grokking-challenge-2.go

package main

import (
	"bufio"
	"encoding/csv"
	"fmt"
	"io"
	"log"
	"os"
	"strconv"
	"strings"

	"github.com/gosimple/slug"
	"github.com/jinzhu/gorm"
	_ "github.com/lib/pq"
)

func main() {
	db, err := gorm.Open("postgres", "user=golang password=123456 dbname=grokking")
	if err != nil {
		log.Fatal(err)
	}
	insertData(db)
}

func insertData(db *gorm.DB) error {
	file, err := os.Open("data.csv")
	if err != nil {
		return err
	}
	defer file.Close()
	r := csv.NewReader(bufio.NewReader(file))
	skipTitle := true
	for {

		record, err := r.Read()
		if err == io.EOF {
			break
		}
		if err != nil {
			log.Fatal(err)
		}
		if skipTitle {
			fmt.Println(record)
			skipTitle = false
			continue
		}

		var age *int
		t, err := strconv.Atoi(record[2])
		if err == nil {
			age = &t
		}

		province := correctProvince(record[3])
		err = db.Exec("INSERT INTO data_samples(date_recruited, gender, age, province, province_normalize, country) VALUES(?, ?, ?, ?, ?, ?)",
			record[0], record[1], age, record[3], province, record[4]).Error
		fmt.Println("Insert: ", record[0], record[1], age, province, record[4])
		if err != nil {
			log.Println(err)
		}
	}
	return nil
}

func correctProvince(province string) string {
	slug.CustomSub = map[string]string{
		"tp":   "",
		"city": "",
	}
	newProvince := strings.ToLower(province)
	newProvince = slug.Make(newProvince)
	newProvince = strings.Replace(newProvince, "-", " ", -1)
	newProvince = strings.Replace(newProvince, "thanh pho", " ", -1)
	newProvince = strings.TrimSpace(newProvince)
	switch newProvince {
	case "sai gon", "hcm", "hcmc", "sg", "hochiminh":
		newProvince = "ho chi minh"
	case "hn", "hanoi":
		newProvince = "ha noi"
	}

	return newProvince
}

q3.csv

q4.csv

SQL-scripts.sql

CREATE TABLE public.data_samples (
	id serial PRIMARY KEY,
	date_recruited timestamp NULL,
	gender text NULL,
	age int4 NULL,
	province_normalize text NULL,
	province text NULL,
	country text NULL
);

create index data_samples_gender_age_province on data_samples(gender, age, province_normalize, country);


---1---
select count(*)
from data_samples
where age > EXTRACT(YEAR from age(date '1984-05-14'));
--Res: 1842


---2---
select count(*)
from data_samples
where province_normalize = 'ho chi minh';
--Res: 2545

---3---
select date_recruited, gender, age, province, country
from data_samples
where province_normalize = 'ho chi minh'
and age < 20 and gender='Female';
--Res: q3.csv

---4---
select date_recruited, gender, age, province, country
from data_samples
where (province_normalize = 'ho chi minh' or province_normalize = 'ha noi')
and EXTRACT(DOW FROM date_recruited)  = 0 and EXTRACT(HOUR FROM date_recruited) between 18 and 22;
--Res: q4.csv