Compare commits

..

No commits in common. "32f83e358b0a72818444abb122bed44bdc9dd10d" and "8abff30b52a57f1eca44fea0e91d1b01519e0dd2" have entirely different histories.

2 changed files with 28 additions and 161 deletions

4
.gitignore vendored
View File

@ -1,4 +1,2 @@
/thehub_cache
/thehub.json
/itjobbank_cache
/it-jobbank.json
/jobs.json

169
main.go
View File

@ -6,7 +6,6 @@ import (
"log"
"os"
"strings"
"time"
"github.com/gocolly/colly"
)
@ -20,8 +19,6 @@ type job struct {
Description string `json:"description"`
Link string `json:"link"`
Skills skills `json:"skills"`
FirstSeen time.Time `json:"firstSeen"`
Source string `json:"source"`
}
type skills struct {
@ -34,9 +31,6 @@ type skills struct {
Tailwind bool `json:"tailwind"`
}
// Utility functions
// Checks if a string contains any of the given keywords
func skillChecker(description string) skills {
return skills{
React: strings.Contains(description, "React"),
@ -48,9 +42,7 @@ func skillChecker(description string) skills {
Tailwind: strings.Contains(description, "Tailwind"),
}
}
// Converts job struct to json
func jobsToJson(file *os.File, jobs []job, fName string) {
func jobsToJson(file *os.File) {
// Encode jobs slice to JSON
encoder := json.NewEncoder(file)
encoder.SetIndent("", " ") // Pretty-print with indentation
@ -61,47 +53,24 @@ func jobsToJson(file *os.File, jobs []job, fName string) {
fmt.Println("Job details successfully written to", fName)
}
func checkIfPaid(description string) {
for _, keyword := range unpaidKeywords {
if strings.Contains(strings.ToLower(description), keyword) {
return
}
}
}
func checkIfStudent(description string) string {
for _, keyword := range studentKeywords {
if strings.Contains(strings.ToLower(description), keyword) {
return "student"
}
}
return "full time"
}
// Slice to store job details
var (
excluded = []string{"senior", "lead", "founder", "cto", "vp of", "erfaren", "arkitekt", "architect", "manager", "ulønnet", "unpaid", "praktik"}
unpaidKeywords = []string{"unpaid", "praktik", "ulønnet"}
studentKeywords = []string{"studerende", "studenter", "student", "medhjælper"}
)
func scrapeHub() {
// declare and initialize variables
var (
jobs []job
jobCount int
fName = "thehub.json"
maxJobs = 30
baseUrl = "https://thehub.io"
searchString = "https://thehub.io/jobs?roles=frontenddeveloper&roles=fullstackdeveloper&roles=backenddeveloper&roles=devops&paid=true&countryCode=DK&sorting=newJobs"
file, err = os.Create(fName)
)
maxJobs int = 30
fName string = "jobs.json"
excluded = []string{"senior", "lead", "founder", "cto", "vp of"}
)
func scrapeJobs() {
file, err := os.Create(fName)
if err != nil {
log.Fatalf("Cannot create file %q: %s", fName, err)
}
defer file.Close()
baseUrl := "https://thehub.io"
searchString := "https://thehub.io/jobs?roles=frontenddeveloper&roles=fullstackdeveloper&roles=backenddeveloper&roles=devops&paid=true&countryCode=DK&sorting=newJobs"
// Instantiate default collector
c := colly.NewCollector(
// visit only the hub
@ -115,9 +84,7 @@ func scrapeHub() {
detailsCollector := c.Clone()
// On every <div> element with class "card__content attribute call callback
c.OnHTML("div[class=card__content]", func(e *colly.HTMLElement) {
if jobCount >= maxJobs {
return
}
//ensure only scrape the amount of jobs specified
// Get the title and ensure it doesn't contain any excluded words
title := e.ChildText("span.card-job-find-list__position")
@ -145,14 +112,14 @@ func scrapeHub() {
cutRight := ");"
trimmedLogo := strings.Trim(logo, cutLeft+cutRight)
// Get the HTML of the description and check to see if it's paid
// Get job description
// Get the HTML of the description
descriptionHTML, err := e.DOM.Find("content.text-block__content > span").Html()
if err != nil {
log.Printf("Error getting HTML of description: %s", err)
return
}
// fill in the job struct
// Get company name
jobDetails := job{
Title: e.ChildText("h2[class=view-job-details__title]"),
Logo: trimmedLogo,
@ -162,15 +129,13 @@ func scrapeHub() {
Description: descriptionHTML,
Link: e.Request.URL.String(),
Skills: skillChecker(e.ChildText("content.text-block__content > span")),
FirstSeen: time.Now(),
Source: baseUrl,
}
jobs = append(jobs, jobDetails)
jobCount++
fmt.Println("Scraped job", jobCount)
if jobCount >= maxJobs {
jobsToJson(file, jobs, fName)
return
if jobCount == maxJobs {
jobsToJson(file)
os.Exit(0)
}
})
// Handle pagination
@ -178,6 +143,7 @@ func scrapeHub() {
nextPage := e.Attr("href")
if nextPage != "" {
fullNextPage := baseUrl + nextPage
fmt.Println("Visiting next page:", fullNextPage)
e.Request.Visit(fullNextPage)
}
})
@ -186,103 +152,6 @@ func scrapeHub() {
}
func scrapeItJobBank() {
// declare and initialize variables
var (
jobs []job
jobCount int
fName = "it-jobbank.json"
maxJobs = 30
baseUrl = "https://www.it-jobbank.dk"
searchString = "https://www.it-jobbank.dk/jobsoegning"
file, err = os.Create(fName)
)
if err != nil {
log.Fatalf("Cannot create file %q: %s", fName, err)
}
defer file.Close()
// Instantiate default collector
c := colly.NewCollector(
// visit only the hub
colly.AllowedDomains("www.it-jobbank.dk", "it-jobbank.dk"),
// Cache responses to prevent multiple requests
colly.CacheDir("./itjobbank_cache"),
)
// Instantiate a new collector to visit the job details page
detailsCollector := c.Clone()
// On every <div> element with class "card__content attribute call callback
c.OnHTML("div[class=result]", func(e *colly.HTMLElement) {
if jobCount >= maxJobs {
return
}
// Get the title and ensure it doesn't contain any excluded words
title := e.ChildText("h3.job-title > a")
for _, excludedWord := range excluded {
if strings.Contains(strings.ToLower(title), excludedWord) {
return
}
}
fullLink := e.ChildAttr("h3.job-title > a", "href")
detailsCollector.Visit(fullLink)
})
detailsCollector.OnRequest(func(r *colly.Request) {
fmt.Println("Visiting", r.URL.String())
})
detailsCollector.OnHTML("section > div", func(e *colly.HTMLElement) {
// get the description as html
descriptionHTML, err := e.DOM.Find("div[id=job_ad]").Html()
if err != nil {
log.Printf("Error getting HTML of description: %s", err)
return
}
// Check if the job is paid
checkIfPaid(descriptionHTML)
// fill in the job struct
title := e.ChildText("h1.title")
if title == "" {
title = e.ChildText("h1[id=jobtitle]")
}
jobDetails := job{
Title: title,
Logo: baseUrl + e.ChildAttr("div.companmony-logo > img", "src"),
Company: e.ChildText("p.published"),
Location: e.ChildText("div.job-location > p.caption"),
Type: checkIfStudent(descriptionHTML),
Description: descriptionHTML,
Link: e.Request.URL.String(),
Skills: skillChecker(descriptionHTML),
FirstSeen: time.Now(),
Source: baseUrl,
}
jobs = append(jobs, jobDetails)
jobCount++
fmt.Println("Scraped job", jobCount)
if jobCount >= maxJobs {
jobsToJson(file, jobs, fName)
return
}
})
// Handle pagination
c.OnHTML("a.page-link", func(e *colly.HTMLElement) {
if jobCount >= maxJobs {
return
}
nextPage := e.Attr("href")
if nextPage != "" {
e.Request.Visit(nextPage)
}
})
c.Visit(searchString)
}
func main() {
scrapeHub()
scrapeItJobBank()
scrapeJobs()
}