Flask Service

[MrHardcode]

Experimental branch with one hardcoded example of how to make the flask server work Working on a proper setup of the flask server on this branch now

• [x] Add try-catch in case we try to plot data, when all data has been filtered away or there simply was no available data to begin with
• [x] Add support for more filters in the endpoint
• [x] Improve readability in code
• [x] Decide how to handle the amount of tweets scraped

[MrHardcode]

I've made some mistakes so don't merge anything

[MrHardcode]

Plotting still seems to be bugged - check out this

Plot was created with the following request data:

{
    "hashtags": [
        "trump",
        "biden"
    ],
    "start_date": "2020-5-3",
    "end_date": "2020-5-12",
    "plot_type": "line",
    "search_for": {"mentions": "@JoeBiden"}
}

[MrHardcode]

Should the amount of tweets scraped be an option or should we just always scrape 500 as standard? (would get more older tweets). I vote for the hardcoded option of course

[Castau]

Either could be an option. For the purpose of the project it's fine to limit it to a specific amount :)

[MalteMagnussen]

Should the amount of tweets scraped be an option or should we just always scrape 500 as standard? (would get more older tweets). I vote for the hardcoded option of course

Depends how long it takes? If it takes barely any time to scrape, lets say, 2000 tweets, compared to 500, then why not let user do that? (Hardcoded option) If its easy to up the amount of tweets scraped, then it should also be relatively straightforward to let user pass a body parameter to take more. Default 500, if no parameter set by user in JSON body?

[MrHardcode]

Okay let's make a default at 300 (takes ca. 10 seconds) and let the user change the amount

[MrHardcode]

Okay now you have the option of controlling the web scraper