marky

commit 49caf6290d92e0bd475f7ad1441364ec1f0ef618
parent ba381a0d8736414280ff50d3f5bdee45ba5d5e97
Author: David Eisinger <[email protected]>
Date:   Mon, 13 Oct 2025 00:18:20 -0400

Fix w3m/Linux/UTF-8 issue

Diffstat:
M
internal/archiver.go
 | 
3
++-

1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/internal/archiver.go b/internal/archiver.go
@@ -80,7 +80,8 @@ func IsHTMLURL(ctx context.Context, rawURL string) bool {
 
 // w3mDump runs `w3m -dump` to render the given URL to plaintext.
 func w3mDump(ctx context.Context, url string) (string, error) {
-	cmd := exec.CommandContext(ctx, "w3m", "-dump", "-T", "text/html", "-o", "display_link_number=1", url)
+	// Force UTF-8 output to avoid mojibake on systems with non-UTF locales.
+	cmd := exec.CommandContext(ctx, "w3m", "-dump", "-T", "text/html", "-O", "UTF-8", "-o", "display_link_number=1", url)
 	// Keep environment as-is; rely on user PATH. Capture stdout only.
 	b, err := cmd.Output()
 	if err != nil {