Implement updateViaCache flag and no-cache by default for main service worker scripts

content/public/browser/service_worker_context.h

 // Copyright 2014 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 CONTENT_PUBLIC_BROWSER_SERVICE_WORKER_CONTEXT_H_
 #define CONTENT_PUBLIC_BROWSER_SERVICE_WORKER_CONTEXT_H_
 
 #include <set>
 #include <string>
 
@@ skipping 34 matching lines @@
   // Registers the header name which should not be passed to the ServiceWorker.
   // Must be called from the IO thread.
   CONTENT_EXPORT static void AddExcludedHeadersForFetchEvent(
       const std::set<std::string>& header_names);
 
   // Returns true if the header name should not be passed to the ServiceWorker.
   // Must be called from the IO thread.
   static bool IsExcludedHeaderNameForFetchEvent(const std::string& header_name);
 
   // Equivalent to calling navigator.serviceWorker.register(script_url, {scope:
-  // pattern}) from a renderer, except that |pattern| is an absolute URL instead
-  // of relative to some current origin.  |callback| is passed true when the JS
-  // promise is fulfilled or false when the JS promise is rejected.
+  // pattern, bool: use_cache}) from a renderer, except that |pattern| is an

[nhiroki, 2017/03/24 00:33:42]

"bool: use_cache" => "useCache: use_cache" ?

[yuryu, 2017/03/28 05:08:56]

Done.

+  // absolute URL instead of relative to some current origin.  |callback| is
+  // passed true when the JS promise is fulfilled or false when the JS promise
+  // is rejected.
   //
   // The registration can fail if:
   //  * |script_url| is on a different origin from |pattern|
   //  * Fetching |script_url| fails.
   //  * |script_url| fails to parse or its top-level execution fails.
   //    TODO: The error message for this needs to be available to developers.
   //  * Something unexpected goes wrong, like a renderer crash or a full disk.
   //
   // This function can be called from any thread, but the callback will always
   // be called on the UI thread.
   virtual void RegisterServiceWorker(const Scope& pattern,
                                      const GURL& script_url,
+                                     bool use_cache,

[falken, 2017/03/24 04:06:19]

Do we need to expose this use_cache to ServiceWorkerContext, or is it possible
to be only on ServiceWorkerContextCore? We may be able to save some callsites
passing false if it's just ContextCore that needs it.

Also, seems better if we can use ServiceWorkerRegistrationOptions here instead
of the individual paramenters.

[yuryu, 2017/03/24 04:45:55]

https://cs.chromium.org/chromium/src/content/browser/service_worker/link_head...

link_header_support.c uses ServiceWorkerContext, and it also needs control of
useCache (and WorkerType in the future). So I think we need to change it to use
ServiceWorkerRegistrationOptions too. Am I correct?

[falken, 2017/03/27 06:15:37]

Yes, you're right.

[yuryu, 2017/03/28 05:08:56]

Acknowledged.

                                      const ResultCallback& callback) = 0;
 
   // Mechanism for embedder to increment/decrement ref count of a service
   // worker.
   // Embedders can call StartingExternalRequest() while it is performing some
   // work with the worker. The worker is considered to be working until embedder
   // calls FinishedExternalRequest(). This ensures that content/ does not
   // shut the worker down while embedder is expecting the worker to be kept
   // alive.
   //
@@ skipping 76 matching lines @@
       const ResultCallback& callback) = 0;
 
  protected:
   ServiceWorkerContext() {}
   virtual ~ServiceWorkerContext() {}
 };
 
 }  // namespace content
 
 #endif  // CONTENT_PUBLIC_BROWSER_SERVICE_WORKER_CONTEXT_H_