fix atom links

src/parser/atom.go

@@ -81,9 +81,16 @@ func ParseAtom(r io.Reader) (*Feed, error) {
 		SiteURL: firstNonEmpty(srcfeed.Links.First("alternate"), srcfeed.Links.First("")),
 	}
 	for _, srcitem := range srcfeed.Entries {
-		link := firstNonEmpty(srcitem.OrigLink, srcitem.Links.First("alternate"), srcitem.Links.First(""))
+		linkFromID := ""
+		guidFromID := ""
+		if htmlutil.IsAPossibleLink(srcitem.ID) {
+			linkFromID = srcitem.ID
+			guidFromID = srcitem.ID + "::" + srcitem.Updated
+		}
+
+		link := firstNonEmpty(srcitem.OrigLink, srcitem.Links.First("alternate"), srcitem.Links.First(""), linkFromID)
 		dstfeed.Items = append(dstfeed.Items, Item{
-			GUID:     firstNonEmpty(srcitem.ID, link),
+			GUID:     firstNonEmpty(guidFromID, srcitem.ID, link),
 			Date:     dateParse(firstNonEmpty(srcitem.Published, srcitem.Updated)),
 			URL:      link,
 			Title:    srcitem.Title.Text(),

src/parser/atom_test.go

@@ -131,3 +131,48 @@ func TestAtomImageLinkDuplicated(t *testing.T) {
 		t.Fatal("item.image_url must be unset if present in the content")
 	}
 }
+
+func TestAtomLinkInID(t *testing.T) {
+	feed, _ := Parse(strings.NewReader(`
+		<?xml version="1.0" encoding="utf-8"?>
+		<feed xmlns="http://www.w3.org/2005/Atom" xmlns:media="http://search.yahoo.com/mrss/">
+			<entry>
+                <title>one updated</title>
+                <id>https://example.com/posts/1</id>
+                <updated>2003-12-13T09:17:51</updated>
+			</entry>
+			<entry>
+                <title>two</title>
+                <id>urn:uuid:60a76c80-d399-11d9-b93C-0003939e0af6</id>
+			</entry>
+			<entry>
+                <title>one</title>
+                <id>https://example.com/posts/1</id>
+			</entry>
+		</feed>
+	`))
+	have := feed.Items
+	want := []Item{
+		Item{
+			GUID:     "https://example.com/posts/1::2003-12-13T09:17:51",
+			Date:     time.Date(2003, time.December, 13, 9, 17, 51, 0, time.UTC),
+			URL:      "https://example.com/posts/1",
+			Title:    "one updated",
+        },
+		Item{
+			GUID: "urn:uuid:60a76c80-d399-11d9-b93C-0003939e0af6",
+            Date: time.Date(1, time.January, 1, 0, 0, 0, 0, time.UTC), URL: "",
+            Title: "two",
+        },
+        Item{
+            GUID: "https://example.com/posts/1::",
+            Date: time.Date(1, time.January, 1, 0, 0, 0, 0, time.UTC),
+            URL: "https://example.com/posts/1",
+            Title: "one",
+            Content: "",
+        },
+	}
+	if !reflect.DeepEqual(want, have) {
+		t.Fatalf("\nwant: %#v\nhave: %#v\n", want, have)
+	}
+}

src/parser/rss.go

@@ -20,7 +20,7 @@ type rssFeed struct {
 }
 
 type rssItem struct {
-	GUID        rssGuid         `xml:"guid"`
+	GUID        rssGuid        `xml:"guid"`
 	Title       string         `xml:"title"`
 	Link        string         `xml:"rss link"`
 	Description string         `xml:"rss description"`
@@ -86,10 +86,10 @@ func ParseRSS(r io.Reader) (*Feed, error) {
 			}
 		}
 
-        permalink := ""
-        if srcitem.GUID.IsPermaLink == "true" {
-            permalink = srcitem.GUID.GUID
-        }
+		permalink := ""
+		if srcitem.GUID.IsPermaLink == "true" {
+			permalink = srcitem.GUID.GUID
+		}
 
 		dstfeed.Items = append(dstfeed.Items, Item{
 			GUID:     firstNonEmpty(srcitem.GUID.GUID, srcitem.Link),

src/parser/rss_test.go

@@ -217,11 +217,11 @@ func TestRSSIsPermalink(t *testing.T) {
 	`))
 	have := feed.Items
 	want := []Item{
-        {
-            GUID:    "http://example.com/posts/1",
-            URL:     "http://example.com/posts/1",
-        },
-    }
+		{
+			GUID: "http://example.com/posts/1",
+			URL:  "http://example.com/posts/1",
+		},
+	}
 	for i := 0; i < len(want); i++ {
 		if want[i] != have[i] {
 			t.Errorf("Failed to handle isPermalink\nwant: %#v\nhave: %#v\n", want[i], have[i])