Change header encoding to UTF8 in DefaultPartHttpMessageReader

This commit changes the encoding used to parse multipart headers from
ISO-8859-1 to UTF-8, in accordance with RFC 7578.

Closes gh-26736

spring-web/src/main/java/org/springframework/http/codec/multipart/DefaultPartHttpMessageReader.java

 /*
- * Copyright 2002-2020 the original author or authors.
+ * Copyright 2002-2021 the original author or authors.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -17,6 +17,7 @@
 package org.springframework.http.codec.multipart;
 
 import java.io.IOException;
+import java.nio.charset.Charset;
 import java.nio.charset.StandardCharsets;
 import java.nio.file.Files;
 import java.nio.file.Path;
@@ -78,6 +79,8 @@ public class DefaultPartHttpMessageReader extends LoggingCodecSupport implements
 
 	private Mono<Path> fileStorageDirectory = Mono.defer(this::defaultFileStorageDirectory).cache();
 
+	private Charset headersCharset = StandardCharsets.UTF_8;
+
 
 	/**
 	 * Configure the maximum amount of memory that is allowed per headers section of each part.
@@ -188,6 +191,18 @@ public class DefaultPartHttpMessageReader extends LoggingCodecSupport implements
 		this.streaming = streaming;
 	}
 
+	/**
+	 * Sets the character set used to decode headers. Defaults to
+	 * UTF-8 as per RFC 7578.
+	 * @param headersCharset the charset to use for decoding headers
+	 * @since 5.3.6
+	 * @see <a href="https://tools.ietf.org/html/rfc7578#section-5.1">RFC-7578 Section 5.2</a>
+	 */
+	public void setHeadersCharset(Charset headersCharset) {
+		Assert.notNull(headersCharset, "HeadersCharset must not be null");
+		this.headersCharset = headersCharset;
+	}
+
 	@Override
 	public List<MediaType> getReadableMediaTypes() {
 		return Collections.singletonList(MediaType.MULTIPART_FORM_DATA);
@@ -214,7 +229,7 @@ public class DefaultPartHttpMessageReader extends LoggingCodecSupport implements
 						message.getHeaders().getContentType() + "\""));
 			}
 			Flux<MultipartParser.Token> tokens = MultipartParser.parse(message.getBody(), boundary,
-					this.maxHeadersSize);
+					this.maxHeadersSize, this.headersCharset);
 
 			return PartGenerator.createParts(tokens, this.maxParts, this.maxInMemorySize, this.maxDiskUsagePerPart,
 					this.streaming, this.fileStorageDirectory, this.blockingOperationScheduler);
@@ -222,7 +237,7 @@ public class DefaultPartHttpMessageReader extends LoggingCodecSupport implements
 	}
 
 	@Nullable
-	private static byte[] boundary(HttpMessage message) {
+	private byte[] boundary(HttpMessage message) {
 		MediaType contentType = message.getHeaders().getContentType();
 		if (contentType != null) {
 			String boundary = contentType.getParameter("boundary");
@@ -231,7 +246,7 @@ public class DefaultPartHttpMessageReader extends LoggingCodecSupport implements
 				if (len > 2 && boundary.charAt(0) == '"' && boundary.charAt(len - 1) == '"') {
 					boundary = boundary.substring(1, len - 1);
 				}
-				return boundary.getBytes(StandardCharsets.ISO_8859_1);
+				return boundary.getBytes(this.headersCharset);
 			}
 		}
 		return null;

spring-web/src/main/java/org/springframework/http/codec/multipart/MultipartParser.java

 /*
- * Copyright 2002-2020 the original author or authors.
+ * Copyright 2002-2021 the original author or authors.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -16,7 +16,7 @@
 
 package org.springframework.http.codec.multipart;
 
-import java.nio.charset.StandardCharsets;
+import java.nio.charset.Charset;
 import java.util.ArrayList;
 import java.util.List;
 import java.util.concurrent.atomic.AtomicBoolean;
@@ -69,11 +69,14 @@ final class MultipartParser extends BaseSubscriber<DataBuffer> {
 
 	private final AtomicBoolean requestOutstanding = new AtomicBoolean();
 
+	private final Charset headersCharset;
 
-	private MultipartParser(FluxSink<Token> sink, byte[] boundary, int maxHeadersSize) {
+
+	private MultipartParser(FluxSink<Token> sink, byte[] boundary, int maxHeadersSize, Charset headersCharset) {
 		this.sink = sink;
 		this.boundary = boundary;
 		this.maxHeadersSize = maxHeadersSize;
+		this.headersCharset = headersCharset;
 		this.state = new AtomicReference<>(new PreambleState());
 	}
 
@@ -82,11 +85,13 @@ final class MultipartParser extends BaseSubscriber<DataBuffer> {
 	 * @param buffers the input buffers
 	 * @param boundary the multipart boundary, as found in the {@code Content-Type} header
 	 * @param maxHeadersSize the maximum buffered header size
+	 * @param headersCharset the charset to use for decoding headers
 	 * @return a stream of parsed tokens
 	 */
-	public static Flux<Token> parse(Flux<DataBuffer> buffers, byte[] boundary, int maxHeadersSize) {
+	public static Flux<Token> parse(Flux<DataBuffer> buffers, byte[] boundary, int maxHeadersSize,
+			Charset headersCharset) {
 		return Flux.create(sink -> {
-			MultipartParser parser = new MultipartParser(sink, boundary, maxHeadersSize);
+			MultipartParser parser = new MultipartParser(sink, boundary, maxHeadersSize, headersCharset);
 			sink.onCancel(parser::onSinkCancel);
 			sink.onRequest(n -> parser.requestBuffer());
 			buffers.subscribe(parser);
@@ -180,7 +185,7 @@ final class MultipartParser extends BaseSubscriber<DataBuffer> {
 
 
 	/**
-	 * Represents the output of {@link #parse(Flux, byte[], int)}.
+	 * Represents the output of {@link #parse(Flux, byte[], int, Charset)}.
 	 */
 	public abstract static class Token {
 
@@ -372,7 +377,6 @@ final class MultipartParser extends BaseSubscriber<DataBuffer> {
 				DataBufferUtils.release(buf);
 
 				emitHeaders(parseHeaders());
-				// TODO: no need to check result of changeState, no further statements
 				changeState(this, new BodyState(), bodyBuf);
 			}
 			else {
@@ -408,7 +412,7 @@ final class MultipartParser extends BaseSubscriber<DataBuffer> {
 			}
 			DataBuffer joined = this.buffers.get(0).factory().join(this.buffers);
 			this.buffers.clear();
-			String string = joined.toString(StandardCharsets.ISO_8859_1);
+			String string = joined.toString(MultipartParser.this.headersCharset);
 			DataBufferUtils.release(joined);
 			String[] lines = string.split(HEADER_ENTRY_SEPARATOR);
 			HttpHeaders result = new HttpHeaders();

spring-web/src/test/java/org/springframework/http/codec/multipart/DefaultPartHttpMessageReaderTests.java

 /*
- * Copyright 2002-2020 the original author or authors.
+ * Copyright 2002-2021 the original author or authors.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -23,6 +23,7 @@ import java.lang.annotation.RetentionPolicy;
 import java.lang.annotation.Target;
 import java.nio.file.Files;
 import java.nio.file.Path;
+import java.util.Collections;
 import java.util.concurrent.CountDownLatch;
 import java.util.concurrent.TimeUnit;
 import java.util.stream.Stream;
@@ -251,6 +252,24 @@ public class DefaultPartHttpMessageReaderTests  {
 		latch.await();
 	}
 
+	@ParameterizedDefaultPartHttpMessageReaderTest
+	public void utf8Headers(String displayName, DefaultPartHttpMessageReader reader) throws InterruptedException {
+		MockServerHttpRequest request = createRequest(
+				new ClassPathResource("utf8.multipart", getClass()), "\"simple-boundary\"");
+
+		Flux<Part> result = reader.read(forClass(Part.class), request, emptyMap());
+
+		CountDownLatch latch = new CountDownLatch(1);
+		StepVerifier.create(result)
+				.consumeNextWith(part -> {
+					assertThat(part.headers()).containsEntry("Føø", Collections.singletonList("Bår"));
+					testPart(part, null, "This is plain ASCII text.", latch);
+				})
+				.verifyComplete();
+
+		latch.await();
+	}
+
 
 	private void testBrowser(DefaultPartHttpMessageReader reader, Resource resource, String boundary)
 			throws InterruptedException {

spring-web/src/test/resources/org/springframework/http/codec/multipart/utf8.multipart

+--simple-boundary
+Føø: Bår
+
+This is plain ASCII text.
+--simple-boundary--