Fix Plainify edge cases

This commit replaces the main part of `helpers.StripHTML` with Go's implementation in its html/template package.

It's a little slower, but correctness is more important:

```bash
BenchmarkStripHTMLOld-10    	  680316	      1764 ns/op	     728 B/op	       4 allocs/op
BenchmarkStripHTMLNew-10    	  384520	      3099 ns/op	    2089 B/op	      10 allocs/op
```

Fixes #9199
Fixes #9909
Closes #9410
This commit is contained in:
Bjørn Erik Pedersen
2022-05-25 10:56:14 +02:00
parent cd0112a05a
commit 3854a6fa6c
10 changed files with 103 additions and 85 deletions

View File

@@ -26,6 +26,7 @@ import (
"github.com/gohugoio/hugo/htesting"
"github.com/gohugoio/hugo/markup/asciidocext"
"github.com/gohugoio/hugo/markup/rst"
"github.com/gohugoio/hugo/tpl"
"github.com/gohugoio/hugo/config"
@@ -40,7 +41,6 @@ import (
qt "github.com/frankban/quicktest"
"github.com/gohugoio/hugo/deps"
"github.com/gohugoio/hugo/helpers"
)
const (
@@ -351,7 +351,7 @@ func normalizeExpected(ext, str string) string {
default:
return str
case "html":
return strings.Trim(helpers.StripHTML(str), " ")
return strings.Trim(tpl.StripHTML(str), " ")
case "ad":
paragraphs := strings.Split(str, "</p>")
expected := ""
@@ -1736,6 +1736,7 @@ Len Summary: {{ len .Summary }}
Len Content: {{ len .Content }}
SUMMARY:{{ .Summary }}:{{ len .Summary }}:END
`}
b := newTestSitesBuilder(t)