WIP add: Dataframe based golang question

platform-engineer/go-dataframes/analysis.csv

@@ -0,0 +1,3 @@
+group_id
+1
+2

platform-engineer/go-dataframes/data.csv

@@ -0,0 +1,9 @@
+group_id,individual_id,attr_1,attr_2
+1,1,abc,def
+1,2,abc,def
+1,3,abc,def
+1,4,abc,def
+2,5,abc,def
+2,6,abc,def
+2,7,abc,def
+2,8,abc,def

platform-engineer/go-dataframes/go.mod

@@ -0,0 +1,5 @@
+module live_coding
+
+go 1.16
+
+require github.com/gocarina/gocsv v0.0.0-20210516172204-ca9e8a8ddea8

platform-engineer/go-dataframes/go.sum

@@ -0,0 +1,2 @@
+github.com/gocarina/gocsv v0.0.0-20210516172204-ca9e8a8ddea8 h1:hp1oqdzmv37vPLYFGjuM/RmUgUMfD9vQfMszc54l55Y=
+github.com/gocarina/gocsv v0.0.0-20210516172204-ca9e8a8ddea8/go.mod h1:5YoVOkjYAQumqlV356Hj3xeYh4BdZuLE0/nRkf2NKkI=

platform-engineer/go-dataframes/main.go

@@ -0,0 +1,72 @@
+package main
+
+import (
+	"github.com/gocarina/gocsv"
+	"live_coding/models"
+	"os"
+)
+
+func main() {
+	// Parse CSV into a grouped dataframe on the basis of the group ID
+	in, groupedDf := parseCSV()
+	defer closeSafely(in)
+
+	// Map with
+	var analyzedGroups []models.AnalyzedGroup
+
+	// Iterating over records to analyse each group
+	for groupId, _ := range groupedDf {
+		analyzedGroup := models.AnalyzedGroup{
+			GroupID: groupId,
+		}
+		// TODO: Insert your code
+		// ...
+		analyzedGroups = append(analyzedGroups, analyzedGroup)
+	}
+
+	// Write results to CSV
+	file, _ := os.OpenFile("analysis.csv", os.O_RDWR|os.O_CREATE, os.ModePerm)
+	// Close file connection
+	defer closeSafely(file)
+	//write the csv file
+	if err := gocsv.MarshalFile(&analyzedGroups, file); err != nil{
+		panic(err)
+	}
+}
+
+// Close connection to a file safely
+func closeSafely(f *os.File) {
+	if err := f.Close(); err != nil {
+		panic(err)
+	}
+}
+
+// Parse the CSV and group the data on the basis of the group ID
+func parseCSV() (*os.File, map[int][]*models.DataFrame) {
+	in, err := os.Open("data.csv")
+	if err != nil {
+		panic(err)
+	}
+
+	var dfs []*models.DataFrame
+
+	if err := gocsv.UnmarshalFile(in, &dfs); err != nil {
+		panic(err)
+	}
+
+	var groupedDf = make(map[int][]*models.DataFrame)
+	var exists bool
+	var df *models.DataFrame
+
+	for i := range dfs {
+		df = dfs[i]
+		// Initializing the map slice
+		if _, exists = groupedDf[df.GroupID]; !exists {
+			groupedDf[df.GroupID] = []*models.DataFrame{}
+		}
+
+		// Map group Ids to all the individual IDs
+		groupedDf[df.GroupID] = append(groupedDf[df.GroupID], df)
+	}
+	return in, groupedDf
+}

platform-engineer/go-dataframes/models/model.go

@@ -0,0 +1,16 @@
+package models
+
+// Frame of data in the CSV
+type DataFrame struct {
+	GroupID      int    `csv:"group_id"` // .csv column headers
+	IndividualID int    `csv:"individual_id"`
+	Attr1        string `csv:"attr_1"`
+	Attr2        string `csv:"attr_2"`
+}
+
+// Analysis on a group
+type AnalyzedGroup struct {
+	GroupID int `csv:"group_id"`
+}
+
+

platform-engineer/go-dataframes/pkg1/analysis.go

@@ -0,0 +1,7 @@
+package pkg1
+
+import "live_coding/models"
+
+func count(df []*models.DataFrame) int {
+	return len(df)
+}

platform-engineer/go-dataframes/pkg2/analysis.go

@@ -0,0 +1 @@
+package pkg2