run crawlers with the webapp

src/cmd/crawler/collectors/collector.go

@@ -1,7 +0,0 @@
-package collectors
-
-import "crowsnest/internal/model/database"
-
-type Collector struct {
-	Articles *database.ArticleModel
-}

src/cmd/crawler/main.go

@@ -1,43 +0,0 @@
-package main
-
-import (
-	"crowsnest/cmd/crawler/collectors"
-	"crowsnest/internal/model/database"
-	"database/sql"
-	"log"
-	"os"
-	"sync"
-
-	_ "github.com/lib/pq"
-)
-
-func main() {
-	// collect environement variables
-	databaseURL := os.Getenv("DB_URL")
-
-	// connect to database
-	db, err := sql.Open("postgres", databaseURL)
-	if err != nil {
-		log.Fatal(err)
-	}
-	defer db.Close()
-
-	// collect websites
-	coll := collectors.Collector{
-		Articles: &database.ArticleModel{DB: db},
-	}
-
-	var wg sync.WaitGroup
-	wg.Add(2)
-
-	go func() {
-		defer wg.Done()
-		coll.CollectSpiegel()
-	}()
-	go func() {
-		defer wg.Done()
-		go coll.CollectZeit()
-	}()
-
-	wg.Wait()
-}

src/cmd/frontend/main.go

@@ -2,10 +2,12 @@ package main
 
 import (
 	"crowsnest/internal/app"
+	"crowsnest/internal/crawler"
 	"crowsnest/internal/middleware"
 	"crowsnest/internal/model/database"
 	"log"
 	"net/http"
+	"time"
 
 	_ "github.com/lib/pq"
 )
@@ -16,6 +18,18 @@ func main() {
 		log.Fatal("failed to connect to database due to", err.Error())
 	}
 
+	// run web crawlers
+	coll := crawler.Crawler{
+		Articles: &database.ArticleModel{DB: db},
+	}
+	go func() {
+		for {
+			coll.ZeitCollectIndex()
+			coll.SpiegelCollectIndex()
+			time.Sleep(5 * time.Minute)
+		}
+	}()
+
 	// define app
 	webapp := app.NewApp(db)
 

src/internal/crawler/crawler.go

@@ -0,0 +1,48 @@
+package crawler
+
+import (
+	"crowsnest/internal/model/database"
+	"fmt"
+	"net/http"
+
+	"github.com/PuerkitoBio/goquery"
+)
+
+type Crawler struct {
+	Articles *database.ArticleModel
+}
+
+// GetAllURLs fetches all URLs from a given web page URL
+func (c *Crawler) GetAllURLs(pageURL string) ([]string, error) {
+	// Send a GET request to the provided URL
+	resp, err := http.Get(pageURL)
+	if err != nil {
+		return nil, fmt.Errorf("failed to fetch URL %s: %w", pageURL, err)
+	}
+	defer resp.Body.Close()
+
+	// Check if the response status is OK
+	if resp.StatusCode != http.StatusOK {
+		return nil, fmt.Errorf("HTTP request failed with status code %d", resp.StatusCode)
+	}
+
+	// Parse the HTML document
+	doc, err := goquery.NewDocumentFromReader(resp.Body)
+	if err != nil {
+		return nil, fmt.Errorf("failed to parse HTML: %w", err)
+	}
+
+	// Slice to store the extracted URLs
+	var urls []string
+
+	// Select all anchor tags and extract the href attribute
+	doc.Find("a").Each(func(index int, element *goquery.Selection) {
+		// Get the href attribute
+		href, exists := element.Attr("href")
+		if exists {
+			urls = append(urls, href)
+		}
+	})
+
+	return urls, nil
+}

src/internal/crawler/spiegel.go

@@ -1,10 +1,9 @@
-package collectors
+package crawler
 
 import (
 	"crowsnest/internal/model"
 	"errors"
 	"fmt"
-	"log"
 	"regexp"
 	"strings"
 	"time"
@@ -13,34 +12,48 @@ import (
 	"github.com/gocolly/colly/v2"
 )
 
-func (c *Collector) CollectSpiegel() {
-	collycollector := colly.NewCollector(
+func (c *Crawler) SpiegelCollector() *colly.Collector {
+	// set cache, domain pattern and max recursion deepth
+	collector := colly.NewCollector(
 		colly.AllowedDomains("www.spiegel.de", "spiegel.de"),
 		colly.CacheDir("./persistence/spiegel_cache"),
-		colly.MaxDepth(3),
+		colly.MaxDepth(5),
 	)
-
 	// store articles
-	collycollector.OnResponse(func(r *colly.Response) {
+	collector.OnResponse(func(r *colly.Response) {
 		url := r.Request.URL.String()
-		err := c.ExtractSpiegel(url, r.Body)
-		if err == nil {
-			log.Println("added article", url)
-		} else {
-			log.Println("failed to add article:", err, "("+url+")")
-		}
+		c.SpiegelExtract(url, r.Body)
 	})
-
 	// cascade
-	collycollector.OnHTML("a[href]", func(e *colly.HTMLElement) {
+	collector.OnHTML("a[href]", func(e *colly.HTMLElement) {
 		url := e.Attr("href")
-
 		if !strings.HasPrefix(url, "http") {
 			return
 		}
 		e.Request.Visit(url)
 	})
 
+	return collector
+}
+
+func (c *Crawler) SpiegelCollectIndex() error {
+	urls, err := c.GetAllURLs("https://www.spiegel.de/")
+	if err != nil {
+		return err
+	}
+
+	collector := c.SpiegelCollector()
+	collector.MaxDepth = 1
+
+	for _, url := range urls {
+		collector.Visit(url)
+	}
+	return nil
+}
+
+func (c *Crawler) SpiegelCollectArchive() {
+	collector := c.SpiegelCollector()
+
 	// go through archive
 	startDate := time.Date(2020, time.January, 1, 0, 0, 0, 0, time.UTC)
 	currentDate := time.Now()
@@ -49,11 +62,11 @@ func (c *Collector) CollectSpiegel() {
 		urlDate := date.Format("02.01.2006")
 		url := fmt.Sprintf("https://www.spiegel.de/nachrichtenarchiv/artikel-%s.html", urlDate)
 
-		collycollector.Visit(url)
+		collector.Visit(url)
 	}
 }
 
-func (c *Collector) ExtractSpiegel(url string, body []byte) error {
+func (c *Crawler) SpiegelExtract(url string, body []byte) error {
 	paywall_pattern := regexp.MustCompile(`"paywall":{"attributes":{"is_active":true`)
 	url_pattern := regexp.MustCompile(`^https://(www\.)?spiegel.de.*`)
 	whitespace := regexp.MustCompile(`\s+`)

src/internal/crawler/zeit.go

@@ -1,10 +1,9 @@
-package collectors
+package crawler
 
 import (
 	"crowsnest/internal/model"
 	"errors"
 	"fmt"
-	"log"
 	"regexp"
 	"strings"
 	"time"
@@ -13,36 +12,50 @@ import (
 	"github.com/gocolly/colly/v2"
 )
 
-// Gets every page of the archive of zeit.de and stores the responses into the
-// database.
-func (c *Collector) CollectZeit() {
-	collycollector := colly.NewCollector(
+func (c *Crawler) ZeitCollector() *colly.Collector {
+	// set cache, domain pattern and max recursion deepth
+	collector := colly.NewCollector(
 		colly.AllowedDomains("www.zeit.de", "zeit.de"),
 		colly.CacheDir("./persistence/zeit_cache"),
-		colly.MaxDepth(3),
+		colly.MaxDepth(5),
 	)
-
 	// store articles
-	collycollector.OnResponse(func(r *colly.Response) {
+	collector.OnResponse(func(r *colly.Response) {
 		url := r.Request.URL.String()
-		err := c.ExtractZeit(url, r.Body)
-		if err == nil {
-			log.Println("added article", url)
-		} else {
-			log.Println("failed to add article:", err, "("+url+")")
-		}
+		c.ZeitExtract(url, r.Body)
 	})
-
 	// cascade
-	collycollector.OnHTML("a[href]", func(e *colly.HTMLElement) {
+	collector.OnHTML("a[href]", func(e *colly.HTMLElement) {
 		url := e.Attr("href")
-
 		if !strings.HasPrefix(url, "http") {
 			return
 		}
 		e.Request.Visit(url)
 	})
 
+	return collector
+}
+
+func (c *Crawler) ZeitCollectIndex() error {
+	urls, err := c.GetAllURLs("https://www.zeit.de/index")
+	if err != nil {
+		return err
+	}
+
+	collector := c.ZeitCollector()
+	collector.MaxDepth = 1
+
+	for _, url := range urls {
+		collector.Visit(url)
+	}
+	return nil
+}
+
+// Gets every page of the archive of zeit.de and stores the responses into the
+// database.
+func (c *Crawler) ZeitCollectArchive() {
+	collector := c.ZeitCollector()
+
 	// go through archive
 	startDate := time.Date(2020, time.January, 1, 0, 0, 0, 0, time.UTC)
 	//startDate := time.Date(1946, time.January, 1, 0, 0, 0, 0, time.UTC)
@@ -52,12 +65,12 @@ func (c *Collector) CollectZeit() {
 		year, week := date.ISOWeek()
 		url := fmt.Sprintf("https://www.zeit.de/%04d/%02d/index", year, week)
 
-		collycollector.Visit(url)
+		collector.Visit(url)
 	}
 }
 
-func (c *Collector) ExtractZeit(url string, body []byte) error {
-	url_pattern := regexp.MustCompile(`^https://(www\.)?zeit.de.*`)
+func (c *Crawler) ZeitExtract(url string, body []byte) error {
+	url_pattern := regexp.MustCompile(`^https://(www\.)?zeit\.de[^#]*$`)
 	whitespace := regexp.MustCompile(`\s+`)
 
 	var exists bool