Merge pull request #5 from EodHistoricalData/function-get-news

implemented get_news()

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: eodhdR2
 Type: Package
 Title: Official R API for Fetching Data from 'EODHD'
-Version: 0.5.0
+Version: 0.5.1
 Authors@R: c(
   person(given = c("Marcelo", "S."),
         family = "Perlin",

NAMESPACE

@@ -4,6 +4,7 @@ export(get_demo_token)
 export(get_dividends)
 export(get_exchanges)
 export(get_fundamentals)
+export(get_news)
 export(get_prices)
 export(get_splits)
 export(get_tickers)

NEWS.md

@@ -1,3 +1,7 @@
-## Version 0.5 (2024-08-08)
+## Version 0.5.1 (2024-09-03)
+
+- implemented `get_news()`
+
+## Version 0.5.0 (2024-08-08)
 
 - first version on CRAN

R/cache.R

@@ -30,7 +30,7 @@ write_cache <- function(x, f_out) {
 
   readr::write_rds(x, f_out)
 
-  cli::cli_alert_info("cache file {f_out} saved")
+  cli::cli_alert_info("cache file {basename(f_out)} saved")
 
   return(invisible(TRUE))
 

R/news.R

@@ -0,0 +1,133 @@
+#' Retrieves news for a given ticker and exchange
+#'
+#' This function will query the news point of eodhd and return all news for a user
+#' supplied time period.
+#'
+#' @inheritParams get_fundamentals
+#' @param first_date the first date to fetch news. The function will keep querying the api
+#'  until this date is reached. Default is previous three months.
+#' @param last_date the last date to fetch news. Default is today.
+#' @param offset_delta how much to change offset in each iterations (higher values will result
+#'  in more query time, but less queries). Default is 500.
+#'
+#' @return A dataframe with news events and sentiments
+#' @export
+#'
+#' @examples
+#' \dontrun{
+#' set_token(get_demo_token())
+#' df_news <- get_news(ticker = "AAPL", exchange = "US")
+#' }
+get_news <- function(ticker = "AAPL",
+                     exchange = "US",
+                     first_date = Sys.Date() - 3*30,
+                     last_date = Sys.Date(),
+                     offset_delta = 500,
+                     cache_folder = get_default_cache(),
+                     check_quota = TRUE) {
+
+  cli::cli_h1("retrieving news data for ticker {ticker}|{exchange}")
+
+  if (check_quota) {
+    get_quota_status()
+  }
+
+  token <- get_token()
+
+  first_date <- as.Date(first_date)
+  last_date <- as.Date(last_date)
+
+  f_out <-get_cache_file(ticker, exchange, cache_folder,
+                         paste0("news_",
+                                first_date, '_',
+                                last_date))
+
+  if (fs::file_exists(f_out)) {
+
+    df_out <- read_cache(f_out)
+
+    return(df_out)
+  }
+
+  i_query <- 1
+  this_offset <- 0
+  l_news <- list()
+  while (TRUE) {
+    cli::cli_alert_info("query #{i_query} | offset = {this_offset}")
+
+    url <- glue::glue(
+      paste0('{get_base_url()}/news?s=',
+             '{ticker}.{exchange}&',
+             'from={as.character(first_date)}&',
+             'to={as.character(last_date)}&',
+             'offset={this_offset}&',
+             'limit={offset_delta}&',
+             'api_token={token}&fmt=json'
+      )
+    )
+
+    content <- query_api(url)
+
+    if (content == "[]") {
+      cli::cli_alert_warning("cant find any more data..")
+      break()
+    }
+
+    this_news <- jsonlite::fromJSON(content) |>
+      dplyr::mutate(
+        ticker = ticker,
+        exchange = exchange
+      )
+
+    vec_symbols <- this_news$symbols |>
+      purrr::map_chr(paste0, collapse = ", ")
+
+    vec_tags <- this_news$tags |>
+      purrr::map_chr(paste0, collapse = ", ")
+
+    sentiment <- this_news$sentiment
+    names(sentiment) <- paste0("sentiment_", names(sentiment) )
+
+    this_news <- jsonlite::fromJSON(content) |>
+      dplyr::select(-sentiment) |>
+      dplyr::mutate(
+        date = as.POSIXct(date),
+        ticker = ticker,
+        exchange = exchange,
+        symbols = vec_symbols,
+        tags = vec_tags
+      ) |>
+      dplyr::bind_cols(sentiment)
+
+    query_last_date <- max(this_news$date)
+    n_rows <- nrow(this_news)
+
+    cli::cli_alert_success(
+      "\tgot {n_rows} news | last date: {query_last_date}"
+      )
+
+    i_query <- i_query + 1
+    this_offset <- this_offset + offset_delta
+
+    l_news[[i_query]] <- this_news
+
+    if (query_last_date <= first_date) {
+      cli::cli_alert_warning("current date is lower than first date.. exiting loop.")
+
+      break()
+
+    }
+
+  }
+
+  df_news <- l_news |>
+    purrr::list_rbind()
+
+  write_cache(df_news, f_out)
+
+  cli::cli_alert_success("got {nrow(df_news)} rows of news from {min(df_news$date)} to {max(df_news$date)}")
+
+  return(df_news)
+
+}
+

tests/testthat/test-news.R

@@ -0,0 +1,27 @@
+test_that("news", {
+
+  # test relies on calling api (we skip it on cran to save network bandwith)
+  skip_if_offline()
+  skip_on_cran() # too heavy for cran
+
+  suppressMessages({
+    set_token()
+
+    df_news1 <- get_news(
+      ticker = "AAPL",
+      exchange = "US"
+    )
+
+    expect_true(nrow(df_news1) > 0)
+
+    # run it again for testing local cache
+    df_news2 <- get_news(
+      ticker = "AAPL",
+      exchange = "US"
+    )
+
+    expect_true(identical(df_news1, df_news2))
+
+
+  })
+})