bric3/HttpClientBug.java

## HttpClientBug.java
import java.io.IOException;
import java.io.InputStream;
import java.io.UncheckedIOException;
import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;
import java.time.Duration;
import java.time.temporal.ChronoUnit;
import java.util.function.Function;
import java.util.function.Supplier;
import java.util.zip.GZIPInputStream;


/**
 * Work around <a href="https://bugs.openjdk.java.net/browse/JDK-8217627">JDK-8217627</a>,
 * <a href="https://bugs.openjdk.java.net/browse/JDK-8217264">JDK-8217264</a>
 * when using {@link java.net.http.HttpResponse.BodySubscribers#mapping(HttpResponse.BodySubscriber, Function)} and a
 * blocking call like {@link GZIPInputStream#GZIPInputStream(InputStream)}.
 *
 * <strong>Note this is fixed since JDK 13.</strong>
 */
public class HttpClientBug {
    public static void main(String[] args) throws IOException, InterruptedException {
        System.getLogger("main").log(System.Logger.Level.INFO, System.getProperty("java.version"));

        var request = HttpRequest.newBuilder(URI.create("https://dog.ceo/api/breeds/image/random"))
                .GET()
                .header("Accept-Encoding", "gzip")
                .build();

        var httpClient = HttpClient.newBuilder()
                .version(HttpClient.Version.HTTP_1_1)
                .followRedirects(HttpClient.Redirect.NORMAL)
                .connectTimeout(Duration.of(5, ChronoUnit.SECONDS))
                .build();

        var httpResponse = httpClient.send(
                request,
                terminalBodyHandler(gzipBodyHandler())
        );

        if (httpResponse.statusCode() == 200) {
            System.out.println(httpResponse.body().get());
        } else {
            System.out.println("Unexpected status code: " + httpResponse.statusCode());
        }
    }

    /**
     * Body handler that returns a Supplier of the response body as a String.
     *
     * <p>
     *     The supplier is necessary for the same reasons stated in {@link #gzipBodyHandler()},
     *     as it relies on this upstream body handler. So the call to initiate the decoding of
     *     the data has to be made after {@link HttpResponse#body()}.
     * </p>
     *
     * @param upstreamHandler The upstream body handler.
     * @return Body handler that decodes the input stream as a String.
     */
    private static HttpResponse.BodyHandler<Supplier<String>> terminalBodyHandler(HttpResponse.BodyHandler<Supplier<InputStream>> upstreamHandler) {
        return responseInfo -> {
            var statusCode = responseInfo.statusCode();
            var upstream = upstreamHandler.apply(responseInfo);

            if (statusCode >= 200 && statusCode < 300) {
                return HttpResponse.BodySubscribers.mapping(upstream, inputStream -> () -> {
                    try (var is = inputStream.get()) {
                        return new String(is.readAllBytes());
                    } catch (IOException e) {
                        e.printStackTrace();
                        throw new UncheckedIOException(e);
                    }
                });
            } else {
                return HttpResponse.BodySubscribers.mapping(upstream, inputStream -> () -> {
                    try (var is = inputStream.get()) {
                        return new String(is.readAllBytes());
                    } catch (IOException e) {
                        e.printStackTrace();
                        throw new UncheckedIOException(e);
                    }
                });
            }
        };
    }

    /**
     * Returns a body handler that returns a Supplier of inputstream able to handle gzip, brotli compression or nothing
     *
     * <p>
     *     The use of the Supplier is necessary on JDK11u, because GZIPInputStream and BrotliInputStream
     *     are reading the stream in their constructor (typically to read the compressed stream headers),
     *     since this operation is blocking, it blocks the executor from performing other HttpClient tasks
     *     And as such, the {@link HttpClient#send(HttpRequest, HttpResponse.BodyHandler)} call is blocked.
     * </p>
     * <p>
     *     In order to prevent that it is necessary to map to a supplier that make the blocking operations
     *     happen <strong>after</strong> the call to {@link HttpResponse#body()}.
     * </p>
     *
     * @see <a href="https://bugs.openjdk.java.net/browse/JDK-8217627">JDK-8217627</a>
     * @see <a href="https://bugs.openjdk.java.net/browse/JDK-8217264">JDK-8217264</a>
     * @return BodyHandler mapping to an input stream supplier
     */
    private static HttpResponse.BodyHandler<Supplier<InputStream>> gzipBodyHandler() {
        return responseInfo -> {
            var gzipped = responseInfo.headers()
                    .firstValue("content-encoding")
                    .orElse("")
                    .equalsIgnoreCase("gzip");

            if (gzipped) {
                System.out.println("Gzipped response");
                return HttpResponse.BodySubscribers.mapping(
                        HttpResponse.BodySubscribers.ofInputStream(),
                        inputStream -> () -> {
                            try {
                                return new GZIPInputStream(inputStream);
                            } catch (IOException e) {
                                e.printStackTrace();
                                throw new UncheckedIOException(e);
                            }
                        }
                );
            } else {
                return HttpResponse.BodySubscribers.mapping(HttpResponse.BodySubscribers.ofInputStream(), inputStream -> () -> inputStream);
            }
        };
    }
}

## without-work-around-thread-dump-stack.adoc

      
    Raw
  

              without-work-around-thread-dump-stack.adoc
            
          
Without the workaround, a thread dump might have stack traces like below.
Note JDK-8217264 is fixed since JDK 13.


"HttpClient-1-Worker-0@2251" daemon prio=5 tid=0x11 nid=NA waiting
  java.lang.Thread.State: WAITING
	  at jdk.internal.misc.Unsafe.park(Unsafe.java:-1)
	  at java.util.concurrent.locks.LockSupport.park(LockSupport.java:194)
	  at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await(AbstractQueuedSynchronizer.java:2081)
	  at java.util.concurrent.ArrayBlockingQueue.take(ArrayBlockingQueue.java:417)
	  at jdk.internal.net.http.ResponseSubscribers$HttpResponseInputStream.current(ResponseSubscribers.java:362)
	  at jdk.internal.net.http.ResponseSubscribers$HttpResponseInputStream.read(ResponseSubscribers.java:418)
	  at java.util.zip.CheckedInputStream.read(CheckedInputStream.java:60)
	  at java.util.zip.GZIPInputStream.readUByte(GZIPInputStream.java:267)
	  at java.util.zip.GZIPInputStream.readUShort(GZIPInputStream.java:259)
	  at java.util.zip.GZIPInputStream.readHeader(GZIPInputStream.java:165)
	  at java.util.zip.GZIPInputStream.<init>(GZIPInputStream.java:80)
	  at java.util.zip.GZIPInputStream.<init>(GZIPInputStream.java:92)
	  at HttpClientBug.lambda$gzipBodyHandler$6(HttpClientBug.java:135)
	  at HttpClientBug$$Lambda$191.1885277244.apply(Unknown Source:-1)
	  at java.util.concurrent.CompletableFuture.uniApplyNow(CompletableFuture.java:680)
	  at java.util.concurrent.CompletableFuture.uniApplyStage(CompletableFuture.java:658)
	  at java.util.concurrent.CompletableFuture.thenApply(CompletableFuture.java:2094)
	  at java.util.concurrent.CompletableFuture$MinimalStage.thenApply(CompletableFuture.java:2820)
	  at jdk.internal.net.http.ResponseSubscribers$MappingSubscriber.getBody(ResponseSubscribers.java:675)
	  at jdk.internal.net.http.ResponseSubscribers$MappingSubscriber.getBody(ResponseSubscribers.java:675)
	  at jdk.internal.net.http.Http1Response.readBody(Http1Response.java:487)
	  at jdk.internal.net.http.Http1Exchange.readBodyAsync(Http1Exchange.java:375)
	  at jdk.internal.net.http.Exchange.readBodyAsync(Exchange.java:175)
	  at jdk.internal.net.http.MultiExchange.lambda$responseAsync0$4(MultiExchange.java:305)
	  at jdk.internal.net.http.MultiExchange$$Lambda$94.1264413185.apply(Unknown Source:-1)
	  at java.util.concurrent.CompletableFuture$UniCompose.tryFire(CompletableFuture.java:1072)
	  at java.util.concurrent.CompletableFuture.postComplete(CompletableFuture.java:506)
	  at java.util.concurrent.CompletableFuture.postFire(CompletableFuture.java:610)
	  at java.util.concurrent.CompletableFuture$UniApply.tryFire(CompletableFuture.java:649)
	  at java.util.concurrent.CompletableFuture$Completion.run(CompletableFuture.java:478)
	  at jdk.internal.net.http.HttpClientImpl$DelegatingExecutor.execute(HttpClientImpl.java:153)
	  at java.util.concurrent.CompletableFuture$UniCompletion.claim(CompletableFuture.java:568)
	  at java.util.concurrent.CompletableFuture$UniApply.tryFire(CompletableFuture.java:638)
	  at java.util.concurrent.CompletableFuture.postComplete(CompletableFuture.java:506)
	  at java.util.concurrent.CompletableFuture.complete(CompletableFuture.java:2073)
	  at jdk.internal.net.http.Http1Response$HeadersReader.handle(Http1Response.java:693)
	  at jdk.internal.net.http.Http1Response$HeadersReader.handle(Http1Response.java:619)
	  at jdk.internal.net.http.Http1Response$Receiver.accept(Http1Response.java:610)
	  at jdk.internal.net.http.Http1Response$HeadersReader.tryAsyncReceive(Http1Response.java:666)
	  at jdk.internal.net.http.Http1AsyncReceiver.flush(Http1AsyncReceiver.java:228)
	  at jdk.internal.net.http.Http1AsyncReceiver$$Lambda$117.1645547422.run(Unknown Source:-1)
	  at jdk.internal.net.http.common.SequentialScheduler$SynchronizedRestartableTask.run(SequentialScheduler.java:175)
	  - locked <0xb74> (a java.lang.Object)
	  at jdk.internal.net.http.common.SequentialScheduler$CompleteRestartableTask.run(SequentialScheduler.java:147)
	  at jdk.internal.net.http.common.SequentialScheduler$SchedulableTask.run(SequentialScheduler.java:198)
	  at jdk.internal.net.http.HttpClientImpl$DelegatingExecutor.execute(HttpClientImpl.java:153)
	  at jdk.internal.net.http.common.SequentialScheduler.runOrSchedule(SequentialScheduler.java:273)
	  at jdk.internal.net.http.common.SequentialScheduler.runOrSchedule(SequentialScheduler.java:242)
	  at jdk.internal.net.http.Http1AsyncReceiver.asyncReceive(Http1AsyncReceiver.java:459)
	  at jdk.internal.net.http.Http1AsyncReceiver$Http1TubeSubscriber.onNext(Http1AsyncReceiver.java:579)
	  at jdk.internal.net.http.Http1AsyncReceiver$Http1TubeSubscriber.onNext(Http1AsyncReceiver.java:536)
	  at jdk.internal.net.http.common.SSLTube$DelegateWrapper.onNext(SSLTube.java:202)
	  at jdk.internal.net.http.common.SSLTube$SSLSubscriberWrapper.onNext(SSLTube.java:484)
	  at jdk.internal.net.http.common.SSLTube$SSLSubscriberWrapper.onNext(SSLTube.java:287)
	  at jdk.internal.net.http.common.SubscriberWrapper$DownstreamPusher.run1(SubscriberWrapper.java:318)
	  at jdk.internal.net.http.common.SubscriberWrapper$DownstreamPusher.run(SubscriberWrapper.java:261)
	  at jdk.internal.net.http.common.SequentialScheduler$SynchronizedRestartableTask.run(SequentialScheduler.java:175)
	  - locked <0xb75> (a java.lang.Object)
	  at jdk.internal.net.http.common.SequentialScheduler$CompleteRestartableTask.run(SequentialScheduler.java:147)
	  at jdk.internal.net.http.common.SequentialScheduler$SchedulableTask.run(SequentialScheduler.java:198)
	  at jdk.internal.net.http.common.SequentialScheduler.runOrSchedule(SequentialScheduler.java:271)
	  at jdk.internal.net.http.common.SequentialScheduler.runOrSchedule(SequentialScheduler.java:224)
	  at jdk.internal.net.http.common.SubscriberWrapper.outgoing(SubscriberWrapper.java:234)
	  at jdk.internal.net.http.common.SubscriberWrapper.outgoing(SubscriberWrapper.java:200)
	  at jdk.internal.net.http.common.SSLFlowDelegate$Reader.processData(SSLFlowDelegate.java:403)
	  at jdk.internal.net.http.common.SSLFlowDelegate$Reader$ReaderDownstreamPusher.run(SSLFlowDelegate.java:264)
	  at jdk.internal.net.http.common.SequentialScheduler$SynchronizedRestartableTask.run(SequentialScheduler.java:175)
	  - locked <0xb76> (a java.lang.Object)
	  at jdk.internal.net.http.common.SequentialScheduler$CompleteRestartableTask.run(SequentialScheduler.java:147)
	  at jdk.internal.net.http.common.SequentialScheduler$SchedulableTask.run(SequentialScheduler.java:198)
	  at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128)
	  at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628)
	  at java.lang.Thread.run(Thread.java:829)
	import java.io.IOException;
	import java.io.InputStream;
	import java.io.UncheckedIOException;
	import java.net.URI;
	import java.net.http.HttpClient;
	import java.net.http.HttpRequest;
	import java.net.http.HttpResponse;
	import java.time.Duration;
	import java.time.temporal.ChronoUnit;
	import java.util.function.Function;
	import java.util.function.Supplier;
	import java.util.zip.GZIPInputStream;


	/**
	* Work around <a href="https://bugs.openjdk.java.net/browse/JDK-8217627">JDK-8217627</a>,
	* <a href="https://bugs.openjdk.java.net/browse/JDK-8217264">JDK-8217264</a>
	* when using {@link java.net.http.HttpResponse.BodySubscribers#mapping(HttpResponse.BodySubscriber, Function)} and a
	* blocking call like {@link GZIPInputStream#GZIPInputStream(InputStream)}.
	*
	* <strong>Note this is fixed since JDK 13.</strong>
	*/
	public class HttpClientBug {
	public static void main(String[] args) throws IOException, InterruptedException {
	System.getLogger("main").log(System.Logger.Level.INFO, System.getProperty("java.version"));

	var request = HttpRequest.newBuilder(URI.create("https://dog.ceo/api/breeds/image/random"))
	.GET()
	.header("Accept-Encoding", "gzip")
	.build();

	var httpClient = HttpClient.newBuilder()
	.version(HttpClient.Version.HTTP_1_1)
	.followRedirects(HttpClient.Redirect.NORMAL)
	.connectTimeout(Duration.of(5, ChronoUnit.SECONDS))
	.build();

	var httpResponse = httpClient.send(
	request,
	terminalBodyHandler(gzipBodyHandler())
	);

	if (httpResponse.statusCode() == 200) {
	System.out.println(httpResponse.body().get());
	} else {
	System.out.println("Unexpected status code: " + httpResponse.statusCode());
	}
	}

	/**
	* Body handler that returns a Supplier of the response body as a String.
	*
	* <p>
	* The supplier is necessary for the same reasons stated in {@link #gzipBodyHandler()},
	* as it relies on this upstream body handler. So the call to initiate the decoding of
	* the data has to be made after {@link HttpResponse#body()}.
	* </p>
	*
	* @param upstreamHandler The upstream body handler.
	* @return Body handler that decodes the input stream as a String.
	*/
	private static HttpResponse.BodyHandler<Supplier<String>> terminalBodyHandler(HttpResponse.BodyHandler<Supplier<InputStream>> upstreamHandler) {
	return responseInfo -> {
	var statusCode = responseInfo.statusCode();
	var upstream = upstreamHandler.apply(responseInfo);

	if (statusCode >= 200 && statusCode < 300) {
	return HttpResponse.BodySubscribers.mapping(upstream, inputStream -> () -> {
	try (var is = inputStream.get()) {
	return new String(is.readAllBytes());
	} catch (IOException e) {
	e.printStackTrace();
	throw new UncheckedIOException(e);
	}
	});
	} else {
	return HttpResponse.BodySubscribers.mapping(upstream, inputStream -> () -> {
	try (var is = inputStream.get()) {
	return new String(is.readAllBytes());
	} catch (IOException e) {
	e.printStackTrace();
	throw new UncheckedIOException(e);
	}
	});
	}
	};
	}

	/**
	* Returns a body handler that returns a Supplier of inputstream able to handle gzip, brotli compression or nothing
	*
	* <p>
	* The use of the Supplier is necessary on JDK11u, because GZIPInputStream and BrotliInputStream
	* are reading the stream in their constructor (typically to read the compressed stream headers),
	* since this operation is blocking, it blocks the executor from performing other HttpClient tasks
	* And as such, the {@link HttpClient#send(HttpRequest, HttpResponse.BodyHandler)} call is blocked.
	* </p>
	* <p>
	* In order to prevent that it is necessary to map to a supplier that make the blocking operations
	* happen <strong>after</strong> the call to {@link HttpResponse#body()}.
	* </p>
	*
	* @see <a href="https://bugs.openjdk.java.net/browse/JDK-8217627">JDK-8217627</a>
	* @see <a href="https://bugs.openjdk.java.net/browse/JDK-8217264">JDK-8217264</a>
	* @return BodyHandler mapping to an input stream supplier
	*/
	private static HttpResponse.BodyHandler<Supplier<InputStream>> gzipBodyHandler() {
	return responseInfo -> {
	var gzipped = responseInfo.headers()
	.firstValue("content-encoding")
	.orElse("")
	.equalsIgnoreCase("gzip");

	if (gzipped) {
	System.out.println("Gzipped response");
	return HttpResponse.BodySubscribers.mapping(
	HttpResponse.BodySubscribers.ofInputStream(),
	inputStream -> () -> {
	try {
	return new GZIPInputStream(inputStream);
	} catch (IOException e) {
	e.printStackTrace();
	throw new UncheckedIOException(e);
	}
	}
	);
	} else {
	return HttpResponse.BodySubscribers.mapping(HttpResponse.BodySubscribers.ofInputStream(), inputStream -> () -> inputStream);
	}
	};
	}
	}