nhtsaslicer.fs

open System.IO
open System
open System.Xml

type ComplaintRecord = {Company:string; Make:string; Model:string; Year:string; Date:DateTime}

let complaintsFile = @"C:\ws\nhtsa\FLAT_CMPL\FLAT_CMPL.txt"

let toDate s = if s="" then DateTime.MinValue else XmlConvert.ToDateTime(s,"yyyyMMdd")

let mapComplaint = (fun (line:string) -> 
                    let fs = line.Split('\t')
                    let date = fs.[7].Trim() |> toDate
                    {Company=fs.[2]; Make=fs.[3]; Model=fs.[4]; Year=fs.[5]; Date=date})

let complaints =  complaintsFile |> File.ReadLines |> Seq.map mapComplaint |> Seq.toArray

let complaintCountByModelYear = 
    complaints
    |> Seq.filter (fun r -> r.Make = "HONDA")
    |> Seq.countBy (fun r -> (r.Model, r.Year)) 
    |> Seq.sortBy (fun (modelYear,count) -> -count)
    |> Seq.toArray
;;
printfn "%A" complaintCountByModelYear

let calcAverage filterFunction = 
    complaints 
    |> Seq.filter filterFunction
    |> Seq.countBy (fun r -> (r.Model, r.Year)) 
    |> Seq.averageBy (fun (_,count) -> (float)count)

let fordBefore2k7 = calcAverage (fun r -> r.Make = "FORD" && r.Date.Year < 2007)
let fordAFter2k7 = calcAverage (fun r -> r.Make = "FORD" && r.Date.Year >= 2007)

let hondaBefore2k7 = calcAverage (fun r -> r.Make = "HONDA" && r.Date.Year < 2007)
let hondaAFter2k7 = calcAverage (fun r -> r.Make = "HONDA" && r.Date.Year >= 2007)

let toyotaBefore2k7 = calcAverage (fun r -> r.Make = "TOYOTA" && r.Date.Year < 2007)
let toyotaAFter2k7 = calcAverage (fun r -> r.Make = "TOYOTA" && r.Date.Year >= 2007)
;;
printfn "Ford %0.2f/%0.2f; Honda %0.2f/%0.2f; Toyota %0.2f/%0.2f" 
        fordBefore2k7 fordAFter2k7 hondaBefore2k7
        hondaAFter2k7 toyotaBefore2k7 toyotaAFter2k7