This is an automated bot for X that scraps horoscopes in a web and translates them to spanish and then posts them.
The job is run daily at 10:00 am UTC and each post has a 15 minute delay between them to avoid too many requests and data in a row.
You can check the .env.template and see what exactly you need for this.
These are the API KEYS needed if you want to do the exact same. I won't specify how to set up the twitter dev account, please refer to their docs here.
DEEPL_API_KEY=__DEEPL_API_KEY__
SCRAP_WEB=__SCRAP_WEB__
SCRAP_CLASS=__SCRAP_CLASS__
X_API_KEY=__X_API_KEY__
X_API_KEY_SECRET=__X_API_KEY_SECRET__
X_ACCESS_TOKEN=__X_ACCESS_TOKEN__
X_ACCESS_TOKEN_SECRET=__X_ACCESS_TOKEN_SECRET__
X_BEARER_TOKEN=__X_BEARER_TOKEN__
This is essentially where do you want to get the data from. You set a web, and the code downloads and parses it as HTML.
Take a look into the lines in scrapper.go
class := config.GetEnvVar("SCRAP_CLASS")
divs := doc.Find(class)
secondParagraph := divs.First().Find("p").Last().Text()
Basically, for my use case, it gets the <div className=".class-here"> and takes the first and last <p>, which I base as the text for the bot.
Then, it just plays with the translation and such.
Notes:
If using this, take care of the deepl api url.
As I'm using the free tier, the url is "https://api-free.deepl.com/v2", but if you opt for another plan, change that.
Also, if you want to change the translator provider, goo into the deepl.go file and focus in the translate function, since it's the one doing the work.