8000 [pull] main from miniflux:main by pull[bot] · Pull Request #342 · suhaoGe/miniflux · GitHub
[go: up one dir, main page]

Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
40 changes: 37 additions & 3 deletions internal/http/response/builder.go
Original file line number Diff line number Diff line change
Expand Up @@ -13,13 +13,38 @@ import (
"mime"
"net/http"
"strings"
"sync"
"time"

"github.com/andybalholm/brotli"
"github.com/andybalholm/brotli/matchfinder"
)

const compressionThreshold = 1024

// Compression writers are pooled so each request reuses their internal
// state (brotli sliding window + hash tables, flate dictionary, etc.)
// instead of allocating it from scratch. Reset(dst) rebinds the
// destination without re-allocating the buffers.
var (
brotliWriterPool = sync.Pool{
New: func() any {
return brotli.NewWriterV2(io.Discard, brotli.DefaultCompression)
},
}
gzipWriterPool = sync.Pool{
New: func() any {
return gzip.NewWriter(io.Discard)
},
}
flateWriterPool = sync.Pool{
New: func() any {
w, _ := flate.NewWriter(io.Discard, flate.DefaultCompression)
return w
},
}
)

// Builder generates HTTP responses.
type Builder struct {
w http.ResponseWriter
Expand Down Expand Up @@ -146,25 +171,34 @@ func (b *Builder) compress(data []byte) {
b.headers.Set("Content-Encoding", "br")
b.writeHeaders()

brotliWriter := brotli.NewWriterV2(b.w, brotli.DefaultCompression)
brotliWriter := brotliWriterPool.Get().(*matchfinder.Writer)
brotliWriter.Reset(b.w)
brotliWriter.Write(data)
brotliWriter.Close()
brotliWriter.Reset(io.Discard)
brotliWriterPool.Put(brotliWriter)
return
case "gzip":
b.headers.Set("Content-Encoding", "gzip")
b.writeHeaders()

gzipWriter := gzip.NewWriter(b.w)
gzipWriter := gzipWriterPool.Get().(*gzip.Writer)
gzipWriter.Reset(b.w)
gzipWriter.Write(data)
gzipWriter.Close()
gzipWriter.Reset(io.Discard)
gzipWriterPool.Put(gzipWriter)
return
case "deflate":
b.headers.Set("Content-Encoding", "deflate")
b.writeHeaders()

flateWriter, _ := flate.NewWriter(b.w, -1)
flateWriter := flateWriterPool.Get().(*flate.Writer)
flateWriter.Reset(b.w)
flateWriter.Write(data)
flateWriter.Close()
flateWriter.Reset(io.Discard)
flateWriterPool.Put(flateWriter)
return
}
}
Expand Down
19 changes: 11 additions & 8 deletions internal/reader/json/adapter.go
Original file line number Diff line number 8000 Diff line change
Expand Up @@ -5,6 +5,7 @@ package json // import "miniflux.app/v2/internal/reader/json"

import (
"cmp"
"html"
"log/slog"
"slices"
"strings"
Expand Down Expand Up @@ -110,14 +111,16 @@ func (j *JSONAdapter) BuildFeed(baseURL string) *model.Feed {
entry.Title = entry.URL
}

// Populate the entry content.
for _, value := range []string{item.ContentHTML, item.ContentText, item.Summary} {
if value = strings.TrimSpace(value); value == "" {
continue
}

entry.Content = value
break
// Populate the entry content. content_html is HTML, but content_text
// and summary are plain text (JSON Feed 1.1), so they must be escaped
// before being stored as HTML, otherwise the sanitizer drops any
// markup-like characters they contain.
if contentHTML := strings.TrimSpace(item.ContentHTML); contentHTML != "" {
entry.Content = contentHTML
} else if contentText := strings.TrimSpace(item.ContentText); contentText != "" {
entry.Content = html.EscapeString(contentText)
} else if summary := strings.TrimSpace(item.Summary); summary != "" {
entry.Content = html.EscapeString(summary)
}

// Populate the entry date.
Expand Down
28 changes: 28 additions & 0 deletions internal/reader/json/parser_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -1203,3 +1203,31 @@ func TestParseItemWithoutLanguageInheritsFeedLanguage(t *testing.T) {
t.Errorf("Expected entry to inherit feed language, got: %q", feed.Entries[0].Language)
}
}

func TestParseItemWithPlainTextContentContainingHTMLCharacters(t *testing.T) {
data := `{
"version": "https://jsonfeed.org/version/1.1",
"title": "Example",
"home_page_url": "https://example.org/",
"feed_url": "https://example.org/feed.json",
"items": [
{
"id": "1",
"url": "https://example.org/1",
"content_text": "Use <div> literally: a & b < c"
}
]
}`

feed, err := Parse("https://example.org/feed.json", bytes.NewBufferString(data))
if err != nil {
t.Fatal(err)
}

// content_text is plain text (JSON Feed 1.1), so it must be HTML-escaped
// before being stored as HTML, otherwise the sanitizer drops the markup.
want := "Use &lt;div&gt; literally: a &amp; b &lt; c"
if feed.Entries[0].Content != want {
t.Errorf("Incorrect entry content, got: %q, want: %q", feed.Entries[0].Content, want)
}
}
Loading
0