Introduction

I wrote something similar before, but again.

Preparing the pyenv environment

Install pyenv and pyenv-virtualenv with homebrew.

$ mkdir qiita-sin
$ cd qiita-sin
$ pyenv virtualenv 2.7.8 qiita-sin
$ pyenv local qiita-sin

Install the required libraries and get ready to run on Heroku

$ pip install flask
$ pip install gunicorn
$ pip install matplotlib
$ echo python-2.7.8 > runtime.txt
$ pip freeze > requirements.txt
$ echo web: gunicorn app:app --log-file=- > Procfile
$ mkdir templates

Template, app preparation

template

`sin.html`


<!DOCTYPE html>
<html lang="ja">
    <head>
        <title>sin curve test</title>
    </head>
    <body>
        <h1>sin curve test</h1>
        {% autoescape false %}{{svgstr}}{% endautoescape %}
    </body>
</html>

App

`app.py`


#!/bin/env python
# coding: utf-8

import os
import StringIO

from flask import Flask, render_template
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np

app = Flask(__name__)
app.debug = True

@app.route('/')
def do_sin():
    x = np.arange(-np.pi, np.pi, 0.1)
    y = np.sin(x)

    fig = plt.figure()
    plt.plot(x, y, label="sin")
    plt.legend(loc="best")

    strio = StringIO.StringIO()
    fig.savefig(strio, format="svg")
    plt.close(fig)

    strio.seek(0)
    svgstr = strio.buf[strio.buf.find("<svg"):]

    return render_template("sin.html", svgstr=svgstr.decode("utf-8"))

if __name__ == "__main__":
    port = int(os.environ.get("PORT", 5000))
    app.run(port=port)

Deploy

Add to git

$ git init
$ echo .python-version > .gitignore
$ git add .
$ git commit -m "initial commit" .

Deploy to heroku

$ heroku create
$ git push heroku master

Coffee break for a while because there is compilation of numpy and matplotlib

Complete

Verification

$ heroku open

https://limitless-garden-3527.herokuapp.com/ opens

point

SVG processing (strio.buf.find ("<svg") and decode ("utf-8 ") are the points
Japanese display will be shown later

comment

For some time, the Cedar-14 stack and https were the defaults.

Write SVG graphs with matplotlib on heroku