Drama generator

Make a drama out of your chats

Drama generator is an easily extensible Python script for generating dramatic texts from chats on social media sites.

Currently, we can generate dramas from chats exported from:

• Facebook Messenger
• Telegram
• WhatsApp

To learn how generate your own dramas, you can read our Wiki page or follow our installation instructions below.

Installation

Before running DramaGenerator script, make sure you have the following requirements:

• XeLaTeX - download MiXTeX and follow this installation guide
• free font Gentium Basics

After you've installed all requirements listed above, run the following commands to clone this repository and install all Python requirements.

git clone https://github.com/drobilc/DramaGenerator.git
cd DramaGenerator
python -m venv venv
.\venv\Scripts\activate
pip install -r requirements.txt

Running

Run the drama generator script using python generate_drama.py <PATH TO INPUT FILE>. The <PATH TO INPUT FILE> is the only mandatory argument.

The drama generator script also accepts the following arguments:

General arguments

These arguments can be applied regardless of the choice of social media, format of exported messages, or format you want generated.

FLAG	SHORTER FLAG	ARGUMENT	DEFAULT	MEANING
--output-file	-o	string	/drama_generator/generated_dramas/<INPUT FILE NAME>	path to output file
--parser	-p	string	FacebookHTMLParser	parser chosen according to what social media the messages are from and in which format they are, values can be choosen from the following options: FacebookHTMLParser, TelegramJSONDataExtractor, WhatsAppParser, PickleParser
--generator	-g	string	LatexGenerator	generator chosen based on what we want our generated file to be: a drama or an infografic with statistics; values can be choosen from the following options: LatexGenerator, PlariLatexGenerator, StatisticsGenerator

Parser specific arguments

USABLE IN PARSER	FLAG	ARGUMENT	DEFAULT	MEANING
FacebookHTMLParser	--date-format	string	%b %d, %Y, %I:%M %p	format of the datetime from the file with your exported messages to be used when parsing
FacebookHTMLParser	--locale	string	english_us	locale from the file with your exported messages to be used when parsing dates*
WhatsAppParser	--date-format	string	%d/%m/%Y, %H:%M	format of the datetime from the file with your exported messages to be used when parsing

* in case date format contains localized date strings such as Marec/March/März/...

Processor specific arguments - filters

These arguments can be applied regardless of the choice of social media, format of exported messages, or format you want generated. They do not have default values.

FLAG	SHORTER FLAG	ARGUMENT	MEANING
--after	-a	string	date in format YYYY-MM-DD-HH:MM:SS or YYYY-MM-DD, only messages sent after this date will be used in generation process
--before	-b	string	date in format YYYY-MM-DD-HH:MM:SS or YYYY-MM-DD, only messages sent before this date will be used in generation process
--exclude	-e	string	list of persons separated by ,, whose messages should not be used in generation process
--shout		none	if this flag is present, all messages in generated drama will be written in capital letters
--remove-emojis		none	if this flag is present, emojis will be removed from the text before generating the desired output, reccomended for generating drama because emojis turn into white squares in generated pdf, not recommended for generating statistics as some statistics are about emojis

Generator specific arguments

USABLE IN GENERATOR	FLAG	SHORTER FLAG	ARGUMENT	DEFAULT	MEANING
LatexGenerator, StatisticsGenerator	--title	-t	string	The Drama	title of the outputed document
LatexGenerator	--no-acts		none		if this flag is present, the generated drama won't be divided into acts
LatexGenerator	--no-scenes		none		if this flag is present, the generated drama won't be divided into scenes, and consequently, won't be divided into acts either
LatexGenerator	--new-scene-time		float		number of hours that need to pass between two consecutive messages for the first to end the last scene and the second to begin a new one
StatisticsGenerator	--stopwords		string		path to file containing a list of stopwords, each in its own line

Example usage

    python generate_drama.py
        drama_generator/chats/test_input
        --output-file drama_generator/generated_dramas/test_output
        --parser FacebookHTMLParser
        --title "The Great Monologue"
        --after 2020-01-01-00:00:00
        --before 2020-07-15

Debugging

For debugging in Visual Studio Code, your launch.json file should look like so:

{
    "configurations": [
        {
            "name": "Python: Generate Drama",
            "type": "python",
            "request": "launch",
            "program": "${file}",
            "console": "integratedTerminal",
            "args": ["drama_generator/chats/test_input",
                    "-o=generated_dramas/test_output",
                    "-p=FacebookHTMLParser",
                    "--title='The Great Monologue'",
                    "-a=2020-03-27-07:31:22",
                    "-b=2020-09-03",
                    "-e=Jane Doe",
                    "--shout"]
        }
    ]
}