betfaiR: R Client for Betfair's API

# #' collect data
# #'
# #' @description helper function to be used within an R script that is part of a
# #' scheduled task.
# #'
# #' @details This function will take a \link{betfair} environment, a
# #' collection of marketIds and priceProjection parameters and retrieve data at
# #' intervals set up by the task, and keep appending new data returned by the
# #' \link{marketBook} method to any previous data returned.  The function will also
# #' add a new variable called \code{collectedAt} to each list, which means cross
# #' market comparisons are possible.  It is a good idea to also store/save data
# #' returned by the \link{marketCatalogue} method which will contain data about the
# #' runners in the markets you are collecting data from.
# #'
# #' @param bf betfair environment, returned from the \link{betfair} function
# #' @param filter \link{marketFilter} object to help retrieve marketIds for markets
# #' whose data is to be collected
# #' @param marketIds ids of markets to retrieve data for
# #' @param maxResults maximum number of markets to retrieve
# #' @param marketIds_file filename for markets collecting data from, default is
# #' "marketIds" which will create a custom file which includes the date on which
# #' the file was created, like so "yyyy-mm-dd_marketIds.RDS"
# #' @param priceProjection what data to return via the \link{marketBook} function
# #' @param marketProjection what data is to be returned via the \link{marketCatalogue}
# #' function, which is used to retrieve metadata about the markets and establish
# #' the live markets
# #'
# #' @return the function saves data returned from the Betfair API into RDS files
# #' named after the market, for example market_1.121212121.RDS.
# #'
# #' @export
# collect_data <- function(bf,
#                          filter = marketFilter(),
#                          marketIds = NULL,
#                          maxResults = 1000,
#                          marketIds_file = "marketIds",
#                          priceProjection = c("EX_ALL_OFFERS", "EX_TRADED"),
#                          marketProjection = c("EVENT", "RUNNER_DESCRIPTION")) {
#
#     # retrieve files from current working directory
#     files <- list.files()
#     today <- Sys.Date()
#     # create marketIds file name using marketIds_file param
#     marketIds_file <- paste0(today, "_", marketIds_file, ".RDS")
#     if(!(marketIds_file %in% files)) {
#         # build filter object
#         filter_obj <- filter
#         # overwrite filter object if marketIds are present in params
#         if(!is.null(marketIds)) filter_obj <- marketFilter(marketIds = marketIds)
#         # retrieve data
#         out <- bf$marketCatalogue(filter = filter_obj,
#                                   maxResults = maxResults,
#                                   marketProjection = c("EVENT", "RUNNER_DESCRIPTION"))
#         saveRDS(out, marketIds_file)
#         marketIds <- unlist(lapply(out, function(i) i$market$marketId))
#     } else {
#         marketIds <- readRDS(marketIds_file)
#         marketIds <- unlist(lapply(marketIds, function(i) i$market$marketId))
#         # ascertain which markets are still live
#         available_markets <- bf$marketCatalogue(filter = marketFilter(marketIds = marketIds),
#                                                 maxResults = length(marketIds))
#         available_markets <- sapply(available_markets, function(i) i$market$marketId)
#         test <- marketIds %in% available_markets
#         marketIds <- marketIds[test]
#     }
#     # save current time, to add to retrieved data
#     currentTime <- Sys.time()
#     # loop through each market and retrieve data, adding currentTime to enable comparison across markets
#     plyr::l_ply(marketIds, .fun = function(i, bf, files, currentTime, priceProjection) {
#
#         filename <- paste0("market_", i, ".RDS")
#         if(filename %in% files) {
#             previous_data <- readRDS(filename)
#         } else {
#             previous_data <- structure(list(), class = c("betfaiR_market", "list"))
#         }
#         # retrieve data from market
#         current_data <- bf$marketBook(marketIds = i,
#                                       priceProjection = priceProjection)
#         # add collectedAt element to data returned
#         current_data[[1]]$collectedAt <- currentTime
#         # append new current_data to previous_data (which is either empty or previously collected data)
#         previous_data <- append(previous_data, current_data)
#         class(previous_data) <- c("betfaiR_market_l", class(previous_data))
#         # save new data
#         saveRDS(previous_data, filename)
#     }, bf = bf,
#        files = files,
#        currentTime = currentTime,
#        priceProjection = priceProjection)
#
#     return(invisible())
# }
#
# #' @export
# as.data.frame.betfaiR_market_l <- function(x, row.names = NULL, optional = FALSE, ...) {
#     tmp <- plyr::ldply(x, function(i) {
#         tmp <- plyr::ldply(i$runners, function(j) {
#             j$basic
#         })
#         tmp$collectedAt <- i$collectedAt
#         return(tmp)
#     })
#     n <- c("selectionId", "lastPriceTraded", "totalMatched", "collectedAt")
#     n <- n[n %in% names(tmp)]
#     tmp <- subset(tmp, select = n)
#     class(tmp) <- c("betfaiR_market_df", class(tmp))
#     return(tmp)
# }
#
# #' @export
# plot.betfaiR_market_df <- function(x, ...) {
#     x$collectedAt <- strptime(x$collectedAt, "%Y-%m-%d %H:%M:%S")
#     p <- ggplot2::ggplot(data = x,
#                          ggplot2::aes(x = x$collectedAt,
#                                       y = 1 / x$lastPriceTraded))
#     p <- p +
#         ggplot2::geom_path(ggplot2::aes(colour = selectionId))
#     p <- p +
#         betfair_theme()
# }
#
# #' @export
# plot.betfaiR_market_l <- function(x, ...) {
#     df <- as.data.frame(x)
#     plot(df)
# }