tpl/tplimpl: Optionally exclude content from sitemap

Define global inclusion/exclusion in site configuration, and override
via front matter. For example, to exclude a page from the sitemap:

    [sitemap]
    disable = true # default is false

Closes #653
Closes #12282

Co-authored-by: kolappannathan <kolappannathan@users.noreply.github.com>
Co-authored-by: felicianotech <FelicianoTech@gmail.com>

config/commonConfig.go

@@ -215,6 +215,8 @@ type SitemapConfig struct {
 	Priority float64
 	// The sitemap filename.
 	Filename string
+	// Whether to disable page inclusion.
+	Disable bool
 }
 
 func DecodeSitemap(prototype SitemapConfig, input map[string]any) (SitemapConfig, error) {

docs/data/docs.yaml

@@ -1649,6 +1649,7 @@ config:
       disableInlineCSS: false
   sitemap:
     changeFreq: ""
+    disable: false
     filename: sitemap.xml
     priority: -1
   social: null
@@ -2797,8 +2798,8 @@ tpl:
             {{ $m.Set "Hugo" "Rocks!" }}
             {{ $m.Values | debug.Dump | safeHTML }}
           - |-
-            map[string]interface {}{
-              "Hugo": "Rocks!",
+            {
+              "Hugo": "Rocks!"
             }
       TestDeprecationErr:
         Aliases: null

hugolib/config_test.go

@@ -490,11 +490,10 @@ name = "menu-theme"
 			got := b.Configs.Base
 
 			if mergeStrategy == "none" {
-				b.Assert(got.Sitemap, qt.DeepEquals, config.SitemapConfig{ChangeFreq: "", Priority: -1, Filename: "sitemap.xml"})
-
+				b.Assert(got.Sitemap, qt.DeepEquals, config.SitemapConfig{ChangeFreq: "", Disable: false, Priority: -1, Filename: "sitemap.xml"})
 				b.AssertFileContent("public/sitemap.xml", "schemas/sitemap")
 			} else {
-				b.Assert(got.Sitemap, qt.DeepEquals, config.SitemapConfig{ChangeFreq: "monthly", Priority: -1, Filename: "sitemap.xml"})
+				b.Assert(got.Sitemap, qt.DeepEquals, config.SitemapConfig{ChangeFreq: "monthly", Disable: false, Priority: -1, Filename: "sitemap.xml"})
 				b.AssertFileContent("public/sitemap.xml", "<changefreq>monthly</changefreq>")
 			}
 		})

hugolib/sitemap_test.go

@@ -108,11 +108,12 @@ outputs: [ "html", "amp" ]
 
 func TestParseSitemap(t *testing.T) {
 	t.Parallel()
-	expected := config.SitemapConfig{Priority: 3.0, Filename: "doo.xml", ChangeFreq: "3"}
+	expected := config.SitemapConfig{ChangeFreq: "3", Disable: true, Filename: "doo.xml", Priority: 3.0}
 	input := map[string]any{
 		"changefreq": "3",
-		"priority":   3.0,
+		"disable":    true,
 		"filename":   "doo.xml",
+		"priority":   3.0,
 		"unknown":    "ignore",
 	}
 	result, err := config.DecodeSitemap(config.SitemapConfig{}, input)

tpl/tplimpl/embedded/templates/_default/sitemap.xml

@@ -1,7 +1,7 @@
 {{ printf "<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"yes\"?>" | safeHTML }}
 <urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
   xmlns:xhtml="http://www.w3.org/1999/xhtml">
-  {{ range .Data.Pages }}
+  {{ range where .Pages "Sitemap.Disable" "ne" true }}
     {{- if .Permalink -}}
   <url>
     <loc>{{ .Permalink }}</loc>{{ if not .Lastmod.IsZero }}

tpl/tplimpl/tplimpl_integration_test.go

@@ -254,3 +254,54 @@ disable = false
 		`s.src = '//' + "foo" + '.disqus.com/embed.js';`,
 	)
 }
+
+func TestSitemap(t *testing.T) {
+	t.Parallel()
+
+	files := `
+-- hugo.toml --
+disableKinds = ['home','rss','section','taxonomy','term']
+[sitemap]
+disable = true
+-- content/p1.md --
+---
+title: p1
+sitemap:
+  p1_disable: foo
+---
+-- content/p2.md --
+---
+title: p2
+
+---
+-- layouts/_default/single.html --
+{{ .Title }}
+`
+
+	// Test A: Exclude all pages via site config.
+	b := hugolib.Test(t, files)
+	b.AssertFileContentExact("public/sitemap.xml",
+		"<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"yes\"?>\n<urlset xmlns=\"http://www.sitemaps.org/schemas/sitemap/0.9\"\n  xmlns:xhtml=\"http://www.w3.org/1999/xhtml\">\n  \n</urlset>\n",
+	)
+
+	// Test B: Include all pages via site config.
+	files_b := strings.ReplaceAll(files, "disable = true", "disable = false")
+	b = hugolib.Test(t, files_b)
+	b.AssertFileContentExact("public/sitemap.xml",
+		"<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"yes\"?>\n<urlset xmlns=\"http://www.sitemaps.org/schemas/sitemap/0.9\"\n  xmlns:xhtml=\"http://www.w3.org/1999/xhtml\">\n  <url>\n    <loc>/p1/</loc>\n  </url><url>\n    <loc>/p2/</loc>\n  </url>\n</urlset>\n",
+	)
+
+	// Test C: Exclude all pages via site config, but include p1 via front matter.
+	files_c := strings.ReplaceAll(files, "p1_disable: foo", "disable: false")
+	b = hugolib.Test(t, files_c)
+	b.AssertFileContentExact("public/sitemap.xml",
+		"<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"yes\"?>\n<urlset xmlns=\"http://www.sitemaps.org/schemas/sitemap/0.9\"\n  xmlns:xhtml=\"http://www.w3.org/1999/xhtml\">\n  <url>\n    <loc>/p1/</loc>\n  </url>\n</urlset>\n",
+	)
+
+	// Test D:  Include all pages via site config, but exclude p1 via front matter.
+	files_d := strings.ReplaceAll(files_b, "p1_disable: foo", "disable: true")
+	b = hugolib.Test(t, files_d)
+	b.AssertFileContentExact("public/sitemap.xml",
+		"<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"yes\"?>\n<urlset xmlns=\"http://www.sitemaps.org/schemas/sitemap/0.9\"\n  xmlns:xhtml=\"http://www.w3.org/1999/xhtml\">\n  <url>\n    <loc>/p2/</loc>\n  </url>\n</urlset>\n",
+	)
+}