<?php
$url = 'https://gakufu.gakki.me/search/?mode=list&word='.$_REQUEST['word'];
$dom = new DOMDocument;
@$dom->loadHTMLFile($url);
$xpath = new DOMXPath($dom);
$entries = [];
foreach ($xpath->query('//div[@class="box"]') as $node) {
	$entries[] = [
		'url'     => $xpath->evaluate('string(.//p[@class="mname"]/a/@href)', $node),
		'mname'   => $xpath->evaluate('string(.//p[@class="mname"]/a)', $node),
	];
}

foreach($entries as $entry) {
	echo $entry['url']."<br />\n";
	echo $entry['mname']."<br />\n";
}
?>

<?php
function getLyrics($url) {
	$dom = new DOMDocument;
	@$dom->loadHTMLFile($url);
	$xpath = new DOMXPath($dom);
	$codes =null;
	$lyrics =null;
	foreach ($xpath->query('//div[@class="cd_1line"]') as $node) {

		foreach ($xpath->query('.//div[@class="cd_pic cd_font"]/div/span',$node) as $subNode) {
			if ($codes) $codes .='>';
			$codes .= $subNode->nodeValue;
		}

		foreach ($xpath->query('.//div[@class="cd_pic blue"]/div',$node) as $subNode) {
			$lyrics .= $subNode->nodeValue;
		}
	}
	// 半角、全角空白どちらも削除
	$lyrics = preg_replace("/( |　)/", "", $lyrics );
	// NO-BREAK SPACE (C2A0)削除
	$lyrics = str_replace( "\xc2\xa0", "", $lyrics );

	return array($codes,$lyrics);
}

$url = $_REQUEST['url'];

list($codes,$lyrics) = getLyrics($url);

echo $codes."<br />\n";
echo $lyrics."<br />\n";
?>

<?php
function getLyrics($url) {
	$dom = new DOMDocument;
	@$dom->loadHTMLFile($url);
	$xpath = new DOMXPath($dom);
	$codes =null;
	$lyrics =null;
	foreach ($xpath->query('//div[@class="cd_1line"]') as $node) {

		foreach ($xpath->query('.//div[@class="cd_pic cd_font"]/div/span',$node) as $subNode) {
			if ($codes) $codes .='>';
			$codes .= $subNode->nodeValue;
		}

		foreach ($xpath->query('.//div[@class="cd_pic blue"]/div',$node) as $subNode) {
			$lyrics .= $subNode->nodeValue;
		}
	}
	// 半角、全角空白どちらも削除
	$lyrics = preg_replace("/( |　)/", "", $lyrics );
	// NO-BREAK SPACE (C2A0)削除
	$lyrics = str_replace( "\xc2\xa0", "", $lyrics );

	return array($codes,$lyrics);
}

$url = 'https://gakufu.gakki.me/search/?mode=list&word='.$_REQUEST['word'];
$dom = new DOMDocument;
@$dom->loadHTMLFile($url);
$xpath = new DOMXPath($dom);
$entries = [];
foreach ($xpath->query('//div[@class="box"]') as $node) {
	$entries[] = [
		'url'     => $xpath->evaluate('string(.//p[@class="mname"]/a/@href)', $node),
		'mname'   => $xpath->evaluate('string(.//p[@class="mname"]/a)', $node),
	];
}

$notes = array();

foreach($entries as $entry) {
	$link  = 'https://gakufu.gakki.me'.$entry['url'];
	$mname = $entry['mname'];
	list($codes,$lyrics) = getLyrics($link);
	$note = array('title'=>$mname,'codes'=>$codes,'lyrics'=>$lyrics);
	$notes[] = $note;
}

$json = json_encode($notes);

$fp = fopen('lyrics.json', 'w');
fwrite($fp,$json);
fclose($fp);
?>

<?php
$filename = 'lyrics.json';
$fp = fopen($filename, 'r');
$json = fread($fp, filesize($filename));
fclose($fp);

$obj = json_decode($json,true);
foreach ($obj as $notes) {
	foreach ($notes as $key=>$value) {
		echo $key."=>".$value."<br>\n";
	}
}
?>

import json
import MeCab

tagger = MeCab.Tagger()

# 文章を単語に分割してリスト出力する
def tokenize(text):
	node = tagger.parseToNode(text)
	tokens = []
	while node:
		if node.surface != '':
			tokens.append(node.surface)
		node = node.next
	return tokens

# JSON形式のテキストファイルを読み込む
path = './lyrics.json'
jsonText = open(path, 'r').read().encode('utf-8')
notes = json.loads(jsonText)

lyrics = notes[0]['lyrics']
words = tokenize(lyrics)
for word in words:
  print(word)

import json
import MeCab
import os
import re

tagger = MeCab.Tagger()

# 文章を単語に分割してリスト出力する
def tokenize(text):
	node = tagger.parseToNode(text)
	tokens = []
	while node:
		if node.surface != '':
			tokens.append(node.surface)
		node = node.next
	return tokens

# 辞書を更新
def updateDic(words):
	tmp = []
	for word in words:
		if word == "" or word == "\r\n" or word == "\n": continue
		tmp.append(word)
		if len(tmp) < 3: continue
		if len(tmp) > 3: tmp = tmp[1:]
		setWord3(tmp)

# 三要素のリストを辞書として登録
def setWord3(s3):
	global dic
	w1, w2, w3 = s3
	if not w1 in dic: dic[w1] = {}
	if not w2 in dic[w1]: dic[w1][w2] = {}
	if not w3 in dic[w1][w2]: dic[w1][w2][w3] = 0
	dic[w1][w2][w3] += 1

def dumpDic(dic):
	for key_1st, value_1st in dic.items():
		for key_2nd, value_2nd in value_1st.items():
			for key_3rd, value_3rd in value_2nd.items():
				print(key_1st,"→",key_2nd,"→",key_3rd,"＝",value_3rd)
	return

# JSON形式の楽曲ファイルを読み込む
path = './lyrics.json'
jsonText = open(path, 'r').read().encode('utf-8')
notes = json.loads(jsonText)

# JSON形式の辞書ファイルを読み込む
path = './markov-lyrics.json'
if(os.path.exists(path)):
	jsonText = open(path, 'r').read().encode('utf-8')
	dic = json.loads(jsonText)
else:
	dic = {}

for note in notes:
	lyrics = note['lyrics']
	lyrics = re.sub(r'[A-z]+', "", lyrics) #アルファベットの削除
	lyrics = re.sub(r'[\r\n]', "", lyrics) #改行などの削除

	words = tokenize(lyrics)
	updateDic(words)

dumpDic(dic)

json.dump(dic, open("markov-lyrics.json", "w", encoding="utf-8"))

import json
import random

def word_choice(sel):
	keys = sel.keys()
	ran = random.choice(list(keys))
	return ran

def make_sentence(dic,keyword):
	ret = []
	if not keyword in dic: return "no dic"
	w1 = keyword
	w2 = word_choice(dic[keyword])
	ret.append(w1)
	ret.append(w2)
	while True:
		w3 = word_choice(dic[w1][w2])
		ret.append(w3)
		if len(ret)>100 : break
		w1, w2 = w2, w3
	return "".join(ret)

dic = open("markov-lyrics.json" , "r")
dic = json.load(dic)

random.seed()
keyword = '見上げ'
s = make_sentence(dic,keyword)
print(s)