AFHTTPSessionManager中的HTTPMaximumConnectionsPerHost被忽略 [英] HTTPMaximumConnectionsPerHost is ignored in AFHTTPSessionManager

查看：283
本文介绍了AFHTTPSessionManager中的HTTPMaximumConnectionsPerHost被忽略的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！

问题描述

我正在尝试将AFNetworking2.0与新的NSURLSession类一起使用以获取一些数据.特别是，我对NSURLSessionConfiguration的新HTTPMaximumConnectionsPerHost属性感兴趣，但不幸的是，它似乎没有按我预期的那样工作.

I am trying to use AFNetworking2.0 with new NSURLSession class to fetch some data. In particular I am interested in new HTTPMaximumConnectionsPerHost property of NSURLSessionConfiguration but unfortunately it doesn't seem to work as I expect.

@interface ViewController ()
@property (nonatomic, strong) AFHTTPSessionManager *httpSessionManager;
@end

@implementation ViewController

- (void)viewDidLoad
{
    [super viewDidLoad];

    NSURLSessionConfiguration *sessionConfig = [[NSURLSessionConfiguration defaultSessionConfiguration] copy];
    sessionConfig.HTTPMaximumConnectionsPerHost = 5;

    self.httpSessionManager = [[AFHTTPSessionManager alloc] initWithSessionConfiguration:sessionConfig];
    self.httpSessionManager.responseSerializer = [AFImageResponseSerializer serializer];

    for (int i = 0; i < 100; ++i) {
        //http://i.dailymail.co.uk/i/pix/2012/07/03/article-2168112-13E70C13000005DC-867_964x946.jpg // 400k picture, the one from deviantart is 2k
        [self.httpSessionManager GET:@"http://a.deviantart.net/avatars/t/o/toaster-kitten.jpg" parameters:nil success:^(NSURLSessionDataTask *task, id responseObject) {
            NSLog(@"has finished task %p, context id %lu", task, task.taskIdentifier);
        } failure:^(NSURLSessionDataTask *task, NSError *error) {
            NSLog(@"has failed task %p, context id %lu", task, task.taskIdentifier);
        }];
    }
}

这是我的示例下载实现.

This is my sample implementation of download.

我还调整了AFHTTPSessionManager来生成日志:

I have also tweaked AFHTTPSessionManager to produce logs:

- (NSURLSessionDataTask *)GET:(NSString *)URLString
                   parameters:(NSDictionary *)parameters
                      success:(void (^)(NSURLSessionDataTask *task, id responseObject))success
                      failure:(void (^)(NSURLSessionDataTask *task, NSError *error))failure
{
    NSMutableURLRequest *request = [self.requestSerializer requestWithMethod:@"GET" URLString:[[NSURL URLWithString:URLString relativeToURL:self.baseURL] absoluteString] parameters:parameters error:nil];

    __block NSURLSessionDataTask *task = [self dataTaskWithRequest:request completionHandler:^(NSURLResponse * __unused response, id responseObject, NSError *error) {
        if (error) {
            if (failure) {
                failure(task, error);
            }
        } else {
            if (success) {
                success(task, responseObject);
            }
        }
    }];

    [task addObserver:self forKeyPath:@"countOfBytesReceived" options:NSKeyValueObservingOptionOld context:(__bridge void*)task];
    NSLog(@"created task %p id %u", task, task.taskIdentifier);

    [task resume];

    return task;
}


- (void)observeValueForKeyPath:(NSString *)keyPath ofObject:(id)object change:(NSDictionary *)change context:(void *)context{
    if([[change objectForKey:@"old"] integerValue] == 0 && [keyPath isEqualToString:@"countOfBytesReceived"]){
        NSLog(@"dtask %p id %u: started", object, ((NSURLSessionDataTask*)object).taskIdentifier);
        @try {
            if ((__bridge void *)object == context) {
                [object removeObserver:self forKeyPath:@"countOfBytesReceived" context:(__bridge void*)object];
            }
        }
        @catch (NSException *exception) {
        }
    }
}

使用此代码，我会收到下一条日志:

With this code I receive the next log: 

2014-04-02 17:10:38.950 test_block[2704:60b] created task 0x1701a4fa0 id 0
2014-04-02 17:10:38.972 test_block[2704:60b] created task 0x1781a5be0 id 1
2014-04-02 17:10:38.974 test_block[2704:60b] created task 0x1701a5160 id 2
2014-04-02 17:10:38.976 test_block[2704:60b] created task 0x1701a5320 id 3
2014-04-02 17:10:38.977 test_block[2704:60b] created task 0x1701a54e0 id 4
2014-04-02 17:10:38.979 test_block[2704:60b] created task 0x1701a56a0 id 5
2014-04-02 17:10:38.980 test_block[2704:60b] created task 0x1701a5940 id 6
2014-04-02 17:10:38.982 test_block[2704:1803] dtask 0x1701a4fa0 id 0: started
2014-04-02 17:10:38.984 test_block[2704:6403] dtask 0x1781a5be0 id 1: started
2014-04-02 17:10:38.986 test_block[2704:4b03] dtask 0x1701a56a0 id 5: started
2014-04-02 17:10:38.987 test_block[2704:5b03] dtask 0x1701a5160 id 2: started
2014-04-02 17:10:38.988 test_block[2704:6703] dtask 0x1701a5320 id 3: started
2014-04-02 17:10:38.990 test_block[2704:60b] created task 0x1781a5da0 id 7
2014-04-02 17:10:38.990 test_block[2704:1803] dtask 0x1701a54e0 id 4: started
2014-04-02 17:10:38.992 test_block[2704:6903] dtask 0x1701a5940 id 6: started
2014-04-02 17:10:38.993 test_block[2704:60b] created task 0x1701a4ec0 id 8
2014-04-02 17:10:38.993 test_block[2704:1803] dtask 0x1781a5da0 id 7: started
2014-04-02 17:10:38.994 test_block[2704:60b] created task 0x1781a5cc0 id 9
2014-04-02 17:10:38.996 test_block[2704:6403] dtask 0x1701a4ec0 id 8: started
2014-04-02 17:10:38.997 test_block[2704:60b] created task 0x1701a5240 id 10
2014-04-02 17:10:38.998 test_block[2704:5103] dtask 0x1781a5cc0 id 9: started
2014-04-02 17:10:38.999 test_block[2704:60b] created task 0x1781a6200 id 11
2014-04-02 17:10:39.391 test_block[2704:4603] dtask 0x1701a5240 id 10: started
2014-04-02 17:10:39.399 test_block[2704:60b] created task 0x1781a62e0 id 12
2014-04-02 17:10:39.406 test_block[2704:3507] dtask 0x1781a6200 id 11: started
2014-04-02 17:10:39.409 test_block[2704:60b] created task 0x1701a63c0 id 13
2014-04-02 17:10:39.411 test_block[2704:60b] created task 0x1701a6580 id 14
2014-04-02 17:10:39.414 test_block[2704:1803] dtask 0x1781a62e0 id 12: started
2014-04-02 17:10:39.418 test_block[2704:3507] dtask 0x1701a63c0 id 13: started
2014-04-02 17:10:39.421 test_block[2704:60b] created task 0x1781a6ac0 id 15
2014-04-02 17:10:39.423 test_block[2704:5b03] dtask 0x1701a6580 id 14: started
2014-04-02 17:10:39.425 test_block[2704:60b] created task 0x1701a5cc0 id 16
2014-04-02 17:10:39.427 test_block[2704:5b03] dtask 0x1781a6ac0 id 15: started
2014-04-02 17:10:39.428 test_block[2704:60b] created task 0x1701a5f60 id 17
2014-04-02 17:10:39.430 test_block[2704:5103] dtask 0x1701a5cc0 id 16: started
2014-04-02 17:10:39.431 test_block[2704:60b] created task 0x1781a64a0 id 18
2014-04-02 17:10:39.433 test_block[2704:3507] dtask 0x1701a5f60 id 17: started
2014-04-02 17:10:39.434 test_block[2704:60b] created task 0x1701a64a0 id 19
2014-04-02 17:10:39.435 test_block[2704:3507] dtask 0x1781a64a0 id 18: started
2014-04-02 17:10:39.437 test_block[2704:60b] created task 0x1701a6820 id 20
2014-04-02 17:10:39.438 test_block[2704:3507] dtask 0x1701a64a0 id 19: started
2014-04-02 17:10:39.440 test_block[2704:60b] created task 0x1781a6580 id 21
2014-04-02 17:10:39.441 test_block[2704:3507] dtask 0x1701a6820 id 20: started
2014-04-02 17:10:39.442 test_block[2704:60b] created task 0x1701a6120 id 22
2014-04-02 17:10:39.444 test_block[2704:3507] dtask 0x1781a6580 id 21: started
2014-04-02 17:10:39.445 test_block[2704:60b] created task 0x1701a7540 id 23
2014-04-02 17:10:39.446 test_block[2704:3507] dtask 0x1701a6120 id 22: started
2014-04-02 17:10:39.447 test_block[2704:60b] created task 0x1781a6c80 id 24
2014-04-02 17:10:39.448 test_block[2704:60b] created task 0x1701a5e80 id 25
2014-04-02 17:10:39.450 test_block[2704:60b] created task 0x1701a70e0 id 26
2014-04-02 17:10:39.450 test_block[2704:5103] dtask 0x1701a7540 id 23: started
2014-04-02 17:10:39.451 test_block[2704:5b03] dtask 0x1781a6c80 id 24: started
2014-04-02 17:10:39.453 test_block[2704:5103] dtask 0x1701a5e80 id 25: started
2014-04-02 17:10:39.454 test_block[2704:60b] created task 0x1701a7a80 id 27
...
2014-04-02 17:10:39.694 test_block[2704:60b] has finished task 0x1781aaaa0, context id 79
2014-04-02 17:10:39.695 test_block[2704:60b] has finished task 0x1701ac320, context id 80
2014-04-02 17:10:39.695 test_block[2704:60b] has finished task 0x1701abfa0, context id 78
2014-04-02 17:10:39.696 test_block[2704:60b] has finished task 0x1701ac4e0, context id 81
2014-04-02 17:10:39.697 test_block[2704:60b] has finished task 0x1701ac6a0, context id 82
2014-04-02 17:10:39.698 test_block[2704:60b] has finished task 0x1701ac860, context id 83
2014-04-02 17:10:39.699 test_block[2704:60b] has finished task 0x1781aac60, context id 84
2014-04-02 17:10:39.699 test_block[2704:60b] has finished task 0x1701aca20, context id 85
2014-04-02 17:10:39.700 test_block[2704:60b] has finished task 0x1781ab1a0, context id 86
2014-04-02 17:10:39.700 test_block[2704:60b] has finished task 0x1781ab360, context id 87
2014-04-02 17:10:39.701 test_block[2704:60b] has finished task 0x1701acbe0, context id 88
2014-04-02 17:10:39.702 test_block[2704:60b] has finished task 0x1701acda0, context id 89
2014-04-02 17:10:39.702 test_block[2704:60b] has finished task 0x1781ab6e0, context id 92
2014-04-02 17:10:39.703 test_block[2704:60b] has finished task 0x1781ab520, context id 90
2014-04-02 17:10:39.705 test_block[2704:60b] has finished task 0x1701ace80, context id 91
2014-04-02 17:10:39.713 test_block[2704:60b] has finished task 0x1781ab8a0, context id 94
2014-04-02 17:10:39.714 test_block[2704:60b] has finished task 0x1781aba60, context id 95
2014-04-02 17:10:39.714 test_block[2704:60b] has finished task 0x1701ad040, context id 93
2014-04-02 17:10:39.715 test_block[2704:60b] has finished task 0x1701ad200, context id 96
2014-04-02 17:10:39.715 test_block[2704:60b] has finished task 0x1701ad3c0, context id 97
2014-04-02 17:10:39.716 test_block[2704:60b] has finished task 0x1701ab520, context id 98
2014-04-02 17:10:39.716 test_block[2704:60b] has finished task 0x1701accc0, context id 99

这不是我期望看到的.如您所见，至少在之前启动的至少一项任务完成之前，已启动了5个以上的数据任务.

Which is not exactly what I expect to see. As you can see more than 5 data tasks are started until at least on of previously started has finished.

我在装有iOS7.1和iOS7.0.3的iPhone上进行了测试

I test on iPhone with iOS7.1 and iOS7.0.3

任何人都可以解释为什么会发生这种情况以及如何使这些任务由NSURLSession排队(或至少使它们在经过一定限制后失败)吗?

Could anyone please explain why this happens and how to get these tasks queued by NSURLSession (or at least make them fail after some limit)?

任何人都可以确认或驳斥这种行为吗?

Could anyone confirm or refute this behavior?

P.S.我已经实现了自己的排队代码，所以这个问题不是关于如何对请求进行排队，而是如何使其与NSURLSession一起使用.

P.S. I have already implemented my own queuing code so this question is not about how to queue requests but rather how to make it work with NSURLSession.

推荐答案

文档指出，它将限制每个主机的连接数.这与创建的任务数量无关.

The documentation states that it will limit the number of connections per host. That has nothing to do with the number of tasks created.

您可以轻松地记录委托回调，这些回调通知增量更新，以查看哪些任务代表实际连接.其他任务排在队列中，等待将其连接到主机.

You can easily log the delegate callbacks which notify of incremental updates to see which tasks represent actual connections. The other tasks are in a queue, waiting for their turn to be connected to the host.

这篇关于AFHTTPSessionManager中的HTTPMaximumConnectionsPerHost被忽略的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！

查看全文
相关文章
移动开发最新文章
热门教程
热门工具
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆