0

I'm trying to capture a data pull from the twelvedata.com webpage. I want to pull historical equity prices for several stocks with a long time frame.

The instructions are located here: https://twelvedata.com/docs#complex-data

This request requires a JSON POST. I'm struggling to make this work. This is what I have thus far.

    url <- "https://api.twelvedata.com/complex_data?apikey=myapikey"
    requestBody <- paste0('{"symbols" : "AAPL",
                            "intervals" : "1day",
                            "start_date" : "2021-05-01",
                            "methods" : "symbol"}')
    
    res <- httr::POST(url = url,
                      body = requestBody,
                      encode = "json") 

    
    test_1 <- content(res, as="text") %>% fromJSON()
    
    test_2 <- as.data.frame(rjson::fromJSON(test_1))

Any help would be greatly appreciated. Thank you for your time.

DougC
  • 3
  • 3

1 Answers1

0

Perhaps use toJSON to create requestBody. Here is an example:

requestBody = toJSON(list(
  symbols=c("AAPL"),
  intervals=c("1day"),
  start_date=unbox("2021-05-01"),
  end_date=unbox("2021-12-31"),
  methods="time_series")
)

res <- httr::POST(url = url,body = requestBody,encode = "json")

The following will then return this type of structure:

test_1 <- httr::content(res, as="text") %>% rjson::fromJSON()
do.call(rbind, lapply(test_1$data[[1]]$values,data.frame)) %>% head()

Output:

    datetime      open      high       low     close   volume
1 2021-12-30 179.47000 180.57001 178.09000 178.20000 59773000
2 2021-12-29 179.33000 180.63000 178.14000 179.38000 62348900
3 2021-12-28 180.16000 181.33000 178.53000 179.28999 79144300
4 2021-12-27 177.09000 180.42000 177.07001 180.33000 74919600
5 2021-12-23 175.85001 176.85001 175.27000 176.28000 68227500
6 2021-12-22 173.03999 175.86000 172.14999 175.64000 92004100

Update: Multiple Symbols.

If you have multiple symbols, requestBody should be updated like this:

requestBody = toJSON(list(
  symbols=c("AAPL", "MSFT"),
  intervals=c("1day"),
  start_date=unbox("2021-05-01"),
  end_date=unbox("2021-12-31"),
  methods="time_series")
)

and you can extract each into its own data frame within a list, like this

lapply(test_1$data, function(s){
  stock = s$meta$symbol
  do.call(rbind, lapply(s$values, data.frame)) %>% mutate(symbol = stock)
})
langtang
  • 22,248
  • 1
  • 12
  • 27
  • This works great. How do you add multiple stocks or methods though? Can you help with that syntax? Thanks for your time. – DougC May 02 '22 at 03:45
  • One last question. How do I pull multiple methods and also set conditions? Let's say I want to pull the EMA with a time period of 7. You have been super helpful langtang. I really appreciate your input. Thank you. – DougC May 02 '22 at 13:47
  • I think it all comes down to writing the json body correctly.. For something like ema with conditions, you can do this : ` methods=data.frame(name="ema", time_period=7)` – langtang May 02 '22 at 14:39