[PYTHON] Web service development technology selection that reminds us of old likes and reminders
What kind of web service?
https://remind-archives.herokuapp.com/main/getting_started
Have you ever laughed or impressed when you looked back at Twilog and favolog and remembered the funny old tweets you forgot? ?? Remind Archives was created to randomly display such past tweets and posts for discovery and serendipity. Currently supports the following. --Twitter likes --Twitter RT --Tumblr's favorite --Tumblr
Which technology (language, framework) to use
- The following is all information as of July 2020.
I've been thinking about creating this Web service for quite some time, but I didn't know how to create it, so I kept the idea for a long time. Even when I actually searched, there were various languages and frameworks, so I had to think about "which one is better" first.
First, I considered Java, which I'm used to, but after seeing Why I'm a Java shop starting Python-Qiita, it takes time to build. Considering this, I excluded it from the candidates.
While searching next, I found out that there are three major frameworks (Rails, Django, Laravel) and decided to choose from these.
But first of all, when I did the tutorial before, Rails often got an error even though I wrote it exactly as it was written in the document, and because it automatically generates code, I often do not know what the problem is when an error occurs. Excluded (for that matter, it's currently declining).
The problem remains with Django (python) and Laravel (PHP), but these are pretty similar in the introductory articles and grammar and I was very confused. Eventually --Larabel has to do various things just by installing PHP and setting up MySQL (so you can proceed with the online development environment by reading the explanation site) --Python seems to have more potential (machine learning, etc.) --Slightly better on Google Trends, but Django is better (especially in the US) --Laravel has few official documents I chose Django for some reason. I especially referred to the article Introduction to Python Django (1) --Qiita.
Where to deploy
First of all, Firebase seems to be unable to set an upper limit on the price, so I stopped it. Next, I considered lollipop, but when I asked about the DB capacity, the answer was only 500MB, so I excluded this as well.
After all, I decided to use Heroku, which is a major player, but I had a lot of trouble installing the library and various errors just by deploying it. However, when I actually tried other services (Linode and Vultr), ** it took a lot of time and effort **, and since there are few documents that can be used as reference, I do not know how to solve the error, so the elimination method After all, it became Heroku.
The following articles were especially helpful. How to deploy the Django app on Heroku-Qiita Other articles such as the following were helpful.
Deploy to Heroku · Django Girls Tutorial: Extensions [Notes around Django settings.py --Qiita](https://qiita.com/jp_ibis/items/0ac1ba8aba1398f2a492#%E8%BF%BD%E5%8A%A0%E3%81%A7%E5%A4% 96% E9% 83% A8% E3% 83% 95% E3% 82% A1% E3% 82% A4% E3% 83% AB% E3% 81% 8B% E3% 82% 89% E8% A8% AD% E5% AE% 9A% E3% 82% 92% E8% AA% AD% E3% 81% BF% E8% BE% BC% E3% 82% 80) [Error handling when deploying Django app to Heroku whitenoise edition --Qiita](https://qiita.com/ymhr1121/items/344c4eb300ab9972d0c2#%E3%81%9D%E3%81%AE2-wsgipy%E3%81 % A7% E3% 81% AF% E3% 81% AA% E3% 81% 8Fsettingspy% E3% 82% 92% E7% B7% A8% E9% 9B% 86% E3% 81% 99% E3% 82% 8B ) What to do about errors when deploying Django apps on Heroku-Qiita
Which ad to choose
I wanted to place an ad because it costs a server fee, but Google AdSense did not pass the examination because "there is little content" (even though it is a service that reads and displays user data!). I couldn't help it, so I searched for something else and decided to include the Japanese Ninja AdMax without any screening. For overseas advertisements, we chose ylliX, which also has no screening.
Finally
Although it is still in the conceptual stage, the technology acquired to create this service will be released on Techpit.