Introduce AppCacheResponseMetadataWriter to support metadata caching.

Introduce AppCacheResponseMetadataWriter to support metadata caching.

We are implementing V8 code caching for ServiceWorker script.

V8 code caching for the normal script resources which are loaded with <script> tag was implemented. (crbug.com/399580)
The code cache is stored in HttpCache as a metadata of the original script.
But ServiceWorker's script is stored not in HttpCache but in ServiceWorkerStorage.

To support V8 code caching for ServiceWorker script, we have to implement metadata caching mechanism in ServiceWorkerStorage.
ServiceWorkerStorage are using AppCache's storage backend.

So this cl introduces metadata caching mechanism in AppCache's storage backend.

To store the metadata we are using "stream 2" of the DiskCache entry.
This is the same way HttpCache uses in HttpCache::Transaction::WriteMetadata().

TEST=content_unittestes --gtest_filter=AppCacheResponseTest.*
BUG=449895
Committed: https://crrev.com/a95000738bc5cfee057bf235c071923c923a1e0f
Cr-Commit-Position: refs/heads/master@{#316772}

[horo, 2015-02-16 18:18:17]

Patchset #1 (id:1) has been deleted

[horo, 2015-02-16 18:19:13]

horo@chromium.org changed reviewers:
+ vogelheim@chromium.org

[horo, 2015-02-16 18:19:14]

vogelheim@
Could you please review this?

[horo, 2015-02-17 08:18:35]

horo@chromium.org changed reviewers:
+ michaeln@chromium.org

[horo, 2015-02-17 08:18:35]

michaeln@
Could you please review this?

[vogelheim, 2015-02-17 10:36:07]

On 2015/02/16 18:19:14, horo wrote:
> vogelheim@
> Could you please review this?

I'd love to help, but I don't think I'm the right reviewer for this. I looked
through the patch and didn't see anything wrong; but since I've never worked on
this part of Chromium I just don't have a good understanding of whether this
does the right thing or not.

[vogelheim, 2015-02-17 10:36:22]

vogelheim@chromium.org changed reviewers:
- vogelheim@chromium.org

[michaeln, 2015-02-18 01:32:25]

On 2015/02/17 08:18:35, horo wrote:
> michaeln@
> Could you please review this?

yup, looking...

[michaeln, 2015-02-18 02:18:59]

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
File content/browser/appcache/appcache_response.cc (right):

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
content/browser/appcache/appcache_response.cc:476: void
AppCacheResponseMetadataWriter::OpenEntry() {
It'd be nice to hoist this method and the following into the base class so they
can be shared with the reader. Wdyt?

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
content/browser/appcache/appcache_response.cc:505:
WriteRaw(kResponseMetadataIndex, 0, buffer_.get(), write_amount_);
and have it call a virtual method here to continue in the success case

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
File content/browser/appcache/appcache_response.h (right):

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
content/browser/appcache/appcache_response.h:298: bool started_;
is this used?

[horo, 2015-02-18 03:33:44]

Thank you!

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
File content/browser/appcache/appcache_response.cc (right):

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
content/browser/appcache/appcache_response.cc:476: void
AppCacheResponseMetadataWriter::OpenEntry() {
On 2015/02/18 02:18:59, michaeln wrote:
> It'd be nice to hoist this method and the following into the base class so
they
> can be shared with the reader. Wdyt?

Sounds good!
done.

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
content/browser/appcache/appcache_response.cc:505:
WriteRaw(kResponseMetadataIndex, 0, buffer_.get(), write_amount_);
On 2015/02/18 02:18:59, michaeln wrote:
> and have it call a virtual method here to continue in the success case

Done.

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
File content/browser/appcache/appcache_response.h (right):

https://codereview.chromium.org/932643002/diff/20001/content/browser/appcache...
content/browser/appcache/appcache_response.h:298: bool started_;
On 2015/02/18 02:18:59, michaeln wrote:
> is this used?

deleted

[michaeln, 2015-02-18 03:48:55]

lgtm, thnx

[horo, 2015-02-18 04:08:56]

The CQ bit was checked by horo@chromium.org

[commit-bot: I haz the power, 2015-02-18 04:09:22]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/932643002/40001

[commit-bot: I haz the power, 2015-02-18 05:08:21]

Committed patchset #2 (id:40001)

[commit-bot: I haz the power, 2015-02-18 05:09:12]

Patchset 2 (id:??) landed as
https://crrev.com/a95000738bc5cfee057bf235c071923c923a1e0f
Cr-Commit-Position: refs/heads/master@{#316772}