Script to scrape the 2015 AJC Peachtree Road Race results

peachtree_2015

#############################################################################
## DOWNLOAD and PARSE data

base_url <- "http://trackshackresults.com/peachtree/results/2015/ptresults.php"

divisions_map <- read.table(text = '
                            
                            Ind   Div     Division              
                            1     1B     "****MEN -- OPEN****"
                            2     1G   "****WOMEN -- OPEN****"
                            3     1M     "****MEN -- MASTERS****"
                            4     1N   "****WOMEN -- MASTERS****"
                            5      C         "MEN -- 14 AND UNDER"
                            6      D         "MEN -- 15 THROUGH 19"
                            7      E         "MEN -- 20 THROUGH 24"
                            8      F         "MEN -- 25 THROUGH 29"
                            9      G         "MEN -- 30 THROUGH 34"
                            10     H         "MEN -- 35 THROUGH 39"
                            11     I         "MEN -- 40 THROUGH 44"
                            12     J         "MEN -- 45 THROUGH 49"
                            13     K         "MEN -- 50 THROUGH 54"
                            14     L         "MEN -- 55 THROUGH 59"
                            15     M         "MEN -- 60 THROUGH 64"
                            16    MA         "MEN -- 65 THROUGH 69"
                            17     N         "MEN -- 70 THOURGH 74"
                            18    NA         "MEN -- 75 THROUGH 79"
                            19    NB         "MEN -- 80 AND OVER"
                            20     Q       "WOMEN -- 14 AND UNDER"
                            21     R       "WOMEN -- 15 THROUGH 19"
                            22     S       "WOMEN -- 20 THROUGH 24"
                            23    SA       "WOMEN -- 25 THROUGH 29"
                            24     U       "WOMEN -- 30 THROUGH 34"
                            25     V       "WOMEN -- 35 THROUGH 39"
                            26     W       "WOMEN -- 40 THROUGH 44"
                            27     X       "WOMEN -- 45 THROUGH 49"
                            28     Y       "WOMEN -- 50 THROUGH 54"
                            29     Z       "WOMEN -- 55 THROUGH 59"
                            30    ZA       "WOMEN -- 60 THROUGH 64"
                            31    ZB       "WOMEN -- 65 THROUGH 69"
                            32    ZC       "WOMEN -- 70 THROUGH 74"
                            33    ZD       "WOMEN -- 75 THROUGH 79"
                            34    ZE       "WOMEN -- 80 AND OVER"
                            
                            ', header = TRUE, na.strings = "")

all_urls <- with(divisions_map,
                 setNames(sprintf("%s?Link=10&Type=2&Div=%s&Ind=%s",
                                  base_url, Div, Ind),
                          Division))

# pull data from website into a list of data.frames (one per Division)
library(XML)
data <- lapply(all_urls, readHTMLTable, which = 6, skip.rows = 1)
# append Division column
data <- Map(cbind, data, Division = names(data))
# collapse everything into one data.frame
data <- do.call(rbind, data)
rownames(data) <- NULL
# rename columns
colnames(data) <- c("DivisionPlace", "FullName", "BibNumber", "Age",
                    "Place", "GenderPlace", "ClockTime", "NetTime",
                    "Hometown", "Division")

#############################################################################
## Data Manipulation

TimeInMinutes <- function(tim) {
   # this prepends sub-hour times with "0:"
   tim <- sub("^(\\d{1,2}:\\d{1,2})$", "0:\\1", tim)
   as.numeric(as.difftime(tim , format = "%H:%M:%S", units = "mins"))
}

data <- within(data, {
   PrintTime = NetTime
   ClockTime = TimeInMinutes(ClockTime)
   NetTime   = TimeInMinutes(NetTime)
   StartTime = ClockTime - NetTime
   Gender    = ifelse(grepl("WOMEN", Division), "F", "M")
})

#############################################################################
## Export

export <- with(data,
  data.frame(Place     = Place,
             FullName  = FullName,
             Age       = Age,
             Gender    = Gender,
             Hometown  = Hometown,
             NetTime   = sprintf("%.3f", NetTime),
             StartTime = sprintf("%.3f", StartTime),
             PrintTime = PrintTime,
             stringsAsFactors = FALSE)
)

write.csv(export, file = "2015.csv", row.names = FALSE)