refactor scraper into modules

2 files changed, 156 insertions, 0 deletions

diff --git a/modules/urlfilter/urlfilter.go b/modules/urlfilter/urlfilter.go
new file mode 100644
index 0000000..14576f0
--- /dev/null
+++ b/modules/urlfilter/urlfilter.go
@@ -0,0 +1,85 @@
+// This Source Code Form is subject to the terms of the Mozilla Public
+// License, v. 2.0. If a copy of the MPL was not distributed with this
+// file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+package urlfilter
+
+import (
+	"regexp"
+
+	"github.com/philippta/flyscrape"
+)
+
+func init() {
+	flyscrape.RegisterModule(new(Module))
+}
+
+type Module struct {
+	URL         string   `json:"url"`
+	AllowedURLs []string `json:"allowedURLs"`
+	BlockedURLs []string `json:"blockedURLs"`
+
+	allowedURLsRE []*regexp.Regexp
+	blockedURLsRE []*regexp.Regexp
+}
+
+func (m *Module) ID() string {
+	return "urlfilter"
+}
+
+func (m *Module) OnLoad(v flyscrape.Visitor) {
+	for _, pat := range m.AllowedURLs {
+		re, err := regexp.Compile(pat)
+		if err != nil {
+			continue
+		}
+		m.allowedURLsRE = append(m.allowedURLsRE, re)
+	}
+
+	for _, pat := range m.BlockedURLs {
+		re, err := regexp.Compile(pat)
+		if err != nil {
+			continue
+		}
+		m.blockedURLsRE = append(m.blockedURLsRE, re)
+	}
+}
+
+func (m *Module) CanRequest(rawurl string, depth int) bool {
+	// allow root url
+	if rawurl == m.URL {
+		return true
+	}
+
+	// allow if no filter is set
+	if len(m.allowedURLsRE) == 0 && len(m.blockedURLsRE) == 0 {
+		return true
+	}
+
+	ok := false
+	if len(m.allowedURLsRE) == 0 {
+		ok = true
+	}
+
+	for _, re := range m.allowedURLsRE {
+		if re.MatchString(rawurl) {
+			ok = true
+			break
+		}
+	}
+
+	for _, re := range m.blockedURLsRE {
+		if re.MatchString(rawurl) {
+			ok = false
+			break
+		}
+	}
+
+	return ok
+}
+
+var (
+	_ flyscrape.Module     = (*Module)(nil)
+	_ flyscrape.CanRequest = (*Module)(nil)
+	_ flyscrape.OnLoad     = (*Module)(nil)
+)
diff --git a/modules/urlfilter/urlfilter_test.go b/modules/urlfilter/urlfilter_test.go
new file mode 100644
index 0000000..e383a32
--- /dev/null
+++ b/modules/urlfilter/urlfilter_test.go
@@ -0,0 +1,71 @@
+// This Source Code Form is subject to the terms of the Mozilla Public
+// License, v. 2.0. If a copy of the MPL was not distributed with this
+// file, You can obtain one at http://mozilla.org/MPL/2.0/.
+
+package urlfilter_test
+
+import (
+	"testing"
+
+	"github.com/philippta/flyscrape"
+	"github.com/philippta/flyscrape/modules/followlinks"
+	"github.com/philippta/flyscrape/modules/starturl"
+	"github.com/philippta/flyscrape/modules/urlfilter"
+	"github.com/stretchr/testify/require"
+)
+
+func TestURLFilterAllowed(t *testing.T) {
+	scraper := flyscrape.NewScraper()
+	scraper.LoadModule(&starturl.Module{URL: "http://www.example.com/"})
+	scraper.LoadModule(&followlinks.Module{})
+	scraper.LoadModule(&urlfilter.Module{
+		URL:         "http://www.example.com/",
+		AllowedURLs: []string{`/foo\?id=\d+`, `/bar$`},
+	})
+
+	scraper.SetTransport(flyscrape.MockTransport(200, `
+        <a href="foo?id=123">123</a>
+        <a href="foo?id=ABC">ABC</a>
+        <a href="/bar">bar</a>
+        <a href="/barz">barz</a>`))
+
+	var urls []string
+	scraper.OnRequest(func(req *flyscrape.Request) {
+		urls = append(urls, req.URL)
+	})
+
+	scraper.Run()
+
+	require.Len(t, urls, 3)
+	require.Contains(t, urls, "http://www.example.com/")
+	require.Contains(t, urls, "http://www.example.com/foo?id=123")
+	require.Contains(t, urls, "http://www.example.com/bar")
+}
+
+func TestURLFilterBlocked(t *testing.T) {
+	scraper := flyscrape.NewScraper()
+	scraper.LoadModule(&starturl.Module{URL: "http://www.example.com/"})
+	scraper.LoadModule(&followlinks.Module{})
+	scraper.LoadModule(&urlfilter.Module{
+		URL:         "http://www.example.com/",
+		BlockedURLs: []string{`/foo\?id=\d+`, `/bar$`},
+	})
+
+	scraper.SetTransport(flyscrape.MockTransport(200, `
+        <a href="foo?id=123">123</a>
+        <a href="foo?id=ABC">ABC</a>
+        <a href="/bar">bar</a>
+        <a href="/barz">barz</a>`))
+
+	var urls []string
+	scraper.OnRequest(func(req *flyscrape.Request) {
+		urls = append(urls, req.URL)
+	})
+
+	scraper.Run()
+
+	require.Len(t, urls, 3)
+	require.Contains(t, urls, "http://www.example.com/")
+	require.Contains(t, urls, "http://www.example.com/foo?id=ABC")
+	require.Contains(t, urls, "http://www.example.com/barz")
+}