Merge pull request #24 from austin1237/scraper-keyword

scraper now passes matched keyword to the job api

go.work.sum

@@ -1,4 +1,6 @@
 github.com/stretchr/testify v1.8.4/go.mod h1:sz/lmYIOXD/1dqDmKjjqLyZ2RngseejIcXlSw2iwfAo=
+golang.org/x/crypto v0.14.0/go.mod h1:MVFd36DqK4CsrnJYDkBA3VC4m2GkXAM0PvzMCn4JQf4=
 golang.org/x/sys v0.13.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/term v0.13.0/go.mod h1:LTmsnFJwVN6bCy1rVCoS+qHT1HhALEFxKncY3WNNh4U=
 golang.org/x/text v0.13.0 h1:ablQoSUd0tRdKxZewP80B+BaqeKJuVhuRxj/dkrun3k=
 golang.org/x/text v0.13.0/go.mod h1:TvPlkZtksWOMsz7fbANvkp4WM8x/WCo/om8BMLbz+aE=

scraper/interest/interest.go

@@ -8,18 +8,18 @@ import (
 	"sync"
 )
 
-func CheckIfInterested(description string) bool {
-	keywords := []string{"typescript", "node", "nodejs", "node.js", "go", "golang"}
+func CheckIfInterested(description string) string {
+	keywords := []string{"go", "golang", "node", "nodejs", "node.js", "typescript"}
 	// Check if keywords are present in the job's text
 	descriptionToLower := strings.ToLower(description)
 	for _, keyword := range keywords {
 		pattern := "\\b" + keyword + "\\b"
 		match, _ := regexp.MatchString(pattern, descriptionToLower)
 		if match {
-			return true
+			return keyword
 		}
 	}
-	return false
+	return ""
 }
 
 type JobInfoGetter func(string, string) (string, error)
@@ -43,7 +43,9 @@ func FilterInterest(proxyUrl string, possibleJobs []job.Job, jobInfoGetter JobIn
 			if err != nil {
 				fmt.Println(err)
 			}
-			if CheckIfInterested(description) {
+
+			possibleJob.Keyword = CheckIfInterested(description)
+			if possibleJob.Keyword != "" {
 				interestingJobs = append(interestingJobs, possibleJob)
 			}
 			goroutineCount--

scraper/interest/interest_test.go

@@ -10,37 +10,37 @@ func TestCheckIfInterested(t *testing.T) {
 	tests := []struct {
 		name        string
 		description string
-		want        bool
+		want        string
 	}{
 		{
 			name:        "contains typescript",
 			description: "This is a job for a TypeScript developer",
-			want:        true,
+			want:        "typescript",
 		},
 		{
 			name:        "contains node",
 			description: "This is a job for a Node.js developer",
-			want:        true,
+			want:        "node",
 		},
 		{
 			name:        "contains go",
 			description: "This is a job for a Go developer",
-			want:        true,
+			want:        "go",
 		},
 		{
 			name:        "contains go",
 			description: "This is a job for go.",
-			want:        true,
+			want:        "go",
 		},
 		{
 			name:        "contains django",
 			description: "This is a job for a django developer in Chicago",
-			want:        false,
+			want:        "",
 		},
 		{
 			name:        "does not contain keyword",
 			description: "This is a job for a Python developer",
-			want:        false,
+			want:        "",
 		},
 	}
 

scraper/job/job.go

@@ -15,6 +15,7 @@ type Job struct {
 	Title   string
 	Company string
 	Link    string
+	Keyword string
 }
 
 type Response struct {

scraper/remotive/remotive.go

@@ -76,11 +76,13 @@ func ScanNewJobs(baseURL string, proxyUrl string) []job.Job {
 	log.Println("Remotive total jobs found", len(remotiveJobs))
 	var interestingJobs []job.Job
 	for _, newJob := range remotiveJobs {
-		if interest.CheckIfInterested(newJob.Description) {
+		keyword := interest.CheckIfInterested(newJob.Description)
+		if keyword != "" {
 			interestingJobs = append(interestingJobs, job.Job{
 				Title:   newJob.Title,
 				Link:    newJob.URL,
 				Company: newJob.CompanyName,
+				Keyword: keyword,
 			})
 		}
 	}