#http://pynote.hatenablog.com/entry/python-wordcloud
#!/usr/bin/env python
# -*- coding: utf-8 -*-

import MeCab
import sys
from sys import argv

import wordcloud as wc 
import codecs
from wordcloud import STOPWORDS, WordCloud
 
import numpy as np
import matplotlib as mpl
import matplotlib.pyplot as plt
FONT="ipaexg.ttf"
'''
#Mecabしない場合
f1 =  open('./wakachidata/output_after_mecab2.txt','r')
#f1 =  open('/home/hashimori/zsugisaki/190730wordcloud/nagaoka/touhoku.txt','r')

# ファイルを読み込む
data1 = f1.read()

f2 =  open('./wakachidata/output_after_mecab2.txt','r')

# ファイルを読み込む
data2 = f2.read()

data1 = data.replace('\n','')
data1 = data.replace('\r','')
'''
#Mecabする場合
with codecs.open(u'./textdata/touhoku_r.txt', "r", "utf-8") as f1:
    data01 = f1.read()
with codecs.open(u'./textdata/yobo3_1_6_r.txt', "r", "utf-8") as f2:
    data02 = f2.read()
with codecs.open(u'./textdata/ds_dourokyousonsyou_r.txt', "r", "utf-8") as f3:
    data03 = f3.read()
with codecs.open(u'./textdata/ks0829_r.txt', "r", "utf-8") as f4:
    data04 = f4.read()
with codecs.open(u'./textdata/ks0381_r.txt', "r", "utf-8") as f5:
    data05 = f5.read()
with codecs.open(u'./textdata/output_america.txt', "r", "utf-8") as f6:
    data06 = f6.read()
    
sys.getsizeof(data01)

data1=data01+data02+data03+data04
data2=data06

        
# 分かち書きのみ出力する設定にする
mecab = MeCab.Tagger("-Owakati")
text1 = mecab.parse(data1)
text1=text1.encode("Shift_jis",'ignore')
text1=text1.decode("Shift_jis")
file_name1 = "./mecab_out/output_after_mecab.txt"
with codecs.open(file_name1,"w",'utf-8') as f1:
    f1.write(text1)

text2 = mecab.parse(data2)
text2=text2.encode("Shift_jis",'ignore')
text2=text2.decode("Shift_jis")

file_name2 = "./mecab_out/output_after_mecab2.txt"
with codecs.open(file_name2,"w",'utf-8') as f2:
    f2.write(text2)

def plot_wordcloud(wordcloud, title="WordCloud", pngfile=None):
    """
    WordCloud描画
    
    parameters
    -------------
    wordcloud: wordcloud object
        描画対象
    tilte: str
        タイトル
    pngfile: str (or None)
        ファイル名
        None出なければここで指定した名のpngファイルに出力
    """
    # 描画できるような形に変換
    awc = wordcloud.to_array()
 
    #描画 （画像を作る）
    fig=plt.figure()   
    left, width = 0.01, 0.98
    bottom, height = 0.01, 0.90
    ax=  plt.axes([left, bottom, width, height])
    
    #plot wordclouds
    ax.imshow(awc, interpolation="bilinear")
    ax.axis("off")
    ax.set_title(title,fontsize=18)
 
 
    # 表示
    plt.show()
    # png
    if pngfile:
        fig.savefig(pngfile)

#fpath = "/System/Library/Fonts/HelveticaNeue-UltraLight.otf"

wc1=wc.WordCloud(
    font_path=FONT,
    width=800,height=600,
    background_color="white",
    max_words=50,
    min_font_size=10,
    max_font_size=96,
    collocations=False,
    stopwords={u'について',u'ある',u'もらい',u'さん',u'たい',u'なっ',u'いる',u'から',u'ます',
               u'その',u'する',u'あり',u'ので',u'まで',u'こと',u'にて',u'でき',u'なかっ',u'ため'}
    
    )

wc1.generate(text1)
wc1.to_file('wordcloud_japan.png')
plot_wordcloud(wc1, title="Japan")

wc2=wc.WordCloud(
    font_path=FONT,
    width=800,height=600,
    background_color="white",
    max_words=50,
    min_font_size=10,
    max_font_size=96,
    collocations=False,
    stopwords={u'and',u'the',u'Figure',u'for',u'of',u'to',u'in',u'that',u'or',
               u'are',u'with',u'a',u'an',u'as',u'may',u'can',u'This',u'also',u'at',
               u'is',u'by',u'be',u'not',u'if',u'it',u'other',u'these',u'when',u'have'}
    
    )

wc2.generate(text2)
wc2.to_file('wordcloud_usa.png')
plot_wordcloud(wc2, title="USA")

#wc2019.generate_from_frequencies(wdict)