// Copyright (c) 2006-2009 The Chromium Authors. All rights reserved. // Use of this source code is governed by a BSD-style license that can be // found in the LICENSE file. #ifndef NET_URL_REQUEST_URL_REQUEST_JOB_H_ #define NET_URL_REQUEST_URL_REQUEST_JOB_H_ #include #include #include "base/ref_counted.h" #include "base/scoped_ptr.h" #include "base/time.h" #include "googleurl/src/gurl.h" #include "net/base/filter.h" #include "net/base/load_states.h" namespace net { class AuthChallengeInfo; class HttpResponseInfo; class IOBuffer; class UploadData; class X509Certificate; } class URLRequest; class URLRequestStatus; class URLRequestJobMetrics; // The URLRequestJob is using RefCounterThreadSafe because some sub classes // can be destroyed on multiple threads. This is the case of the // UrlRequestFileJob. class URLRequestJob : public base::RefCountedThreadSafe, public FilterContext { public: // When histogramming results related to SDCH and/or an SDCH latency test, the // number of packets for which we need to record arrival times so as to // calculate interpacket latencies. We currently are only looking at the // first few packets, as we're monitoring the impact of the initial TCP // congestion window on stalling of transmissions. static const size_t kSdchPacketHistogramCount = 5; explicit URLRequestJob(URLRequest* request); // Returns the request that owns this job. THIS POINTER MAY BE NULL if the // request was destroyed. URLRequest* request() const { return request_; } // Sets the upload data, most requests have no upload data, so this is a NOP. // Job types supporting upload data will override this. virtual void SetUpload(net::UploadData* upload) { } // Sets extra request headers for Job types that support request headers. virtual void SetExtraRequestHeaders(const std::string& headers) { } // If any error occurs while starting the Job, NotifyStartError should be // called. // This helps ensure that all errors follow more similar notification code // paths, which should simplify testing. virtual void Start() = 0; // This function MUST somehow call NotifyDone/NotifyCanceled or some requests // will get leaked. Certain callers use that message to know when they can // delete their URLRequest object, even when doing a cancel. The default Kill // implementation calls NotifyCanceled, so it is recommended that subclasses // call URLRequestJob::Kill() after doing any additional work. // // The job should endeavor to stop working as soon as is convenient, but must // not send and complete notifications from inside this function. Instead, // complete notifications (including "canceled") should be sent from a // callback run from the message loop. // // The job is not obliged to immediately stop sending data in response to // this call, nor is it obliged to fail with "canceled" unless not all data // was sent as a result. A typical case would be where the job is almost // complete and can succeed before the canceled notification can be // dispatched (from the message loop). // // The job should be prepared to receive multiple calls to kill it, but only // one notification must be issued. virtual void Kill(); // Called to detach the request from this Job. Results in the Job being // killed off eventually. The job must not use the request pointer any more. void DetachRequest(); // Called to read post-filtered data from this Job, returning the number of // bytes read, 0 when there is no more data, or -1 if there was an error. // This is just the backend for URLRequest::Read, see that function for more // info. bool Read(net::IOBuffer* buf, int buf_size, int *bytes_read); // Called to fetch the current load state for the job. virtual net::LoadState GetLoadState() const { return net::LOAD_STATE_IDLE; } // Called to get the upload progress in bytes. virtual uint64 GetUploadProgress() const { return 0; } // Called to fetch the charset for this request. Only makes sense for some // types of requests. Returns true on success. Calling this on a type that // doesn't have a charset will return false. virtual bool GetCharset(std::string* charset) { return false; } // Called to get response info. virtual void GetResponseInfo(net::HttpResponseInfo* info) {} // Returns the cookie values included in the response, if applicable. // Returns true if applicable. // NOTE: This removes the cookies from the job, so it will only return // useful results once per job. virtual bool GetResponseCookies(std::vector* cookies) { return false; } // Called to fetch the encoding types for this request. Only makes sense for // some types of requests. Returns true on success. Calling this on a request // that doesn't have or specify an encoding type will return false. // Returns a array of strings showing the sequential encodings used on the // content. // For example, encoding_types[0] = FILTER_TYPE_SDCH and encoding_types[1] = // FILTER_TYPE_GZIP, means the content was first encoded by sdch, and then // result was encoded by gzip. To decode, a series of filters must be applied // in the reverse order (in the above example, ungzip first, and then sdch // expand). virtual bool GetContentEncodings( std::vector* encoding_types) { return false; } // Find out if this is a download. virtual bool IsDownload() const; // Find out if this is a response to a request that advertised an SDCH // dictionary. Only makes sense for some types of requests. virtual bool IsSdchResponse() const { return false; } // Called to setup stream filter for this request. An example of filter is // content encoding/decoding. void SetupFilter(); // Called to determine if this response is a redirect. Only makes sense // for some types of requests. This method returns true if the response // is a redirect, and fills in the location param with the URL of the // redirect. The HTTP status code (e.g., 302) is filled into // |*http_status_code| to signify the type of redirect. // // The caller is responsible for following the redirect by setting up an // appropriate replacement Job. Note that the redirected location may be // invalid, the caller should be sure it can handle this. // // The default implementation inspects the response_info_. virtual bool IsRedirectResponse(GURL* location, int* http_status_code); // Called to determine if it is okay to redirect this job to the specified // location. This may be used to implement protocol-specific restrictions. // If this function returns false, then the URLRequest will fail reporting // net::ERR_UNSAFE_REDIRECT. virtual bool IsSafeRedirect(const GURL& location) { return true; } // Called to determine if this response is asking for authentication. Only // makes sense for some types of requests. The caller is responsible for // obtaining the credentials passing them to SetAuth. virtual bool NeedsAuth() { return false; } // Fills the authentication info with the server's response. virtual void GetAuthChallengeInfo( scoped_refptr* auth_info); // Resend the request with authentication credentials. virtual void SetAuth(const std::wstring& username, const std::wstring& password); // Display the error page without asking for credentials again. virtual void CancelAuth(); virtual void ContinueWithCertificate(net::X509Certificate* client_cert); // Continue processing the request ignoring the last error. virtual void ContinueDespiteLastError(); void FollowDeferredRedirect(); // Returns true if the Job is done producing response data and has called // NotifyDone on the request. bool is_done() const { return done_; } // Returns true if the job is doing performance profiling bool is_profiling() const { return is_profiling_; } // Retrieve the performance measurement of the job. The data is encapsulated // with a URLRequestJobMetrics object. The caller owns this object from now // on. URLRequestJobMetrics* RetrieveMetrics(); // Get/Set expected content size int64 expected_content_size() const { return expected_content_size_; } void set_expected_content_size(const int64& size) { expected_content_size_ = size; } // Whether we have processed the response for that request yet. bool has_response_started() const { return has_handled_response_; } // FilterContext methods: // These methods are not applicable to all connections. virtual bool GetMimeType(std::string* mime_type) const { return false; } virtual bool GetURL(GURL* gurl) const; virtual base::Time GetRequestTime() const; virtual bool IsCachedContent() const { return false; } virtual int64 GetByteReadCount() const; virtual int GetResponseCode() const { return -1; } virtual int GetInputStreamBufferSize() const { return kFilterBufSize; } virtual void RecordPacketStats(StatisticSelector statistic) const; protected: friend class base::RefCountedThreadSafe; virtual ~URLRequestJob(); // Notifies the job that headers have been received. void NotifyHeadersComplete(); // Notifies the request that the job has completed a Read operation. void NotifyReadComplete(int bytes_read); // Notifies the request that a start error has occurred. void NotifyStartError(const URLRequestStatus& status); // NotifyDone marks when we are done with a request. It is really // a glorified set_status, but also does internal state checking and // job tracking. It should be called once per request, when the job is // finished doing all IO. void NotifyDone(const URLRequestStatus& status); // Some work performed by NotifyDone must be completed on a separate task // so as to avoid re-entering the delegate. This method exists to perform // that work. void CompleteNotifyDone(); // Used as an asynchronous callback for Kill to notify the URLRequest that // we were canceled. void NotifyCanceled(); // Notifies the job the request should be restarted. // Should only be called if the job has not started a resposne. void NotifyRestartRequired(); // Called to read raw (pre-filtered) data from this Job. // If returning true, data was read from the job. buf will contain // the data, and bytes_read will receive the number of bytes read. // If returning true, and bytes_read is returned as 0, there is no // additional data to be read. // If returning false, an error occurred or an async IO is now pending. // If async IO is pending, the status of the request will be // URLRequestStatus::IO_PENDING, and buf must remain available until the // operation is completed. See comments on URLRequest::Read for more info. virtual bool ReadRawData(net::IOBuffer* buf, int buf_size, int *bytes_read); // Informs the filter that data has been read into its buffer void FilteredDataRead(int bytes_read); // Reads filtered data from the request. Returns true if successful, // false otherwise. Note, if there is not enough data received to // return data, this call can issue a new async IO request under // the hood. bool ReadFilteredData(int *bytes_read); // Facilitate histogramming by turning on packet counting. // If called more than once, the largest value will be used. void EnablePacketCounting(size_t max_packets_timed); // At or near destruction time, a derived class may request that the filters // be destroyed so that statistics can be gathered while the derived class is // still present to assist in calculations. This is used by URLRequestHttpJob // to get SDCH to emit stats. void DestroyFilters() { filter_.reset(); } // The request that initiated this job. This value MAY BE NULL if the // request was released by DetachRequest(). URLRequest* request_; // The status of the job. const URLRequestStatus GetStatus(); // Set the status of the job. void SetStatus(const URLRequestStatus& status); // Whether the job is doing performance profiling bool is_profiling_; // Contains IO performance measurement when profiling is enabled. scoped_ptr metrics_; private: // Size of filter input buffers used by this class. static const int kFilterBufSize; // When data filtering is enabled, this function is used to read data // for the filter. Returns true if raw data was read. Returns false if // an error occurred (or we are waiting for IO to complete). bool ReadRawDataForFilter(int *bytes_read); // Called in response to a redirect that was not canceled to follow the // redirect. The current job will be replaced with a new job loading the // given redirect destination. void FollowRedirect(const GURL& location, int http_status_code); // Updates the profiling info and notifies observers that bytes_read bytes // have been read. void RecordBytesRead(int bytes_read); // Called to query whether there is data available in the filter to be read // out. bool FilterHasData(); // Record packet arrival times for possible use in histograms. void UpdatePacketReadTimes(); // Indicates that the job is done producing data, either it has completed // all the data or an error has been encountered. Set exclusively by // NotifyDone so that it is kept in sync with the request. bool done_; // Cache the load flags from request_ because it might go away. int load_flags_; // The data stream filter which is enabled on demand. scoped_ptr filter_; // If the filter filled its output buffer, then there is a change that it // still has internal data to emit, and this flag is set. bool filter_needs_more_output_space_; // When we filter data, we receive data into the filter buffers. After // processing the filtered data, we return the data in the caller's buffer. // While the async IO is in progress, we save the user buffer here, and // when the IO completes, we fill this in. net::IOBuffer *read_buffer_; int read_buffer_len_; // Used by HandleResponseIfNecessary to track whether we've sent the // OnResponseStarted callback and potentially redirect callbacks as well. bool has_handled_response_; // Expected content size int64 expected_content_size_; // Set when a redirect is deferred. GURL deferred_redirect_url_; int deferred_redirect_status_code_; //---------------------------------------------------------------------------- // Data used for statistics gathering in some instances. This data is only // used for histograms etc., and is not required. It is optionally gathered // based on the settings of several control variables. // Enable recording of packet arrival times for histogramming. bool packet_timing_enabled_; // TODO(jar): improve the quality of the gathered info by gathering most times // at a lower point in the network stack, assuring we have actual packet // boundaries, rather than approximations. Also note that input byte count // as gathered here is post-SSL, and post-cache-fetch, and does not reflect // true packet arrival times in such cases. // Total number of bytes read from network (or cache) and and typically handed // to filter to process. Used to histogram compression ratios, and error // recovery scenarios in filters. int64 filter_input_byte_count_; // The number of bytes that have been accounted for in packets (where some of // those packets may possibly have had their time of arrival recorded). int64 bytes_observed_in_packets_; // Limit on the size of the array packet_times_. This can be set to // zero, and then no packet times will be gathered. size_t max_packets_timed_; // Arrival times for some of the first few packets. std::vector packet_times_; // The request time may not be available when we are being destroyed, so we // snapshot it early on. base::Time request_time_snapshot_; // Since we don't save all packet times in packet_times_, we save the // last time for use in histograms. base::Time final_packet_time_; // The count of the number of packets, some of which may not have been timed. // We're ignoring overflow, as 1430 x 2^31 is a LOT of bytes. int observed_packet_count_; DISALLOW_COPY_AND_ASSIGN(URLRequestJob); }; #endif // NET_URL_REQUEST_URL_REQUEST_JOB_H_