Add helper to get labels from arXiv terms

internal/utils/helpers.go

@@ -5,6 +5,7 @@ import (
 	"io/ioutil"
 	"net/http"
 	"regexp"
+	"unicode/utf8"
 )
 
 func FetchEmails(url string) ([]string, error) {
@@ -30,3 +31,18 @@ func FetchEmails(url string) ([]string, error) {
 
 	return emails, nil
 }
+
+func TrimToMaxLen(s string, maxLen int) string {
+	// Check if the string length exceeds the maximum length
+	if utf8.RuneCountInString(s) > maxLen {
+		// Convert the string to a slice of runes
+		runes := []rune(s)
+
+		// Truncate the slice to the maximum length
+		runes = runes[:maxLen]
+
+		// Convert the slice of runes back to a string
+		return string(runes)
+	}
+	return s
+}

pkg/arxiv/category.go

@@ -1,6 +1,40 @@
 package arxiv
 
+import (
+	"fmt"
+	"io"
+	"net/http"
+	"regexp"
+)
+
 type Category struct {
 	Term   string `xml:"term,attr"`
 	Scheme string `xml:"scheme,attr"`
 }
+
+func GetCategoryLabels() map[string]string {
+	categories := map[string]string{}
+
+	url := "https://arxiv.org/category_taxonomy"
+	resp, err := http.Get(url)
+	if err != nil {
+		fmt.Println("Error fetching URL: ", err)
+		return categories
+	}
+	defer resp.Body.Close()
+	body, err := io.ReadAll(resp.Body)
+	if err != nil {
+		fmt.Println("Error reading response body: ", err)
+		return categories
+	}
+
+	pattern := `<h4>([a-z\-]+\.[A-Z]+) <span>\(([^)]+)\)</span></h4>`
+	re := regexp.MustCompile(pattern)
+	matches := re.FindAllStringSubmatch(string(body), -1)
+
+	for _, match := range matches {
+		categories[match[1]] = match[2]
+	}
+
+	return categories
+}

pkg/arxiv/entry.go

@@ -20,6 +20,7 @@ type Entry struct {
 	JournalRef      string     `xml:"http://arxiv.org/schemas/atom journal_ref"`
 	PrimaryCategory Category   `xml:"http://arxiv.org/schemas/atom primary_category"`
 	Categories      []Category `xml:"category"`
+	License         string     `xml:"license"`
 	PDF             string
 }