The Bukowski Sequence:

Like any other 17 year old guy, Bukowski was a really fun read. Different era, different perspective. Slums, women, booze, fights, all make for a potent youth catnip. It got me back into reading so I have been grateful for that. More recently, and partially due to a location change, I've chased down another Bukowski fixture: the horses. As I like uncertainty and probalistic estimates, I figured I'd throw my hat into the ring.

First step was to acquire the dataset. Copy, pastes and the following script helped out with that:

	direc <- direcHere
	realFile <- textFileContainingInfo
	require(stringr)
	resultsAttr <- c("raceID", "lastRaceDT","lastRaceNum", "lastTrack", "lastFin",
	"pgm","horseNM","jockeyNM","Wgt","ME","PP","fin","odds","cmmt")
	resultsFrame <- as.data.frame(matrix(NA,nrow=1,ncol=length(resultsAttr)))
	colnames(resultsFrame) <- resultsAttr
	resFrame <- resultsFrame
	notesFrame <- as.data.frame(matrix(NA,nrow=1,ncol=2))
	raceAttr <- c("Track","Date","Race Number","Race Type","HorseType",
	"ClaimPriceHigh","ClaimPriceLow","Race Length","Record Time",
	"Record Date", "Purse","Plus",
	"Available Money","Valueist","Value2nd","Value3rd","Weather",
	"Track","Time","Start",
	"2FTime1","2FTime2","3FTime1","3FTime2","3FTime3",
	"4FTime1","4FTime2","4FTime3","4FTime4",
	"t_Final",
	"2SplT1","2SplT2","3SplT1","3SplT2","3SplT3",
	"4SplT1","4SplT2","4SplT3","4SplT4",
	"Run-up","WPS_Pool", "ID")
	raceFrame <- as.data.frame(matrix(NA,nrow=1,ncol=length(raceAttr)))
	colnames(raceFrame) <- raceAttr
	fileName <- realFile
	testRun <- readLines( paste(direc,fileName,sep="")
	)
	chunkSplits <- c(0,grep("Copyright", testRun))
	for (i in seq( 1 , length( chunkSplits ) - 1 ) ) {
	print(i)
	## Sets the current chunk to be parsed.
	chunk <- testRun[ chunkSplits[i]:chunkSplits[i+1] ]
	testGrep <- grep("GOLDEN GATE FIELDS", chunk)
	testGreg <- gregexpr("([^-])+", chunk[testGrep])
	raceFrame[i,1:3] <- str_trim(regmatches(chunk[testGrep],testGreg)[[1]]
	,side="both")
	raceFrame[i,2] <- as.character( as.Date(raceFrame[i,2],
	format='%B %d, %Y') )
	raceFrame[i,4:5] <- str_trim(
	regmatches(chunk[testGrep+1]
	,gregexpr("([^-])+", chunk[testGrep+1] ) )[[1]],
	side="both" )
	priceGrep <- grep("Claiming Price:", chunk)+1
	if (length(priceGrep)!=0){
	raceFrame[i,6:7] <- str_trim(regmatches(chunk[priceGrep]
	,gregexpr("([^-])+",chunk[priceGrep]) )[[1]]
	, side = "both" )
	raceFrame[i,6:7] <- gsub('\\$','',raceFrame[i,6:7])
	} else{
	raceFrame[i,6:7] <- NA
	}
	lengthGrep <- grep("[Ff]urlong | [Mm]ile | [Ff]urlongs | [Mm]iles",chunk)[1]
	raceFrame[i,8] <- str_trim(chunk[lengthGrep],side='both')
	## raceFrame[i,8] <- str_trim(chunk[priceGrep+2],side='both')
	recordGrep <- grep("Track Record",chunk)
	raceFrame[i,9:10] <- str_trim(
	regmatches(chunk[recordGrep+1]
	,gregexpr("([^-])+", chunk[recordGrep+1] ) )[[1]][2:3],
	side="both" )
	raceFrame[i,10] <- substr(raceFrame[i,10],1,nchar(raceFrame[i,10])-1)
	raceFrame[i,10] <- as.character(as.Date(raceFrame[i,10],
	format='%B %d, %Y' ) )
	raceFrame[i,11] <- gsub('\\$','',str_trim(chunk[grep("Purse", chunk)+1]))
	raceFrame[i,12] <- 0
	## str_trim(chunk[grep("Plus:",chunk)+1] )
	raceFrame[i,13] <- gsub('\\$','',
	str_trim(chunk[grep("Available Money:",chunk)+1] ) )
	raceValGrep <- grep("Value of Race:", chunk)
	raceFrame[i,14:16] <-regmatches(chunk[raceValGrep+1],
	gregexpr("[0-9^\\,0-9^]+",
	chunk[raceValGrep+1]) )[[1]][c(1,3,5)]
	raceFrame[i,14:16] <-regmatches(chunk[raceValGrep+1],
	gregexpr("[0-9^\\,]+[0-9]{3}",
	chunk[raceValGrep+1]) )[[1]][c(1,2,3)]
	raceFrame[i,17] <- str_trim(chunk[grep("Weather:",chunk)+1] )
	raceFrame[i,18] <- str_trim(chunk[grep("Track:",chunk)+1] )
	raceFrame[i,19] <- str_trim(chunk[grep("Off at:",chunk)+1] )
	raceFrame[i,20] <- str_trim(chunk[grep("Start:",chunk)+1] )
	fTimes <- str_trim(
	strsplit(chunk[grep("Fractional Times", chunk)+1], " ")[[1]],
	side="both" )
	if (length(fTimes)==2){
	raceFrame[i,21:22] <- fTimes
	}
	if (length(fTimes)==3){
	raceFrame[i,23:25] <- fTimes
	}
	if (length(fTimes)==4){
	raceFrame[i,26:29] <- fTimes
	}
	raceFrame[i,30] <- str_trim(chunk[grep("Final Time:",chunk)+1] )
	sTimes <- gsub('\\)','',
	str_trim(
	gsub('\\(', '',
	strsplit(chunk[grep("Split Times", chunk)+1], " ")[[1]]
	),
	side="both" ) )
	if (length(sTimes)==2){
	raceFrame[i,31:32] <- sTimes
	}
	if (length(sTimes)==3){
	raceFrame[i,33:35] <- sTimes
	}
	if (length(sTimes)==4){
	raceFrame[i,36:39] <- sTimes
	}
	raceFrame[i,40] <- str_trim(chunk[grep("Run-Up:",chunk)+1] )
	raceFrame[i,41] <- str_trim(chunk[grep("Total WPS Pool:",chunk)+1] )
	raceID <- i
	raceFrame[i,42] <- raceID
	output <- "C:\\Users\\Coleman\\Desktop\\Projects\\Horses\\footNoteFile.txt"
	notesRange <- c(grep("Footnotes",chunk)+1,
	grep("Equibase Company LLC\\.", chunk)-1)
	cat(paste("Race ID:",raceID,sep=''),file=output,append=TRUE)
	cat("\n",file=output,append=TRUE)
	if (length(notesRange!=2)){
	finalNote <- tail(notesRange,n=1) }
	cat(paste(chunk[notesRange[1]:finalNote]),file=output,append=TRUE)
	cat("\n \n \n",file=output,append=TRUE)
	##--------Begins result information.
	smaller <- c( grep( "Last Raced",chunk ),
	grep( "Fractional Times",chunk ) )
	subChunk <- chunk[smaller[1]:smaller[2]]
	grepName <- grep( "\\((.*)\\)", subChunk )
	grepName <- grepName[2:length(grepName)]
	for (k in 1:length(grepName)){
	tryCatch({
	resultsFrame[1,] <- NA
	resultsFrame[1,1:5]<- c(raceID
	,subChunk[grepName[k]-4]
	,subChunk[grepName[k]-3]
	,subChunk[grepName[k]-2]
	,subChunk[grepName[k]-1] )
	resultsFrame[1,6] <- substr(subChunk[grepName[k]],1,1)
	if (resultsFrame[1,6]=="-"){resultsFrame[1,2:5] <- NA}
	resultsFrame[1,7] <- str_trim(
	regmatches( subChunk[grepName[k]],
	gregexpr("[A-Za-z](.*) \\(",
	subChunk[grepName[k]])[[1]]-1 ) )
	resultsFrame[1,8] <- str_trim(
	gsub("\\(|\\)","",
	regmatches( subChunk[grepName[k]],
	gregexpr("\\((.*)\\)",
	subChunk[grepName[k]])[[1]] ) ) )
	resultsFrame[1,9] <- str_trim( gsub( "\\)","",
	regmatches( subChunk[grepName[k]],
	gregexpr("\\) [0-9]* ",
	subChunk[grepName[k]])[[1]] ) ) )
	resultsFrame[1,10] <- str_trim(
	regmatches( subChunk[grepName[k]],
	gregexpr(paste(resultsFrame[1,9],"([A-Za-z]* )*",sep=" "),
	subChunk[grepName[k]])[[1]] ) )
	resultsFrame[1,10] <- str_trim( gsub(
	"[0-9]","",resultsFrame[1,10] ) )
	if (resultsFrame[1,10]==""){
	resultsFrame[1,11] <- str_trim(
	regmatches( subChunk[grepName[k]],
	gregexpr(paste(resultsFrame[1,9],"[0-9]",sep=" "),
	subChunk[grepName[k]])[[1]] ) )
	resultsFrame[1,11] <- str_trim( strsplit (
	resultsFrame[1,11], " " )[[1]][2] )
	} else {
	resultsFrame[1,11] <- str_trim( gsub("[A-Za-z]","",
	regmatches( subChunk[grepName[k]],
	gregexpr(paste(resultsFrame[1,10]," [0-9]",sep=""),
	subChunk[grepName[k]])[[1]] ) ) )
	}
	resultsFrame[1,12] <- k
	decGrep <- grep("[0-9]*\\.[0-9]*", subChunk)
	resultsFrame[1,13] <- str_trim(
	regmatches ( subChunk[decGrep[k]],
	gregexpr("[0-9]*\\.[0-9]*",
	subChunk[decGrep[k]])[[1]] ) )
	resultsFrame[1,14] <- str_trim(
	gsub("[0-9]*\\.[0-9]*", "", subChunk[decGrep[k]]) )
	resFrame <- rbind(resFrame,resultsFrame)
	}, warning=function(war){
	print("problem")
	}, error=function(err){
	print("error")
	}, finally={})
	}
	}

view raw gistfile1.txt hosted with ❤ by GitHub