Visualising line messages

line_visualizer.py

#coding: utf-8
import csv
import MeCab
import matplotlib.pyplot as plt
from wordcloud import WordCloud
import re

STOP_WORDS = "の もの こと よう これ 一 http:// の 笑 ω 物 とき ら 画像 Sticker Photo".split()
STOP_NOUN_DETAILS = "数 接続助詞 接尾 代名詞 非自立 副詞可能".split()

def get_nouns(string):
    tagger = MeCab.Tagger()
    text = str(string)
    node = tagger.parseToNode(text)

    nouns = []
    while node:
        word_detail = node.feature.split(",")
        pos = node.feature.split(",")[0]
        word = node.surface
        noun_detail = word_detail[1]
        if pos == "名詞":
            if noun_detail not in STOP_NOUN_DETAILS:
                if word not in STOP_WORDS:
                    nouns.append(word)
        node = node.next
    return nouns

def main():
    fpath = "/Library/Fonts/Yu Gothic Medium.otf"
    lines = open('talk.txt','r').readlines()

    posts = []
    for line in lines:
        usual = re.match("\d\d:\d\d\t.*\t", line)
        if usual:
            posts.append(line[usual.end():].rstrip())
        elif re.match('.*"', line):
            posts.append(line.rstrip())
        else:
            continue

    nouns_all = []

    for post in posts:
        nouns_in_row = get_nouns(post)
        nouns_all.extend(nouns_in_row)

    lst_wordcloud = " ".join(nouns_all).decode('utf-8')

    wordcloud = WordCloud(
        background_color = "white",
        font_path = fpath,
        width = 1280,
        height = 720).generate(lst_wordcloud)

    plt.figure(figsize=(16,9))
    plt.imshow(wordcloud)
    plt.axis("off")
    plt.show()


## Execute
if __name__ == "__main__":
    main()