brokenlinks: store the anchor or image source in link

In the struct Link, we add field Value that store the href from A element
or src from IMG element.
This allow us to debug any error during scan, especially joining path
and link.

1 files changed, 7 insertions, 5 deletions

diff --git a/brokenlinks/worker.go b/brokenlinks/worker.go
index 387cc05..79ae796 100644
--- a/brokenlinks/worker.go
+++ b/brokenlinks/worker.go
@@ -89,8 +89,10 @@ func newWorker(opts Options) (wrk *worker, err error) {
 	}
 
 	wrk.baseUrl = &url.URL{
-		Scheme: wrk.opts.scanUrl.Scheme,
-		Host:   wrk.opts.scanUrl.Host,
+		Scheme:  wrk.opts.scanUrl.Scheme,
+		Host:    wrk.opts.scanUrl.Host,
+		Path:    `/`,
+		RawPath: `/`,
 	}
 
 	if opts.PastResultFile == "" {
@@ -304,8 +306,6 @@ func (wrk *worker) scan(linkq jarink.Link) (resultq map[string]jarink.Link) {
 		return resultq
 	}
 
-	var parentUrl *url.URL
-
 	// Check and get the redirect location or use the original URL.
 	location := httpResp.Header.Get(`Location`)
 	if location == `` {
@@ -314,6 +314,7 @@ func (wrk *worker) scan(linkq jarink.Link) (resultq map[string]jarink.Link) {
 		linkq.Url = location
 	}
 
+	var parentUrl *url.URL
 	parentUrl, err = url.Parse(location)
 	if err != nil {
 		linkq.StatusCode = StatusBadLink
@@ -415,6 +416,7 @@ func (wrk *worker) processLink(parentUrl *url.URL, val string, kind int) (
 
 	linkq = &jarink.Link{
 		ParentUrl: parentUrl,
+		Value:     val,
 		Kind:      kind,
 	}
 
@@ -452,7 +454,7 @@ func (wrk *worker) processLink(parentUrl *url.URL, val string, kind int) (
 		// "parent/page.ext" + "val" => "parent/val"
 		tmp, _ := url.Parse(parentUrl.String())
 		tmp.Path = path.Dir(tmp.Path)
-		linkq.Url = parentUrl.JoinPath(tmp.Path).String()
+		linkq.Url = tmp.JoinPath(val).String()
 	}
 	return linkq
 }