом</W>, \n",
       " <W DOM=\"4\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"7\" KSNAME=\"ЗАСЫПАТЬ1\" LEMMA=\"ЗАСЫПАТЬ\" LINK=\"сочин\">засыпает</W>\n",
       " <W DOM=\"9\" FEAT=\"PART\" ID=\"8\" LEMMA=\"ЛИШЬ\" LINK=\"огранич\">лишь</W>\n",
       " <W DOM=\"7\" FEAT=\"CONJ\" ID=\"9\" LEMMA=\"ПРИ УСЛОВИИ ЧТО\" LINK=\"обст\">при условии, что</W>\n",
       " <W DOM=\"11\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"10\" KSNAME=\"ФИКТ-ЛИЧИМЯ\" LEMMA=\"ЮМА\" LINK=\"предик\">Юма</W>\n",
       " <W DOM=\"9\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"11\" LEMMA=\"ЛЕЖАТЬ\" LINK=\"подч-союзн\">лежит</W>\n",
       " <W DOM=\"11\" FEAT=\"PR\" ID=\"12\" KSNAME=\"У1\" LEMMA=\"У\" LINK=\"неакт-компл\">у</W>\n",
       " <W DOM=\"12\" FEAT=\"S ЕД МУЖ РОД ОД\" ID=\"13\" KSNAME=\"НЕГО1\" LEMMA=\"ОН\" LINK=\"предл\">него</W>\n",
       " <W DOM=\"11\" FEAT=\"PR\" ID=\"14\" KSNAME=\"ПОД2\" LEMMA=\"ПОД\" LINK=\"1-компл\">под</W>\n",
       " <W DOM=\"14\" FEAT=\"S ЕД МУЖ ТВОР НЕОД\" ID=\"15\" LEMMA=\"БОК\" LINK=\"предл\">боком</W>.\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"80\">\n",
       " <W DOM=\"2\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"1\" LEMMA=\"ПЕРВЫЙ\" LINK=\"опред\">Первая</W>\n",
       " <W DOM=\"_root\" FEAT=\"S ЕД ЖЕН ИМ НЕОД\" ID=\"2\" LEMMA=\"ОШИБКА\">ошибка</W>, \n",
       " <W DOM=\"5\" FEAT=\"S ЕД ЖЕН ВИН\" ID=\"3\" LEMMA=\"КОТОРЫЙ\" LINK=\"1-компл\">которую</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"4\" LEMMA=\"Я\" LINK=\"предик\">я</W>\n",
       " <W DOM=\"2\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"5\" KSNAME=\"ДОПУСКАТЬ1\" LEMMA=\"ДОПУСКАТЬ\" LINK=\"релят\">допустила</W>: \n",
       " <W DOM=\"2\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"6\" LEMMA=\"ПОЯВЛЯТЬСЯ\" LINK=\"разъяснит\">появилась</W>\n",
       " <W DOM=\"6\" FEAT=\"PR\" ID=\"7\" KSNAME=\"НА3\" LEMMA=\"НА\" LINK=\"2-компл\">на</W>\n",
       " <W DOM=\"7\" FEAT=\"S ЕД МУЖ МЕСТН НЕОД\" ID=\"8\" LEMMA=\"ЛУГ\" LINK=\"предл\">лугу</W>\n",
       " <W DOM=\"6\" FEAT=\"PR\" ID=\"9\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"обст\">в</W>\n",
       " <W DOM=\"11\" FEAT=\"A ЕД ЖЕН ПР\" ID=\"10\" KSNAME=\"ПРОСТОЙ1\" LEMMA=\"ПРОСТОЙ\" LINK=\"опред\">простой</W>\n",
       " <W DOM=\"9\" FEAT=\"S ЕД ЖЕН ПР НЕОД\" ID=\"11\" LEMMA=\"ОДЕЖДА\" LINK=\"предл\">одежде</W>\n",
       " <W DOM=\"9\" FEAT=\"CONJ\" ID=\"12\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"12\" FEAT=\"PR\" ID=\"13\" LEMMA=\"БЕЗ\" LINK=\"соч-союзн\">без</W>\n",
       " <W DOM=\"13\" FEAT=\"S ЕД МУЖ РОД НЕОД\" ID=\"14\" LEMMA=\"МАКИЯЖ\" LINK=\"предл\">макияжа</W>.\n",
       " <LF LFARG=\"2\" LFFUNC=\"_OPER1\" LFVAL=\"5\"/>\n",
       " <LF LFARG=\"8\" LFFUNC=\"_LOC\" LFVAL=\"7\"/>\n",
       " </S>,\n",
       " <S ID=\"84\">\n",
       " <W DOM=\"3\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"1\" LEMMA=\"ВТОРОЙ\" LINK=\"опред\">Вторая</W>\n",
       " <W DOM=\"3\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"2\" LEMMA=\"МОЙ\" LINK=\"опред\">моя</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД ЖЕН ИМ НЕОД\" ID=\"3\" LEMMA=\"ОШИБКА\" LINK=\"предик\">ошибка</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"4\" KSNAME=\"БЫТЬ\" LEMMA=\"БЫТЬ\">была</W>\n",
       " <W DOM=\"6\" FEAT=\"ADV\" ID=\"5\" LEMMA=\"СОВСЕМ\" LINK=\"огранич\">совсем</W>\n",
       " <W DOM=\"4\" FEAT=\"A ЕД ЖЕН ТВОР\" ID=\"6\" LEMMA=\"НЕПРОСТИТЕЛЬНЫЙ\" LINK=\"присвяз\">непростительной</W>: \n",
       " <W DOM=\"8\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"7\" LEMMA=\"Я\" LINK=\"предик\">я</W>\n",
       " <W DOM=\"4\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"8\" LEMMA=\"ПЫТАТЬСЯ\" LINK=\"разъяснит\">попыталась</W>\n",
       " <W DOM=\"14\" FEAT=\"A ЕД ЖЕН ВИН\" ID=\"9\" KSNAME=\"ОДИН2\" LEMMA=\"ОДИН\" LINK=\"опред\">одну</W>\n",
       " <W DOM=\"14\" FEAT=\"A ЕД ЖЕН ВИН\" ID=\"10\" LEMMA=\"МИЛЫЙ\" LINK=\"опред\">милую</W>, \n",
       " <W DOM=\"10\" FEAT=\"CONJ\" ID=\"11\" KSNAME=\"КАК1\" LEMMA=\"КАК\" LINK=\"вводн\">как</W>\n",
       " <W DOM=\"13\" FEAT=\"S ЕД ЖЕН ДАТ ОД\" ID=\"12\" LEMMA=\"Я\" LINK=\"2-компл\">мне</W>\n",
       " <W DOM=\"11\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД СРЕД\" ID=\"13\" KSNAME=\"КАЗАТЬСЯ\" LEMMA=\"КАЗАТЬСЯ\" LINK=\"сравн-союзн\">показалось</W>, \n",
       " <W DOM=\"15\" FEAT=\"S ЕД ЖЕН ВИН ОД\" ID=\"14\" LEMMA=\"ДАМА\" LINK=\"1-компл\">даму</W>\n",
       " <W DOM=\"8\" FEAT=\"V СОВ ИНФ\" ID=\"15\" LEMMA=\"ПРИГЛАШАТЬ\" LINK=\"1-компл\">пригласить</W>\n",
       " <W DOM=\"15\" FEAT=\"PR\" ID=\"16\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"3-компл\">на</W>\n",
       " <W DOM=\"18\" FEAT=\"A ЕД ЖЕН ВИН\" ID=\"17\" LEMMA=\"СВОЙ\" LINK=\"опред\">свою</W>\n",
       " <W DOM=\"16\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"18\" LEMMA=\"ВЫСТАВКА\" LINK=\"предл\">выставку</W>.\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"89\">\n",
       " <W DOM=\"4\" FEAT=\"ADV\" ID=\"1\" LEMMA=\"ЧАСТЕНЬКО\" LINK=\"обст\">Частенько</W>\n",
       " <W DOM=\"4\" FEAT=\"PR\" ID=\"2\" KSNAME=\"НА3\" LEMMA=\"НА\" LINK=\"2-компл\">на</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ МЕСТН НЕОД\" ID=\"3\" LEMMA=\"ЛУГ\" LINK=\"предл\">лугу</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"4\" LEMMA=\"ПОЯВЛЯТЬСЯ\">появляется</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"5\" LEMMA=\"ДАМА\" LINK=\"предик\">дама</W>\n",
       " <W DOM=\"5\" FEAT=\"PR\" ID=\"6\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"атриб\">с</W>\n",
       " <W DOM=\"8\" FEAT=\"A ЕД МУЖ ТВОР\" ID=\"7\" KSNAME=\"ФИКТ-ПРИЛ\" LEMMA=\"70-КИЛОГРАММОВЫЙ\" LINK=\"опред\">70-килограммовым</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД МУЖ ТВОР ОД\" ID=\"8\" LEMMA=\"МАСТИФ\" LINK=\"предл\">мастифом</W>, \n",
       " <W DOM=\"12\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"9\" LEMMA=\"КОТОРЫЙ\" LINK=\"1-компл\">которого</W>\n",
       " <W DOM=\"12\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"10\" LEMMA=\"ОНА\" LINK=\"предик\">она</W>\n",
       " <W DOM=\"12\" FEAT=\"ADV\" ID=\"11\" LEMMA=\"НЕЖНО\" LINK=\"обст\">нежно</W>\n",
       " <W DOM=\"8\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"12\" LEMMA=\"НАЗЫВАТЬ\" LINK=\"релят\">называет</W> \n",
       " \"<W DOM=\"12\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"13\" KSNAME=\"ФИКТ-ЛИЧИМЯ\" LEMMA=\"МОЙЗЕ\" LINK=\"2-компл\">Мойзе</W>-\n",
       " <W DOM=\"13\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"14\" KSNAME=\"МЫШКА1\" LEMMA=\"МЫШКА\" LINK=\"аппоз\">мышка</W>\", \n",
       " <W DOM=\"16\" FEAT=\"PART\" ID=\"15\" KSNAME=\"НЕ1\" LEMMA=\"НЕ\" LINK=\"огранич\">не</W>\n",
       " <W DOM=\"12\" FEAT=\"V НЕСОВ ДЕЕПР НЕПРОШ\" ID=\"16\" KSNAME=\"ОБРАЩАТЬ2\" LEMMA=\"ОБРАЩАТЬ\" LINK=\"обст\">обращая</W>\n",
       " <W DOM=\"16\" FEAT=\"S ЕД СРЕД РОД НЕОД\" ID=\"17\" LEMMA=\"ВНИМАНИЕ\" LINK=\"1-компл\">внимания</W>\n",
       " <W DOM=\"19\" FEAT=\"PART\" ID=\"18\" KSNAME=\"НИ2\" LEMMA=\"НИ\" LINK=\"огранич\">ни</W>\n",
       " <W DOM=\"16\" FEAT=\"PR\" ID=\"19\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"2-компл\">на</W>\n",
       " <W DOM=\"19\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"20\" LEMMA=\"КТО\" LINK=\"предл\">кого</W>\n",
       " <W DOM=\"20\" FEAT=\"ADV\" ID=\"21\" KSNAME=\"ВОКРУГ2\" LEMMA=\"ВОКРУГ\" LINK=\"атриб\">вокруг</W>.\n",
       " <LF LFARG=\"17\" LFFUNC=\"_OPER1\" LFVAL=\"16\"/>\n",
       " <LF LFARG=\"3\" LFFUNC=\"_LOC\" LFVAL=\"2\"/>\n",
       " <LF LFARG=\"17\" LFFUNC=\"_CAUSFUNC0\" LFVAL=\"16\"/>\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"98\">\n",
       " <W DOM=\"9\" FEAT=\"ADV\" ID=\"1\" LEMMA=\"ОДНАЖДЫ\" LINK=\"обст\">Однажды</W>, \n",
       " <W DOM=\"9\" FEAT=\"PR\" ID=\"2\" KSNAME=\"ПО1\" LEMMA=\"ПО\" LINK=\"вводн\">по</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ДАТ НЕОД\" ID=\"3\" KSNAME=\"СОВЕТ1\" LEMMA=\"СОВЕТ\" LINK=\"предл\">совету</W>\n",
       " <W DOM=\"3\" FEAT=\"S МН МУЖ РОД ОД\" ID=\"4\" LEMMA=\"СОСЕД\" LINK=\"квазиагент\">соседей</W>, \n",
       " <W DOM=\"8\" FEAT=\"A ЕД СРЕД ТВОР\" ID=\"5\" LEMMA=\"РАННИЙ\" LINK=\"опред\">ранним</W>\n",
       " <W DOM=\"8\" FEAT=\"A ЕД СРЕД ТВОР\" ID=\"6\" LEMMA=\"СОЛНЕЧНЫЙ\" LINK=\"опред\">солнечным</W>\n",
       " <W DOM=\"8\" FEAT=\"A ЕД СРЕД ТВОР\" ID=\"7\" LEMMA=\"МАЙСКИЙ\" LINK=\"опред\">майским</W>\n",
       " <W DOM=\"9\" FEAT=\"S ЕД СРЕД ТВОР НЕОД\" ID=\"8\" LEMMA=\"УТРО\" LINK=\"обст\">утром</W>\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ МН\" ID=\"9\" LEMMA=\"ОТПРАВЛЯТЬСЯ\">отправились</W>\n",
       " <W DOM=\"9\" FEAT=\"S МН ИМ ОД\" ID=\"10\" LEMMA=\"МЫ\" LINK=\"предик\">мы</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"11\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"атриб\">с</W>\n",
       " <W DOM=\"11\" FEAT=\"S ЕД МУЖ ТВОР ОД\" ID=\"12\" KSNAME=\"ФИКТ-ЛИЧИМЯ\" LEMMA=\"ЮМА\" LINK=\"предл\">Юмой</W>\n",
       " <W DOM=\"9\" FEAT=\"PR\" ID=\"13\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"17\" FEAT=\"V СОВ СТРАД ПРИЧ ПРОШ ЕД МУЖ ВИН НЕОД\" ID=\"14\" KSNAME=\"ЗАБРАСЫВАТЬ3\" LEMMA=\"ЗАБРАСЫВАТЬ\" LINK=\"опред\">заброшенный</W>\n",
       " <W DOM=\"17\" FEAT=\"A ЕД МУЖ ВИН НЕОД\" ID=\"15\" LEMMA=\"ГОРОДСКОЙ\" LINK=\"опред\">городской</W>\n",
       " <W DOM=\"17\" FEAT=\"A ЕД МУЖ ВИН НЕОД\" ID=\"16\" LEMMA=\"ЯБЛОНЕВЫЙ\" LINK=\"опред\">яблоневый</W>\n",
       " <W DOM=\"13\" FEAT=\"S ЕД МУЖ ВИН НЕОД\" ID=\"17\" LEMMA=\"САД\" LINK=\"предл\">сад</W>, \n",
       " <W DOM=\"22\" FEAT=\"ADV\" ID=\"18\" LEMMA=\"ГДЕ\" LINK=\"2-компл\">где</W>\n",
       " <W DOM=\"21\" FEAT=\"ADV\" ID=\"19\" KSNAME=\"ЯКОБЫ1\" LEMMA=\"ЯКОБЫ\" LINK=\"огранич\">якобы</W>\n",
       " <W DOM=\"21\" FEAT=\"ADV\" ID=\"20\" LEMMA=\"МОЖНО\" LINK=\"присвяз\">можно</W>\n",
       " <W DOM=\"17\" FEAT=\"V НЕСОВ ИЗЪЯВ ПРОШ ЕД СРЕД\" ID=\"21\" KSNAME=\"БЫТЬ\" LEMMA=\"БЫТЬ\" LINK=\"релят\">было</W>\n",
       " <W DOM=\"21\" FEAT=\"V СОВ ИНФ\" ID=\"22\" LEMMA=\"БРАТЬ\" LINK=\"предик\">взять</W>\n",
       " <W DOM=\"22\" FEAT=\"PR\" ID=\"23\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"3-компл\">в</W>\n",
       " <W DOM=\"23\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"24\" LEMMA=\"АРЕНДА\" LINK=\"предл\">аренду</W>\n",
       " <W DOM=\"22\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"25\" KSNAME=\"ПАРА1\" LEMMA=\"ПАРА\" LINK=\"1-компл\">пару</W>\n",
       " <W DOM=\"25\" FEAT=\"S МН ЖЕН РОД НЕОД\" ID=\"26\" LEMMA=\"ЯБЛОНЯ\" LINK=\"квазиагент\">яблонь</W> \n",
       " (<W DOM=\"26\" FEAT=\"V НЕСОВ ИНФ\" ID=\"27\" LEMMA=\"УХАЖИВАТЬ\" LINK=\"примыкат\">ухаживать</W>\n",
       " <W DOM=\"27\" FEAT=\"PR\" ID=\"28\" KSNAME=\"ЗА2\" LEMMA=\"ЗА\" LINK=\"1-компл\">за</W>\n",
       " <W DOM=\"28\" FEAT=\"S МН ТВОР ОД\" ID=\"29\" LEMMA=\"ОНИ\" LINK=\"предл\">ними</W>\n",
       " <W DOM=\"27\" FEAT=\"CONJ\" ID=\"30\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"32\" FEAT=\"ADV\" ID=\"31\" LEMMA=\"ПОТОМ\" LINK=\"обст\">потом</W>\n",
       " <W DOM=\"30\" FEAT=\"V СОВ ИНФ\" ID=\"32\" KSNAME=\"СОБИРАТЬ1\" LEMMA=\"СОБИРАТЬ\" LINK=\"соч-союзн\">собрать</W>\n",
       " <W DOM=\"32\" FEAT=\"PR\" ID=\"33\" KSNAME=\"С1\" LEMMA=\"С\" LINK=\"2-компл\">с</W>\n",
       " <W DOM=\"33\" FEAT=\"S МН РОД ОД\" ID=\"34\" LEMMA=\"ОНИ\" LINK=\"предл\">них</W>\n",
       " <W DOM=\"32\" FEAT=\"S ЕД МУЖ ВИН НЕОД\" ID=\"35\" LEMMA=\"УРОЖАЙ\" LINK=\"1-компл\">урожай</W>).\n",
       " <LF LFARG=\"35\" LFFUNC=\"_REAL1\" LFVAL=\"32\"/>\n",
       " <LF LFARG=\"3\" LFFUNC=\"_ADV2-UN\" LFVAL=\"2\"/>\n",
       " </S>,\n",
       " <S ID=\"103\">\n",
       " <W DOM=\"3\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"1\" LEMMA=\"Я\" LINK=\"предик\">Я</W>\n",
       " <W DOM=\"3\" FEAT=\"ADV\" ID=\"2\" LEMMA=\"НЕМНОГО\" LINK=\"огранич\">немного</W>\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"3\" KSNAME=\"ОТХОДИТЬ1\" LEMMA=\"ОТХОДИТЬ\">отошла</W>\n",
       " <W DOM=\"3\" FEAT=\"PR\" ID=\"4\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"5\" LEMMA=\"СТОРОНА\" LINK=\"предл\">сторону</W>, \n",
       " <W DOM=\"3\" FEAT=\"V СОВ ДЕЕПР ПРОШ\" ID=\"6\" KSNAME=\"РЕШАТЬ2\" LEMMA=\"РЕШАТЬ\" LINK=\"обст\">решив</W>\n",
       " <W DOM=\"6\" FEAT=\"V СОВ ИНФ\" ID=\"7\" KSNAME=\"ПРОПУСКАТЬ1\" LEMMA=\"ПРОПУСКАТЬ\" LINK=\"1-компл\">пропустить</W>\n",
       " <W DOM=\"7\" FEAT=\"S МН ВИН ОД\" ID=\"8\" LEMMA=\"ОНИ\" LINK=\"1-компл\">их</W> - \n",
       " <W DOM=\"10\" FEAT=\"S МН ИМ ОД\" ID=\"9\" LEMMA=\"ОНИ\" LINK=\"предик\">они</W>\n",
       " <W DOM=\"3\" FEAT=\"V НЕСОВ ИЗЪЯВ ПРОШ МН\" ID=\"10\" LEMMA=\"НАПРАВЛЯТЬСЯ\" LINK=\"разъяснит\">направлялись</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"11\" LEMMA=\"К\" LINK=\"2-компл\">к</W>\n",
       " <W DOM=\"11\" FEAT=\"S ЕД МУЖ ДАТ НЕОД\" ID=\"12\" KSNAME=\"ВЫХОД1\" LEMMA=\"ВЫХОД\" LINK=\"предл\">выходу</W>.\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"104\">\n",
       " <W DOM=\"2\" FEAT=\"S МН ИМ ОД\" ID=\"1\" LEMMA=\"ОНИ\" LINK=\"предик\">Они</W>\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ МН\" ID=\"2\" KSNAME=\"ПРОХОДИТЬ1\" LEMMA=\"ПРОХОДИТЬ\">прошли</W>\n",
       " <W DOM=\"2\" FEAT=\"ADV\" ID=\"3\" KSNAME=\"МИМО2\" LEMMA=\"МИМО\" LINK=\"4-компл\">мимо</W>, \n",
       " <W DOM=\"5\" FEAT=\"ADV\" ID=\"4\" KSNAME=\"ПРЯМО1\" LEMMA=\"ПРЯМО\" LINK=\"огранич\">прямо</W>\n",
       " <W DOM=\"7\" FEAT=\"PR\" ID=\"5\" KSNAME=\"ПЕРЕД1\" LEMMA=\"ПЕРЕД\" LINK=\"обст\">передо</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД ЖЕН ТВОР ОД\" ID=\"6\" LEMMA=\"Я\" LINK=\"предл\">мной</W>\n",
       " <W DOM=\"2\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ МН\" ID=\"7\" LEMMA=\"ЗАХЛОПЫВАТЬ\" LINK=\"сочин\">захлопнули</W>\n",
       " <W DOM=\"7\" FEAT=\"S МН СРЕД ВИН НЕОД\" ID=\"8\" LEMMA=\"ВОРОТА\" LINK=\"1-компл\">ворота</W>\n",
       " <W DOM=\"7\" FEAT=\"CONJ\" ID=\"9\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"9\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ МН\" ID=\"10\" LEMMA=\"ЗАКРЫВАТЬ\" LINK=\"соч-союзн\">закрыли</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"11\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"2-компл\">на</W>\n",
       " <W DOM=\"11\" FEAT=\"S ЕД МУЖ ВИН НЕОД\" ID=\"12\" KSNAME=\"ЗАМОК1\" LEMMA=\"ЗАМОК\" LINK=\"предл\">замок</W>.\n",
       " <LF LFARG=\"12\" LFFUNC=\"_LABREAL1-2\" LFPREP=\"11\" LFVAL=\"10\"/>\n",
       " </S>,\n",
       " <S ID=\"107\">\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"1\" LEMMA=\"НАЧИНАТЬ\">Начала</W>\n",
       " <W DOM=\"1\" FEAT=\"V НЕСОВ ИНФ\" ID=\"2\" LEMMA=\"ИСКАТЬ\" LINK=\"1-компл\">искать</W>\n",
       " <W DOM=\"2\" FEAT=\"PR\" ID=\"3\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД МУЖ ПР ОД\" ID=\"4\" KSNAME=\"НЕГО1\" LEMMA=\"ОН\" LINK=\"предл\">нем</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"5\" LEMMA=\"ДЫРКА\" LINK=\"1-компл\">дырку</W>, \n",
       " <W DOM=\"13\" FEAT=\"PR\" ID=\"6\" LEMMA=\"ЧЕРЕЗ\" LINK=\"обст\">через</W>\n",
       " <W DOM=\"8\" FEAT=\"NUM\" ID=\"7\" KSNAME=\"ЧИСЛ-БОЛЬШ\" LEMMA=\"15\" LINK=\"количест\">15</W>\n",
       " <W DOM=\"6\" FEAT=\"S МН ЖЕН РОД НЕОД\" ID=\"8\" LEMMA=\"МИНУТА\" LINK=\"предл\">минут</W>\n",
       " <W DOM=\"8\" FEAT=\"S ЕД СРЕД РОД НЕОД\" ID=\"9\" LEMMA=\"РАЗВИНЧИВАНИЕ\" LINK=\"квазиагент\">развинчивания</W>\n",
       " <W DOM=\"11\" FEAT=\"A ЕД ЖЕН РОД\" ID=\"10\" LEMMA=\"КОЛЮЧИЙ\" LINK=\"опред\">колючей</W>\n",
       " <W DOM=\"9\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"11\" LEMMA=\"ПРОВОЛОКА\" LINK=\"1-компл\">проволоки</W>\n",
       " <W DOM=\"13\" FEAT=\"S МН ЖЕН ИМ НЕОД\" ID=\"12\" LEMMA=\"РУКА\" LINK=\"предик\">руки</W>\n",
       " <W DOM=\"1\" FEAT=\"V НЕСОВ ИЗЪЯВ ПРОШ МН\" ID=\"13\" KSNAME=\"БЫТЬ\" LEMMA=\"БЫТЬ\" LINK=\"сент-соч\">были</W>\n",
       " <W DOM=\"13\" FEAT=\"A МН ИМ\" ID=\"14\" KSNAME=\"ВЕСЬ\" LEMMA=\"ВЕСЬ\" LINK=\"суб-копр\">все</W>\n",
       " <W DOM=\"13\" FEAT=\"PR\" ID=\"15\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"присвяз\">в</W>\n",
       " <W DOM=\"15\" FEAT=\"S ЕД ЖЕН ПР НЕОД\" ID=\"16\" LEMMA=\"КРОВЬ\" LINK=\"предл\">крови</W>, \n",
       " <W DOM=\"13\" FEAT=\"CONJ\" ID=\"17\" LEMMA=\"НО\" LINK=\"сент-соч\">но</W>\n",
       " <W DOM=\"24\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"18\" LEMMA=\"ДЫРКА\" LINK=\"1-компл\">дырку</W>\n",
       " <W DOM=\"18\" FEAT=\"S ЕД МУЖ ТВОР НЕОД\" ID=\"19\" LEMMA=\"РАЗМЕР\" LINK=\"атриб\">размером</W>\n",
       " <W DOM=\"19\" FEAT=\"PR\" ID=\"20\" KSNAME=\"С2\" LEMMA=\"С\" LINK=\"компл-аппоз\">с</W>\n",
       " <W DOM=\"20\" FEAT=\"S ЕД ЖЕН ВИН ОД\" ID=\"21\" LEMMA=\"СОБАКА\" LINK=\"предл\">собаку</W>\n",
       " <W DOM=\"24\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"22\" LEMMA=\"Я\" LINK=\"предик\">я</W>\n",
       " <W DOM=\"24\" FEAT=\"PART\" ID=\"23\" KSNAME=\"УЖЕ\" LEMMA=\"УЖЕ\" LINK=\"огранич\">уже</W>\n",
       " <W DOM=\"17\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"24\" KSNAME=\"ПРОДЕЛЫВАТЬ2\" LEMMA=\"ПРОДЕЛЫВАТЬ\" LINK=\"соч-союзн\">проделала</W>.\n",
       " </S>,\n",
       " <S ID=\"109\">\n",
       " <W DOM=\"2\" FEAT=\"ADV\" ID=\"1\" KSNAME=\"ТУТ1\" LEMMA=\"ТУТ\" LINK=\"обст\">Тут</W>\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД МУЖ\" ID=\"2\" LEMMA=\"ПОДЪЕЗЖАТЬ\">подъехал</W>\n",
       " <W DOM=\"4\" FEAT=\"A ЕД МУЖ ИМ\" ID=\"3\" LEMMA=\"НЕКИЙ\" LINK=\"опред\">некий</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"4\" LEMMA=\"МУЖЧИНА\" LINK=\"предик\">мужчина</W>\n",
       " <W DOM=\"4\" FEAT=\"PR\" ID=\"5\" KSNAME=\"НА2\" LEMMA=\"НА\" LINK=\"атриб\">на</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД МУЖ ПР НЕОД\" ID=\"6\" LEMMA=\"ВЕЛОСИПЕД\" LINK=\"предл\">велосипеде</W>, \n",
       " <W DOM=\"2\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД МУЖ\" ID=\"7\" KSNAME=\"ОТКРЫВАТЬ1\" LEMMA=\"ОТКРЫВАТЬ\" LINK=\"сочин\">открыл</W>\n",
       " <W DOM=\"7\" FEAT=\"S МН СРЕД ВИН НЕОД\" ID=\"8\" LEMMA=\"ВОРОТА\" LINK=\"1-компл\">ворота</W>\n",
       " <W DOM=\"7\" FEAT=\"CONJ\" ID=\"9\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>, \n",
       " <W DOM=\"15\" FEAT=\"V НЕСОВ ДЕЕПР НЕПРОШ\" ID=\"10\" LEMMA=\"ПРЯТАТЬ\" LINK=\"обст\">пряча</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"11\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"3-компл\">в</W>\n",
       " <W DOM=\"11\" FEAT=\"S ЕД МУЖ ВИН НЕОД\" ID=\"12\" LEMMA=\"ВОРОТНИК\" LINK=\"предл\">воротник</W>\n",
       " <W DOM=\"14\" FEAT=\"A ЕД ЖЕН ВИН\" ID=\"13\" LEMMA=\"УДОВЛЕТВОРЕННЫЙ\" LINK=\"опред\">удовлетворенную</W>\n",
       " <W DOM=\"10\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"14\" LEMMA=\"УЛЫБКА\" LINK=\"1-компл\">улыбку</W>, \n",
       " <W DOM=\"9\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД МУЖ\" ID=\"15\" KSNAME=\"СПРАШИВАТЬ1\" LEMMA=\"СПРАШИВАТЬ\" LINK=\"соч-союзн\">спросил</W>: \n",
       " \"<W DOM=\"15\" FEAT=\"CONJ\" ID=\"16\" KSNAME=\"А1\" LEMMA=\"А\" LINK=\"2-компл\">А</W>\n",
       " <W DOM=\"21\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"17\" KSNAME=\"ЧТО2\" LEMMA=\"ЧТО\" LINK=\"обст\">что</W>\n",
       " <W DOM=\"17\" FEAT=\"PART\" ID=\"18\" KSNAME=\"ЭТО2\" LEMMA=\"ЭТО\" LINK=\"огранич\">это</W>\n",
       " <W DOM=\"21\" FEAT=\"S МН ИМ ОД\" ID=\"19\" LEMMA=\"ВЫ\" LINK=\"предик\">вы</W>\n",
       " <W DOM=\"21\" FEAT=\"S ЕД МУЖ ВИН НЕОД\" ID=\"20\" KSNAME=\"ЗАБОР1\" LEMMA=\"ЗАБОР\" LINK=\"1-компл\">забор</W>\n",
       " <W DOM=\"16\" FEAT=\"V НЕСОВ ПОВ МН 2-Л\" ID=\"21\" KSNAME=\"ПОРТИТЬ\" LEMMA=\"ПОРТИТЬ\" LINK=\"соч-союзн\">портите</W>?\"\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"119\">\n",
       " <W DOM=\"3\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"1\" LEMMA=\"КТО-ТО\" LINK=\"предик\">Кто-то</W>\n",
       " <W DOM=\"3\" FEAT=\"PART\" ID=\"2\" LEMMA=\"ДАЖЕ\" LINK=\"огранич\">даже</W>\n",
       " <W DOM=\"_root\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД МУЖ\" ID=\"3\" LEMMA=\"ПРЕДУПРЕЖДАТЬ\">предупредил</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД ЖЕН ВИН ОД\" ID=\"4\" LEMMA=\"Я\" LINK=\"1-компл\">меня</W>, \n",
       " <W DOM=\"3\" FEAT=\"CONJ\" ID=\"5\" KSNAME=\"ЧТО1\" LEMMA=\"ЧТО\" LINK=\"2-компл\">что</W>\n",
       " <W DOM=\"5\" FEAT=\"A КР ЕД СРЕД\" ID=\"6\" LEMMA=\"НУЖНЫЙ\" LINK=\"подч-союзн\">нужно</W>\n",
       " <W DOM=\"6\" FEAT=\"V СОВ ИНФ\" ID=\"7\" LEMMA=\"ПОЛУЧАТЬ\" LINK=\"предик\">получить</W>\n",
       " <W DOM=\"7\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"8\" KSNAME=\"РАЗРЕШЕНИЕ2\" LEMMA=\"РАЗРЕШЕНИЕ\" LINK=\"1-компл\">разрешение</W>, \n",
       " <W DOM=\"8\" FEAT=\"CONJ\" ID=\"9\" LEMMA=\"ЧТОБЫ\" LINK=\"1-компл\">чтобы</W>\n",
       " <W DOM=\"9\" FEAT=\"V НЕСОВ ИНФ\" ID=\"10\" LEMMA=\"РИСОВАТЬ\" LINK=\"инф-союзн\">рисовать</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"11\" KSNAME=\"НА2\" LEMMA=\"НА\" LINK=\"обст\">на</W>\n",
       " <W DOM=\"11\" FEAT=\"S ЕД ЖЕН ПР НЕОД\" ID=\"12\" LEMMA=\"УЛИЦА\" LINK=\"предл\">улице</W>.\n",
       " <LF LFARG=\"8\" LFFUNC=\"_INCEPOPER3\" LFVAL=\"7\"/>\n",
       " <LF LFARG=\"12\" LFFUNC=\"_LOC\" LFVAL=\"11\"/>\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"122\">\n",
       " <W DOM=\"10\" FEAT=\"ADV СРАВ\" ID=\"1\" KSNAME=\"МНОГО1\" LEMMA=\"МНОГО\" LINK=\"пролепт\">Больше</W>\n",
       " <W DOM=\"1\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"2\" KSNAME=\"ПОЛОВИНА1\" LEMMA=\"ПОЛОВИНА\" LINK=\"сравнит\">половины</W>\n",
       " <W DOM=\"2\" FEAT=\"S МН МУЖ РОД ОД\" ID=\"3\" LEMMA=\"ЧЕЛОВЕК\" LINK=\"квазиагент\">людей</W>, \n",
       " <W DOM=\"6\" FEAT=\"S МН ВИН ОД\" ID=\"4\" LEMMA=\"КОТОРЫЙ\" LINK=\"1-компл\">которых</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"5\" LEMMA=\"Я\" LINK=\"предик\">я</W>\n",
       " <W DOM=\"3\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 1-Л\" ID=\"6\" LEMMA=\"ВСТРЕЧАТЬ\" LINK=\"релят\">встречаю</W>\n",
       " <W DOM=\"6\" FEAT=\"PR\" ID=\"7\" KSNAME=\"НА2\" LEMMA=\"НА\" LINK=\"2-компл\">на</W>\n",
       " <W DOM=\"7\" FEAT=\"S МН ЖЕН ПР НЕОД\" ID=\"8\" LEMMA=\"УЛИЦА\" LINK=\"предл\">улицах</W>\n",
       " <W DOM=\"8\" FEAT=\"S ЕД МУЖ РОД НЕОД\" ID=\"9\" LEMMA=\"БАДЕН\" LINK=\"атриб\">Бадена</W> - \n",
       " <W DOM=\"13\" FEAT=\"S ЕД СРЕД ИМ НЕОД\" ID=\"10\" KSNAME=\"ЭТО\" LEMMA=\"ЭТО\" LINK=\"предик\">это</W>\n",
       " <W DOM=\"13\" FEAT=\"A МН ИМ\" ID=\"11\" LEMMA=\"ТОЩИЙ\" LINK=\"опред\">тощие</W>\n",
       " <W DOM=\"13\" FEAT=\"A МН ИМ\" ID=\"12\" LEMMA=\"НЕМЕЦКИЙ\" LINK=\"опред\">немецкие</W>\n",
       " <W DOM=\"_root\" FEAT=\"S МН ЖЕН ИМ ОД\" ID=\"13\" LEMMA=\"СТАРУХА\">старухи</W> \n",
       " (<W DOM=\"13\" FEAT=\"A МН ИМ\" ID=\"14\" LEMMA=\"ТОЩИЙ\" LINK=\"примыкат\">тощие</W>\n",
       " <W DOM=\"16\" FEAT=\"PART\" ID=\"15\" KSNAME=\"НЕ1\" LEMMA=\"НЕ\" LINK=\"огранич\">не</W>\n",
       " <W DOM=\"14\" FEAT=\"PR\" ID=\"16\" LEMMA=\"ОТ\" LINK=\"атриб\">от</W>\n",
       " <W DOM=\"16\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"17\" LEMMA=\"БОЛЕЗНЬ\" LINK=\"предл\">болезни</W>, \n",
       " <W DOM=\"19\" FEAT=\"S ЕД СРЕД ИМ НЕОД\" ID=\"18\" KSNAME=\"ДЕЛО1\" LEMMA=\"ДЕЛО\" LINK=\"предик\">дело</W>\n",
       " <W DOM=\"14\" FEAT=\"PR\" ID=\"19\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"сент-соч\">в</W>\n",
       " <W DOM=\"19\" FEAT=\"S ЕД ЖЕН ПР НЕОД\" ID=\"20\" LEMMA=\"МОДА\" LINK=\"предл\">моде</W>) \n",
       " <W DOM=\"13\" FEAT=\"PR\" ID=\"21\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"атриб\">в</W>\n",
       " <W DOM=\"23\" FEAT=\"A ЕД ЖЕН ПР\" ID=\"22\" LEMMA=\"СПОРТИВНЫЙ\" LINK=\"опред\">спортивной</W>\n",
       " <W DOM=\"21\" FEAT=\"S ЕД ЖЕН ПР НЕОД\" ID=\"23\" LEMMA=\"ОДЕЖДА\" LINK=\"предл\">одежде</W>\n",
       " <W DOM=\"25\" FEAT=\"A МН РОД\" ID=\"24\" KSNAME=\"САМЫЙ\" LEMMA=\"САМЫЙ\" LINK=\"опред\">самых</W>\n",
       " <W DOM=\"26\" FEAT=\"A МН РОД\" ID=\"25\" KSNAME=\"ДОРОГОЙ1\" LEMMA=\"ДОРОГОЙ\" LINK=\"опред\">дорогих</W>\n",
       " <W DOM=\"23\" FEAT=\"S МН ЖЕН РОД НЕОД\" ID=\"26\" KSNAME=\"МАРКА1\" LEMMA=\"МАРКА\" LINK=\"атриб\">марок</W>, \n",
       " <W DOM=\"31\" FEAT=\"S МН ИМ ОД\" ID=\"27\" KSNAME=\"ВСЕ3\" LEMMA=\"ВСЕ\" LINK=\"предик\">все</W>\n",
       " <W DOM=\"29\" FEAT=\"ADV\" ID=\"28\" LEMMA=\"КОРОТКО\" LINK=\"обст\">коротко</W>\n",
       " <W DOM=\"31\" FEAT=\"V НЕСОВ СТРАД ПРИЧ ПРОШ МН ИМ\" ID=\"29\" LEMMA=\"СТРИЧЬ\" LINK=\"опред\">стриженные</W>\n",
       " <W DOM=\"31\" FEAT=\"A МН ИМ\" ID=\"30\" LEMMA=\"КРАШЕНЫЙ\" LINK=\"опред\">крашеные</W>\n",
       " <W DOM=\"13\" FEAT=\"S МН ЖЕН ИМ ОД\" ID=\"31\" LEMMA=\"БЛОНДИНКА\" LINK=\"сент-соч\">блондинки</W>, \n",
       " <W DOM=\"31\" FEAT=\"PR\" ID=\"32\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"атриб\">с</W>\n",
       " <W DOM=\"34\" FEAT=\"A МН ТВОР\" ID=\"33\" LEMMA=\"СИЛИКОНОВЫЙ\" LINK=\"опред\">силиконовыми</W>\n",
       " <W DOM=\"32\" FEAT=\"S МН ЖЕН ТВОР НЕОД\" ID=\"34\" LEMMA=\"ГУБА\" LINK=\"предл\">губами</W>-\n",
       " <W DOM=\"34\" FEAT=\"S МН МУЖ ТВОР НЕОД\" ID=\"35\" LEMMA=\"ПЕЛЬМЕНЬ\" LINK=\"аппоз\">пельменями</W>, \n",
       " <W DOM=\"37\" FEAT=\"A МН ТВОР\" ID=\"36\" LEMMA=\"ПЛАСТИКОВЫЙ\" LINK=\"опред\">пластиковыми</W>\n",
       " <W DOM=\"34\" FEAT=\"S МН МУЖ ТВОР НЕОД\" ID=\"37\" KSNAME=\"ЗУБ1\" LEMMA=\"ЗУБ\" LINK=\"сочин\">зубами</W>\n",
       " <W DOM=\"39\" FEAT=\"A ЕД ЖЕН РОД\" ID=\"38\" LEMMA=\"НЕИМОВЕРНЫЙ\" LINK=\"опред\">неимоверной</W>\n",
       " <W DOM=\"37\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"39\" LEMMA=\"БЕЛИЗНА\" LINK=\"атриб\">белизны</W>, \n",
       " <W DOM=\"32\" FEAT=\"PR\" ID=\"40\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"сочин\">в</W>\n",
       " <W DOM=\"42\" FEAT=\"A МН ПР\" ID=\"41\" LEMMA=\"ОГРОМНЫЙ\" LINK=\"опред\">огромных</W>\n",
       " <W DOM=\"40\" FEAT=\"S МН МУЖ ПР НЕОД\" ID=\"42\" LEMMA=\"ОЧКИ\" LINK=\"предл\">очках</W>\n",
       " <W DOM=\"42\" FEAT=\"CONJ\" ID=\"43\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"43\" FEAT=\"S МН ЖЕН ПР НЕОД\" ID=\"44\" KSNAME=\"ШЛЯПА1\" LEMMA=\"ШЛЯПА\" LINK=\"соч-союзн\">шляпах</W>, \n",
       " <W DOM=\"40\" FEAT=\"PR\" ID=\"45\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"сочин\">с</W>\n",
       " <W DOM=\"48\" FEAT=\"A МН ТВОР\" ID=\"46\" LEMMA=\"УЖАСНЫЙ\" LINK=\"опред\">ужасными</W>\n",
       " <W DOM=\"48\" FEAT=\"A МН ТВОР\" ID=\"47\" LEMMA=\"МОРЩИНИСТЫЙ\" LINK=\"опред\">морщинистыми</W>\n",
       " <W DOM=\"45\" FEAT=\"S МН ЖЕН ТВОР НЕОД\" ID=\"48\" LEMMA=\"РУКА\" LINK=\"предл\">руками</W>\n",
       " <W DOM=\"48\" FEAT=\"PR\" ID=\"49\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"атриб\">в</W>\n",
       " <W DOM=\"51\" FEAT=\"A МН ПР\" ID=\"50\" LEMMA=\"БРИЛЛИАНТОВЫЙ\" LINK=\"опред\">бриллиантовых</W>\n",
       " <W DOM=\"49\" FEAT=\"S МН СРЕД ПР НЕОД\" ID=\"51\" LEMMA=\"КОЛЬЦО\" LINK=\"предл\">кольцах</W>, \n",
       " <W DOM=\"45\" FEAT=\"PR\" ID=\"52\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"сочин\">со</W>\n",
       " <W DOM=\"54\" FEAT=\"NUM ВИН\" ID=\"53\" LEMMA=\"СТО\" LINK=\"количест\">сто</W>\n",
       " <W DOM=\"55\" FEAT=\"S МН МУЖ РОД НЕОД\" ID=\"54\" KSNAME=\"РАЗ1\" LEMMA=\"РАЗ\" LINK=\"обст\">раз</W>\n",
       " <W DOM=\"58\" FEAT=\"V СОВ СТРАД ПРИЧ ПРОШ МН ТВОР\" ID=\"55\" KSNAME=\"ОПЕРИРОВАТЬ1\" LEMMA=\"ОПЕРИРОВАТЬ\" LINK=\"опред\">прооперированными</W>\n",
       " <W DOM=\"55\" FEAT=\"PR\" ID=\"56\" KSNAME=\"НА2\" LEMMA=\"НА\" LINK=\"обст\">на</W>\n",
       " <W DOM=\"56\" FEAT=\"S МН ЖЕН ПР НЕОД\" ID=\"57\" LEMMA=\"НОГА\" LINK=\"предл\">ногах</W>\n",
       " <W DOM=\"52\" FEAT=\"S МН ЖЕН ТВОР НЕОД\" ID=\"58\" KSNAME=\"ВЕНА1\" LEMMA=\"ВЕНА\" LINK=\"предл\">венами</W>\n",
       " <W DOM=\"52\" FEAT=\"CONJ\" ID=\"59\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"59\" FEAT=\"PR\" ID=\"60\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"соч-союзн\">в</W>\n",
       " <W DOM=\"62\" FEAT=\"A МН ПР\" ID=\"61\" LEMMA=\"БЕЛОСНЕЖНЫЙ\" LINK=\"опред\">белоснежных</W>\n",
       " <W DOM=\"60\" FEAT=\"S МН ЖЕН ПР НЕОД\" ID=\"62\" LEMMA=\"КРОССОВКА\" LINK=\"предл\">кроссовках</W>.\n",
       " <LF LFARG=\"8\" LFFUNC=\"_LOC\" LFVAL=\"7\"/>\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"124\">\n",
       " <W DOM=\"13\" FEAT=\"V НЕСОВ ДЕЕПР НЕПРОШ\" ID=\"1\" LEMMA=\"СИДЕТЬ\" LINK=\"обст\">Сидя</W>\n",
       " <W DOM=\"1\" FEAT=\"S МН МУЖ ВИН НЕОД\" ID=\"2\" LEMMA=\"ДЕНЬ\" LINK=\"длительн\">дни</W>\n",
       " <W DOM=\"2\" FEAT=\"ADV\" ID=\"3\" KSNAME=\"НАПРОЛЕТ\" LEMMA=\"НАПРОЛЕТ\" LINK=\"атриб\">напролет</W>\n",
       " <W DOM=\"1\" FEAT=\"PR\" ID=\"4\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"1-компл\">в</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД СРЕД ПР НЕОД\" ID=\"5\" LEMMA=\"КАФЕ\" LINK=\"предл\">кафе</W>, \n",
       " <W DOM=\"13\" FEAT=\"S МН ИМ ОД\" ID=\"6\" LEMMA=\"ОНИ\" LINK=\"предик\">они</W>, \n",
       " <W DOM=\"13\" FEAT=\"PR\" ID=\"7\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"вводн\">в</W>\n",
       " <W DOM=\"7\" FEAT=\"A ЕД СРЕД ПР\" ID=\"8\" LEMMA=\"ОБЩИЙ\" LINK=\"предл\">общем</W>, \n",
       " <W DOM=\"10\" FEAT=\"PART\" ID=\"9\" KSNAME=\"НИ2\" LEMMA=\"НИ\" LINK=\"огранич\">ни</W>\n",
       " <W DOM=\"13\" FEAT=\"PR\" ID=\"10\" KSNAME=\"О2\" LEMMA=\"О\" LINK=\"2-компл\">о</W>\n",
       " <W DOM=\"10\" FEAT=\"S ЕД СРЕД ПР НЕОД\" ID=\"11\" KSNAME=\"ЧТО2\" LEMMA=\"ЧТО\" LINK=\"предл\">чем</W>\n",
       " <W DOM=\"13\" FEAT=\"PART\" ID=\"12\" KSNAME=\"НЕ1\" LEMMA=\"НЕ\" LINK=\"огранич\">не</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ МН 3-Л\" ID=\"13\" KSNAME=\"ГОВОРИТЬ2\" LEMMA=\"ГОВОРИТЬ\">говорят</W>, \n",
       " <W DOM=\"16\" FEAT=\"PART\" ID=\"14\" KSNAME=\"ТОЛЬКО1\" LEMMA=\"ТОЛЬКО\" LINK=\"огранич\">только</W>\n",
       " <W DOM=\"16\" FEAT=\"ADV\" ID=\"15\" LEMMA=\"ЗЛОБНО\" LINK=\"обст\">злобно</W>\n",
       " <W DOM=\"13\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ МН 3-Л\" ID=\"16\" LEMMA=\"ОГЛЯДЫВАТЬ\" LINK=\"сочин\">оглядывают</W>\n",
       " <W DOM=\"16\" FEAT=\"S МН ВИН ОД\" ID=\"17\" LEMMA=\"ОКРУЖАЮЩИЕ\" LINK=\"1-компл\">окружающих</W>.\n",
       " <LF LFARG=\"5\" LFFUNC=\"_REAL1\" LFPREP=\"4\" LFVAL=\"1\"/>\n",
       " </S>,\n",
       " <S ID=\"126\">\n",
       " <W DOM=\"11\" FEAT=\"PR\" ID=\"1\" KSNAME=\"ПРИ\" LEMMA=\"ПРИ\" LINK=\"обст\">При</W> \n",
       " \"<W DOM=\"1\" FEAT=\"S ЕД СРЕД ПР НЕОД\" ID=\"2\" LEMMA=\"ПОХОЛОДАНИЕ\" LINK=\"предл\">похолодании</W>\", \n",
       " <W DOM=\"5\" FEAT=\"ADV\" ID=\"3\" KSNAME=\"КОГДА1\" LEMMA=\"КОГДА\" LINK=\"обст\">когда</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД ЖЕН ИМ НЕОД\" ID=\"4\" LEMMA=\"ТЕМПЕРАТУРА\" LINK=\"предик\">температура</W>\n",
       " <W DOM=\"2\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"5\" KSNAME=\"ПАДАТЬ2\" LEMMA=\"ПАДАТЬ\" LINK=\"релят\">падает</W>\n",
       " <W DOM=\"5\" FEAT=\"PR\" ID=\"6\" KSNAME=\"ДО1\" LEMMA=\"ДО\" LINK=\"2-компл\">до</W>\n",
       " <W DOM=\"8\" FEAT=\"S ЕД МУЖ ИМ НЕОД\" ID=\"7\" KSNAME=\"ПЛЮС1\" LEMMA=\"ПЛЮС\" LINK=\"огранич\">плюс</W>\n",
       " <W DOM=\"6\" FEAT=\"NUM\" ID=\"8\" KSNAME=\"ЧИСЛ-БОЛЬШ\" LEMMA=\"10\" LINK=\"предл\">10</W>, \n",
       " <W DOM=\"11\" FEAT=\"S МН ИМ ОД\" ID=\"9\" LEMMA=\"ОНИ\" LINK=\"предик\">они</W>\n",
       " <W DOM=\"11\" FEAT=\"ADV\" ID=\"10\" LEMMA=\"СРОЧНО\" LINK=\"обст\">срочно</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ МН 3-Л\" ID=\"11\" LEMMA=\"ВОДРУЖАТЬ\">водружают</W>\n",
       " <W DOM=\"13\" FEAT=\"A МН ВИН НЕОД\" ID=\"12\" LEMMA=\"ПЕСЦОВЫЙ\" LINK=\"опред\">песцовые</W>\n",
       " <W DOM=\"11\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"13\" LEMMA=\"ШУБА\" LINK=\"1-компл\">шубы</W>\n",
       " <W DOM=\"17\" FEAT=\"A МН РОД\" ID=\"14\" LEMMA=\"РОЗОВЫЙ\" LINK=\"опред\">розовых</W>\n",
       " <W DOM=\"14\" FEAT=\"CONJ\" ID=\"15\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"15\" FEAT=\"A МН РОД\" ID=\"16\" LEMMA=\"БИРЮЗОВЫЙ\" LINK=\"соч-союзн\">бирюзовых</W>\n",
       " <W DOM=\"13\" FEAT=\"S МН МУЖ РОД НЕОД\" ID=\"17\" KSNAME=\"ТОН1\" LEMMA=\"ТОН\" LINK=\"атриб\">тонов</W>\n",
       " <W DOM=\"11\" FEAT=\"PR\" ID=\"18\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"2-компл\">на</W>\n",
       " <W DOM=\"21\" FEAT=\"A МН ВИН НЕОД\" ID=\"19\" LEMMA=\"СВОЙ\" LINK=\"опред\">свои</W> \n",
       " \"<W DOM=\"21\" FEAT=\"A МН ВИН НЕОД\" ID=\"20\" LEMMA=\"ХРУПКИЙ\" LINK=\"опред\">хрупкие</W>\" \n",
       " <W DOM=\"18\" FEAT=\"S МН СРЕД ВИН НЕОД\" ID=\"21\" KSNAME=\"ПЛЕЧО1\" LEMMA=\"ПЛЕЧО\" LINK=\"предл\">плечи</W>.\n",
       " </S>,\n",
       " <S ID=\"130\">\n",
       " <W DOM=\"2\" FEAT=\"S МН ВИН ОД\" ID=\"1\" LEMMA=\"ОНИ\" LINK=\"1-компл\">Их</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ МН 3-Л\" ID=\"2\" KSNAME=\"ВЫПИСЫВАТЬ2\" LEMMA=\"ВЫПИСЫВАТЬ\">выписывают</W>\n",
       " <W DOM=\"2\" FEAT=\"PR\" ID=\"3\" LEMMA=\"ИЗ-ЗА\" LINK=\"2-компл\">из-за</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"4\" LEMMA=\"ГРАНИЦА\" LINK=\"предл\">границы</W>\n",
       " <W DOM=\"2\" FEAT=\"PR\" ID=\"5\" LEMMA=\"В КАЧЕСТВЕ\" LINK=\"обст\">в качестве</W>\n",
       " <W DOM=\"7\" FEAT=\"V НЕСОВ ПРИЧ НЕПРОШ ЕД МУЖ РОД\" ID=\"6\" LEMMA=\"ОБСЛУЖИВАТЬ\" LINK=\"опред\">обслуживающего</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД МУЖ РОД НЕОД\" ID=\"7\" LEMMA=\"ПЕРСОНАЛ\" LINK=\"предл\">персонала</W>.\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"146\">\n",
       " <W DOM=\"6\" FEAT=\"PR\" ID=\"1\" KSNAME=\"С1\" LEMMA=\"С\" LINK=\"обст\">С</W>\n",
       " <W DOM=\"1\" FEAT=\"S ЕД СРЕД РОД НЕОД\" ID=\"2\" KSNAME=\"НАЧАЛО1\" LEMMA=\"НАЧАЛО\" LINK=\"предл\">начала</W>\n",
       " <W DOM=\"4\" FEAT=\"NUM\" ID=\"3\" LEMMA=\"XVIII\" LINK=\"опред\">XVIII</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ РОД НЕОД\" ID=\"4\" KSNAME=\"ВЕК\" LEMMA=\"ВЕК\" LINK=\"квазиагент\">века</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД МУЖ ИМ НЕОД\" ID=\"5\" LEMMA=\"БАДЕН\" LINK=\"предик\">Баден</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"6\" LEMMA=\"ПРИОБРЕТАТЬ\">приобретает</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"7\" LEMMA=\"ПОПУЛЯРНОСТЬ\" LINK=\"1-компл\">популярность</W>\n",
       " <W DOM=\"6\" FEAT=\"PR\" ID=\"8\" KSNAME=\"У1\" LEMMA=\"У\" LINK=\"1-несобст-компл\">у</W>\n",
       " <W DOM=\"10\" FEAT=\"A ЕД ЖЕН РОД\" ID=\"9\" KSNAME=\"РУССКИЙ1\" LEMMA=\"РУССКИЙ\" LINK=\"опред\">русской</W>\n",
       " <W DOM=\"8\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"10\" KSNAME=\"ЗНАТЬ2\" LEMMA=\"ЗНАТЬ\" LINK=\"предл\">знати</W>, \n",
       " <W DOM=\"6\" FEAT=\"CONJ\" ID=\"11\" LEMMA=\"ТАК КАК\" LINK=\"обст\">так как</W>\n",
       " <W DOM=\"13\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"12\" LEMMA=\"БАДЕНСКИЙ\" LINK=\"опред\">баденская</W>\n",
       " <W DOM=\"15\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"13\" LEMMA=\"ПРИНЦЕССА\" LINK=\"предик\">принцесса</W>\n",
       " <W DOM=\"13\" FEAT=\"S ЕД ЖЕН ИМ ОД\" ID=\"14\" LEMMA=\"ЛУИЗА\" LINK=\"аппоз\">Луиза</W>\n",
       " <W DOM=\"11\" FEAT=\"V СОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"15\" KSNAME=\"ВЫХОДИТЬ1\" LEMMA=\"ВЫХОДИТЬ\" LINK=\"подч-союзн\">вышла</W>\n",
       " <W DOM=\"15\" FEAT=\"ADV\" ID=\"16\" LEMMA=\"ЗАМУЖ\" LINK=\"2-компл\">замуж</W>\n",
       " <W DOM=\"16\" FEAT=\"PR\" ID=\"17\" KSNAME=\"ЗА1\" LEMMA=\"ЗА\" LINK=\"1-компл\">за</W>\n",
       " <W DOM=\"19\" FEAT=\"A ЕД МУЖ ВИН ОД\" ID=\"18\" KSNAME=\"РУССКИЙ1\" LEMMA=\"РУССКИЙ\" LINK=\"опред\">русского</W>\n",
       " <W DOM=\"17\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"19\" LEMMA=\"ЦАРЬ\" LINK=\"предл\">царя</W>\n",
       " <W DOM=\"19\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"20\" KSNAME=\"АЛЕКСАНДР\" LEMMA=\"АЛЕКСАНДР\" LINK=\"аппоз\">Александра</W>\n",
       " <W DOM=\"20\" FEAT=\"NUM\" ID=\"21\" LEMMA=\"I\" LINK=\"вспом\">I</W>.\n",
       " <LF LFARG=\"7\" LFFUNC=\"_INCEPOPER1\" LFVAL=\"6\"/>\n",
       " </S>,\n",
       " <S ID=\"159\">\n",
       " <W DOM=\"12\" FEAT=\"CONJ\" ID=\"1\" LEMMA=\"ЕСЛИ\" LINK=\"обст\">Если</W>\n",
       " <W DOM=\"1\" FEAT=\"V СОВ ИНФ\" ID=\"2\" LEMMA=\"ВСПОМИНАТЬ\" LINK=\"инф-союзн\">вспомнить</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"3\" KSNAME=\"ГОГОЛЬ1\" LEMMA=\"ГОГОЛЬ\" LINK=\"1-компл\">Гоголя</W>, \n",
       " <W DOM=\"3\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"4\" KSNAME=\"ДОСТОЕВСКИЙ\" LEMMA=\"ДОСТОЕВСКИЙ\" LINK=\"сочин\">Достоевского</W>, \n",
       " <W DOM=\"6\" FEAT=\"PART\" ID=\"5\" KSNAME=\"ДА1\" LEMMA=\"ДА\" LINK=\"огранич\">да</W>\n",
       " <W DOM=\"9\" FEAT=\"S ЕД СРЕД ИМ НЕОД\" ID=\"6\" KSNAME=\"ЧТО2\" LEMMA=\"ЧТО\" LINK=\"огранич\">что</W>\n",
       " <W DOM=\"6\" FEAT=\"ADV\" ID=\"7\" LEMMA=\"ТАМ\" LINK=\"вспом\">там</W> - \n",
       " \"<W DOM=\"9\" FEAT=\"A ЕД ЖЕН ВИН\" ID=\"8\" KSNAME=\"ПИКОВЫЙ2\" LEMMA=\"ПИКОВЫЙ\" LINK=\"опред\">Пиковую</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД ЖЕН ВИН ОД\" ID=\"9\" LEMMA=\"ДАМА\" LINK=\"сочин\">даму</W>\", \n",
       " <W DOM=\"12\" FEAT=\"S МН ИМ ОД\" ID=\"10\" LEMMA=\"ВЫ\" LINK=\"предик\">вы</W>\n",
       " <W DOM=\"12\" FEAT=\"ADV\" ID=\"11\" LEMMA=\"СРАЗУ\" LINK=\"обст\">сразу</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ МН 2-Л\" ID=\"12\" KSNAME=\"УЗНАВАТЬ1\" LEMMA=\"УЗНАВАТЬ\">узнаете</W>\n",
       " <W DOM=\"14\" FEAT=\"A МН ВИН ОД\" ID=\"13\" KSNAME=\"ВЕСЬ\" LEMMA=\"ВЕСЬ\" LINK=\"опред\">всех</W>\n",
       " <W DOM=\"12\" FEAT=\"S МН МУЖ ВИН ОД\" ID=\"14\" KSNAME=\"ПЕРСОНАЖ1\" LEMMA=\"ПЕРСОНАЖ\" LINK=\"1-компл\">персонажей</W>, \n",
       " <W DOM=\"12\" FEAT=\"A КР ЕД СРЕД\" ID=\"15\" LEMMA=\"ДОСТАТОЧНЫЙ\" LINK=\"сент-соч\">достаточно</W>\n",
       " <W DOM=\"15\" FEAT=\"V СОВ ИНФ\" ID=\"16\" KSNAME=\"ЗАХОДИТЬ1\" LEMMA=\"ЗАХОДИТЬ\" LINK=\"предик\">зайти</W>\n",
       " <W DOM=\"16\" FEAT=\"PR\" ID=\"17\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"17\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"18\" LEMMA=\"КАЗИНО\" LINK=\"предл\">казино</W>\n",
       " <W DOM=\"16\" FEAT=\"PR\" ID=\"19\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"обст\">на</W>\n",
       " <W DOM=\"21\" FEAT=\"NUM СЛ\" ID=\"20\" KSNAME=\"ПОЛ3\" LEMMA=\"ПОЛ\" LINK=\"количест\">пол</W>\n",
       " <W DOM=\"19\" FEAT=\"S ЕД МУЖ РОД НЕОД\" ID=\"21\" KSNAME=\"ЧАС\" LEMMA=\"ЧАС\" LINK=\"предл\">часа</W>!\n",
       " </S>,\n",
       " <S ID=\"173\">\n",
       " <W DOM=\"2\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"1\" LEMMA=\"ПРЕСЫЩЕННЫЙ\" LINK=\"опред\">Пресыщенная</W>\n",
       " <W DOM=\"8\" FEAT=\"S ЕД ЖЕН ИМ НЕОД\" ID=\"2\" LEMMA=\"ПУБЛИКА\" LINK=\"предик\">публика</W>, \n",
       " <W DOM=\"8\" FEAT=\"V НЕСОВ ДЕЕПР НЕПРОШ\" ID=\"3\" LEMMA=\"ПОПИВАТЬ\" LINK=\"обст\">попивая</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"4\" LEMMA=\"ШАМПАНСКОЕ\" LINK=\"1-компл\">шампанское</W>\n",
       " <W DOM=\"4\" FEAT=\"CONJ\" ID=\"5\" KSNAME=\"ИЛИ1\" LEMMA=\"ИЛИ\" LINK=\"сочин\">или</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"6\" LEMMA=\"ВИНО\" LINK=\"соч-союзн\">вино</W>, \n",
       " <W DOM=\"8\" FEAT=\"ADV\" ID=\"7\" LEMMA=\"ЛЕНИВО\" LINK=\"обст\">лениво</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"8\" LEMMA=\"СЛУШАТЬ\">слушает</W>, \n",
       " <W DOM=\"10\" FEAT=\"ADV\" ID=\"9\" LEMMA=\"БЛАГОРОДНО\" LINK=\"обст\">благородно</W>\n",
       " <W DOM=\"8\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"10\" KSNAME=\"КЛАСТЬ\" LEMMA=\"КЛАСТЬ\" LINK=\"сочин\">кладет</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"11\" KSNAME=\"ПО4\" LEMMA=\"ПО\" LINK=\"1-компл\">по</W>\n",
       " <W DOM=\"13\" FEAT=\"NUM\" ID=\"12\" KSNAME=\"ЧИСЛ-БОЛЬШ\" LEMMA=\"5\" LINK=\"количест\">5</W>\n",
       " <W DOM=\"11\" FEAT=\"S МН МУЖ РОД НЕОД\" ID=\"13\" KSNAME=\"ЕВРО2\" LEMMA=\"ЕВРО\" LINK=\"предл\">евро</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"14\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"16\" FEAT=\"A ЕД ЖЕН ВИН\" ID=\"15\" LEMMA=\"ОБЩИЙ\" LINK=\"опред\">общую</W>\n",
       " <W DOM=\"14\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"16\" LEMMA=\"КАССА\" LINK=\"предл\">кассу</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"17\" KSNAME=\"ЗА1\" LEMMA=\"ЗА\" LINK=\"обст\">за</W>\n",
       " <W DOM=\"19\" FEAT=\"A ЕД СРЕД ВИН\" ID=\"18\" KSNAME=\"ФИКТ-ПРИЛ\" LEMMA=\"2-ЧАСОВОЙ\" LINK=\"опред\">2-часовое</W>\n",
       " <W DOM=\"17\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"19\" KSNAME=\"ВЫСТУПЛЕНИЕ1\" LEMMA=\"ВЫСТУПЛЕНИЕ\" LINK=\"предл\">выступление</W>\n",
       " <W DOM=\"10\" FEAT=\"CONJ\" ID=\"20\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"20\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"21\" KSNAME=\"ПЕРЕХОДИТЬ1\" LEMMA=\"ПЕРЕХОДИТЬ\" LINK=\"соч-союзн\">переходит</W>, \n",
       " <W DOM=\"23\" FEAT=\"ADV\" ID=\"22\" LEMMA=\"ОЧЕНЬ\" LINK=\"огранич\">очень</W>\n",
       " <W DOM=\"21\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"23\" LEMMA=\"ДОВОЛЬНЫЙ\" LINK=\"суб-копр\">довольная</W>, \n",
       " <W DOM=\"21\" FEAT=\"PR\" ID=\"24\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"24\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"25\" LEMMA=\"КАЗИНО\" LINK=\"предл\">казино</W>\n",
       " <W DOM=\"21\" FEAT=\"V НЕСОВ ИНФ\" ID=\"26\" LEMMA=\"ИГРАТЬ\" LINK=\"обст\">играть</W>\n",
       " <W DOM=\"26\" FEAT=\"PR\" ID=\"27\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"обст\">на</W>\n",
       " <W DOM=\"27\" FEAT=\"S МН МУЖ ВИН НЕОД\" ID=\"28\" LEMMA=\"МИЛЛИОН\" LINK=\"предл\">миллионы</W>.\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"186\">\n",
       " <W DOM=\"5\" FEAT=\"PR\" ID=\"1\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"обст\">В</W>\n",
       " <W DOM=\"1\" FEAT=\"S ЕД МУЖ ПР НЕОД\" ID=\"2\" LEMMA=\"КОНЕЦ\" LINK=\"предл\">конце</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"3\" LEMMA=\"ПРОЦЕДУРА\" LINK=\"квазиагент\">процедуры</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"4\" LEMMA=\"ТЫ\" LINK=\"1-компл\">тебя</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ МН 3-Л\" ID=\"5\" KSNAME=\"ЗАВОРАЧИВАТЬ2\" LEMMA=\"ЗАВОРАЧИВАТЬ\">заворачивают</W>\n",
       " <W DOM=\"5\" FEAT=\"PR\" ID=\"6\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"7\" LEMMA=\"ПРОСТЫНЯ\" LINK=\"предл\">простыню</W>, \n",
       " <W DOM=\"5\" FEAT=\"CONJ\" ID=\"8\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сент-соч\">и</W>\n",
       " <W DOM=\"10\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"9\" LEMMA=\"ТЫ\" LINK=\"предик\">ты</W>\n",
       " <W DOM=\"8\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 2-Л\" ID=\"10\" KSNAME=\"СПАТЬ\" LEMMA=\"СПАТЬ\" LINK=\"соч-союзн\">спишь</W>\n",
       " <W DOM=\"12\" FEAT=\"NUM СЛ\" ID=\"11\" KSNAME=\"ПОЛ3\" LEMMA=\"ПОЛ\" LINK=\"количест\">пол</W>\n",
       " <W DOM=\"10\" FEAT=\"S ЕД МУЖ РОД НЕОД\" ID=\"12\" KSNAME=\"ЧАС\" LEMMA=\"ЧАС\" LINK=\"длительн\">часа</W>.\n",
       " <LF LFARG=\"2\" LFFUNC=\"_LOC\" LFVAL=\"1\"/>\n",
       " </S>,\n",
       " <S ID=\"193\">\n",
       " <W DOM=\"2\" FEAT=\"A МН ИМ\" ID=\"1\" LEMMA=\"ВОСТОЧНЫЙ\" LINK=\"опред\">Восточные</W>\n",
       " <W DOM=\"10\" FEAT=\"S МН МУЖ ИМ ОД\" ID=\"2\" LEMMA=\"МУЖЧИНА\" LINK=\"предик\">мужчины</W>\n",
       " <W DOM=\"2\" FEAT=\"PART\" ID=\"3\" LEMMA=\"ЖЕ\" LINK=\"огранич\">же</W>, \n",
       " <W DOM=\"10\" FEAT=\"V СОВ ДЕЕПР ПРОШ\" ID=\"4\" LEMMA=\"ОСТАВЛЯТЬ\" LINK=\"обст\">оставив</W>\n",
       " <W DOM=\"7\" FEAT=\"A МН ВИН ОД\" ID=\"5\" LEMMA=\"СВОЙ\" LINK=\"опред\">своих</W>\n",
       " <W DOM=\"7\" FEAT=\"A МН ВИН ОД\" ID=\"6\" LEMMA=\"СМУГЛЫЙ\" LINK=\"опред\">смуглых</W>\n",
       " <W DOM=\"4\" FEAT=\"S МН ЖЕН ВИН ОД\" ID=\"7\" LEMMA=\"ЖЕНА\" LINK=\"1-компл\">жен</W>\n",
       " <W DOM=\"7\" FEAT=\"CONJ\" ID=\"8\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"8\" FEAT=\"S МН МУЖ ВИН ОД\" ID=\"9\" LEMMA=\"РЕБЕНОК\" LINK=\"соч-союзн\">детей</W>, \n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ МН 3-Л\" ID=\"10\" LEMMA=\"СТРЕМИТЬСЯ\">стремятся</W>\n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"11\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"2-компл\">в</W>\n",
       " <W DOM=\"11\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"12\" LEMMA=\"САУНА\" LINK=\"предл\">сауны</W> - \n",
       " <W DOM=\"14\" FEAT=\"ADV\" ID=\"13\" LEMMA=\"ВДОВОЛЬ\" LINK=\"обст\">вдоволь</W>\n",
       " <W DOM=\"10\" FEAT=\"V СОВ ИНФ\" ID=\"14\" LEMMA=\"НАСМОТРЕТЬСЯ\" LINK=\"разъяснит\">насмотреться</W>\n",
       " <W DOM=\"14\" FEAT=\"PR\" ID=\"15\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"1-компл\">на</W>\n",
       " <W DOM=\"17\" FEAT=\"A МН ВИН ОД\" ID=\"16\" LEMMA=\"ЕВРОПЕЙСКИЙ\" LINK=\"опред\">европейских</W>\n",
       " <W DOM=\"15\" FEAT=\"S МН ЖЕН ВИН ОД\" ID=\"17\" LEMMA=\"ЖЕНЩИНА\" LINK=\"предл\">женщин</W>\n",
       " <W DOM=\"17\" FEAT=\"PR\" ID=\"18\" KSNAME=\"В2\" LEMMA=\"В\" LINK=\"атриб\">в</W>\n",
       " <W DOM=\"18\" FEAT=\"S ЕД СРЕД ПР НЕОД\" ID=\"19\" KSNAME=\"НЕГЛИЖЕ1\" LEMMA=\"НЕГЛИЖЕ\" LINK=\"предл\">неглиже</W>.\n",
       " </S>,\n",
       " <S ID=\"197\">\n",
       " <W DOM=\"2\" FEAT=\"A ЕД ЖЕН ИМ\" ID=\"1\" LEMMA=\"ПОСЛЕДНИЙ\" LINK=\"опред\">Последняя</W>\n",
       " <W DOM=\"19\" FEAT=\"S ЕД ЖЕН ИМ НЕОД\" ID=\"2\" LEMMA=\"ВЫСТАВКА\" LINK=\"предик\">выставка</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ РОД ОД\" ID=\"3\" KSNAME=\"РИХТЕР1\" LEMMA=\"РИХТЕР\" LINK=\"1-компл\">Рихтера</W>, \n",
       " <W DOM=\"2\" FEAT=\"V СОВ СТРАД ПРИЧ ПРОШ ЕД ЖЕН ИМ\" ID=\"4\" KSNAME=\"ПОСВЯЩАТЬ1\" LEMMA=\"ПОСВЯЩАТЬ\" LINK=\"опред\">посвященная</W>\n",
       " <W DOM=\"4\" FEAT=\"S МН МУЖ ДАТ НЕОД\" ID=\"5\" LEMMA=\"ЛАГЕРЬ\" LINK=\"2-компл\">лагерям</W>\n",
       " <W DOM=\"5\" FEAT=\"CONJ\" ID=\"6\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"6\" FEAT=\"S ЕД МУЖ ДАТ НЕОД\" ID=\"7\" LEMMA=\"ФАШИЗМ\" LINK=\"соч-союзн\">фашизму</W>, \n",
       " <W DOM=\"2\" FEAT=\"PR\" ID=\"8\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"атриб\">с</W>\n",
       " <W DOM=\"14\" FEAT=\"V СОВ СТРАД ПРИЧ ПРОШ МН ТВОР\" ID=\"9\" KSNAME=\"РАСКАТЫВАТЬ2\" LEMMA=\"РАСКАТЫВАТЬ\" LINK=\"опред\">раскатанными</W>\n",
       " <W DOM=\"9\" FEAT=\"PR\" ID=\"10\" KSNAME=\"ПО1\" LEMMA=\"ПО\" LINK=\"обст\">по</W>\n",
       " <W DOM=\"12\" FEAT=\"A МН ДАТ\" ID=\"11\" LEMMA=\"ОГРОМНЫЙ\" LINK=\"опред\">огромным</W>\n",
       " <W DOM=\"10\" FEAT=\"S МН МУЖ ДАТ НЕОД\" ID=\"12\" LEMMA=\"ХОЛСТ\" LINK=\"предл\">холстам</W>\n",
       " <W DOM=\"14\" FEAT=\"A ЕД ЖЕН ТВОР\" ID=\"13\" LEMMA=\"БОРДОВЫЙ\" LINK=\"опред\">бордовой</W>\n",
       " <W DOM=\"8\" FEAT=\"S ЕД ЖЕН ТВОР НЕОД\" ID=\"14\" LEMMA=\"КРАСКА\" LINK=\"предл\">краской</W>\n",
       " <W DOM=\"14\" FEAT=\"CONJ\" ID=\"15\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"15\" FEAT=\"S ЕД МУЖ ТВОР НЕОД\" ID=\"16\" LEMMA=\"ПЕСОК\" LINK=\"соч-союзн\">песком</W>, \n",
       " <W DOM=\"14\" FEAT=\"V НЕСОВ ПРИЧ НЕПРОШ МН ТВОР\" ID=\"17\" LEMMA=\"ИМИТИРОВАТЬ\" LINK=\"опред\">имитирующими</W>\n",
       " <W DOM=\"17\" FEAT=\"S ЕД ЖЕН ВИН НЕОД\" ID=\"18\" LEMMA=\"КРОВЬ\" LINK=\"1-компл\">кровь</W>, \n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ ПРОШ ЕД ЖЕН\" ID=\"19\" KSNAME=\"БЫТЬ\" LEMMA=\"БЫТЬ\">была</W>\n",
       " <W DOM=\"19\" FEAT=\"A КР ЕД ЖЕН\" ID=\"20\" LEMMA=\"ЧУДОВИЩНЫЙ\" LINK=\"присвяз\">чудовищна</W>.\n",
       " </S>,\n",
       " <S CLASS=\"LF\" ID=\"223\">\n",
       " <W DOM=\"_root\" FEAT=\"CONJ\" ID=\"1\" LEMMA=\"НО\">Но</W>\n",
       " <W DOM=\"5\" FEAT=\"PART\" ID=\"2\" LEMMA=\"ВСЕ ЖЕ\" LINK=\"огранич\">все же</W>\n",
       " <W DOM=\"5\" FEAT=\"PR\" ID=\"3\" KSNAME=\"В1\" LEMMA=\"В\" LINK=\"обст\">в</W>\n",
       " <W DOM=\"3\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"4\" KSNAME=\"ЗАКЛЮЧЕНИЕ2\" LEMMA=\"ЗАКЛЮЧЕНИЕ\" LINK=\"предл\">заключение</W>\n",
       " <W DOM=\"1\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 1-Л\" ID=\"5\" LEMMA=\"ХОТЕТЬ\" LINK=\"соч-союзн\">хочу</W>\n",
       " <W DOM=\"5\" FEAT=\"V СОВ ИНФ\" ID=\"6\" LEMMA=\"СКАЗАТЬ\" LINK=\"1-компл\">сказать</W>, \n",
       " <W DOM=\"6\" FEAT=\"CONJ\" ID=\"7\" KSNAME=\"ЧТО1\" LEMMA=\"ЧТО\" LINK=\"1-компл\">что</W>\n",
       " <W DOM=\"9\" FEAT=\"ADV\" ID=\"8\" LEMMA=\"ЧАСТО\" LINK=\"обст\">часто</W>\n",
       " <W DOM=\"7\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 1-Л\" ID=\"9\" LEMMA=\"СМОТРЕТЬ\" LINK=\"подч-союзн\">смотрю</W>\n",
       " <W DOM=\"9\" FEAT=\"PR\" ID=\"10\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"1-компл\">на</W>\n",
       " <W DOM=\"12\" FEAT=\"A ЕД СРЕД ВИН\" ID=\"11\" LEMMA=\"ДИВНЫЙ\" LINK=\"опред\">дивное</W>\n",
       " <W DOM=\"10\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"12\" KSNAME=\"НЕБО1\" LEMMA=\"НЕБО\" LINK=\"предл\">небо</W>\n",
       " <W DOM=\"12\" FEAT=\"PR\" ID=\"13\" KSNAME=\"НАД\" LEMMA=\"НАД\" LINK=\"атриб\">над</W>\n",
       " <W DOM=\"13\" FEAT=\"S ЕД МУЖ ТВОР НЕОД\" ID=\"14\" LEMMA=\"БАДЕН\" LINK=\"предл\">Баденом</W>, \n",
       " <W DOM=\"12\" FEAT=\"V СОВ СТРАД ПРИЧ ПРОШ ЕД СРЕД ВИН\" ID=\"15\" LEMMA=\"ПОКРЫВАТЬ\" LINK=\"опред\">покрытое</W>\n",
       " <W DOM=\"18\" FEAT=\"A МН ТВОР\" ID=\"16\" LEMMA=\"ЛЕГКИЙ\" LINK=\"опред\">легкими</W>\n",
       " <W DOM=\"18\" FEAT=\"A МН ТВОР\" ID=\"17\" LEMMA=\"ПЕРИСТЫЙ\" LINK=\"опред\">перистыми</W>\n",
       " <W DOM=\"15\" FEAT=\"S МН СРЕД ТВОР НЕОД\" ID=\"18\" LEMMA=\"ОБЛАКО\" LINK=\"2-компл\">облаками</W>, \n",
       " <W DOM=\"10\" FEAT=\"PR\" ID=\"19\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"сочин\">на</W>\n",
       " <W DOM=\"21\" FEAT=\"A МН ВИН НЕОД\" ID=\"20\" LEMMA=\"РОЗОВЫЙ\" LINK=\"опред\">розовые</W>\n",
       " <W DOM=\"19\" FEAT=\"S МН МУЖ ВИН НЕОД\" ID=\"21\" LEMMA=\"ЗАКАТ\" LINK=\"предл\">закаты</W>, \n",
       " <W DOM=\"19\" FEAT=\"PR\" ID=\"22\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"сочин\">на</W>\n",
       " <W DOM=\"24\" FEAT=\"A ЕД ЖЕН ВИН\" ID=\"23\" LEMMA=\"СЕРЕБРЯНЫЙ\" LINK=\"опред\">серебряную</W>\n",
       " <W DOM=\"22\" FEAT=\"S ЕД ЖЕН ВИН ОД\" ID=\"24\" KSNAME=\"ЗМЕЙКА1\" LEMMA=\"ЗМЕЙКА\" LINK=\"предл\">змейку</W>\n",
       " <W DOM=\"24\" FEAT=\"S ЕД ЖЕН РОД НЕОД\" ID=\"25\" LEMMA=\"РЕКА\" LINK=\"атриб\">реки</W>, \n",
       " <W DOM=\"22\" FEAT=\"PR\" ID=\"26\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"сочин\">на</W>\n",
       " <W DOM=\"28\" FEAT=\"A МН ВИН НЕОД\" ID=\"27\" LEMMA=\"ВЕКОВОЙ\" LINK=\"опред\">вековые</W>\n",
       " <W DOM=\"26\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"28\" LEMMA=\"СЕКВОЙЯ\" LINK=\"предл\">секвойи</W>\n",
       " <W DOM=\"28\" FEAT=\"CONJ\" ID=\"29\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"29\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"30\" LEMMA=\"СОСНА\" LINK=\"соч-союзн\">сосны</W>, \n",
       " <W DOM=\"9\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 1-Л\" ID=\"31\" LEMMA=\"ВСПОМИНАТЬ\" LINK=\"сочин\">вспоминаю</W>\n",
       " <W DOM=\"34\" FEAT=\"A ЕД МУЖ ВИН ОД\" ID=\"32\" LEMMA=\"СВОЙ\" LINK=\"опред\">своего</W>\n",
       " <W DOM=\"34\" FEAT=\"A ЕД МУЖ ВИН ОД\" ID=\"33\" KSNAME=\"ЛЮБИМЫЙ1\" LEMMA=\"ЛЮБИМЫЙ\" LINK=\"опред\">любимого</W>\n",
       " <W DOM=\"31\" FEAT=\"S ЕД МУЖ ВИН ОД\" ID=\"34\" KSNAME=\"ОТЕЦ1\" LEMMA=\"ОТЕЦ\" LINK=\"1-компл\">отца</W>\n",
       " <W DOM=\"31\" FEAT=\"CONJ\" ID=\"35\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сочин\">и</W>\n",
       " <W DOM=\"35\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 1-Л\" ID=\"36\" LEMMA=\"ЗАБЫВАТЬ\" LINK=\"соч-союзн\">забываю</W>\n",
       " <W DOM=\"36\" FEAT=\"PR\" ID=\"37\" KSNAME=\"О2\" LEMMA=\"О\" LINK=\"2-компл\">о</W>\n",
       " <W DOM=\"37\" FEAT=\"S ЕД ЖЕН ПР НЕОД\" ID=\"38\" LEMMA=\"ГРУСТЬ\" LINK=\"предл\">грусти</W>, \n",
       " <W DOM=\"37\" FEAT=\"PR\" ID=\"39\" KSNAME=\"О2\" LEMMA=\"О\" LINK=\"сочин\">о</W>\n",
       " <W DOM=\"39\" FEAT=\"S МН СРЕД ПР НЕОД\" ID=\"40\" LEMMA=\"СОЖАЛЕНИЕ\" LINK=\"предл\">сожалениях</W>, \n",
       " <W DOM=\"39\" FEAT=\"PR\" ID=\"41\" KSNAME=\"О2\" LEMMA=\"О\" LINK=\"сочин\">об</W>\n",
       " <W DOM=\"43\" FEAT=\"V СОВ ПРИЧ ПРОШ МН ПР\" ID=\"42\" KSNAME=\"УХОДИТЬ1\" LEMMA=\"УХОДИТЬ\" LINK=\"опред\">ушедших</W>\n",
       " <W DOM=\"41\" FEAT=\"S МН МУЖ ПР ОД\" ID=\"43\" LEMMA=\"ЧЕЛОВЕК\" LINK=\"предл\">людях</W>, \n",
       " <W DOM=\"41\" FEAT=\"PR\" ID=\"44\" KSNAME=\"О2\" LEMMA=\"О\" LINK=\"сочин\">о</W>\n",
       " <W DOM=\"46\" FEAT=\"A МН ПР\" ID=\"45\" LEMMA=\"ПОТЕРЯННЫЙ\" LINK=\"опред\">потерянных</W>\n",
       " <W DOM=\"44\" FEAT=\"S МН СРЕД ПР НЕОД\" ID=\"46\" KSNAME=\"ОТНОШЕНИЕ1\" LEMMA=\"ОТНОШЕНИЕ\" LINK=\"предл\">отношениях</W>.\n",
       " <LF LFARG=\"18\" LFFUNC=\"_ANTIMAGN\" LFVAL=\"16\"/>\n",
       " </S>,\n",
       " <S ID=\"225\">\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ИМ ОД\" ID=\"1\" LEMMA=\"Я\" LINK=\"предик\">Я</W>\n",
       " <W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 1-Л\" ID=\"2\" LEMMA=\"ВСПОМИНАТЬ\">вспоминаю</W>\n",
       " <W DOM=\"4\" FEAT=\"A ЕД МУЖ ВИН НЕОД\" ID=\"3\" LEMMA=\"ЗНАМЕНИТЫЙ\" LINK=\"опред\">знаменитый</W>\n",
       " <W DOM=\"2\" FEAT=\"S ЕД МУЖ ВИН НЕОД\" ID=\"4\" LEMMA=\"АКАФИСТ\" LINK=\"1-компл\">Акафист</W>\n",
       " <W DOM=\"4\" FEAT=\"S ЕД МУЖ РОД ОД\" ID=\"5\" LEMMA=\"МИТРОПОЛИТ\" LINK=\"квазиагент\">митрополита</W>\n",
       " <W DOM=\"5\" FEAT=\"S ЕД МУЖ РОД ОД\" ID=\"6\" LEMMA=\"ТРИФОН\" LINK=\"аппоз\">Трифона</W> \n",
       " \"<W DOM=\"4\" FEAT=\"S ЕД ЖЕН ИМ НЕОД\" ID=\"7\" KSNAME=\"СЛАВА1\" LEMMA=\"СЛАВА\" LINK=\"аппоз\">Слава</W>\n",
       " <W DOM=\"7\" FEAT=\"S ЕД МУЖ ДАТ ОД\" ID=\"8\" LEMMA=\"БОГ\" LINK=\"дат-сент\">Богу</W>\n",
       " <W DOM=\"7\" FEAT=\"PR\" ID=\"9\" KSNAME=\"ЗА1\" LEMMA=\"ЗА\" LINK=\"атриб\">за</W>\n",
       " <W DOM=\"9\" FEAT=\"S ЕД СРЕД ВИН НЕОД\" ID=\"10\" KSNAME=\"ВСЕ1\" LEMMA=\"ВСЕ\" LINK=\"предл\">все</W>!\", \n",
       " <W DOM=\"2\" FEAT=\"CONJ\" ID=\"11\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сент-соч\">и</W>\n",
       " <W DOM=\"13\" FEAT=\"S ЕД СРЕД ИМ НЕОД\" ID=\"12\" KSNAME=\"ВСЕ1\" LEMMA=\"ВСЕ\" LINK=\"предик\">все</W>\n",
       " <W DOM=\"11\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"13\" LEMMA=\"ВСТАВАТЬ\" LINK=\"соч-союзн\">встает</W>\n",
       " <W DOM=\"13\" FEAT=\"PR\" ID=\"14\" KSNAME=\"НА1\" LEMMA=\"НА\" LINK=\"2-компл\">на</W>\n",
       " <W DOM=\"16\" FEAT=\"A МН ВИН НЕОД\" ID=\"15\" LEMMA=\"СВОЙ\" LINK=\"опред\">свои</W>\n",
       " <W DOM=\"14\" FEAT=\"S МН СРЕД ВИН НЕОД\" ID=\"16\" LEMMA=\"МЕСТО\" LINK=\"предл\">места</W>, \n",
       " <W DOM=\"13\" FEAT=\"CONJ\" ID=\"17\" KSNAME=\"И1\" LEMMA=\"И\" LINK=\"сент-соч\">и</W>\n",
       " <W DOM=\"17\" FEAT=\"ADV\" ID=\"18\" LEMMA=\"МОЖНО\" LINK=\"соч-союзн\">можно</W>\n",
       " <W DOM=\"18\" FEAT=\"V НЕСОВ ИНФ\" ID=\"19\" LEMMA=\"ЖИТЬ\" LINK=\"предик\">жить</W>\n",
       " <W DOM=\"19\" FEAT=\"ADV\" ID=\"20\" LEMMA=\"ДАЛЬШЕ\" LINK=\"обст\">дальше</W>…\n",
       " </S>]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "another-forum",
   "metadata": {},
   "outputs": [],
   "source": [
    "cur_sent = sents[1]\n",
    "num_left = 2\n",
    "num_right = 2\n",
    "\n",
    "left = []\n",
    "right = []\n",
    "for i, prev_ in enumerate(cur_sent.previous_siblings):\n",
    "    if prev_ is not None and not isinstance(prev_, str):\n",
    "        left.append(prev_)\n",
    "        \n",
    "    if len(left) >= num_left or prev_ is None:\n",
    "        break\n",
    "\n",
    "for i, next_ in enumerate(cur_sent.next_siblings):\n",
    "    if next_ is not None and not isinstance(next_, str):\n",
    "        right.append(next_)\n",
    "        \n",
    "    if len(right) >= num_right or next_ is None:\n",
    "        break\n",
    "\n",
    "left_context = '' if not left else ' '.join(get_text_from_sentence(prev_) for prev_ in left)\n",
    "right_context = '' if not right else ' '.join(get_text_from_sentence(next_) for next_ in right)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "million-excellence",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2'"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cur_sent['ID']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "collect-alarm",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Мария Орлова'"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "left_context"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "transsexual-tower",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Вот уже полтора года живу я в этом красивом, зеленом курортном, провинциальном, холодном и чопорном городишке. \"Здесь русский дух…\"'"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "right_context"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "automatic-basement",
   "metadata": {},
   "outputs": [],
   "source": [
    "res, tokens, text = parse_sent(sents[45], get_pos_to_feats_format(), return_tokens_list_text=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "legendary-subscription",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "rocky-worth",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sentence([Word({'form': 'лошадей', 'id': 1, 'dom': 2, 'link': 'предик', 'lemma': 'лошадь', 'feat': {'gender': 'жен', 'number': 'мн', 'case': 'род', 'animacy': 'од'}, 'pos': 'S'}, children={1: [Word({'form': 'породистых', 'id': 3, 'dom': 1, 'link': 'оп-опред', 'lemma': 'породистый', 'feat': {'number': 'мн', 'case': 'род'}, 'pos': 'A'}, children={1: [Word({'form': 'и', 'id': 4, 'dom': 3, 'link': 'сочин', 'lemma': 'и', 'feat': {}, 'pos': 'CONJ'}, children={1: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 3: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'и', 'id': 4, 'dom': 3, 'link': 'сочин', 'lemma': 'и', 'feat': {}, 'pos': 'CONJ'}, children={1: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 3: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 4: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]}), Word({'form': 'много', 'id': 2, 'dom': _root, 'link': '_root', 'lemma': 'много', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'лошадей', 'id': 1, 'dom': 2, 'link': 'предик', 'lemma': 'лошадь', 'feat': {'gender': 'жен', 'number': 'мн', 'case': 'род', 'animacy': 'од'}, 'pos': 'S'}, children={1: [Word({'form': 'породистых', 'id': 3, 'dom': 1, 'link': 'оп-опред', 'lemma': 'породистый', 'feat': {'number': 'мн', 'case': 'род'}, 'pos': 'A'}, children={1: [Word({'form': 'и', 'id': 4, 'dom': 3, 'link': 'сочин', 'lemma': 'и', 'feat': {}, 'pos': 'CONJ'}, children={1: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 3: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'и', 'id': 4, 'dom': 3, 'link': 'сочин', 'lemma': 'и', 'feat': {}, 'pos': 'CONJ'}, children={1: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 3: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 4: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]}), Word({'form': 'отводят', 'id': 10, 'dom': 2, 'link': 'сент-соч', 'lemma': 'отводить', 'feat': {'mood': 'изъяв', 'aspect': 'несов', 'tense': 'непрош', 'person': '3-л', 'number': 'мн'}, 'pos': 'V'}, children={1: [Word({'form': 'утро', 'id': 8, 'dom': 10, 'link': 'обст', 'lemma': 'утро', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={1: [Word({'form': 'каждое', 'id': 7, 'dom': 8, 'link': 'опред', 'lemma': 'каждый', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин'}, 'pos': 'PRON'}, children={})]}), Word({'form': 'их', 'id': 9, 'dom': 10, 'link': '1-компл', 'lemma': 'они', 'feat': {'number': 'мн', 'case': 'вин', 'animacy': 'од'}, 'pos': 'PRON'}, children={}), Word({'form': 'на', 'id': 11, 'dom': 10, 'link': '3-компл', 'lemma': 'на', 'feat': {}, 'pos': 'PR'}, children={1: [Word({'form': 'выпас', 'id': 12, 'dom': 11, 'link': 'предл', 'lemma': 'выпас', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={})]})], 2: [Word({'form': 'каждое', 'id': 7, 'dom': 8, 'link': 'опред', 'lemma': 'каждый', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин'}, 'pos': 'PRON'}, children={}), Word({'form': 'выпас', 'id': 12, 'dom': 11, 'link': 'предл', 'lemma': 'выпас', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={})]})], 2: [Word({'form': 'породистых', 'id': 3, 'dom': 1, 'link': 'оп-опред', 'lemma': 'породистый', 'feat': {'number': 'мн', 'case': 'род'}, 'pos': 'A'}, children={1: [Word({'form': 'и', 'id': 4, 'dom': 3, 'link': 'сочин', 'lemma': 'и', 'feat': {}, 'pos': 'CONJ'}, children={1: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 3: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]}), Word({'form': 'утро', 'id': 8, 'dom': 10, 'link': 'обст', 'lemma': 'утро', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={1: [Word({'form': 'каждое', 'id': 7, 'dom': 8, 'link': 'опред', 'lemma': 'каждый', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин'}, 'pos': 'PRON'}, children={})]}), Word({'form': 'их', 'id': 9, 'dom': 10, 'link': '1-компл', 'lemma': 'они', 'feat': {'number': 'мн', 'case': 'вин', 'animacy': 'од'}, 'pos': 'PRON'}, children={}), Word({'form': 'на', 'id': 11, 'dom': 10, 'link': '3-компл', 'lemma': 'на', 'feat': {}, 'pos': 'PR'}, children={1: [Word({'form': 'выпас', 'id': 12, 'dom': 11, 'link': 'предл', 'lemma': 'выпас', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={})]})], 3: [Word({'form': 'и', 'id': 4, 'dom': 3, 'link': 'сочин', 'lemma': 'и', 'feat': {}, 'pos': 'CONJ'}, children={1: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]}), Word({'form': 'каждое', 'id': 7, 'dom': 8, 'link': 'опред', 'lemma': 'каждый', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин'}, 'pos': 'PRON'}, children={}), Word({'form': 'выпас', 'id': 12, 'dom': 11, 'link': 'предл', 'lemma': 'выпас', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={})], 4: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 5: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]}), Word({'form': 'породистых', 'id': 3, 'dom': 1, 'link': 'оп-опред', 'lemma': 'породистый', 'feat': {'number': 'мн', 'case': 'род'}, 'pos': 'A'}, children={1: [Word({'form': 'и', 'id': 4, 'dom': 3, 'link': 'сочин', 'lemma': 'и', 'feat': {}, 'pos': 'CONJ'}, children={1: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 3: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]}), Word({'form': 'и', 'id': 4, 'dom': 3, 'link': 'сочин', 'lemma': 'и', 'feat': {}, 'pos': 'CONJ'}, children={1: [Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]})], 2: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]}), Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={}), Word({'form': 'очень', 'id': 6, 'dom': 4, 'link': 'соч-союзн', 'lemma': 'очень', 'feat': {}, 'pos': 'ADV'}, children={1: [Word({'form': 'не', 'id': 5, 'dom': 6, 'link': 'огранич', 'lemma': 'не', 'feat': {}, 'pos': 'PART'}, children={})]}), Word({'form': 'каждое', 'id': 7, 'dom': 8, 'link': 'опред', 'lemma': 'каждый', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин'}, 'pos': 'PRON'}, children={}), Word({'form': 'утро', 'id': 8, 'dom': 10, 'link': 'обст', 'lemma': 'утро', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={1: [Word({'form': 'каждое', 'id': 7, 'dom': 8, 'link': 'опред', 'lemma': 'каждый', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин'}, 'pos': 'PRON'}, children={})]}), Word({'form': 'их', 'id': 9, 'dom': 10, 'link': '1-компл', 'lemma': 'они', 'feat': {'number': 'мн', 'case': 'вин', 'animacy': 'од'}, 'pos': 'PRON'}, children={}), Word({'form': 'отводят', 'id': 10, 'dom': 2, 'link': 'сент-соч', 'lemma': 'отводить', 'feat': {'mood': 'изъяв', 'aspect': 'несов', 'tense': 'непрош', 'person': '3-л', 'number': 'мн'}, 'pos': 'V'}, children={1: [Word({'form': 'утро', 'id': 8, 'dom': 10, 'link': 'обст', 'lemma': 'утро', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={1: [Word({'form': 'каждое', 'id': 7, 'dom': 8, 'link': 'опред', 'lemma': 'каждый', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин'}, 'pos': 'PRON'}, children={})]}), Word({'form': 'их', 'id': 9, 'dom': 10, 'link': '1-компл', 'lemma': 'они', 'feat': {'number': 'мн', 'case': 'вин', 'animacy': 'од'}, 'pos': 'PRON'}, children={}), Word({'form': 'на', 'id': 11, 'dom': 10, 'link': '3-компл', 'lemma': 'на', 'feat': {}, 'pos': 'PR'}, children={1: [Word({'form': 'выпас', 'id': 12, 'dom': 11, 'link': 'предл', 'lemma': 'выпас', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={})]})], 2: [Word({'form': 'каждое', 'id': 7, 'dom': 8, 'link': 'опред', 'lemma': 'каждый', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'вин'}, 'pos': 'PRON'}, children={}), Word({'form': 'выпас', 'id': 12, 'dom': 11, 'link': 'предл', 'lemma': 'выпас', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={})]}), Word({'form': 'на', 'id': 11, 'dom': 10, 'link': '3-компл', 'lemma': 'на', 'feat': {}, 'pos': 'PR'}, children={1: [Word({'form': 'выпас', 'id': 12, 'dom': 11, 'link': 'предл', 'lemma': 'выпас', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={})]}), Word({'form': 'выпас', 'id': 12, 'dom': 11, 'link': 'предл', 'lemma': 'выпас', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'вин', 'animacy': 'неод'}, 'pos': 'S'}, children={})])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokens"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "presidential-digest",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_text_from_sentence(sent: bs4.element.Tag) -> str:\n",
    "    return (''.join(fragment for fragment in sent.strings).strip()\n",
    "        .replace('\\n', ' ').replace('  ', ' ')\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "whole-coordinator",
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_feat(\n",
    "    feat: str,\n",
    "    pos_to_feats_formats: Dict[str, Dict[str, List[str]]],\n",
    ") -> Dict[str, str]:\n",
    "    result = {}\n",
    "    \n",
    "    features = feat.split()\n",
    "    pos = features[0]\n",
    "    \n",
    "    result['pos'] = pos\n",
    "    if pos not in pos_to_feats_formats:\n",
    "        if len(features) > 1:\n",
    "            print(f'`{pos}` not in pos_to_feats_formats, '\n",
    "                  f'but there are {len(features) - 1} values unparsed: '\n",
    "                  f'`{features[1:]}`'\n",
    "            )\n",
    "        \n",
    "        return result\n",
    "    \n",
    "    feats_left_str = ' '.join(features[1:]).lower()\n",
    "    for param, possible_values in pos_to_feats_formats[pos].items():\n",
    "\n",
    "        for value in possible_values:\n",
    "            value_regex = rf'\\b{value}\\b'\n",
    "            value_match = re.search(value_regex, feats_left_str)\n",
    "\n",
    "            if value_match:\n",
    "                result[param] = value_match.group(0)\n",
    "                feats_left_str = feats_left_str.replace(value_match.group(0), '')\n",
    "                break\n",
    "                \n",
    "    if feats_left_str.replace(' ', '') != '':  # are there only spaces left?\n",
    "        print(f\"all params from pos_to_feats_formats were parsed, \"\n",
    "              f\"but there is something left in string `{feats_left_str.strip()}` \"\n",
    "              f\"orig `{feat}`\"\n",
    "        )\n",
    "    \n",
    "    return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "reliable-finish",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Word(object):\n",
    "    def __init__(\n",
    "        self, word_dict: Dict[str, str], children: Optional[Dict[int, List['Word']]] = None,\n",
    "        mandatory_attributes = ['dom', 'id']\n",
    "    ):\n",
    "        if any(attr not in word_dict for attr in mandatory_attributes):\n",
    "            raise ValueError(f'no `dom` or `id` in word_dict `{word_dict}`')\n",
    "        \n",
    "        self.attrs = ['dom', 'feat', 'id', 'lemma', 'link', 'form']\n",
    "        self.id = word_dict['id']\n",
    "        self.dom = word_dict['dom']\n",
    "        link = word_dict.get('link')\n",
    "        self.link = link if link else '_root'\n",
    "        self.form = word_dict['form']\n",
    "        self.lemma = word_dict['lemma']\n",
    "        \n",
    "        # TODO: add condition to only do the search for non-verbs?\n",
    "        #   (to prevent `bylo` in `(kto) by to ni bylo` from messing up `bylo` verb)\n",
    "        if (any(\n",
    "                re.search(rf'\\b{modif}\\b', self.lemma)\n",
    "                for modif in PRON_ONLY_MODIFIERS\n",
    "            ) or any(self.lemma == pron for pron in PRONS)\n",
    "        ):\n",
    "            self.pos = 'PRON'\n",
    "        else:\n",
    "            self.pos = word_dict['pos'] \n",
    "        \n",
    "        self.feat = word_dict['feat']\n",
    "        if 'ksname' in word_dict:\n",
    "            self.ksname = word_dict['ksname']\n",
    "            self.attrs.append('ksname')\n",
    "        if 'nodetype'in word_dict:\n",
    "            self.nodetype = word_dict['nodetype']\n",
    "            \n",
    "        if any(attr not in self.__dict__ for attr in word_dict):\n",
    "            print(f'word_dict not parsed fully: {word_dict}')\n",
    "        \n",
    "        self.children: Dict[int, List['Word']] = {}\n",
    "        if children:\n",
    "            self.children = children\n",
    "        self.parent = None\n",
    "    \n",
    "    \n",
    "    def add_child(self, child: 'Word', depth: int):\n",
    "        self.children.setdefault(depth, []).append(child)\n",
    "        if depth == 1:\n",
    "            child.parent = self\n",
    "        if self.parent:\n",
    "            self.parent.add_child(child, depth+1) \n",
    "       \n",
    "    \n",
    "    def filter_out_appos_groups(self) -> Dict[int, List['Word']]:\n",
    "        # TODO: exclude groups that depend on target word but are linearly away\n",
    "        full_group = self.group\n",
    "        children = self.children \n",
    "        \n",
    "        words_in_appos_groups = list(chain(\n",
    "            *[appos_head.group\n",
    "              for appos_head in full_group\n",
    "              if appos_head.link == 'оп-опред']\n",
    "        ))\n",
    "        \n",
    "        # NEW\n",
    "#         potential_distant_dependents = list(chain(\n",
    "#             *[dep_head.group\n",
    "#               for dep_head in direct_dependents[0]\n",
    "#               if dep_head.id != self.id + 1\n",
    "#              ]\n",
    "#         ))\n",
    "#         potential_distant_dependents = []\n",
    "#         for pot_dist_dep in potential_distant_dependents:\n",
    "#             if pot_dist\n",
    "        # NEW: save whether word group has appositive dependents\n",
    "        if not words_in_appos_groups:\n",
    "            self.has_appos = 0\n",
    "            return children\n",
    "\n",
    "        else:\n",
    "            self.has_appos=1\n",
    "            filtered_children_dict = {}\n",
    "            for depth, children in self.children.items():\n",
    "                for child in children:\n",
    "                    if child in words_in_appos_groups:\n",
    "                        continue\n",
    "                    filtered_children_dict.setdefault(depth, []).append(child)\n",
    "\n",
    "            return filtered_children_dict\n",
    "    \n",
    "    \n",
    "    def get_children_depth(self, exclude_appos_groups=False):\n",
    "        if exclude_appos_groups:\n",
    "            children = self.filter_out_appos_groups()\n",
    "        else:\n",
    "            children = self.children\n",
    "        return max(children.keys(), default=0)    \n",
    "    children_depth = property(get_children_depth)\n",
    "    \n",
    "    def get_children_depth_no_appos(self):\n",
    "        return self.get_children_depth(exclude_appos_groups=True)\n",
    "    no_appos_children_depth = property(get_children_depth_no_appos)\n",
    "    \n",
    "    \n",
    "    def get_group(self, depth=None, add_root=True, exclude_appos_groups=False):\n",
    "        if exclude_appos_groups:\n",
    "            children_dct = self.filter_out_appos_groups()\n",
    "        else:\n",
    "            children_dct = self.children\n",
    "        if depth:\n",
    "            children = list(chain(*[children_dct[d] for d in children_dct.keys() if d <= depth]))\n",
    "        else:\n",
    "            children = list(chain(*children_dct.values()))        \n",
    "        if add_root:\n",
    "            children += [self]\n",
    "        return children\n",
    "    group = property(get_group)\n",
    "    \n",
    "    \n",
    "    def get_group_str(self, **kwargs):\n",
    "        group = self.get_group(**kwargs)\n",
    "        try:\n",
    "            group_str = ' '.join([\n",
    "                child.form\n",
    "                for child in sorted(group, key = lambda word: int(getattr(word, 'id')))\n",
    "            ])\n",
    "        except (TypeError) as e:\n",
    "            print(self, *group, sep='\\n')\n",
    "            raise(e)\n",
    "            \n",
    "        return group_str\n",
    "    group_str = property(get_group_str) \n",
    "    \n",
    "    def get_group_no_appos(self, **kwargs):\n",
    "        return self.get_group(exclude_appos_groups=True, **kwargs)\n",
    "    no_appos_group = property(get_group_no_appos)\n",
    "    \n",
    "    def get_group_str_no_appos(self, **kwargs):\n",
    "        return self.get_group_str(exclude_appos_groups=True, **kwargs)\n",
    "    no_appos_group_str = property(get_group_str_no_appos)\n",
    "    \n",
    "    \n",
    "    def to_dict(self):\n",
    "        # можно бы и word_dict...\n",
    "        return {attr: getattr(self, attr) for attr in self.attrs}\n",
    "    \n",
    "    \n",
    "    def __repr__(self):\n",
    "        return (\n",
    "            f\"\"\"Word({{'form': '{self.form}', 'id': {self.id}, \"\"\"\n",
    "            f\"\"\"'dom': {self.dom}, 'link': '{self.link}', \"\"\"\n",
    "            f\"\"\"'lemma': '{self.lemma}', 'feat': {self.feat}, \"\"\"\n",
    "            f\"\"\"'pos': '{self.pos}'}}, children={self.children})\"\"\"\n",
    "        )\n",
    "    \n",
    "    \n",
    "    def __str__(self):\n",
    "        children = chain(*self.children.values())\n",
    "        parent = self.parent\n",
    "        return (\n",
    "            f\"\"\"<Word({{'form': '{self.form}', 'id': {self.id}, \"\"\"\n",
    "            f\"\"\"'dom': {self.dom}, 'link': '{self.link}', 'pos': '{self.pos}' \"\"\"\n",
    "            f\"\"\"'lemma': '{self.lemma}', 'feat': {self.feat}}}, \"\"\"\n",
    "            f\"\"\"children={[(child.id, child.form) for child in children]}), \"\"\"\n",
    "            f\"\"\"parent={None if not parent else (parent.id, parent.form)}>\"\"\"\n",
    "        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "earlier-disaster",
   "metadata": {},
   "outputs": [],
   "source": [
    "class TokensList(list):\n",
    "    def filter(self, **kwargs):\n",
    "        if not kwargs:\n",
    "            return \n",
    "        \n",
    "        res = TokensList()\n",
    "        \n",
    "        def get_val(attr, word):\n",
    "            if 'feat__' in attr:\n",
    "                return getattr(word, 'feat').get(attr.split('__')[1])\n",
    "            else:\n",
    "                return None or getattr(word, attr)\n",
    "\n",
    "        for word in self:\n",
    "            if all(get_val(attr, word) == value for attr, value in kwargs.items()):\n",
    "                res.append(word)        \n",
    "        return res    \n",
    "    \n",
    "    def __repr__(self):\n",
    "        list_repr = super().__repr__()\n",
    "        return f'{self.__class__.__name__}({list_repr})'\n",
    "    \n",
    "\n",
    "class Sentence(TokensList):\n",
    "    pass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "continued-rolling",
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_tree(\n",
    "    root: Word,\n",
    "    depth: int,\n",
    "    words: List[Word],    \n",
    ") -> None:    \n",
    "    children = [word for word in words if word.dom == root.id]\n",
    "\n",
    "    for word in children:\n",
    "        root.add_child(word, 1)\n",
    "        make_tree(word, 1, words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "martial-arthur",
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_full_tree(\n",
    "    words: List[WordDict],\n",
    ") -> None:  \n",
    "    \n",
    "#     for word in words:\n",
    "#         id_to_word[word.id] = word\n",
    "    \n",
    "    root = next(word for word in words if word.dom == '_root')\n",
    "    words.root = root\n",
    "    \n",
    "    depth = 1\n",
    "    make_tree(root, depth, words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "colonial-melbourne",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_pos_chain(\n",
    "    words: List, include_feat=False, include_feat_name=False,\n",
    "    word_sep=' ', pos_feat_sep='+', feat_sep='|',\n",
    "    feat_name_val_sep = '=',\n",
    ") -> str:\n",
    "    \n",
    "    word_strs = []\n",
    "    for word in words:\n",
    "        word_str = word.pos\n",
    "        if include_feat and include_feat_name:\n",
    "            feat_str = feat_sep.join(\n",
    "                [f'{feat}{feat_name_val_sep}{val}'\n",
    "                for feat, val in word.feat.items()]\n",
    "            )\n",
    "        elif include_feat:\n",
    "            feat_str = feat_sep.join(word.feat.values())\n",
    "        else:\n",
    "            feat_str = ''\n",
    "            \n",
    "        word_str += f'{pos_feat_sep}{feat_str}' if feat_str else ''\n",
    "        word_strs.append(word_str)\n",
    "        \n",
    "    return word_sep.join(word_strs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "married-eugene",
   "metadata": {},
   "outputs": [],
   "source": [
    "def feats_dict_to_str(feats_dict: Dict[str, Union[str, int]]) -> str:\n",
    "    return '|'.join([f'{feat}={val}' for feat, val in feats_dict.items()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "northern-kidney",
   "metadata": {},
   "outputs": [],
   "source": [
    "def has_relative_clause(word_group: List[Word]) -> bool:\n",
    "    has_wh_relcl = any(word.link == 'релят' for word in word_group)\n",
    "    has_participle = False\n",
    "    \n",
    "    # тоже можно изменить на any\n",
    "    for word in word_group:\n",
    "        if word.pos == 'V' and word.feat.get('representation') == 'прич':\n",
    "            has_participle = True\n",
    "            break\n",
    "    has_relcl = has_wh_relcl or has_participle\n",
    "    return has_relcl, has_wh_relcl, has_participle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "framed-tulsa",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_data_from_sent(\n",
    "    full_analysis: str, text: str, words: Sentence,\n",
    "    sent_number: int, context: Tuple[str],\n",
    "    exclude_appos_descriptions=True\n",
    ") -> List[Res]:\n",
    "    res = []\n",
    "    \n",
    "    sent_chain = get_pos_chain(words)\n",
    "    \n",
    "    for verb in words.filter(pos='V'):\n",
    "        dos = words.filter(dom=verb.id, feat__case='вин', link='1-компл')\n",
    "        secos = words.filter(dom=verb.id, link='2-компл')\n",
    "        \n",
    "        if not (dos and secos):\n",
    "            continue\n",
    "        \n",
    "        has_many_dos, has_many_ios = 0, 0\n",
    "        if len(dos) >= 2:\n",
    "            has_many_dos = 1\n",
    "            print(len(dos))\n",
    "        if len(secos) >= 2:\n",
    "            has_many_ios = 1\n",
    "            print(len(secos))\n",
    "        do = dos[0]\n",
    "        seco = secos[0]\n",
    "        \n",
    "        if exclude_appos_descriptions:\n",
    "            do_group, do_group_str = do.no_appos_group, do.no_appos_group_str\n",
    "            seco_group, seco_group_str = seco.no_appos_group, seco.no_appos_group_str\n",
    "            do_children_depth = do.no_appos_children_depth\n",
    "            seco_children_depth = seco.no_appos_children_depth\n",
    "        else:\n",
    "            do_group, do_group_str = do.group, do.group_str\n",
    "            seco_group, seco_group_str = seco.group, seco.group_str\n",
    "            do_children_depth = do.children_depth\n",
    "            seco_children_depth = seco.children_depth\n",
    "        \n",
    "        do_had_appos = do.__dict__.get('has_appos', '')\n",
    "        seco_had_appos = seco.__dict__.get('has_appos', '')\n",
    "        \n",
    "        (do_has_relcl, do_has_wh_relcl, \n",
    "         do_has_participle) = has_relative_clause(do_group)\n",
    "        (seco_has_relcl, seco_has_wh_relcl, \n",
    "         seco_has_participle) = has_relative_clause(seco_group)\n",
    "        \n",
    "        do_pos_feat_chain = get_pos_chain(\n",
    "            do_group, include_feat=True, include_feat_name=True\n",
    "        )\n",
    "        seco_pos_feat_chain = get_pos_chain(\n",
    "            seco_group, include_feat=True, include_feat_name=True\n",
    "        )\n",
    "        \n",
    "        verb_feats = feats_dict_to_str(verb.feat)\n",
    "        verb_has_neg = 1 if words.filter(dom=verb.id, pos='PART', lemma='не') else 0\n",
    "        \n",
    "        seco_head_pos = 'PR' if seco.pos == 'PR' else 'V'\n",
    "        if seco_head_pos == 'PR':\n",
    "            seco_head = seco.form\n",
    "            seco = words.filter(dom=seco.id, link='предл')[0]\n",
    "        else:\n",
    "            seco_head = verb.form\n",
    "        \n",
    "        id_to_role = {do.id: 'DO', verb.id: 'V', seco.id: 'IO'}\n",
    "        order = sorted(\n",
    "            [verb] + [do, seco],\n",
    "            key=lambda token: token.id,\n",
    "        )\n",
    "        roles_order = ' '.join(id_to_role[token.id] for token in order)\n",
    "        \n",
    "        has_nodetype_in_our_forms = 0\n",
    "        if any('nodetype' in word.__dict__ for word in chain(*[[verb], do_group, seco_group])):\n",
    "            has_nodetype_in_our_forms = 1\n",
    "        \n",
    "        res.append(Res(\n",
    "            sent_number, context[0], text, context[1],\n",
    "                full_analysis, sent_chain,\n",
    "            has_many_dos, has_many_ios,\n",
    "            has_nodetype_in_our_forms,\n",
    "            verb.lemma, verb.form,\n",
    "            verb_feats, verb_has_neg,\n",
    "            do_group_str, seco_group_str, \n",
    "                do_had_appos, seco_had_appos,\n",
    "                do_has_relcl, seco_has_relcl,\n",
    "                    do_has_wh_relcl, seco_has_wh_relcl,\n",
    "                    do_has_participle, seco_has_participle,\n",
    "                do_children_depth, seco_children_depth,\n",
    "                    len(do_group), len(seco_group),\n",
    "                do_pos_feat_chain, seco_pos_feat_chain,\n",
    "            do.form,\n",
    "            do.pos, do.lemma, do.feat.get('animacy', ''),\n",
    "                do.feat.get('case', ''), do.feat.get('number', ''),\n",
    "            seco.form,\n",
    "            seco.pos, seco_head, seco_head_pos, seco.lemma,\n",
    "                seco.feat.get('animacy', ''), seco.feat.get('case', ''),\n",
    "                seco.feat.get('number', ''),\n",
    "            roles_order\n",
    "        ))\n",
    "        \n",
    "    return res "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "christian-chorus",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_context(\n",
    "    sent: bs4.element.Tag,\n",
    "    left_context_len = 2,\n",
    "    right_context_len = 2,\n",
    "):\n",
    "    left = []\n",
    "    right = []\n",
    "    for i, prev_ in enumerate(sent.previous_siblings):\n",
    "        if prev_ is not None and not isinstance(prev_, str):\n",
    "            left.append(prev_)\n",
    "\n",
    "        if len(left) >= num_left or prev_ is None:\n",
    "            break\n",
    "\n",
    "    for i, next_ in enumerate(sent.next_siblings):\n",
    "        if next_ is not None and not isinstance(next_, str):\n",
    "            right.append(next_)\n",
    "\n",
    "        if len(right) >= num_right or next_ is None:\n",
    "            break\n",
    "\n",
    "    left_context = '' if not left else ' '.join(get_text_from_sentence(prev_) for prev_ in left)\n",
    "    right_context = '' if not right else ' '.join(get_text_from_sentence(next_) for next_ in right) \n",
    "    \n",
    "    return left_context, right_context"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "lightweight-walnut",
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_sent(\n",
    "    sent: bs4.element.Tag,\n",
    "    pos_to_feats_formats: Dict[str, Dict[str, List[str]]],\n",
    "    return_tokens_list_text: bool = False,\n",
    "    return_context: bool = True,\n",
    "    left_context_len: int = 2, right_context_len: int = 2,\n",
    "    word_tag: str = 'W',\n",
    "    data_from_sent_kwargs={},\n",
    "):\n",
    "    \n",
    "    words = Sentence()\n",
    "    for word in sent.find_all(word_tag, recursive=False):\n",
    "        word_res = {}\n",
    "        for attr, value in word.attrs.items():\n",
    "            attr = attr.lower()\n",
    "            if attr not in ['dom', 'id'] or value == '_root':\n",
    "                word_res[attr] = value\n",
    "            else:\n",
    "                word_res[attr] = int(value)\n",
    "                \n",
    "#         word_res = {\n",
    "#             attr.lower(): (value if (attr not in ['DOM', 'ID'] or value == '_root') else int(value))\n",
    "#             for attr, value in word.attrs.items()\n",
    "#         }\n",
    "        word_res['lemma'] = word_res.get('lemma', '').lower()\n",
    "        word_res['feat'] = parse_feat(word_res['feat'], pos_to_feats_formats)\n",
    "        word_res['pos'] = word_res['feat'].pop('pos')\n",
    "        word_form = word.string or ''\n",
    "        word_res['form'] = word_form.lower()\n",
    "        word_obj = Word(word_res)\n",
    "        \n",
    "        words.append(word_obj)\n",
    "    \n",
    "    make_full_tree(words)\n",
    "    sent_number = sent.get('ID')\n",
    "    text = get_text_from_sentence(sent)\n",
    "    \n",
    "    left_context, right_context = '', ''\n",
    "    if return_context:\n",
    "        left_context, right_context = get_context(sent, left_context_len, right_context_len)\n",
    "    context = (left_context, right_context)\n",
    "\n",
    "    res = get_data_from_sent(str(sent), text, words, sent_number, context, **data_from_sent_kwargs)\n",
    "    if not return_tokens_list_text:\n",
    "        return res\n",
    "    else:\n",
    "        return res, words, text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "proud-decrease",
   "metadata": {},
   "outputs": [],
   "source": [
    "file = syntagrus_dir / Path('2019/baden-baden.tgt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "diverse-suspect",
   "metadata": {},
   "outputs": [],
   "source": [
    "file_soup = get_soup(file).find('body')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "cross-boulder",
   "metadata": {},
   "outputs": [],
   "source": [
    "sents = file_soup.find_all('S')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "generic-capital",
   "metadata": {},
   "outputs": [],
   "source": [
    "res, tokens, text = parse_sent(sents[9], get_pos_to_feats_format(), return_tokens_list_text=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "suspected-upgrade",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[res(sent_num_file='10', left_context='Он чувствовал себя здесь русским, потерянным и оторванным от корней… Тургенев в течение семи лет бродил тут с собакой по окрестным лесам и горам, искал, думал, страдал.', text='Это чувство родства с ним дает мне силы жить здесь.', right_context='Элементы русской архитектуры - резные деревянные балкончики и наличники - напоминают Абрамцево, белоснежные и бледно-розовые усадьбы, заросшие сиренью, возвращают меня в нашу тихую подмосковную жизнь. Из влажной туманной зимы, из голубой мглы и одиночества Баден в апреле возрождается безграничными полями огромных сиреневых крокусов и пением птиц.', full_analysis='<S ID=\"10\">\\n<W DOM=\"2\" FEAT=\"A ЕД СРЕД ИМ\" ID=\"1\" KSNAME=\"ЭТОТ\" LEMMA=\"ЭТОТ\" LINK=\"опред\">Это</W>\\n<W DOM=\"6\" FEAT=\"S ЕД СРЕД ИМ НЕОД\" ID=\"2\" LEMMA=\"ЧУВСТВО\" LINK=\"предик\">чувство</W>\\n<W DOM=\"2\" FEAT=\"S ЕД СРЕД РОД НЕОД\" ID=\"3\" LEMMA=\"РОДСТВО\" LINK=\"1-компл\">родства</W>\\n<W DOM=\"3\" FEAT=\"PR\" ID=\"4\" KSNAME=\"С3\" LEMMA=\"С\" LINK=\"1-компл\">с</W>\\n<W DOM=\"4\" FEAT=\"S ЕД МУЖ ТВОР ОД\" ID=\"5\" KSNAME=\"НЕГО1\" LEMMA=\"ОН\" LINK=\"предл\">ним</W>\\n<W DOM=\"_root\" FEAT=\"V НЕСОВ ИЗЪЯВ НЕПРОШ ЕД 3-Л\" ID=\"6\" KSNAME=\"ДАВАТЬ1\" LEMMA=\"ДАВАТЬ\">дает</W>\\n<W DOM=\"6\" FEAT=\"S ЕД ЖЕН ДАТ ОД\" ID=\"7\" LEMMA=\"Я\" LINK=\"2-компл\">мне</W>\\n<W DOM=\"6\" FEAT=\"S МН ЖЕН ВИН НЕОД\" ID=\"8\" LEMMA=\"СИЛА\" LINK=\"1-компл\">силы</W>\\n<W DOM=\"8\" FEAT=\"V НЕСОВ ИНФ\" ID=\"9\" LEMMA=\"ЖИТЬ\" LINK=\"1-компл\">жить</W>\\n<W DOM=\"9\" FEAT=\"ADV\" ID=\"10\" LEMMA=\"ЗДЕСЬ\" LINK=\"1-компл\">здесь</W>.\\n</S>', sent_pos_chain='PRON S S PR PRON V PRON S V ADV', has_many_dos=0, has_many_ios=0, has_nodetype_in_our_forms=0, verb_inf='давать', verb='дает', verb_feats='mood=изъяв|aspect=несов|tense=непрош|person=3-л|number=ед', verb_has_neg=0, DO_group='силы жить здесь', IO_group='мне', DO_had_appos=0, IO_had_appos=0, DO_has_relcl=False, IO_has_relcl=False, DO_has_wh_relcl=False, IO_has_wh_relcl=False, DO_has_participle=False, IO_has_participle=False, DO_depth=2, IO_depth=0, DO_length=3, IO_length=1, DO_pos_feat_chain='V+representation=инф|aspect=несов ADV S+gender=жен|number=мн|case=вин|animacy=неод', IO_pos_feat_chain='PRON+gender=жен|number=ед|case=дат|animacy=од', DO='силы', DO_pos='S', DO_lemma='сила', DO_animacy='неод', DO_case='вин', DO_number='мн', IO='мне', IO_pos='PRON', IO_head='дает', IO_head_pos='V', IO_lemma='я', IO_animacy='од', IO_case='дат', IO_number='ед', order='V IO DO')]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "acceptable-cricket",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<Word({'form': 'это', 'id': 1, 'dom': 2, 'link': 'опред', 'pos': 'PRON' 'lemma': 'этот', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'им'}}, children=[]), parent=(2, 'чувство')>\n",
      "<Word({'form': 'чувство', 'id': 2, 'dom': 6, 'link': 'предик', 'pos': 'S' 'lemma': 'чувство', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'им', 'animacy': 'неод'}}, children=[(1, 'это'), (3, 'родства'), (4, 'с'), (5, 'ним')]), parent=(6, 'дает')>\n",
      "<Word({'form': 'родства', 'id': 3, 'dom': 2, 'link': '1-компл', 'pos': 'S' 'lemma': 'родство', 'feat': {'gender': 'сред', 'number': 'ед', 'case': 'род', 'animacy': 'неод'}}, children=[(4, 'с'), (5, 'ним')]), parent=(2, 'чувство')>\n",
      "<Word({'form': 'с', 'id': 4, 'dom': 3, 'link': '1-компл', 'pos': 'PR' 'lemma': 'с', 'feat': {}}, children=[(5, 'ним')]), parent=(3, 'родства')>\n",
      "<Word({'form': 'ним', 'id': 5, 'dom': 4, 'link': 'предл', 'pos': 'PRON' 'lemma': 'он', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'твор', 'animacy': 'од'}}, children=[]), parent=(4, 'с')>\n",
      "<Word({'form': 'дает', 'id': 6, 'dom': _root, 'link': '_root', 'pos': 'V' 'lemma': 'давать', 'feat': {'mood': 'изъяв', 'aspect': 'несов', 'tense': 'непрош', 'person': '3-л', 'number': 'ед'}}, children=[(2, 'чувство'), (7, 'мне'), (8, 'силы'), (1, 'это'), (3, 'родства'), (9, 'жить'), (4, 'с'), (10, 'здесь'), (5, 'ним')]), parent=None>\n",
      "<Word({'form': 'мне', 'id': 7, 'dom': 6, 'link': '2-компл', 'pos': 'PRON' 'lemma': 'я', 'feat': {'gender': 'жен', 'number': 'ед', 'case': 'дат', 'animacy': 'од'}}, children=[]), parent=(6, 'дает')>\n",
      "<Word({'form': 'силы', 'id': 8, 'dom': 6, 'link': '1-компл', 'pos': 'S' 'lemma': 'сила', 'feat': {'gender': 'жен', 'number': 'мн', 'case': 'вин', 'animacy': 'неод'}}, children=[(9, 'жить'), (10, 'здесь')]), parent=(6, 'дает')>\n",
      "<Word({'form': 'жить', 'id': 9, 'dom': 8, 'link': '1-компл', 'pos': 'V' 'lemma': 'жить', 'feat': {'representation': 'инф', 'aspect': 'несов'}}, children=[(10, 'здесь')]), parent=(8, 'силы')>\n",
      "<Word({'form': 'здесь', 'id': 10, 'dom': 9, 'link': '1-компл', 'pos': 'ADV' 'lemma': 'здесь', 'feat': {}}, children=[]), parent=(9, 'жить')>\n"
     ]
    }
   ],
   "source": [
    "print(*tokens, sep='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "compliant-infrastructure",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Это чувство родства с ним дает мне силы жить здесь.\n",
      "\n",
      "group_str: это чувство родства с ним\n",
      "no_appos_group_str: это чувство родства с ним\n",
      "\n",
      "children_depth: 3\n",
      "no_appos_children_depth: 3\n",
      "\n",
      "len_group: 5\n",
      "len_no_appos_group: 5\n",
      "\n"
     ]
    }
   ],
   "source": [
    "token = tokens[1]\n",
    "print(text, end='\\n\\n')\n",
    "for attr in ('group_str', 'children_depth'):\n",
    "    print(f\"{attr}: {getattr(token, attr)}\\n\"\n",
    "          f\"{f'no_appos_{attr}'}: {getattr(token, f'no_appos_{attr}')}\\n\"\n",
    "    )\n",
    "\n",
    "attr = 'group'\n",
    "print(\n",
    "    f\"len_{attr}: {len(getattr(token, attr))}\\n\"\n",
    "    f\"len_{f'no_appos_{attr}'}: {len(getattr(token, f'no_appos_{attr}'))}\\n\"\n",
    " )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "recreational-basis",
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_sent_in_file_by_num(short_filename, sent_num, return_tokens_list=True):\n",
    "    file = syntagrus_dir / Path(short_filename)\n",
    "    file_soup = get_soup(file).find('body')\n",
    "    \n",
    "    sents = file_soup.find_all('S', recursive=False)\n",
    "    sent = sents[sent_num-1]\n",
    "    \n",
    "    res, tokens = parse_sent(\n",
    "        sents[182], get_pos_to_feats_format(),\n",
    "        return_tokens_list=return_tokens_list\n",
    "    )\n",
    "    \n",
    "    return res, tokens"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "silent-material",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ним\n",
      "PRON+gender=муж|number=ед|case=твор|animacy=од\n"
     ]
    }
   ],
   "source": [
    "gr = tokens.filter(id=5)[0].group_str\n",
    "r = get_pos_chain(tokens.filter(id=5)[0].group, include_feat=True, include_feat_name=True)\n",
    "\n",
    "print(gr, r, sep='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "prescription-election",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Word({'form': 'ним', 'id': 5, 'dom': 4, 'link': 'предл', 'lemma': 'он', 'feat': {'gender': 'муж', 'number': 'ед', 'case': 'твор', 'animacy': 'од'}, 'pos': 'PRON'}, children={})"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokens[4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "heavy-genesis",
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'sentences' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-66-ba03659c1619>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mwords\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mparse_sent\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msentences\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m3\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mget_pos_to_feats_format\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m: name 'sentences' is not defined"
     ]
    }
   ],
   "source": [
    "words = parse_sent(sentences[3], get_pos_to_feats_format())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "classified-device",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(*words.filter(link='1-компл'), sep='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "moderate-indicator",
   "metadata": {},
   "outputs": [],
   "source": [
    "words[0].group_str"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "vanilla-detroit",
   "metadata": {},
   "outputs": [],
   "source": [
    "def do_search(files, num_first_files=None):\n",
    "    total_passing, total_regex = 0, 0\n",
    "    \n",
    "    if num_first_files:\n",
    "        files = files[:num_first_files]\n",
    "    \n",
    "    examples = {}\n",
    "    files_dfs = []\n",
    "    for file in tqdm(files):\n",
    "        soup = get_soup(file)\n",
    "        \n",
    "        doc_body = soup.find('body')\n",
    "        \n",
    "        sents = find_sentences_do_io(doc_body)\n",
    "        num_regex_sents = len(sents)\n",
    "        \n",
    "        unique_file_path = file.relative_to(syntagrus_dir)\n",
    "        path_str = str(unique_file_path).replace('\\\\', '/')\n",
    "        examples[path_str] = [str(sent) for sent in sents]\n",
    "                        \n",
    "        good_sents = []\n",
    "        for sent in sents:\n",
    "            parsed_sent = parse_sent(sent, get_pos_to_feats_format())\n",
    "            if parsed_sent:\n",
    "                good_sents.extend(parsed_sent)\n",
    "                \n",
    "        num_passing_sents = len(good_sents)\n",
    "        print(f'file is `{unique_file_path}`, there are {num_passing_sents} passing sentences '\n",
    "              f'(were {num_regex_sents} by regex)')\n",
    "        total_passing += num_passing_sents\n",
    "        total_regex += num_regex_sents\n",
    "        \n",
    "        good_sents_table = pd.DataFrame(good_sents)\n",
    "        good_sents_table.insert(0, 'source', path_str)\n",
    "        files_dfs.append(good_sents_table)\n",
    "    \n",
    "    print(f'passing {total_passing} among {total_regex} found by regex')\n",
    "    return pd.concat(files_dfs), total_passing, total_regex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "surprising-crowd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "dc33228daa054171b18589147587b365",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(FloatProgress(value=0.0, max=729.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "file is `2003\\anketa.tgt`, there are 16 passing sentences (were 26 by regex)\n",
      "file is `2003\\armeniya.tgt`, there are 6 passing sentences (were 15 by regex)\n",
      "file is `2003\\artist_mimansa.tgt`, there are 39 passing sentences (were 75 by regex)\n",
      "file is `2003\\ataka.tgt`, there are 1 passing sentences (were 3 by regex)\n",
      "file is `2003\\atlanty_i_atlantologi.tgt`, there are 15 passing sentences (were 22 by regex)\n",
      "file is `2003\\avtomatizatsiya.tgt`, there are 6 passing sentences (were 15 by regex)\n",
      "file is `2003\\a_on_myatezhnyi.tgt`, there are 20 passing sentences (were 35 by regex)\n",
      "file is `2003\\baklanov.tgt`, there are 54 passing sentences (were 82 by regex)\n",
      "file is `2003\\batareika.tgt`, there are 2 passing sentences (were 8 by regex)\n",
      "file is `2003\\bessonnitsa.tgt`, there are 6 passing sentences (were 13 by regex)\n",
      "file is `2003\\bez_epokhi.tgt`, there are 11 passing sentences (were 15 by regex)\n",
      "file is `2003\\biologiya.tgt`, there are 15 passing sentences (were 20 by regex)\n",
      "file is `2003\\bionika.tgt`, there are 10 passing sentences (were 22 by regex)\n",
      "file is `2003\\bolshie_peremeny.tgt`, there are 15 passing sentences (were 24 by regex)\n",
      "file is `2003\\chelovek_oshibka_prirody.tgt`, there are 7 passing sentences (were 14 by regex)\n",
      "file is `2003\\chto_doktor_propisal.tgt`, there are 17 passing sentences (were 25 by regex)\n",
      "file is `2003\\delit_na_vosem.tgt`, there are 10 passing sentences (were 17 by regex)\n",
      "file is `2003\\dom_pod_solntsem.tgt`, there are 7 passing sentences (were 9 by regex)\n",
      "file is `2003\\dvoe_v_dekabre.tgt`, there are 10 passing sentences (were 33 by regex)\n",
      "file is `2003\\dvortsovyi_makiyazh.tgt`, there are 13 passing sentences (were 15 by regex)\n",
      "file is `2003\\ekspluatatsiya_elity.tgt`, there are 12 passing sentences (were 23 by regex)\n",
      "file is `2003\\ezhednevnaya_simfoniya.tgt`, there are 10 passing sentences (were 11 by regex)\n",
      "file is `2003\\faraony_sobstvennoi_personoi.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2003\\fitoterapiya.tgt`, there are 11 passing sentences (were 11 by regex)\n",
      "file is `2003\\gorodskoe_slonovodstvo.tgt`, there are 9 passing sentences (were 20 by regex)\n",
      "file is `2003\\igrushki_bogov.tgt`, there are 9 passing sentences (were 16 by regex)\n",
      "file is `2003\\interaktiv.tgt`, there are 6 passing sentences (were 10 by regex)\n",
      "file is `2003\\internet-zavisimost.tgt`, there are 8 passing sentences (were 12 by regex)\n",
      "file is `2003\\istoriya_tualeta.tgt`, there are 5 passing sentences (were 7 by regex)\n",
      "file is `2003\\i_evro_takoi_molodoi.tgt`, there are 10 passing sentences (were 17 by regex)\n",
      "file is `2003\\i_k_nim_ne_zarastet_narodnaya_tropa.tgt`, there are 13 passing sentences (were 25 by regex)\n",
      "file is `2003\\i_slepye_prozreyut.tgt`, there are 4 passing sentences (were 11 by regex)\n",
      "file is `2003\\khranilishcha.tgt`, there are 17 passing sentences (were 17 by regex)\n",
      "file is `2003\\komu_dostanetsya_chingiskhan.tgt`, there are 11 passing sentences (were 14 by regex)\n",
      "file is `2003\\krupnaya_kala.tgt`, there are 5 passing sentences (were 5 by regex)\n",
      "file is `2003\\lyubit_drakona.tgt`, there are 14 passing sentences (were 21 by regex)\n",
      "file is `2003\\manekeny.tgt`, there are 5 passing sentences (were 7 by regex)\n",
      "file is `2003\\metis.tgt`, there are 6 passing sentences (were 11 by regex)\n",
      "file is `2003\\nalog_na_tainu.tgt`, there are 14 passing sentences (were 23 by regex)\n",
      "file is `2003\\nelegalnaya_perepis.tgt`, there are 10 passing sentences (were 13 by regex)\n",
      "file is `2003\\nelzya_sebya_delit.tgt`, there are 13 passing sentences (were 22 by regex)\n",
      "file is `2003\\nepotoplyaemye.tgt`, there are 6 passing sentences (were 12 by regex)\n",
      "file is `2003\\nobelevskie_premii.tgt`, there are 4 passing sentences (were 10 by regex)\n",
      "file is `2003\\obratnaya_reaktsiya.tgt`, there are 4 passing sentences (were 4 by regex)\n",
      "file is `2003\\ochishchenie_olkhona.tgt`, there are 10 passing sentences (were 16 by regex)\n",
      "file is `2003\\okhota.tgt`, there are 11 passing sentences (were 14 by regex)\n",
      "file is `2003\\opasnaya_blizost.tgt`, there are 6 passing sentences (were 11 by regex)\n",
      "file is `2003\\optimizm.tgt`, there are 12 passing sentences (were 22 by regex)\n",
      "file is `2003\\perekrestok.tgt`, there are 11 passing sentences (were 17 by regex)\n",
      "file is `2003\\perestanovka_mebeli.tgt`, there are 7 passing sentences (were 15 by regex)\n",
      "file is `2003\\piramida.tgt`, there are 5 passing sentences (were 10 by regex)\n",
      "file is `2003\\plastikovaya_krepost.tgt`, there are 16 passing sentences (were 21 by regex)\n",
      "file is `2003\\platno.tgt`, there are 8 passing sentences (were 14 by regex)\n",
      "file is `2003\\plesnite_koldovstva.tgt`, there are 12 passing sentences (were 19 by regex)\n",
      "file is `2003\\portret_fridy.tgt`, there are 4 passing sentences (were 5 by regex)\n",
      "file is `2003\\pravilo_75.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2003\\prilet_ptits.tgt`, there are 18 passing sentences (were 37 by regex)\n",
      "file is `2003\\privatizatsiya_istorii.tgt`, there are 10 passing sentences (were 17 by regex)\n",
      "file is `2003\\propoved_na_zadannuyu_temu.tgt`, there are 15 passing sentences (were 23 by regex)\n",
      "file is `2003\\razvedka.tgt`, there are 2 passing sentences (were 13 by regex)\n",
      "file is `2003\\robototekhnika.tgt`, there are 10 passing sentences (were 15 by regex)\n",
      "file is `2003\\roboty.tgt`, there are 3 passing sentences (were 8 by regex)\n",
      "file is `2003\\rtut.tgt`, there are 2 passing sentences (were 4 by regex)\n",
      "file is `2003\\shok_bez_trepeta.tgt`, there are 17 passing sentences (were 30 by regex)\n",
      "file is `2003\\somnambula_v_tumane.tgt`, there are 33 passing sentences (were 56 by regex)\n",
      "file is `2003\\son.tgt`, there are 4 passing sentences (were 8 by regex)\n",
      "file is `2003\\spam.tgt`, there are 3 passing sentences (were 4 by regex)\n",
      "file is `2003\\sputniki.tgt`, there are 4 passing sentences (were 12 by regex)\n",
      "file is `2003\\stolknovenie_vermeer.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2003\\taezhnyi_ofshor.tgt`, there are 14 passing sentences (were 25 by regex)\n",
      "file is `2003\\taina_zakovannaya_v_led.tgt`, there are 2 passing sentences (were 9 by regex)\n",
      "file is `2003\\to_v_zhar_to_v_kholod.tgt`, there are 7 passing sentences (were 9 by regex)\n",
      "file is `2003\\tyurma_britanskoi_korony.tgt`, there are 7 passing sentences (were 13 by regex)\n",
      "file is `2003\\tyurma_dlya_svekrovei.tgt`, there are 12 passing sentences (were 11 by regex)\n",
      "file is `2003\\uidya_iz_skazki.tgt`, there are 9 passing sentences (were 17 by regex)\n",
      "file is `2003\\uroki_nord-osta.tgt`, there are 5 passing sentences (were 14 by regex)\n",
      "file is `2003\\uteshenie.tgt`, there are 0 passing sentences (were 3 by regex)\n",
      "file is `2003\\vash_personalnyi_nimb.tgt`, there are 5 passing sentences (were 17 by regex)\n",
      "file is `2003\\vremya_torgovli.tgt`, there are 5 passing sentences (were 6 by regex)\n",
      "file is `2003\\vyzhivshii_kamikadze.tgt`, there are 25 passing sentences (were 49 by regex)\n",
      "file is `2003\\v_zimu_bez_grippa.tgt`, there are 3 passing sentences (were 5 by regex)\n",
      "file is `2003\\zdorovie_na_zavtra.tgt`, there are 6 passing sentences (were 14 by regex)\n",
      "file is `2003\\zemlya_naiznanku.tgt`, there are 14 passing sentences (were 18 by regex)\n",
      "file is `2003\\zhivotnye_mukhi.tgt`, there are 8 passing sentences (were 8 by regex)\n",
      "file is `2003\\zhores.tgt`, there are 28 passing sentences (were 57 by regex)\n",
      "file is `2004\\andrei_ashkerov.tgt`, there are 11 passing sentences (were 15 by regex)\n",
      "file is `2004\\tarifnaya_viselitsa.tgt`, there are 18 passing sentences (were 23 by regex)\n",
      "file is `2005\\blesnuli_masterstvom.tgt`, there are 0 passing sentences (were 0 by regex)\n",
      "file is `2005\\bolshe_kandidatov.tgt`, there are 7 passing sentences (were 10 by regex)\n",
      "file is `2005\\byt_modnym.tgt`, there are 11 passing sentences (were 13 by regex)\n",
      "file is `2005\\delitsya_vlastiyu.tgt`, there are 2 passing sentences (were 4 by regex)\n",
      "file is `2005\\dvukhprotsentnaya_reforma.tgt`, there are 9 passing sentences (were 14 by regex)\n",
      "file is `2005\\gubarev.tgt`, there are 12 passing sentences (were 23 by regex)\n",
      "file is `2005\\istochniki_rosta.tgt`, there are 4 passing sentences (were 12 by regex)\n",
      "file is `2005\\kaprossiya.tgt`, there are 4 passing sentences (were 6 by regex)\n",
      "file is `2005\\kompartii.tgt`, there are 2 passing sentences (were 3 by regex)\n",
      "file is `2005\\kruglyi_stol.tgt`, there are 0 passing sentences (were 2 by regex)\n",
      "file is `2005\\kvartira_na_depozit.tgt`, there are 5 passing sentences (were 15 by regex)\n",
      "file is `2005\\mamleev_charli.tgt`, there are 38 passing sentences (were 66 by regex)\n",
      "file is `2005\\mamleev_svadba.tgt`, there are 17 passing sentences (were 22 by regex)\n",
      "file is `2005\\monopolizatsiya_kanalizatsii.tgt`, there are 12 passing sentences (were 19 by regex)\n",
      "file is `2005\\muzhei_ne_obeshchal.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `2005\\na_levom_flange.tgt`, there are 6 passing sentences (were 10 by regex)\n",
      "file is `2005\\neftyanye_kacheli.tgt`, there are 1 passing sentences (were 3 by regex)\n",
      "file is `2005\\nichego_chelovecheskogo.tgt`, there are 4 passing sentences (were 5 by regex)\n",
      "file is `2005\\novyi_levyi.tgt`, there are 7 passing sentences (were 19 by regex)\n",
      "file is `2005\\otkaz_ot_grefa.tgt`, there are 5 passing sentences (were 14 by regex)\n",
      "file is `2005\\parlamentskaya_komissiya.tgt`, there are 1 passing sentences (were 2 by regex)\n",
      "file is `2005\\po-nashemu.tgt`, there are 4 passing sentences (were 4 by regex)\n",
      "file is `2005\\prodazha_zemli.tgt`, there are 4 passing sentences (were 5 by regex)\n",
      "file is `2005\\shokovaya_terapiya.tgt`, there are 2 passing sentences (were 5 by regex)\n",
      "file is `2005\\sluzhit_by_rad.tgt`, there are 4 passing sentences (were 19 by regex)\n",
      "file is `2005\\stranitsy_voennoi_istorii.tgt`, there are 46 passing sentences (were 75 by regex)\n",
      "file is `2005\\sudu_pomogayut.tgt`, there are 2 passing sentences (were 5 by regex)\n",
      "file is `2005\\tokarskaya.tgt`, there are 0 passing sentences (were 1 by regex)\n",
      "file is `2005\\voennaya_doktrina.tgt`, there are 15 passing sentences (were 24 by regex)\n",
      "file is `2005\\voennoe_bessilie.tgt`, there are 21 passing sentences (were 24 by regex)\n",
      "file is `2005\\yakovlev.tgt`, there are 10 passing sentences (were 17 by regex)\n",
      "file is `2005\\yanin.tgt`, there are 15 passing sentences (were 22 by regex)\n",
      "file is `2005\\zhenshchin_uravnyayut.tgt`, there are 6 passing sentences (were 6 by regex)\n",
      "file is `2006\\antiterror.tgt`, there are 4 passing sentences (were 12 by regex)\n",
      "file is `2006\\byudzhet.tgt`, there are 14 passing sentences (were 16 by regex)\n",
      "file is `2006\\demonstratsiya_flaga.tgt`, there are 7 passing sentences (were 16 by regex)\n",
      "file is `2006\\ditya_20_siezda.tgt`, there are 24 passing sentences (were 34 by regex)\n",
      "file is `2006\\dobretsov.tgt`, there are 23 passing sentences (were 40 by regex)\n",
      "file is `2006\\dvesti_let.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `2006\\gaidar.tgt`, there are 15 passing sentences (were 16 by regex)\n",
      "file is `2006\\informtekhnologii.tgt`, there are 9 passing sentences (were 15 by regex)\n",
      "file is `2006\\levyi_povorot.tgt`, there are 21 passing sentences (were 30 by regex)\n",
      "file is `2006\\lunnye_kamni.tgt`, there are 30 passing sentences (were 59 by regex)\n",
      "file is `2006\\metallovedenie.tgt`, there are 10 passing sentences (were 19 by regex)\n",
      "file is `2006\\misha.tgt`, there are 49 passing sentences (were 77 by regex)\n",
      "file is `2006\\muzei.tgt`, there are 3 passing sentences (were 3 by regex)\n",
      "file is `2006\\neiron.tgt`, there are 20 passing sentences (were 27 by regex)\n",
      "file is `2006\\piterskaya_model.tgt`, there are 15 passing sentences (were 33 by regex)\n",
      "file is `2006\\predsedatel.tgt`, there are 12 passing sentences (were 18 by regex)\n",
      "file is `2006\\rentgen.tgt`, there are 17 passing sentences (were 36 by regex)\n",
      "file is `2006\\sluzhivoe_pravo.tgt`, there are 38 passing sentences (were 52 by regex)\n",
      "file is `2006\\stylaya_krov.tgt`, there are 24 passing sentences (were 36 by regex)\n",
      "file is `2006\\syuzhet.tgt`, there are 8 passing sentences (were 11 by regex)\n",
      "file is `2006\\uroki.tgt`, there are 16 passing sentences (were 28 by regex)\n",
      "file is `2006\\veter.tgt`, there are 15 passing sentences (were 22 by regex)\n",
      "file is `2006\\vinnyi_kamen.tgt`, there are 11 passing sentences (were 21 by regex)\n",
      "file is `2006\\vse_vozrastu_pokorny.tgt`, there are 16 passing sentences (were 32 by regex)\n",
      "file is `2006\\vtoraya_popytka.tgt`, there are 16 passing sentences (were 23 by regex)\n",
      "file is `2006\\v_pogone.tgt`, there are 15 passing sentences (were 29 by regex)\n",
      "file is `2007\\beg_po_krugu.tgt`, there are 7 passing sentences (were 18 by regex)\n",
      "file is `2007\\bezrybie.tgt`, there are 23 passing sentences (were 25 by regex)\n",
      "file is `2007\\bob.tgt`, there are 16 passing sentences (were 26 by regex)\n",
      "file is `2007\\buran.tgt`, there are 6 passing sentences (were 17 by regex)\n",
      "file is `2007\\chelovek_na_tribune.tgt`, there are 15 passing sentences (were 26 by regex)\n",
      "file is `2007\\chuvstvo_spravedlivosti.tgt`, there are 20 passing sentences (were 22 by regex)\n",
      "file is `2007\\darts.tgt`, there are 10 passing sentences (were 22 by regex)\n",
      "file is `2007\\dvigatel.tgt`, there are 13 passing sentences (were 26 by regex)\n",
      "file is `2007\\ekzamen.tgt`, there are 4 passing sentences (were 5 by regex)\n",
      "file is `2007\\foto.tgt`, there are 20 passing sentences (were 23 by regex)\n",
      "file is `2007\\gematologiya.tgt`, there are 10 passing sentences (were 10 by regex)\n",
      "file is `2007\\grechko.tgt`, there are 31 passing sentences (were 60 by regex)\n",
      "file is `2007\\kak_politikov.tgt`, there are 9 passing sentences (were 15 by regex)\n",
      "file is `2007\\katastrofa.tgt`, there are 0 passing sentences (were 4 by regex)\n",
      "file is `2007\\kodeks_molchaniya.tgt`, there are 8 passing sentences (were 10 by regex)\n",
      "file is `2007\\kovka.tgt`, there are 14 passing sentences (were 15 by regex)\n",
      "file is `2007\\kto_ubil.tgt`, there are 10 passing sentences (were 14 by regex)\n",
      "file is `2007\\led.tgt`, there are 21 passing sentences (were 30 by regex)\n",
      "file is `2007\\lesorub.tgt`, there are 16 passing sentences (were 21 by regex)\n",
      "file is `2007\\lunokhod.tgt`, there are 7 passing sentences (were 13 by regex)\n",
      "file is `2007\\lyzhi.tgt`, there are 11 passing sentences (were 26 by regex)\n",
      "file is `2007\\mars.tgt`, there are 15 passing sentences (were 20 by regex)\n",
      "file is `2007\\mech.tgt`, there are 26 passing sentences (were 34 by regex)\n",
      "file is `2007\\okhota.tgt`, there are 3 passing sentences (were 9 by regex)\n",
      "file is `2007\\parovoz.tgt`, there are 7 passing sentences (were 14 by regex)\n",
      "file is `2007\\pirat.tgt`, there are 8 passing sentences (were 12 by regex)\n",
      "file is `2007\\podkormlennyi_natsizm.tgt`, there are 39 passing sentences (were 52 by regex)\n",
      "file is `2007\\porazhenie.tgt`, there are 9 passing sentences (were 12 by regex)\n",
      "file is `2007\\pravda_zalozhnikov.tgt`, there are 11 passing sentences (were 16 by regex)\n",
      "file is `2007\\prevratnosti_razvitiya.tgt`, there are 20 passing sentences (were 37 by regex)\n",
      "file is `2007\\pushka.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2007\\pylesos.tgt`, there are 10 passing sentences (were 14 by regex)\n",
      "file is `2007\\raspad_sng.tgt`, there are 5 passing sentences (were 13 by regex)\n",
      "file is `2007\\remni.tgt`, there are 13 passing sentences (were 15 by regex)\n",
      "file is `2007\\rumba.tgt`, there are 9 passing sentences (were 19 by regex)\n",
      "file is `2007\\skulachev.tgt`, there are 20 passing sentences (were 29 by regex)\n",
      "file is `2007\\sluchai_dlya_zhyuri.tgt`, there are 9 passing sentences (were 11 by regex)\n",
      "file is `2007\\soldatiki.tgt`, there are 15 passing sentences (were 20 by regex)\n",
      "file is `2007\\suverennaya_mutatsiya.tgt`, there are 18 passing sentences (were 35 by regex)\n",
      "file is `2007\\tele.tgt`, there are 7 passing sentences (were 10 by regex)\n",
      "file is `2007\\terpenie_lopnulo.tgt`, there are 13 passing sentences (were 16 by regex)\n",
      "file is `2007\\transport.tgt`, there are 2 passing sentences (were 7 by regex)\n",
      "file is `2007\\tsennosti.tgt`, there are 3 passing sentences (were 9 by regex)\n",
      "file is `2007\\ukho.tgt`, there are 9 passing sentences (were 16 by regex)\n",
      "file is `2007\\underground.tgt`, there are 4 passing sentences (were 12 by regex)\n",
      "file is `2007\\ustar.tgt`, there are 10 passing sentences (were 15 by regex)\n",
      "file is `2007\\zamoroz.tgt`, there are 3 passing sentences (were 2 by regex)\n",
      "file is `2007\\zheldor.tgt`, there are 10 passing sentences (were 13 by regex)\n",
      "file is `2008\\bankovskii_krizis.tgt`, there are 4 passing sentences (were 6 by regex)\n",
      "file is `2008\\bezrabotnye.tgt`, there are 4 passing sentences (were 9 by regex)\n",
      "file is `2008\\bronya.tgt`, there are 10 passing sentences (were 14 by regex)\n",
      "file is `2008\\chem_kumushek_schitat.tgt`, there are 7 passing sentences (were 9 by regex)\n",
      "file is `2008\\chernaya_osen.tgt`, there are 10 passing sentences (were 18 by regex)\n",
      "file is `2008\\chto_v_imeni.tgt`, there are 7 passing sentences (were 14 by regex)\n",
      "file is `2008\\drakony.tgt`, there are 11 passing sentences (were 14 by regex)\n",
      "file is `2008\\ekonomicheskii_rost.tgt`, there are 5 passing sentences (were 10 by regex)\n",
      "file is `2008\\galopom_ot_trollei.tgt`, there are 13 passing sentences (were 18 by regex)\n",
      "file is `2008\\kobzon.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `2008\\kollaider.tgt`, there are 0 passing sentences (were 3 by regex)\n",
      "file is `2008\\korrida.tgt`, there are 5 passing sentences (were 6 by regex)\n",
      "file is `2008\\koshki.tgt`, there are 20 passing sentences (were 23 by regex)\n",
      "file is `2008\\lego.tgt`, there are 11 passing sentences (were 15 by regex)\n",
      "file is `2008\\mars_1.tgt`, there are 12 passing sentences (were 24 by regex)\n",
      "file is `2008\\mars_2.tgt`, there are 12 passing sentences (were 22 by regex)\n",
      "file is `2008\\molodezh.tgt`, there are 2 passing sentences (were 2 by regex)\n",
      "file is `2008\\narkotiki.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2008\\nedouchenie.tgt`, there are 3 passing sentences (were 6 by regex)\n",
      "file is `2008\\nekommercheskie_organizatsii.tgt`, there are 15 passing sentences (were 26 by regex)\n",
      "file is `2008\\nepodiemnyi_kredit.tgt`, there are 3 passing sentences (were 10 by regex)\n",
      "file is `2008\\olenina.tgt`, there are 7 passing sentences (were 9 by regex)\n",
      "file is `2008\\opasnye_prazdniki.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `2008\\osoboe_mnenie.tgt`, there are 5 passing sentences (were 9 by regex)\n",
      "file is `2008\\otboinyi_molotok.tgt`, there are 3 passing sentences (were 8 by regex)\n",
      "file is `2008\\o_boikote_pekina.tgt`, there are 6 passing sentences (were 10 by regex)\n",
      "file is `2008\\pessimist_platit_dvazhdy.tgt`, there are 12 passing sentences (were 20 by regex)\n",
      "file is `2008\\poslednee_vystuplenie_prezidenta.tgt`, there are 1 passing sentences (were 5 by regex)\n",
      "file is `2008\\posledstviya_krizisa.tgt`, there are 0 passing sentences (were 1 by regex)\n",
      "file is `2008\\pritok_kapitala.tgt`, there are 1 passing sentences (were 4 by regex)\n",
      "file is `2008\\programma_alfa-shans.tgt`, there are 2 passing sentences (were 5 by regex)\n",
      "file is `2008\\rossiya_i_vto.tgt`, there are 4 passing sentences (were 6 by regex)\n",
      "file is `2008\\rynok.tgt`, there are 10 passing sentences (were 12 by regex)\n",
      "file is `2008\\tolkovateli_snov.tgt`, there are 16 passing sentences (were 22 by regex)\n",
      "file is `2008\\trudnosti_rosta.tgt`, there are 5 passing sentences (were 11 by regex)\n",
      "file is `2008\\tseny_na_produkty.tgt`, there are 9 passing sentences (were 12 by regex)\n",
      "file is `2008\\vneshnyaya_politika.tgt`, there are 21 passing sentences (were 29 by regex)\n",
      "file is `2008\\vse_molchat.tgt`, there are 1 passing sentences (were 4 by regex)\n",
      "file is `2008\\v_redaktorskom_kresle.tgt`, there are 16 passing sentences (were 19 by regex)\n",
      "file is `2008\\zhurnalistskoe_obrazovanie.tgt`, there are 10 passing sentences (were 20 by regex)\n",
      "file is `2009\\aprelskie_tezisy_surkova.tgt`, there are 10 passing sentences (were 12 by regex)\n",
      "file is `2009\\auditory.tgt`, there are 7 passing sentences (were 16 by regex)\n",
      "file is `2009\\avraam_linkoln.tgt`, there are 4 passing sentences (were 6 by regex)\n",
      "file is `2009\\biznes-obrazovanie.tgt`, there are 6 passing sentences (were 8 by regex)\n",
      "file is `2009\\chernye_dyry.tgt`, there are 3 passing sentences (were 3 by regex)\n",
      "file is `2009\\chlenstvo_v_vto.tgt`, there are 4 passing sentences (were 10 by regex)\n",
      "file is `2009\\dolgovaya_vyshka.tgt`, there are 7 passing sentences (were 15 by regex)\n",
      "file is `2009\\ege.tgt`, there are 7 passing sentences (were 10 by regex)\n",
      "file is `2009\\eto_tolko_nachalo.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2009\\fantasticheskie_perspektivy.tgt`, there are 2 passing sentences (were 4 by regex)\n",
      "file is `2009\\final_ligi_chempionov.tgt`, there are 12 passing sentences (were 14 by regex)\n",
      "file is `2009\\finansovyi_shtorm.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `2009\\firma_ne_garantiruet.tgt`, there are 2 passing sentences (were 6 by regex)\n",
      "file is `2009\\gadzhety-neudachniki.tgt`, there are 10 passing sentences (were 18 by regex)\n",
      "file is `2009\\gde_budet_vlast.tgt`, there are 3 passing sentences (were 6 by regex)\n",
      "file is `2009\\informatsionnoe_obshchestvo.tgt`, there are 9 passing sentences (were 21 by regex)\n",
      "file is `2009\\innovatsionnaya_ekonomika.tgt`, there are 6 passing sentences (were 14 by regex)\n",
      "file is `2009\\interviyu_medvedeva.tgt`, there are 3 passing sentences (were 6 by regex)\n",
      "file is `2009\\iz-pod_kontrolya.tgt`, there are 8 passing sentences (were 12 by regex)\n",
      "file is `2009\\izobreteno_vo_sne.tgt`, there are 10 passing sentences (were 18 by regex)\n",
      "file is `2009\\kholodnaya_voina.tgt`, there are 15 passing sentences (were 19 by regex)\n",
      "file is `2009\\krizis_blizhe.tgt`, there are 10 passing sentences (were 17 by regex)\n",
      "file is `2009\\kukla_gitlera.tgt`, there are 11 passing sentences (were 17 by regex)\n",
      "file is `2009\\letayushchaya_tarelka.tgt`, there are 17 passing sentences (were 24 by regex)\n",
      "file is `2009\\mgu.tgt`, there are 10 passing sentences (were 21 by regex)\n",
      "file is `2009\\molodtsy.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `2009\\moskva_privlekaet_menshe.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `2009\\nalogovaya_sistema.tgt`, there are 13 passing sentences (were 28 by regex)\n",
      "file is `2009\\natsionalnaya_assambleya.tgt`, there are 8 passing sentences (were 14 by regex)\n",
      "file is `2009\\nebesnye_formatsii.tgt`, there are 6 passing sentences (were 16 by regex)\n",
      "file is `2009\\nekhvatka_spetsialistov.tgt`, there are 2 passing sentences (were 8 by regex)\n",
      "file is `2009\\nelishnyaya_formalnost.tgt`, there are 11 passing sentences (were 19 by regex)\n",
      "file is `2009\\neudachnye_vremena.tgt`, there are 8 passing sentences (were 14 by regex)\n",
      "file is `2009\\nezhilye_pomeshcheniya.tgt`, there are 9 passing sentences (were 13 by regex)\n",
      "file is `2009\\novaya_model_obrazovaniya.tgt`, there are 5 passing sentences (were 5 by regex)\n",
      "file is `2009\\novaya_russkaya_ruletka.tgt`, there are 6 passing sentences (were 12 by regex)\n",
      "file is `2009\\novye_neformaly.tgt`, there are 9 passing sentences (were 15 by regex)\n",
      "file is `2009\\obrazovanie_-_novaya_model.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `2009\\obrazovatelnoe_kreditovanie.tgt`, there are 3 passing sentences (were 5 by regex)\n",
      "file is `2009\\obrushenie_shkoly.tgt`, there are 6 passing sentences (were 6 by regex)\n",
      "file is `2009\\oka.tgt`, there are 10 passing sentences (were 13 by regex)\n",
      "file is `2009\\ot_vlasti.tgt`, there are 15 passing sentences (were 28 by regex)\n",
      "file is `2009\\pastukh.tgt`, there are 9 passing sentences (were 13 by regex)\n",
      "file is `2009\\peremeny_v_pravitelstve.tgt`, there are 5 passing sentences (were 9 by regex)\n",
      "file is `2009\\pereselenie_na_yug.tgt`, there are 2 passing sentences (were 5 by regex)\n",
      "file is `2009\\podstava_na_ekzamene.tgt`, there are 2 passing sentences (were 5 by regex)\n",
      "file is `2009\\politicheskaya_zavisimost.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `2009\\pora_stavit_zadachi.tgt`, there are 5 passing sentences (were 13 by regex)\n",
      "file is `2009\\posledstviya_krizisa.tgt`, there are 5 passing sentences (were 12 by regex)\n",
      "file is `2009\\posledstviya_nezavisimosti.tgt`, there are 9 passing sentences (were 10 by regex)\n",
      "file is `2009\\poslevkusie.tgt`, there are 9 passing sentences (were 12 by regex)\n",
      "file is `2009\\povyshenie_tarifov.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2009\\pravila_igry.tgt`, there are 3 passing sentences (were 8 by regex)\n",
      "file is `2009\\prazdnik_obshchei_bedy.tgt`, there are 5 passing sentences (were 10 by regex)\n",
      "file is `2009\\predkrizisnyi_perekur.tgt`, there are 17 passing sentences (were 22 by regex)\n",
      "file is `2009\\predvybornaya_situatsiya.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `2009\\prizyvy_pravitelstva_zubkova.tgt`, there are 2 passing sentences (were 4 by regex)\n",
      "file is `2009\\problemy_obrazovaniya.tgt`, there are 10 passing sentences (were 18 by regex)\n",
      "file is `2009\\problemy_s_khlebom.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `2009\\prognoz_inflyatsii.tgt`, there are 13 passing sentences (were 20 by regex)\n",
      "file is `2009\\proryvy_goda.tgt`, there are 14 passing sentences (were 27 by regex)\n",
      "file is `2009\\protorgovavshiesya.tgt`, there are 10 passing sentences (were 18 by regex)\n",
      "file is `2009\\rabotat_nekomu.tgt`, there are 8 passing sentences (were 9 by regex)\n",
      "file is `2009\\razvilka.tgt`, there are 0 passing sentences (were 5 by regex)\n",
      "file is `2009\\reforma_nds.tgt`, there are 9 passing sentences (were 16 by regex)\n",
      "file is `2009\\reforma_obrazovaniya.tgt`, there are 11 passing sentences (were 15 by regex)\n",
      "file is `2009\\rossiya_nuzhdaetsya_v_migrantakh.tgt`, there are 4 passing sentences (were 11 by regex)\n",
      "file is `2009\\sanktsii.tgt`, there are 9 passing sentences (were 19 by regex)\n",
      "file is `2009\\sdavaisya_kto_mozhet.tgt`, there are 9 passing sentences (were 15 by regex)\n",
      "file is `2009\\shkola_kapitalizma.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `2009\\sinie_vorotnichki.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `2009\\skiltoi.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `2009\\srednii_klass.tgt`, there are 2 passing sentences (were 6 by regex)\n",
      "file is `2009\\status_vshe.tgt`, there are 14 passing sentences (were 20 by regex)\n",
      "file is `2009\\stipendiat.tgt`, there are 9 passing sentences (were 19 by regex)\n",
      "file is `2009\\svechenie_venery.tgt`, there are 2 passing sentences (were 3 by regex)\n",
      "file is `2009\\taktika_melkikh_shchelchkov.tgt`, there are 4 passing sentences (were 10 by regex)\n",
      "file is `2009\\tranzit_14.tgt`, there are 7 passing sentences (were 15 by regex)\n",
      "file is `2009\\tranzit_15.tgt`, there are 13 passing sentences (were 16 by regex)\n",
      "file is `2009\\tranzit_16.tgt`, there are 15 passing sentences (were 18 by regex)\n",
      "file is `2009\\tranzit_17.tgt`, there are 6 passing sentences (were 12 by regex)\n",
      "file is `2009\\tranzit_18.tgt`, there are 14 passing sentences (were 26 by regex)\n",
      "file is `2009\\trud_i_kapital.tgt`, there are 8 passing sentences (were 10 by regex)\n",
      "file is `2009\\tseny_na_zerno.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `2009\\tseny_rastut.tgt`, there are 2 passing sentences (were 6 by regex)\n",
      "file is `2009\\ugroza_deflyatsii.tgt`, there are 6 passing sentences (were 8 by regex)\n",
      "file is `2009\\velikaya_pochinka.tgt`, there are 9 passing sentences (were 17 by regex)\n",
      "file is `2009\\voina_vozrastov.tgt`, there are 5 passing sentences (were 9 by regex)\n",
      "file is `2009\\vozvrashchenie_dollara.tgt`, there are 1 passing sentences (were 2 by regex)\n",
      "file is `2009\\vysokie_tekhnologii.tgt`, there are 5 passing sentences (were 14 by regex)\n",
      "file is `2009\\vysshaya_shkola_ekonomiki.tgt`, there are 5 passing sentences (were 9 by regex)\n",
      "file is `2009\\vyzov_i_otvet.tgt`, there are 8 passing sentences (were 13 by regex)\n",
      "file is `2009\\v_ozhidanii_peremen.tgt`, there are 14 passing sentences (were 31 by regex)\n",
      "file is `2009\\v_zashchitu_torgashei.tgt`, there are 4 passing sentences (were 10 by regex)\n",
      "file is `2009\\zanimatelnye_tsifry.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `2009\\zapretnoe_iskusstvo_2006.tgt`, there are 24 passing sentences (were 32 by regex)\n",
      "file is `2009\\zavaly_rosobrnadzora.tgt`, there are 6 passing sentences (were 12 by regex)\n",
      "file is `2009\\za_pravoe_delo.tgt`, there are 3 passing sentences (were 11 by regex)\n",
      "file is `2009\\zhdet_li_nas_golod.tgt`, there are 4 passing sentences (were 8 by regex)\n",
      "file is `2009\\zhivaya_zemlya.tgt`, there are 7 passing sentences (were 12 by regex)\n",
      "file is `2010\\bolezn_rosta.tgt`, there are 20 passing sentences (were 30 by regex)\n",
      "file is `2010\\chistka_v_sporte.tgt`, there are 12 passing sentences (were 15 by regex)\n",
      "file is `2010\\ekonomisty.tgt`, there are 6 passing sentences (were 13 by regex)\n",
      "file is `2010\\ekstremizm.tgt`, there are 9 passing sentences (were 14 by regex)\n",
      "file is `2010\\filmy_timura_bekmambetova.tgt`, there are 8 passing sentences (were 18 by regex)\n",
      "file is `2010\\fizkultura_i_okolo.tgt`, there are 11 passing sentences (were 16 by regex)\n",
      "file is `2010\\gibridomobil.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2010\\karpov-kasparov.tgt`, there are 18 passing sentences (were 25 by regex)\n",
      "file is `2010\\khudozhestvennaya_gimnastika.tgt`, there are 17 passing sentences (were 21 by regex)\n",
      "file is `2010\\konkurs_otmenyaetsya.tgt`, there are 6 passing sentences (were 9 by regex)\n",
      "file is `2010\\korrektsiya_mifov.tgt`, there are 11 passing sentences (were 12 by regex)\n",
      "file is `2010\\kovcheg_zhana_vanie.tgt`, there are 23 passing sentences (were 30 by regex)\n",
      "file is `2010\\liberaly_sdayutsya.tgt`, there are 18 passing sentences (were 26 by regex)\n",
      "file is `2010\\lyzhi.tgt`, there are 7 passing sentences (were 10 by regex)\n",
      "file is `2010\\mars_3.tgt`, there are 20 passing sentences (were 29 by regex)\n",
      "file is `2010\\matematika_probok.tgt`, there are 2 passing sentences (were 6 by regex)\n",
      "file is `2010\\mezhdunarodnaya_olimpiada.tgt`, there are 1 passing sentences (were 5 by regex)\n",
      "file is `2010\\mozgi_naprokat.tgt`, there are 19 passing sentences (were 30 by regex)\n",
      "file is `2010\\nadpisi_iz_doliny_inda.tgt`, there are 10 passing sentences (were 14 by regex)\n",
      "file is `2010\\paraolimpiitsy.tgt`, there are 11 passing sentences (were 19 by regex)\n",
      "file is `2010\\peregrev_ekonomiki.tgt`, there are 8 passing sentences (were 11 by regex)\n",
      "file is `2010\\prichiny_krizisa.tgt`, there are 7 passing sentences (were 12 by regex)\n",
      "file is `2010\\profilnoe_obrazovanie.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `2010\\pryamoi_ukol.tgt`, there are 3 passing sentences (were 5 by regex)\n",
      "file is `2010\\psevdoobrazovanie.tgt`, there are 14 passing sentences (were 22 by regex)\n",
      "file is `2010\\realnaya_politika.tgt`, there are 9 passing sentences (were 18 by regex)\n",
      "file is `2010\\rodnye_i_priemnye.tgt`, there are 7 passing sentences (were 9 by regex)\n",
      "file is `2010\\rossiiskie_biznes-shkoly.tgt`, there are 17 passing sentences (were 27 by regex)\n",
      "file is `2010\\russkii_dom.tgt`, there are 9 passing sentences (were 23 by regex)\n",
      "file is `2010\\sokrashchenie_vuzov.tgt`, there are 1 passing sentences (were 5 by regex)\n",
      "file is `2010\\spasenie_evro.tgt`, there are 8 passing sentences (were 20 by regex)\n",
      "file is `2010\\tsilindry_faraona.tgt`, there are 19 passing sentences (were 34 by regex)\n",
      "file is `2010\\v_perevode_s_nebesnogo.tgt`, there are 13 passing sentences (were 18 by regex)\n",
      "file is `2011\\alpinizm.tgt`, there are 17 passing sentences (were 37 by regex)\n",
      "file is `2011\\apraushev.tgt`, there are 14 passing sentences (were 19 by regex)\n",
      "file is `2011\\banan.tgt`, there are 6 passing sentences (were 11 by regex)\n",
      "file is `2011\\chto_takoe_ukhomor.tgt`, there are 2 passing sentences (were 4 by regex)\n",
      "file is `2011\\facebook.tgt`, there are 19 passing sentences (were 27 by regex)\n",
      "file is `2011\\formula-1.tgt`, there are 36 passing sentences (were 89 by regex)\n",
      "file is `2011\\gde_zhe_shliman.tgt`, there are 9 passing sentences (were 12 by regex)\n",
      "file is `2011\\grisha_perelman_devyatyi_genii.tgt`, there are 20 passing sentences (were 23 by regex)\n",
      "file is `2011\\interviyu_mariny_astvatsaturyan.tgt`, there are 4 passing sentences (were 9 by regex)\n",
      "file is `2011\\lekarstva_ot_ustalosti.tgt`, there are 15 passing sentences (were 20 by regex)\n",
      "file is `2011\\mariam_petrosyan.tgt`, there are 36 passing sentences (were 53 by regex)\n",
      "file is `2011\\nano.tgt`, there are 31 passing sentences (were 53 by regex)\n",
      "file is `2011\\otkrytoe_pismo.tgt`, there are 7 passing sentences (were 10 by regex)\n",
      "file is `2011\\petrushka.tgt`, there are 27 passing sentences (were 40 by regex)\n",
      "file is `2011\\planerizm.tgt`, there are 13 passing sentences (were 37 by regex)\n",
      "file is `2011\\pochemu_my_ne_s_nimi.tgt`, there are 19 passing sentences (were 23 by regex)\n",
      "file is `2011\\pri_popustitelstve_robotov.tgt`, there are 6 passing sentences (were 16 by regex)\n",
      "file is `2011\\radioastron.tgt`, there are 2 passing sentences (were 3 by regex)\n",
      "file is `2011\\shakhmaty.tgt`, there are 34 passing sentences (were 59 by regex)\n",
      "file is `2011\\sinkhronnoe_plavanie.tgt`, there are 7 passing sentences (were 12 by regex)\n",
      "file is `2011\\sverkhkorotkoe_vremya.tgt`, there are 33 passing sentences (were 60 by regex)\n",
      "file is `2011\\tekhnologii.tgt`, there are 10 passing sentences (were 14 by regex)\n",
      "file is `2011\\traektoriya_ekologicheskoi_mysli.tgt`, there are 29 passing sentences (were 52 by regex)\n",
      "file is `2011\\troitsk_kak_tsentr_novoi_moskvy.tgt`, there are 10 passing sentences (were 13 by regex)\n",
      "file is `2012\\chto_delat_posle_24_dekabrya.tgt`, there are 4 passing sentences (were 11 by regex)\n",
      "file is `2012\\fobos-grunt_gibel_mechty.tgt`, there are 22 passing sentences (were 40 by regex)\n",
      "file is `2012\\fotograf_vladimir_mishukov_sindrom_platona.tgt`, there are 14 passing sentences (were 23 by regex)\n",
      "file is `2012\\galileo_galilei.tgt`, there are 87 passing sentences (were 124 by regex)\n",
      "file is `2012\\gosduma_prinyala_zakon_o_kastratsii_pedofilov.tgt`, there are 7 passing sentences (were 13 by regex)\n",
      "file is `2012\\itogi_subboty.tgt`, there are 5 passing sentences (were 14 by regex)\n",
      "file is `2012\\iz_pochtovykh_yashchikov.tgt`, there are 15 passing sentences (were 17 by regex)\n",
      "file is `2012\\kak_sokhranit_gumanitarnuyu_nauku.tgt`, there are 4 passing sentences (were 10 by regex)\n",
      "file is `2012\\kak_voevat_s_lzhenaukoi.tgt`, there are 22 passing sentences (were 26 by regex)\n",
      "file is `2012\\kak_zakupat_dorogostoyashchee_oborudovanie.tgt`, there are 9 passing sentences (were 14 by regex)\n",
      "file is `2012\\kolchak.tgt`, there are 65 passing sentences (were 101 by regex)\n",
      "file is `2012\\kooperativ_u_ozera.tgt`, there are 12 passing sentences (were 24 by regex)\n",
      "file is `2012\\lozh_i_legitimnost.tgt`, there are 25 passing sentences (were 32 by regex)\n",
      "file is `2012\\marko_polo.tgt`, there are 15 passing sentences (were 22 by regex)\n",
      "file is `2012\\ne_mogu_bolshe_molchat.tgt`, there are 36 passing sentences (were 47 by regex)\n",
      "file is `2012\\okhotnik_na_palachei.tgt`, there are 57 passing sentences (were 87 by regex)\n",
      "file is `2012\\on_boitsya_konkurentsii.tgt`, there are 5 passing sentences (were 12 by regex)\n",
      "file is `2012\\otluchenie_tserkvi.tgt`, there are 30 passing sentences (were 43 by regex)\n",
      "file is `2012\\plan_razvitiya_vmf.tgt`, there are 0 passing sentences (were 0 by regex)\n",
      "file is `2012\\poslednie_russkie.tgt`, there are 23 passing sentences (were 40 by regex)\n",
      "file is `2012\\problema_vybora.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `2012\\putin_vzyal_kurs_na_agressivnoe_uderzhanie_status-kvo.tgt`, there are 17 passing sentences (were 32 by regex)\n",
      "file is `2012\\repressii.tgt`, there are 13 passing sentences (were 12 by regex)\n",
      "file is `2012\\sluzhivaya_kvazinauka.tgt`, there are 9 passing sentences (were 18 by regex)\n",
      "file is `2012\\sovety_oppozitsii.tgt`, there are 14 passing sentences (were 22 by regex)\n",
      "file is `2012\\stalin_vozvrashchalsya.tgt`, there are 8 passing sentences (were 17 by regex)\n",
      "file is `2012\\teorema_stolypina.tgt`, there are 20 passing sentences (were 27 by regex)\n",
      "file is `2012\\v_konstitutsii_ne_dolzhno_byt_mesta_dlya_vozhdya.tgt`, there are 35 passing sentences (were 43 by regex)\n",
      "file is `2013\\algoritm.tgt`, there are 26 passing sentences (were 34 by regex)\n",
      "file is `2013\\ataman_vikhr.tgt`, there are 26 passing sentences (were 40 by regex)\n",
      "file is `2013\\budushchee_medvedeva.tgt`, there are 4 passing sentences (were 6 by regex)\n",
      "file is `2013\\davaite_zhit_smelo.tgt`, there are 10 passing sentences (were 17 by regex)\n",
      "file is `2013\\doktor_z.tgt`, there are 28 passing sentences (were 30 by regex)\n",
      "file is `2013\\doroga_v_nikuda.tgt`, there are 22 passing sentences (were 25 by regex)\n",
      "file is `2013\\dvadtsat_let_spustya.tgt`, there are 22 passing sentences (were 40 by regex)\n",
      "file is `2013\\es-ofshor.tgt`, there are 11 passing sentences (were 17 by regex)\n",
      "file is `2013\\esli_by_k_vlasti.tgt`, there are 9 passing sentences (were 9 by regex)\n",
      "file is `2013\\kamennyi_roi.tgt`, there are 6 passing sentences (were 10 by regex)\n",
      "file is `2013\\katyn.tgt`, there are 12 passing sentences (were 30 by regex)\n",
      "file is `2013\\khidzhaby_ne_zakazyvali.tgt`, there are 4 passing sentences (were 11 by regex)\n",
      "file is `2013\\kitaiskii_krossvord.tgt`, there are 7 passing sentences (were 10 by regex)\n",
      "file is `2013\\kooperativ_lavka.tgt`, there are 12 passing sentences (were 27 by regex)\n",
      "file is `2013\\literatura.tgt`, there are 17 passing sentences (were 22 by regex)\n",
      "file is `2013\\martovskaya_revolyutsiya.tgt`, there are 23 passing sentences (were 36 by regex)\n",
      "file is `2013\\maski.tgt`, there are 26 passing sentences (were 26 by regex)\n",
      "file is `2013\\minfin_protiv_glazieva.tgt`, there are 4 passing sentences (were 5 by regex)\n",
      "file is `2013\\nauka.tgt`, there are 41 passing sentences (were 64 by regex)\n",
      "file is `2013\\plody_alternativnogo_prosveshcheniya.tgt`, there are 31 passing sentences (were 53 by regex)\n",
      "file is `2013\\podarok_dlya_vsekh.tgt`, there are 17 passing sentences (were 24 by regex)\n",
      "file is `2013\\poslednii_dovod_issledovatelya.tgt`, there are 36 passing sentences (were 57 by regex)\n",
      "file is `2013\\poslednyaya_taina.tgt`, there are 3 passing sentences (were 5 by regex)\n",
      "file is `2013\\povedenie.tgt`, there are 12 passing sentences (were 23 by regex)\n",
      "file is `2013\\pravo_na_dukhovnuyu_pomoshch.tgt`, there are 1 passing sentences (were 8 by regex)\n",
      "file is `2013\\pravo_na_obidu.tgt`, there are 10 passing sentences (were 20 by regex)\n",
      "file is `2013\\preodolet_kompleks_liliputa.tgt`, there are 8 passing sentences (were 10 by regex)\n",
      "file is `2013\\putin_idet_na_ty.tgt`, there are 11 passing sentences (were 13 by regex)\n",
      "file is `2013\\rossiya_na_evropeiskom_fone.tgt`, there are 15 passing sentences (were 26 by regex)\n",
      "file is `2013\\stalingrad.tgt`, there are 26 passing sentences (were 38 by regex)\n",
      "file is `2013\\strast_dushi.tgt`, there are 15 passing sentences (were 19 by regex)\n",
      "file is `2013\\sumerki_rossiiskoi_korruptsii.tgt`, there are 9 passing sentences (were 9 by regex)\n",
      "file is `2013\\sverkhnovaya_ekonomika.tgt`, there are 15 passing sentences (were 26 by regex)\n",
      "file is `2013\\sverzhenie_svetskogo_stroya.tgt`, there are 9 passing sentences (were 14 by regex)\n",
      "file is `2013\\transplantatsiya_organov.tgt`, there are 11 passing sentences (were 17 by regex)\n",
      "file is `2013\\uchebnik_istorii.tgt`, there are 7 passing sentences (were 16 by regex)\n",
      "file is `2013\\ukroshchenie_stroptivogo_naukograda.tgt`, there are 20 passing sentences (were 24 by regex)\n",
      "file is `2013\\zhivut_ne_dlya_radosti.tgt`, there are 5 passing sentences (were 12 by regex)\n",
      "file is `2014\\bogi_menyayutsya.tgt`, there are 14 passing sentences (were 24 by regex)\n",
      "file is `2014\\chto_budet_posle.tgt`, there are 59 passing sentences (were 97 by regex)\n",
      "file is `2014\\chto_takoe_interesno.tgt`, there are 11 passing sentences (were 15 by regex)\n",
      "file is `2014\\geroi.tgt`, there are 77 passing sentences (were 91 by regex)\n",
      "file is `2014\\granin.tgt`, there are 21 passing sentences (were 26 by regex)\n",
      "file is `2014\\kulturnye_olimpiitsy.tgt`, there are 19 passing sentences (were 36 by regex)\n",
      "file is `2014\\na_dvukh_voinakh_1.tgt`, there are 74 passing sentences (were 116 by regex)\n",
      "file is `2014\\na_dvukh_voinakh_2.tgt`, there are 44 passing sentences (were 68 by regex)\n",
      "file is `2014\\o_chem_rech.tgt`, there are 11 passing sentences (were 26 by regex)\n",
      "file is `2014\\sverkhnovaya_2.tgt`, there are 37 passing sentences (were 70 by regex)\n",
      "file is `2014\\travin_1.tgt`, there are 54 passing sentences (were 90 by regex)\n",
      "file is `2014\\travin_2.tgt`, there are 48 passing sentences (were 71 by regex)\n",
      "file is `2014\\vladimir_vladimirovich.tgt`, there are 32 passing sentences (were 46 by regex)\n",
      "file is `2014\\vspolokhi.tgt`, there are 65 passing sentences (were 107 by regex)\n",
      "file is `2014\\vysota.tgt`, there are 20 passing sentences (were 36 by regex)\n",
      "file is `2015\\blokadnoe_detstvo.tgt`, there are 67 passing sentences (were 90 by regex)\n",
      "file is `2015\\donbass.tgt`, there are 8 passing sentences (were 13 by regex)\n",
      "file is `2015\\glavnaya_taina_nachala_voiny.tgt`, there are 20 passing sentences (were 31 by regex)\n",
      "file is `2015\\gorbachev_kak_menya_khoronili.tgt`, there are 18 passing sentences (were 25 by regex)\n",
      "file is `2015\\grekhi_ukrainskie.tgt`, there are 11 passing sentences (were 15 by regex)\n",
      "file is `2015\\interviyu_afanasieva.tgt`, there are 13 passing sentences (were 17 by regex)\n",
      "file is `2015\\interviyu_dvorkina.tgt`, there are 13 passing sentences (were 18 by regex)\n",
      "file is `2015\\interviyu_yavlinskogo.tgt`, there are 12 passing sentences (were 23 by regex)\n",
      "file is `2015\\interviyu_yurgensa.tgt`, there are 13 passing sentences (were 21 by regex)\n",
      "file is `2015\\iz_semeinykh_predanii.tgt`, there are 30 passing sentences (were 46 by regex)\n",
      "file is `2015\\kriptografiya.tgt`, there are 12 passing sentences (were 19 by regex)\n",
      "file is `2015\\nelegalnye_kholodilniki.tgt`, there are 12 passing sentences (were 21 by regex)\n",
      "file is `2015\\ne_nado_vozvrashchatsya_k_yalte.tgt`, there are 13 passing sentences (were 17 by regex)\n",
      "file is `2015\\oppozitsiyu_ne_pustyat.tgt`, there are 13 passing sentences (were 20 by regex)\n",
      "file is `2015\\opravdanie_televideniya.tgt`, there are 30 passing sentences (were 49 by regex)\n",
      "file is `2015\\pervyi_avtomobil.tgt`, there are 13 passing sentences (were 28 by regex)\n",
      "file is `2015\\poroshenko_dolzhen_pobedit.tgt`, there are 31 passing sentences (were 46 by regex)\n",
      "file is `2015\\pust_dolgim_budet_razgovor.tgt`, there are 24 passing sentences (were 33 by regex)\n",
      "file is `2015\\razvorot_nad_atlantidoi.tgt`, there are 9 passing sentences (were 18 by regex)\n",
      "file is `2015\\rezhimnyi_subiekt.tgt`, there are 14 passing sentences (were 22 by regex)\n",
      "file is `2015\\spiski_shindlerov.tgt`, there are 50 passing sentences (were 78 by regex)\n",
      "file is `2015\\strakhovanie_vkladov_1.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `2015\\strakhovanie_vkladov_2.tgt`, there are 6 passing sentences (were 9 by regex)\n",
      "file is `2015\\velikaya_neftegazovaya.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `2016\\chelovek_chitayushchii.tgt`, there are 10 passing sentences (were 13 by regex)\n",
      "file is `2016\\gumanisticheskaya_pedagogika.tgt`, there are 29 passing sentences (were 54 by regex)\n",
      "file is `2016\\migranty_ekspaty_kosmopolity.tgt`, there are 16 passing sentences (were 20 by regex)\n",
      "file is `2016\\okna_zhizni.tgt`, there are 20 passing sentences (were 36 by regex)\n",
      "file is `2016\\slovo_ne_to_chem_kazhetsya.tgt`, there are 28 passing sentences (were 39 by regex)\n",
      "file is `2016\\tselina_glazami_pervokursnika.tgt`, there are 22 passing sentences (were 35 by regex)\n",
      "file is `2016\\ugroza_ravnopraviem.tgt`, there are 7 passing sentences (were 17 by regex)\n",
      "file is `2016\\v_chem_vinovat_putin.tgt`, there are 32 passing sentences (were 48 by regex)\n",
      "file is `2017\\apresyan_1.tgt`, there are 12 passing sentences (were 20 by regex)\n",
      "file is `2017\\chto_gubit_nauku.tgt`, there are 9 passing sentences (were 19 by regex)\n",
      "file is `2017\\eltsin.tgt`, there are 24 passing sentences (were 33 by regex)\n",
      "file is `2017\\gde_zhe_vykhod.tgt`, there are 48 passing sentences (were 75 by regex)\n",
      "file is `2017\\kulturnaya_revolyutsiya.tgt`, there are 35 passing sentences (were 43 by regex)\n",
      "file is `2017\\nad_sinaem.tgt`, there are 3 passing sentences (were 6 by regex)\n",
      "file is `2017\\poet_i_prezident.tgt`, there are 12 passing sentences (were 20 by regex)\n",
      "file is `2017\\siriiskaya_lovushka.tgt`, there are 2 passing sentences (were 7 by regex)\n",
      "file is `2017\\ubiistvo_pavla_sheremeta.tgt`, there are 8 passing sentences (were 12 by regex)\n",
      "file is `2017\\vsevidyashchee_oko_armii.tgt`, there are 2 passing sentences (were 4 by regex)\n",
      "file is `2017\\v_poiskakh_buntovshchikov.tgt`, there are 14 passing sentences (were 25 by regex)\n",
      "file is `2018\\akademicheskii_schet.tgt`, there are 9 passing sentences (were 14 by regex)\n",
      "file is `2018\\boris_titov_i_1000_advokatov.tgt`, there are 16 passing sentences (were 24 by regex)\n",
      "file is `2018\\budet_chto_vspomnit.tgt`, there are 4 passing sentences (were 9 by regex)\n",
      "file is `2018\\byt_li_nam_vmeste.tgt`, there are 15 passing sentences (were 27 by regex)\n",
      "file is `2018\\dissidenty.tgt`, there are 28 passing sentences (were 46 by regex)\n",
      "file is `2018\\dvukhpartiinaya_rossiya.tgt`, there are 5 passing sentences (were 11 by regex)\n",
      "file is `2018\\gorod-sad_singapur.tgt`, there are 7 passing sentences (were 16 by regex)\n",
      "file is `2018\\interviyu_granina_raznykh_let.tgt`, there are 12 passing sentences (were 19 by regex)\n",
      "file is `2018\\kto_za_trampa.tgt`, there are 19 passing sentences (were 27 by regex)\n",
      "file is `2018\\legkoe_dykhanie.tgt`, there are 9 passing sentences (were 18 by regex)\n",
      "file is `2018\\nachalo.tgt`, there are 66 passing sentences (were 80 by regex)\n",
      "file is `2018\\novye_maiskie_ukazy.tgt`, there are 8 passing sentences (were 21 by regex)\n",
      "file is `2018\\odnazhdy_v_taruse.tgt`, there are 11 passing sentences (were 17 by regex)\n",
      "file is `2018\\pensionnyi_vozrast.tgt`, there are 1 passing sentences (were 3 by regex)\n",
      "file is `2018\\pontii_pilat.tgt`, there are 88 passing sentences (were 139 by regex)\n",
      "file is `2018\\poseshchenie_muzeya.tgt`, there are 25 passing sentences (were 51 by regex)\n",
      "file is `2018\\puteshestvie_v_armeniyu.tgt`, there are 80 passing sentences (were 123 by regex)\n",
      "file is `2018\\pyatichasovoi_skoryi.tgt`, there are 38 passing sentences (were 70 by regex)\n",
      "file is `2018\\rech_zaliznyaka.tgt`, there are 19 passing sentences (were 19 by regex)\n",
      "file is `2018\\roman_pak_u_granina.tgt`, there are 16 passing sentences (were 23 by regex)\n",
      "file is `2018\\v_strane_mormonov.tgt`, there are 50 passing sentences (were 71 by regex)\n",
      "file is `2018\\zvukovidy.tgt`, there are 20 passing sentences (were 34 by regex)\n",
      "file is `2019\\baden-baden.tgt`, there are 15 passing sentences (were 32 by regex)\n",
      "file is `2019\\dvadtsat_let_defoltu.tgt`, there are 12 passing sentences (were 23 by regex)\n",
      "file is `2019\\interviyu_nazarova.tgt`, there are 17 passing sentences (were 27 by regex)\n",
      "file is `2019\\kartinki_imperii.tgt`, there are 47 passing sentences (were 72 by regex)\n",
      "file is `2019\\lingvistika_i_morzhevanie.tgt`, there are 4 passing sentences (were 9 by regex)\n",
      "file is `2019\\medvedev_ostanetsya_s_partiei.tgt`, there are 6 passing sentences (were 11 by regex)\n",
      "file is `2019\\obraztsovyi_oppozitsionnyi_kandidat.tgt`, there are 34 passing sentences (were 53 by regex)\n",
      "file is `2019\\ochen_affektnoe_ubiistvo.tgt`, there are 20 passing sentences (were 24 by regex)\n",
      "file is `2019\\podkidysh.tgt`, there are 84 passing sentences (were 125 by regex)\n",
      "file is `2019\\pokhozhdeniya_ryadovogo_podpisanta.tgt`, there are 18 passing sentences (were 29 by regex)\n",
      "file is `2019\\poslednyaya_prosba.tgt`, there are 24 passing sentences (were 36 by regex)\n",
      "file is `2019\\publichnye_meropriyatiya.tgt`, there are 23 passing sentences (were 26 by regex)\n",
      "file is `2019\\rasstrel_v_kerchi.tgt`, there are 10 passing sentences (were 17 by regex)\n",
      "file is `2019\\reforma_ran.tgt`, there are 20 passing sentences (were 35 by regex)\n",
      "file is `2019\\sekta_radikalnogo_feminizma.tgt`, there are 15 passing sentences (were 20 by regex)\n",
      "file is `2019\\sotvori_sebe_bankira.tgt`, there are 20 passing sentences (were 35 by regex)\n",
      "file is `2019\\vtoraya_mirovaya_voina.tgt`, there are 88 passing sentences (were 160 by regex)\n",
      "file is `2019\\zanimatelnaya_estetika.tgt`, there are 46 passing sentences (were 77 by regex)\n",
      "file is `2019\\znaet_li_ob_etom_vozhd.tgt`, there are 16 passing sentences (were 33 by regex)\n",
      "file is `2020_corpus2_0\\chelovek_okazalsya_ne_nuzhen.tgt`, there are 3 passing sentences (were 6 by regex)\n",
      "file is `2020_corpus2_0\\fenomen_antichnosti.tgt`, there are 27 passing sentences (were 40 by regex)\n",
      "file is `2020_corpus2_0\\gastronomicheskie_khroniki.tgt`, there are 63 passing sentences (were 100 by regex)\n",
      "file is `2020_corpus2_0\\iz_zhizni_odnogo_ezha.tgt`, there are 31 passing sentences (were 52 by regex)\n",
      "file is `2020_corpus2_0\\kakie_nashi_gody_1.tgt`, there are 75 passing sentences (were 122 by regex)\n",
      "file is `2020_corpus2_0\\khochu_byt_negrom.tgt`, there are 28 passing sentences (were 35 by regex)\n",
      "file is `2020_corpus2_0\\konets_ekonomicheskogo_rosta.tgt`, there are 20 passing sentences (were 30 by regex)\n",
      "file is `2020_corpus2_0\\kto_nasledil.tgt`, there are 24 passing sentences (were 30 by regex)\n",
      "file is `2020_corpus2_0\\malenkii_prints.tgt`, there are 84 passing sentences (were 137 by regex)\n",
      "file is `2020_corpus2_0\\milaya_reptiliya.tgt`, there are 29 passing sentences (were 46 by regex)\n",
      "file is `2020_corpus2_0\\muraviinyi_kosmos.tgt`, there are 39 passing sentences (were 50 by regex)\n",
      "file is `2020_corpus2_0\\papa-yumorist.tgt`, there are 40 passing sentences (were 65 by regex)\n",
      "file is `2020_corpus2_0\\parallelnyi_mir.tgt`, there are 28 passing sentences (were 47 by regex)\n",
      "file is `2020_corpus2_0\\peredacha.tgt`, there are 33 passing sentences (were 56 by regex)\n",
      "file is `2020_corpus2_0\\pisma_schastiya.tgt`, there are 31 passing sentences (were 43 by regex)\n",
      "file is `2020_corpus2_0\\poslannik_iz_proshlogo.tgt`, there are 29 passing sentences (were 44 by regex)\n",
      "file is `2020_corpus2_0\\velikoe_pereselenie.tgt`, there are 24 passing sentences (were 46 by regex)\n",
      "file is `2020_rffi\\gryadet_li_apokalipsis.tgt`, there are 9 passing sentences (were 26 by regex)\n",
      "file is `2020_rffi\\indoneziya_ot_ada_do_raya.tgt`, there are 53 passing sentences (were 97 by regex)\n",
      "file is `2020_rffi\\instrument_yazyka_1.tgt`, there are 61 passing sentences (were 96 by regex)\n",
      "file is `2020_rffi\\instrument_yazyka_2.tgt`, there are 80 passing sentences (were 118 by regex)\n",
      "file is `2020_rffi\\moskva_2042.tgt`, there are 73 passing sentences (were 122 by regex)\n",
      "file is `2020_rffi\\nyuta_federmesser.tgt`, there are 31 passing sentences (were 47 by regex)\n",
      "file is `2020_rffi\\ognev.tgt`, there are 51 passing sentences (were 82 by regex)\n",
      "file is `2020_rffi\\trudnosti_ponimaniya.tgt`, there are 31 passing sentences (were 65 by regex)\n",
      "file is `news\\meinert-ranks.tgt`, there are 17 passing sentences (were 31 by regex)\n",
      "file is `news\\panyushkin.tgt`, there are 6 passing sentences (were 13 by regex)\n",
      "file is `news\\ya_01.tgt`, there are 8 passing sentences (were 13 by regex)\n",
      "file is `news\\ya_02.tgt`, there are 5 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_03.tgt`, there are 6 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_04.tgt`, there are 2 passing sentences (were 6 by regex)\n",
      "file is `news\\ya_05.tgt`, there are 2 passing sentences (were 7 by regex)\n",
      "file is `news\\ya_06.tgt`, there are 5 passing sentences (were 12 by regex)\n",
      "file is `news\\ya_09.tgt`, there are 7 passing sentences (were 10 by regex)\n",
      "file is `news\\ya_10.tgt`, there are 4 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_11.tgt`, there are 3 passing sentences (were 4 by regex)\n",
      "file is `news\\ya_12.tgt`, there are 7 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_13.tgt`, there are 1 passing sentences (were 10 by regex)\n",
      "file is `news\\ya_14.tgt`, there are 2 passing sentences (were 12 by regex)\n",
      "file is `news\\ya_15.tgt`, there are 4 passing sentences (were 13 by regex)\n",
      "file is `news\\ya_16.tgt`, there are 2 passing sentences (were 12 by regex)\n",
      "file is `news\\ya_17.tgt`, there are 9 passing sentences (were 20 by regex)\n",
      "file is `news\\ya_18.tgt`, there are 9 passing sentences (were 13 by regex)\n",
      "file is `news\\ya_19.tgt`, there are 4 passing sentences (were 10 by regex)\n",
      "file is `news\\ya_20.tgt`, there are 6 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_21.tgt`, there are 2 passing sentences (were 8 by regex)\n",
      "file is `news\\ya_22.tgt`, there are 15 passing sentences (were 16 by regex)\n",
      "file is `news\\ya_23.tgt`, there are 3 passing sentences (were 8 by regex)\n",
      "file is `news\\ya_24.tgt`, there are 7 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_25.tgt`, there are 6 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_26.tgt`, there are 11 passing sentences (were 16 by regex)\n",
      "file is `news\\ya_27.tgt`, there are 2 passing sentences (were 8 by regex)\n",
      "file is `news\\ya_30.tgt`, there are 2 passing sentences (were 12 by regex)\n",
      "file is `news\\ya_31.tgt`, there are 2 passing sentences (were 8 by regex)\n",
      "file is `news\\ya_32.tgt`, there are 7 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_33.tgt`, there are 1 passing sentences (were 6 by regex)\n",
      "file is `news\\ya_34.tgt`, there are 0 passing sentences (were 4 by regex)\n",
      "file is `news\\ya_35.tgt`, there are 1 passing sentences (were 4 by regex)\n",
      "file is `news\\ya_36.tgt`, there are 7 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_37.tgt`, there are 6 passing sentences (were 12 by regex)\n",
      "file is `news\\ya_38.tgt`, there are 4 passing sentences (were 10 by regex)\n",
      "file is `news\\ya_39.tgt`, there are 3 passing sentences (were 8 by regex)\n",
      "file is `news\\ya_40.tgt`, there are 4 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_41.tgt`, there are 6 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_42.tgt`, there are 6 passing sentences (were 15 by regex)\n",
      "file is `news\\ya_43.tgt`, there are 5 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_44.tgt`, there are 5 passing sentences (were 6 by regex)\n",
      "file is `news\\ya_45.tgt`, there are 6 passing sentences (were 12 by regex)\n",
      "file is `news\\ya_46.tgt`, there are 1 passing sentences (were 2 by regex)\n",
      "file is `news\\ya_47.tgt`, there are 0 passing sentences (were 4 by regex)\n",
      "file is `news\\ya_48.tgt`, there are 4 passing sentences (were 8 by regex)\n",
      "file is `news\\ya_49.tgt`, there are 6 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_50.tgt`, there are 4 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_51.tgt`, there are 5 passing sentences (were 13 by regex)\n",
      "file is `news\\ya_52.tgt`, there are 1 passing sentences (were 7 by regex)\n",
      "file is `news\\ya_53.tgt`, there are 4 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_54.tgt`, there are 3 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_55.tgt`, there are 8 passing sentences (were 10 by regex)\n",
      "file is `news\\ya_56.tgt`, there are 1 passing sentences (were 3 by regex)\n",
      "file is `news\\ya_58.tgt`, there are 6 passing sentences (were 14 by regex)\n",
      "file is `news\\ya_59.tgt`, there are 7 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_60.tgt`, there are 8 passing sentences (were 16 by regex)\n",
      "file is `news\\ya_70.tgt`, there are 4 passing sentences (were 16 by regex)\n",
      "file is `news\\ya_72.tgt`, there are 10 passing sentences (were 17 by regex)\n",
      "file is `news\\ya_73.tgt`, there are 2 passing sentences (were 9 by regex)\n",
      "file is `news\\ya_74.tgt`, there are 7 passing sentences (were 11 by regex)\n",
      "file is `news\\ya_75.tgt`, there are 14 passing sentences (were 21 by regex)\n",
      "file is `news\\ya_76.tgt`, there are 3 passing sentences (were 6 by regex)\n",
      "file is `news\\ya_77.tgt`, there are 3 passing sentences (were 4 by regex)\n",
      "file is `news\\ya_78.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `news\\ya_79.tgt`, there are 6 passing sentences (were 17 by regex)\n",
      "file is `news\\ya_90.tgt`, there are 5 passing sentences (were 12 by regex)\n",
      "file is `news\\ya_91.tgt`, there are 0 passing sentences (were 0 by regex)\n",
      "file is `news\\ya_92.tgt`, there are 3 passing sentences (were 6 by regex)\n",
      "file is `news\\ya_93.tgt`, there are 10 passing sentences (were 21 by regex)\n",
      "file is `news\\ya_94.tgt`, there are 7 passing sentences (were 10 by regex)\n",
      "file is `news\\ya_96.tgt`, there are 1 passing sentences (were 4 by regex)\n",
      "file is `news\\ya_97.tgt`, there are 6 passing sentences (were 9 by regex)\n",
      "file is `uppsala\\bitov_1.tgt`, there are 37 passing sentences (were 66 by regex)\n",
      "file is `uppsala\\bitov_2.tgt`, there are 28 passing sentences (were 54 by regex)\n",
      "file is `uppsala\\bitov_3.tgt`, there are 37 passing sentences (were 53 by regex)\n",
      "file is `uppsala\\grekova_1.tgt`, there are 29 passing sentences (were 42 by regex)\n",
      "file is `uppsala\\grekova_2.tgt`, there are 45 passing sentences (were 68 by regex)\n",
      "file is `uppsala\\grekova_3.tgt`, there are 32 passing sentences (were 54 by regex)\n",
      "file is `uppsala\\korp_210.tgt`, there are 22 passing sentences (were 34 by regex)\n",
      "file is `uppsala\\korp_220.tgt`, there are 12 passing sentences (were 25 by regex)\n",
      "file is `uppsala\\korp_230.tgt`, there are 3 passing sentences (were 3 by regex)\n",
      "file is `uppsala\\korp_240.tgt`, there are 22 passing sentences (were 37 by regex)\n",
      "file is `uppsala\\korp_250.tgt`, there are 5 passing sentences (were 12 by regex)\n",
      "file is `uppsala\\korp_261.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_262.tgt`, there are 3 passing sentences (were 4 by regex)\n",
      "file is `uppsala\\korp_263.tgt`, there are 3 passing sentences (were 5 by regex)\n",
      "file is `uppsala\\korp_264.tgt`, there are 7 passing sentences (were 12 by regex)\n",
      "file is `uppsala\\korp_265.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `uppsala\\korp_266.tgt`, there are 8 passing sentences (were 13 by regex)\n",
      "file is `uppsala\\korp_409.tgt`, there are 5 passing sentences (were 9 by regex)\n",
      "file is `uppsala\\korp_601.tgt`, there are 5 passing sentences (were 14 by regex)\n",
      "file is `uppsala\\korp_602.tgt`, there are 6 passing sentences (were 14 by regex)\n",
      "file is `uppsala\\korp_603.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `uppsala\\korp_604.tgt`, there are 2 passing sentences (were 9 by regex)\n",
      "file is `uppsala\\korp_605.tgt`, there are 2 passing sentences (were 2 by regex)\n",
      "file is `uppsala\\korp_606.tgt`, there are 11 passing sentences (were 13 by regex)\n",
      "file is `uppsala\\korp_607.tgt`, there are 5 passing sentences (were 8 by regex)\n",
      "file is `uppsala\\korp_608.tgt`, there are 2 passing sentences (were 3 by regex)\n",
      "file is `uppsala\\korp_609.tgt`, there are 8 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_610.tgt`, there are 5 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_611.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_612.tgt`, there are 4 passing sentences (were 11 by regex)\n",
      "file is `uppsala\\korp_613.tgt`, there are 5 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_614.tgt`, there are 4 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_615.tgt`, there are 8 passing sentences (were 11 by regex)\n",
      "file is `uppsala\\korp_616.tgt`, there are 11 passing sentences (were 14 by regex)\n",
      "file is `uppsala\\korp_617.tgt`, there are 2 passing sentences (were 8 by regex)\n",
      "file is `uppsala\\korp_618.tgt`, there are 10 passing sentences (were 21 by regex)\n",
      "file is `uppsala\\korp_619.tgt`, there are 18 passing sentences (were 25 by regex)\n",
      "file is `uppsala\\korp_620.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_621.tgt`, there are 20 passing sentences (were 28 by regex)\n",
      "file is `uppsala\\korp_622.tgt`, there are 13 passing sentences (were 18 by regex)\n",
      "file is `uppsala\\korp_623.tgt`, there are 5 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_624.tgt`, there are 1 passing sentences (were 4 by regex)\n",
      "file is `uppsala\\korp_625.tgt`, there are 2 passing sentences (were 3 by regex)\n",
      "file is `uppsala\\korp_626.tgt`, there are 0 passing sentences (were 3 by regex)\n",
      "file is `uppsala\\korp_627.tgt`, there are 14 passing sentences (were 18 by regex)\n",
      "file is `uppsala\\korp_628.tgt`, there are 1 passing sentences (were 3 by regex)\n",
      "file is `uppsala\\korp_701.tgt`, there are 1 passing sentences (were 4 by regex)\n",
      "file is `uppsala\\korp_702.tgt`, there are 3 passing sentences (were 8 by regex)\n",
      "file is `uppsala\\korp_703.tgt`, there are 3 passing sentences (were 6 by regex)\n",
      "file is `uppsala\\korp_704.tgt`, there are 3 passing sentences (were 9 by regex)\n",
      "file is `uppsala\\korp_705.tgt`, there are 2 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_706.tgt`, there are 3 passing sentences (were 8 by regex)\n",
      "file is `uppsala\\korp_707.tgt`, there are 1 passing sentences (were 4 by regex)\n",
      "file is `uppsala\\korp_708.tgt`, there are 9 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_709.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_710.tgt`, there are 4 passing sentences (were 8 by regex)\n",
      "file is `uppsala\\korp_711.tgt`, there are 9 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_712.tgt`, there are 6 passing sentences (were 12 by regex)\n",
      "file is `uppsala\\korp_713.tgt`, there are 10 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_714.tgt`, there are 5 passing sentences (were 10 by regex)\n",
      "file is `uppsala\\korp_716.tgt`, there are 5 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_717.tgt`, there are 8 passing sentences (were 9 by regex)\n",
      "file is `uppsala\\korp_718.tgt`, there are 5 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_719.tgt`, there are 2 passing sentences (were 6 by regex)\n",
      "file is `uppsala\\korp_720.tgt`, there are 4 passing sentences (were 6 by regex)\n",
      "file is `uppsala\\korp_721.tgt`, there are 6 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_722.tgt`, there are 3 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_723.tgt`, there are 4 passing sentences (were 7 by regex)\n",
      "file is `uppsala\\korp_724.tgt`, there are 7 passing sentences (were 9 by regex)\n",
      "file is `uppsala\\korp_725.tgt`, there are 8 passing sentences (were 9 by regex)\n",
      "file is `uppsala\\korp_726.tgt`, there are 15 passing sentences (were 19 by regex)\n",
      "file is `uppsala\\nagibin_1.tgt`, there are 36 passing sentences (were 69 by regex)\n",
      "file is `uppsala\\nagibin_2.tgt`, there are 36 passing sentences (were 49 by regex)\n",
      "file is `uppsala\\nagibin_3.tgt`, there are 12 passing sentences (were 15 by regex)\n",
      "file is `uppsala\\nagibin_4.tgt`, there are 14 passing sentences (were 19 by regex)\n",
      "\n",
      "passing 9612 among 15697 found by regex\n",
      "Wall time: 2min 45s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "corpus_base, total_passing, total_regex = do_search(files)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "massive-highway",
   "metadata": {},
   "outputs": [],
   "source": [
    "def cast_bool_cols_to_int(df: pd.DataFrame) -> pd.DataFrame:\n",
    "    subword_bool_cols_to_int = ['has', 'is']\n",
    "    pat_bool_cols_to_int = [rf'(\\b|_){sub}(\\b|_)' for sub in subword_bool_cols_to_int]\n",
    "\n",
    "    for colname in df.columns:\n",
    "        if any(re.search(pat, colname) for pat in pat_bool_cols_to_int):\n",
    "            df.loc[:, colname] = df[colname].astype(int)\n",
    "             "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "narrative-arrow",
   "metadata": {},
   "outputs": [],
   "source": [
    "cast_bool_cols_to_int(corpus_base)\n",
    "\n",
    "corpus = corpus_base.rename_axis('file_idx').reset_index()\n",
    "corpus.loc[:, 'full_analysis'] = corpus['full_analysis'].str.replace('\\n', ' | ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "suspended-ghost",
   "metadata": {},
   "outputs": [],
   "source": [
    "path = Path('./syntagrus/result/syntagrus_6_1+context.csv')\n",
    "path.parent.mkdir(parents=True, exist_ok=True)\n",
    "\n",
    "#ID!\n",
    "corpus.to_csv(path, decimal=',')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "assured-museum",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ### ТОЛЬКО НОВЫЕ КОЛОНКИ\n",
    "\n",
    "# upd_cols = ['file_idx', 'source'] + NEW_COLS\n",
    "# corpus_upd = corpus[upd_cols]\n",
    "# corpus_upd.loc[:, 'file_source_idx'] = corpus_upd.apply(lambda row: f'{row.file_idx}_{row.source}', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "stone-consideration",
   "metadata": {},
   "outputs": [],
   "source": [
    "# path = Path('./syntagrus/result/syntagrus_4_newonly.csv')\n",
    "# path.parent.mkdir(parents=True, exist_ok=True)\n",
    "# corpus_upd.to_csv(path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "daily-contribution",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>file_idx</th>\n",
       "      <th>source</th>\n",
       "      <th>sent_num_file</th>\n",
       "      <th>left_context</th>\n",
       "      <th>text</th>\n",
       "      <th>right_context</th>\n",
       "      <th>full_analysis</th>\n",
       "      <th>sent_pos_chain</th>\n",
       "      <th>has_many_dos</th>\n",
       "      <th>has_many_ios</th>\n",
       "      <th>...</th>\n",
       "      <th>DO_number</th>\n",
       "      <th>IO</th>\n",
       "      <th>IO_pos</th>\n",
       "      <th>IO_head</th>\n",
       "      <th>IO_head_pos</th>\n",
       "      <th>IO_lemma</th>\n",
       "      <th>IO_animacy</th>\n",
       "      <th>IO_case</th>\n",
       "      <th>IO_number</th>\n",
       "      <th>order</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0</td>\n",
       "      <td>2003/artist_mimansa.tgt</td>\n",
       "      <td>4</td>\n",
       "      <td>Это случилось вечером, в третьем акте. Илью Ил...</td>\n",
       "      <td>Будучи одет в малиновую ливрею с галунами, бел...</td>\n",
       "      <td>Он мирно стоял с подносом в кулисе № 2, ожидая...</td>\n",
       "      <td>&lt;S ID=\"4\"&gt; | &lt;W DOM=\"17\" FEAT=\"V НЕСОВ ДЕЕПР Н...</td>\n",
       "      <td>V V PR A S PR S A S CONJ S PR S S S V V V PR S...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>мн</td>\n",
       "      <td>балеринам</td>\n",
       "      <td>S</td>\n",
       "      <td>раздать</td>\n",
       "      <td>V</td>\n",
       "      <td>балерина</td>\n",
       "      <td>од</td>\n",
       "      <td>дат</td>\n",
       "      <td>мн</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>20</td>\n",
       "      <td>2003/artist_mimansa.tgt</td>\n",
       "      <td>237</td>\n",
       "      <td>От такого кошмара у Ильи Ильича выступила испа...</td>\n",
       "      <td>Он пересиливал себя, вставал и ходил по комнат...</td>\n",
       "      <td>И все это было так ужасно, просто конец света....</td>\n",
       "      <td>&lt;S ID=\"237\"&gt; | &lt;W DOM=\"2\" FEAT=\"S ЕД МУЖ ИМ ОД...</td>\n",
       "      <td>PRON V PRON V CONJ V PR S V PR S ADV V PRON S ...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>ед</td>\n",
       "      <td>одеяло</td>\n",
       "      <td>S</td>\n",
       "      <td>под</td>\n",
       "      <td>PR</td>\n",
       "      <td>одеяло</td>\n",
       "      <td>неод</td>\n",
       "      <td>вин</td>\n",
       "      <td>ед</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>33</td>\n",
       "      <td>2003/artist_mimansa.tgt</td>\n",
       "      <td>368</td>\n",
       "      <td>Еще двое вышли, но один вошел… Один вышел, ост...</td>\n",
       "      <td>Уж совсем было освободился директор - и тут на...</td>\n",
       "      <td>Гости ушли через час, когда приемная уже ломил...</td>\n",
       "      <td>&lt;S ID=\"368\"&gt; | &lt;W DOM=\"4\" FEAT=\"PART\" ID=\"1\" K...</td>\n",
       "      <td>PART ADV PRON V S CONJ ADV V S A PR S PR S PR ...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>ед</td>\n",
       "      <td>кабинет</td>\n",
       "      <td>S</td>\n",
       "      <td>в</td>\n",
       "      <td>PR</td>\n",
       "      <td>кабинет</td>\n",
       "      <td>неод</td>\n",
       "      <td>вин</td>\n",
       "      <td>ед</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64</th>\n",
       "      <td>2</td>\n",
       "      <td>2003/atlanty_i_atlantologi.tgt</td>\n",
       "      <td>33</td>\n",
       "      <td>В мае 2001 года в заливе Гуанаакабибес, на зап...</td>\n",
       "      <td>При горизонтальном сканировании глубоководные ...</td>\n",
       "      <td>Руководитель экспедиции Полина Зелицкая заявил...</td>\n",
       "      <td>&lt;S ID=\"33\"&gt; | &lt;W DOM=\"6\" FEAT=\"PR\" ID=\"1\" KSNA...</td>\n",
       "      <td>PR A S A S V PR S S A A S A PR A S PR ADV A CO...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>ед</td>\n",
       "      <td>дне</td>\n",
       "      <td>S</td>\n",
       "      <td>на</td>\n",
       "      <td>PR</td>\n",
       "      <td>дно</td>\n",
       "      <td>неод</td>\n",
       "      <td>пр</td>\n",
       "      <td>ед</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>68</th>\n",
       "      <td>6</td>\n",
       "      <td>2003/atlanty_i_atlantologi.tgt</td>\n",
       "      <td>54</td>\n",
       "      <td>Ищут следы Атлантиды не только в Западном полу...</td>\n",
       "      <td>Так, в 1973 году американская подводная лодка ...</td>\n",
       "      <td>Американскому правительству было тотчас доложе...</td>\n",
       "      <td>&lt;S ID=\"54\"&gt; | &lt;W DOM=\"9\" FEAT=\"ADV\" ID=\"1\" LEM...</td>\n",
       "      <td>ADV PR NUM S A A S ADV V PR S S S V S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>мн</td>\n",
       "      <td>берегов</td>\n",
       "      <td>S</td>\n",
       "      <td>у</td>\n",
       "      <td>PR</td>\n",
       "      <td>берег</td>\n",
       "      <td>неод</td>\n",
       "      <td>род</td>\n",
       "      <td>мн</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9508</th>\n",
       "      <td>9</td>\n",
       "      <td>uppsala/korp_726.tgt</td>\n",
       "      <td>35</td>\n",
       "      <td>Следовательно, наши предки на Волыни и в Галич...</td>\n",
       "      <td>Святой равноапостольный князь Владимир Великий...</td>\n",
       "      <td>Следовательно, и духовная власть упомянутого в...</td>\n",
       "      <td>&lt;S ID=\"35\"&gt; | &lt;W DOM=\"3\" FEAT=\"A ЕД МУЖ ИМ\" ID...</td>\n",
       "      <td>A A S S A V S PR S PRON S S PRON V PART PART S...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>ед</td>\n",
       "      <td>князем</td>\n",
       "      <td>S</td>\n",
       "      <td>назначил</td>\n",
       "      <td>V</td>\n",
       "      <td>князь</td>\n",
       "      <td>од</td>\n",
       "      <td>твор</td>\n",
       "      <td>ед</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9519</th>\n",
       "      <td>5</td>\n",
       "      <td>uppsala/nagibin_1.tgt</td>\n",
       "      <td>46</td>\n",
       "      <td>Он вернулся, вернулся сам, не дождавшись меня,...</td>\n",
       "      <td>- Обедать!.. - послышался голос мамы, и, груст...</td>\n",
       "      <td>Путешествие так и не состоялось. Но зато сколь...</td>\n",
       "      <td>&lt;S ID=\"46\"&gt; | - &lt;W DOM=\"2\" FEAT=\"V НЕСОВ ИНФ\" ...</td>\n",
       "      <td>V V S S CONJ ADV V S PRON ADV V PR V S PR A S ...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>ед</td>\n",
       "      <td>взглядом</td>\n",
       "      <td>S</td>\n",
       "      <td>окинув</td>\n",
       "      <td>V</td>\n",
       "      <td>взгляд</td>\n",
       "      <td>неод</td>\n",
       "      <td>твор</td>\n",
       "      <td>ед</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9549</th>\n",
       "      <td>35</td>\n",
       "      <td>uppsala/nagibin_1.tgt</td>\n",
       "      <td>270</td>\n",
       "      <td>Но почему же началось это с Тенненбаума, котор...</td>\n",
       "      <td>Очень сильная, очень настоящая любовь делает п...</td>\n",
       "      <td>Вопреки очевидности, я где-то на дне души не в...</td>\n",
       "      <td>&lt;S ID=\"270\"&gt; | &lt;W DOM=\"2\" FEAT=\"ADV\" ID=\"1\" LE...</td>\n",
       "      <td>ADV A ADV A S V S PRON S PART PRON A PRON V PR...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>ед</td>\n",
       "      <td>провидцем</td>\n",
       "      <td>S</td>\n",
       "      <td>делает</td>\n",
       "      <td>V</td>\n",
       "      <td>провидец</td>\n",
       "      <td>од</td>\n",
       "      <td>твор</td>\n",
       "      <td>ед</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9552</th>\n",
       "      <td>2</td>\n",
       "      <td>uppsala/nagibin_2.tgt</td>\n",
       "      <td>15</td>\n",
       "      <td>Все знали, что оно будет, это наступление, и ч...</td>\n",
       "      <td>Ради этого младший лейтенант Павлов скрывал от...</td>\n",
       "      <td>Он был преступно недальновиден: думал, что рас...</td>\n",
       "      <td>&lt;S ID=\"15\"&gt; | &lt;W DOM=\"6\" FEAT=\"PR\" ID=\"1\" LEMM...</td>\n",
       "      <td>PR PRON A S S V PR PRON CONJ PR PRON S V PRON S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>ед</td>\n",
       "      <td>себя</td>\n",
       "      <td>PRON</td>\n",
       "      <td>от</td>\n",
       "      <td>PR</td>\n",
       "      <td>себя</td>\n",
       "      <td>од</td>\n",
       "      <td>род</td>\n",
       "      <td>ед</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9555</th>\n",
       "      <td>5</td>\n",
       "      <td>uppsala/nagibin_2.tgt</td>\n",
       "      <td>31</td>\n",
       "      <td>Жена все приняла на себя. В трудную минуту жиз...</td>\n",
       "      <td>С непостижимой быстротой и неприметной затрато...</td>\n",
       "      <td>Возле поселка находился санаторий, там шло стр...</td>\n",
       "      <td>&lt;S ID=\"31\"&gt; | &lt;W DOM=\"8\" FEAT=\"PR\" ID=\"1\" KSNA...</td>\n",
       "      <td>PR A S CONJ A S S V PRON PR NUM S PR S S PR S ...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>ед</td>\n",
       "      <td>километрах</td>\n",
       "      <td>S</td>\n",
       "      <td>в</td>\n",
       "      <td>PR</td>\n",
       "      <td>километр</td>\n",
       "      <td>неод</td>\n",
       "      <td>пр</td>\n",
       "      <td>мн</td>\n",
       "      <td>V IO DO</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>732 rows × 44 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      file_idx                          source sent_num_file  \\\n",
       "22           0         2003/artist_mimansa.tgt             4   \n",
       "42          20         2003/artist_mimansa.tgt           237   \n",
       "55          33         2003/artist_mimansa.tgt           368   \n",
       "64           2  2003/atlanty_i_atlantologi.tgt            33   \n",
       "68           6  2003/atlanty_i_atlantologi.tgt            54   \n",
       "...        ...                             ...           ...   \n",
       "9508         9            uppsala/korp_726.tgt            35   \n",
       "9519         5           uppsala/nagibin_1.tgt            46   \n",
       "9549        35           uppsala/nagibin_1.tgt           270   \n",
       "9552         2           uppsala/nagibin_2.tgt            15   \n",
       "9555         5           uppsala/nagibin_2.tgt            31   \n",
       "\n",
       "                                           left_context  \\\n",
       "22    Это случилось вечером, в третьем акте. Илью Ил...   \n",
       "42    От такого кошмара у Ильи Ильича выступила испа...   \n",
       "55    Еще двое вышли, но один вошел… Один вышел, ост...   \n",
       "64    В мае 2001 года в заливе Гуанаакабибес, на зап...   \n",
       "68    Ищут следы Атлантиды не только в Западном полу...   \n",
       "...                                                 ...   \n",
       "9508  Следовательно, наши предки на Волыни и в Галич...   \n",
       "9519  Он вернулся, вернулся сам, не дождавшись меня,...   \n",
       "9549  Но почему же началось это с Тенненбаума, котор...   \n",
       "9552  Все знали, что оно будет, это наступление, и ч...   \n",
       "9555  Жена все приняла на себя. В трудную минуту жиз...   \n",
       "\n",
       "                                                   text  \\\n",
       "22    Будучи одет в малиновую ливрею с галунами, бел...   \n",
       "42    Он пересиливал себя, вставал и ходил по комнат...   \n",
       "55    Уж совсем было освободился директор - и тут на...   \n",
       "64    При горизонтальном сканировании глубоководные ...   \n",
       "68    Так, в 1973 году американская подводная лодка ...   \n",
       "...                                                 ...   \n",
       "9508  Святой равноапостольный князь Владимир Великий...   \n",
       "9519  - Обедать!.. - послышался голос мамы, и, груст...   \n",
       "9549  Очень сильная, очень настоящая любовь делает п...   \n",
       "9552  Ради этого младший лейтенант Павлов скрывал от...   \n",
       "9555  С непостижимой быстротой и неприметной затрато...   \n",
       "\n",
       "                                          right_context  \\\n",
       "22    Он мирно стоял с подносом в кулисе № 2, ожидая...   \n",
       "42    И все это было так ужасно, просто конец света....   \n",
       "55    Гости ушли через час, когда приемная уже ломил...   \n",
       "64    Руководитель экспедиции Полина Зелицкая заявил...   \n",
       "68    Американскому правительству было тотчас доложе...   \n",
       "...                                                 ...   \n",
       "9508  Следовательно, и духовная власть упомянутого в...   \n",
       "9519  Путешествие так и не состоялось. Но зато сколь...   \n",
       "9549  Вопреки очевидности, я где-то на дне души не в...   \n",
       "9552  Он был преступно недальновиден: думал, что рас...   \n",
       "9555  Возле поселка находился санаторий, там шло стр...   \n",
       "\n",
       "                                          full_analysis  \\\n",
       "22    <S ID=\"4\"> | <W DOM=\"17\" FEAT=\"V НЕСОВ ДЕЕПР Н...   \n",
       "42    <S ID=\"237\"> | <W DOM=\"2\" FEAT=\"S ЕД МУЖ ИМ ОД...   \n",
       "55    <S ID=\"368\"> | <W DOM=\"4\" FEAT=\"PART\" ID=\"1\" K...   \n",
       "64    <S ID=\"33\"> | <W DOM=\"6\" FEAT=\"PR\" ID=\"1\" KSNA...   \n",
       "68    <S ID=\"54\"> | <W DOM=\"9\" FEAT=\"ADV\" ID=\"1\" LEM...   \n",
       "...                                                 ...   \n",
       "9508  <S ID=\"35\"> | <W DOM=\"3\" FEAT=\"A ЕД МУЖ ИМ\" ID...   \n",
       "9519  <S ID=\"46\"> | - <W DOM=\"2\" FEAT=\"V НЕСОВ ИНФ\" ...   \n",
       "9549  <S ID=\"270\"> | <W DOM=\"2\" FEAT=\"ADV\" ID=\"1\" LE...   \n",
       "9552  <S ID=\"15\"> | <W DOM=\"6\" FEAT=\"PR\" ID=\"1\" LEMM...   \n",
       "9555  <S ID=\"31\"> | <W DOM=\"8\" FEAT=\"PR\" ID=\"1\" KSNA...   \n",
       "\n",
       "                                         sent_pos_chain  has_many_dos  \\\n",
       "22    V V PR A S PR S A S CONJ S PR S S S V V V PR S...             0   \n",
       "42    PRON V PRON V CONJ V PR S V PR S ADV V PRON S ...             0   \n",
       "55    PART ADV PRON V S CONJ ADV V S A PR S PR S PR ...             0   \n",
       "64    PR A S A S V PR S S A A S A PR A S PR ADV A CO...             0   \n",
       "68                ADV PR NUM S A A S ADV V PR S S S V S             0   \n",
       "...                                                 ...           ...   \n",
       "9508  A A S S A V S PR S PRON S S PRON V PART PART S...             0   \n",
       "9519  V V S S CONJ ADV V S PRON ADV V PR V S PR A S ...             0   \n",
       "9549  ADV A ADV A S V S PRON S PART PRON A PRON V PR...             0   \n",
       "9552    PR PRON A S S V PR PRON CONJ PR PRON S V PRON S             0   \n",
       "9555  PR A S CONJ A S S V PRON PR NUM S PR S S PR S ...             0   \n",
       "\n",
       "      has_many_ios  ...  DO_number          IO IO_pos   IO_head  IO_head_pos  \\\n",
       "22               0  ...         мн   балеринам      S   раздать            V   \n",
       "42               0  ...         ед      одеяло      S       под           PR   \n",
       "55               0  ...         ед     кабинет      S         в           PR   \n",
       "64               0  ...         ед         дне      S        на           PR   \n",
       "68               0  ...         мн     берегов      S         у           PR   \n",
       "...            ...  ...        ...         ...    ...       ...          ...   \n",
       "9508             0  ...         ед      князем      S  назначил            V   \n",
       "9519             0  ...         ед    взглядом      S    окинув            V   \n",
       "9549             0  ...         ед   провидцем      S    делает            V   \n",
       "9552             0  ...         ед        себя   PRON        от           PR   \n",
       "9555             0  ...         ед  километрах      S         в           PR   \n",
       "\n",
       "      IO_lemma IO_animacy  IO_case  IO_number    order  \n",
       "22    балерина         од      дат         мн  V IO DO  \n",
       "42      одеяло       неод      вин         ед  V IO DO  \n",
       "55     кабинет       неод      вин         ед  V IO DO  \n",
       "64         дно       неод       пр         ед  V IO DO  \n",
       "68       берег       неод      род         мн  V IO DO  \n",
       "...        ...        ...      ...        ...      ...  \n",
       "9508     князь         од     твор         ед  V IO DO  \n",
       "9519    взгляд       неод     твор         ед  V IO DO  \n",
       "9549  провидец         од     твор         ед  V IO DO  \n",
       "9552      себя         од      род         ед  V IO DO  \n",
       "9555  километр       неод       пр         мн  V IO DO  \n",
       "\n",
       "[732 rows x 44 columns]"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corpus[corpus['DO_has_relcl']==1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "id": "needed-directive",
   "metadata": {},
   "outputs": [],
   "source": [
    "corp = pd.read_csv('./syntagrus/result/syntagrus.csv')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pandas",
   "language": "python",
   "name": "pandas"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}