Encode Unicode download filenames in RFC 5987 response headers

src/main/java/lion/Http/FileResponseFactory.java

@@ -1,6 +1,8 @@
 package lion.Http;
 
 import java.io.File;
+import java.net.URLEncoder;
+import java.nio.charset.StandardCharsets;
 import java.util.LinkedHashMap;
 import java.util.Map;
 
@@ -11,9 +13,6 @@ import java.util.Map;
  */
 public final class FileResponseFactory {
 
-    /** 不能作为 Content-Disposition 文件名的字符（引号、反斜杠与控制字符）统一替换。 */
-    private static final String UNSAFE_FILENAME_CHARS = "[\r\n\"\\\\]";
-
     private FileResponseFactory() {
     }
 
@@ -72,11 +71,22 @@ public final class FileResponseFactory {
     }
 
     /**
-     * 文件名在响应头里是 {@code attachment; filename="..."} 形式，必须剔除可能断开
-     * 头部行或提前闭合引号的字符。仅影响响应头，不改变磁盘上的文件名。
+     * HTTP 响应头不能直接包含中文等 Unicode 字符。保留 ASCII 兼容名，并用
+     * RFC 5987 的 UTF-8 filename* 传递完整文件名；控制字符、引号和反斜杠仍会清理。
      */
     static String contentDisposition(String fileName) {
-        String safe = fileName == null ? "" : fileName.replaceAll(UNSAFE_FILENAME_CHARS, "_");
-        return "attachment; filename=\"" + safe + "\"";
+        StringBuilder safe = new StringBuilder();
+        StringBuilder ascii = new StringBuilder();
+        if (fileName != null) fileName.codePoints().forEach(codePoint -> {
+            int sanitized = Character.isISOControl(codePoint) || codePoint == '"' || codePoint == '\\'
+                    ? '_' : codePoint;
+            safe.appendCodePoint(sanitized);
+            ascii.append(sanitized <= 0x7e ? (char) sanitized : '_');
+        });
+        String disposition = "attachment; filename=\"" + ascii + "\"";
+        if (safe.toString().equals(ascii.toString())) return disposition;
+        String encoded = URLEncoder.encode(safe.toString(), StandardCharsets.UTF_8)
+                .replace("+", "%20").replace("*", "%2A");
+        return disposition + "; filename*=UTF-8''" + encoded;
     }
 }

src/test/java/lion/Http/UnicodeDownloadTest.java

@@ -0,0 +1,73 @@
+package lion.Http;
+
+import org.junit.jupiter.api.Test;
+import org.junit.jupiter.api.io.TempDir;
+
+import java.net.URI;
+import java.net.URLDecoder;
+import java.net.http.HttpClient;
+import java.net.http.HttpRequest;
+import java.net.http.HttpResponse;
+import java.nio.charset.StandardCharsets;
+import java.nio.file.Files;
+import java.nio.file.Path;
+import java.time.Duration;
+import java.util.Arrays;
+
+import static org.junit.jupiter.api.Assertions.*;
+
+class UnicodeDownloadTest {
+    @Test void asciiNamesKeepTheExistingHeader() {
+        assertEquals("attachment; filename=\"archive + name.zip\"",
+                FileResponseFactory.contentDisposition("archive + name.zip"));
+    }
+
+    @Test void unicodeNamesHaveAsciiHeadersAndLosslessUtf8Filename() {
+        for (String filename : new String[]{"中文 日本語.zip", "café + 100%*.zip", "附件 📚.zip"}) {
+            String header = FileResponseFactory.contentDisposition(filename);
+            assertTrue(header.chars().allMatch(c -> c >= 32 && c < 127));
+            assertEquals(filename, decodeFilename(header));
+            assertFalse(header.contains("+100"));
+        }
+    }
+
+    @Test void allControlsAndQuotedFilenameEscapesAreRemoved() {
+        String header = FileResponseFactory.contentDisposition("标题\r\n\u0000\u007f\u0085\"\\.zip");
+        assertTrue(header.chars().allMatch(c -> c >= 32 && c < 127));
+        assertEquals("标题_______.zip", decodeFilename(header));
+    }
+
+    @Test void realHttpServerDownloadsUnicodeFileAndSupportsHeadAndResume(@TempDir Path directory) throws Exception {
+        String filename = "[中文 (日本語)] 下载 + 100% 📚.zip";
+        Path file = directory.resolve(filename);
+        byte[] content = new byte[128 * 1024 + 73];
+        new java.util.Random(73).nextBytes(content);
+        Files.write(file, content);
+        try (HttpFileServer server = HttpFileServer.create(0, "/", request ->
+                FileResponseFactory.forFile(request.rangeHeader(), file.toFile(), filename), 2);
+             HttpClient client = HttpClient.newBuilder().connectTimeout(Duration.ofSeconds(2)).build()) {
+            server.start();
+            URI uri = URI.create("http://127.0.0.1:" + server.port() + "/archive.zip");
+            var full = client.send(HttpRequest.newBuilder(uri).GET().build(), HttpResponse.BodyHandlers.ofByteArray());
+            assertEquals(200, full.statusCode());
+            assertArrayEquals(content, full.body());
+            assertEquals(filename, decodeFilename(full.headers().firstValue("Content-Disposition").orElseThrow()));
+            var head = client.send(HttpRequest.newBuilder(uri).method("HEAD", HttpRequest.BodyPublishers.noBody()).build(),
+                    HttpResponse.BodyHandlers.ofByteArray());
+            assertEquals(200, head.statusCode());
+            assertEquals(Integer.toString(content.length), head.headers().firstValue("Content-Length").orElseThrow());
+            assertEquals(0, head.body().length);
+            var range = client.send(HttpRequest.newBuilder(uri).header("Range", "bytes=4096-8191").GET().build(),
+                    HttpResponse.BodyHandlers.ofByteArray());
+            assertEquals(206, range.statusCode());
+            assertArrayEquals(Arrays.copyOfRange(content, 4096, 8192), range.body());
+            assertEquals("bytes 4096-8191/" + content.length, range.headers().firstValue("Content-Range").orElseThrow());
+            assertEquals(filename, decodeFilename(range.headers().firstValue("Content-Disposition").orElseThrow()));
+        }
+    }
+
+    private static String decodeFilename(String header) {
+        return URLDecoder.decode(header.substring(header.indexOf("filename*=UTF-8''") + "filename*=UTF-8''".length()),
+                StandardCharsets.UTF_8);
+    }
+}