[fsp] Refactor FakeProvidedFileSystem to be more flexible.

[fsp] Refactor FakeProvidedFileSystem to be more flexible.

Before, the FakeProvidedFileSystem had some hardcoded responses. However, since
FileStreamWriter is coming soon, we will need to verify if the file on the
file system is changed correctly.

For that, FakeProvidedFileSystem has been refactored, so it is possible to add
and get fake files and verify their values.

TEST=unit_tests: *FileSystemProvider*
BUG=391362
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=283538

[mtomasz, 2014-07-16 06:08:17]

@hirono: PTAL.

This is just a refactoring task. Previously responses in FakeProvidedFileSystem
were hardcoded. This patch allows to add fake entries and check their values.

Changes in file_stream_reader.cc are just replaces of removed constants with
values from FakeProvidedFileSystem::GetFile().
Eg.

Old:
kFakeFileSize

New:
FakeEntry entry;
FakeProvidedFileSystem::GetFile(kFakeFilePath, &entry)
entry.metadata.size

This feature will be especially needed in the upcoming FileStreamWriter, where
we need to verify if contents of the file on the FakeProvidedFileSystem are
changed after performing a write operation. Also, it will allow to increase test
coverage in provider_async_file_util.cc.

Thanks.

[hirono, 2014-07-16 06:30:00]

lgtm with nits.

https://codereview.chromium.org/393233002/diff/1/chrome/browser/chromeos/file...
File chrome/browser/chromeos/file_system_provider/fake_provided_file_system.cc
(right):

https://codereview.chromium.org/393233002/diff/1/chrome/browser/chromeos/file...
chrome/browser/chromeos/file_system_provider/fake_provided_file_system.cc:70:
Entries::const_iterator entry_it = entries_.find(entry_path);
nit: const Entries::const_iterator

https://codereview.chromium.org/393233002/diff/1/chrome/browser/chromeos/file...
chrome/browser/chromeos/file_system_provider/fake_provided_file_system.cc:228:
// TODO(mtomasz): Implement it once needed.
Cannot we use NOTIMPLEMENTED()?

https://codereview.chromium.org/393233002/diff/1/chrome/browser/chromeos/file...
File
chrome/browser/chromeos/file_system_provider/fileapi/file_stream_reader_unittest.cc
(right):

https://codereview.chromium.org/393233002/diff/1/chrome/browser/chromeos/file...
chrome/browser/chromeos/file_system_provider/fileapi/file_stream_reader_unittest.cc:104:
FakeProvidedFileSystem* provided_file_system =
nit: FakeProvidedFileSystem* const

[mtomasz, 2014-07-16 07:29:15]

https://codereview.chromium.org/393233002/diff/1/chrome/browser/chromeos/file...
File chrome/browser/chromeos/file_system_provider/fake_provided_file_system.cc
(right):

https://codereview.chromium.org/393233002/diff/1/chrome/browser/chromeos/file...
chrome/browser/chromeos/file_system_provider/fake_provided_file_system.cc:70:
Entries::const_iterator entry_it = entries_.find(entry_path);
On 2014/07/16 06:30:00, hirono wrote:
> nit: const Entries::const_iterator

Done.

https://codereview.chromium.org/393233002/diff/1/chrome/browser/chromeos/file...
chrome/browser/chromeos/file_system_provider/fake_provided_file_system.cc:228:
// TODO(mtomasz): Implement it once needed.
On 2014/07/16 06:30:00, hirono wrote:
> Cannot we use NOTIMPLEMENTED()?

This is implemented, but the implementation is super simplified. We use it, but
we don't verify if the directory got actually created.
https://code.google.com/p/chromium/codesearch#chromium/src/chrome/browser/chr...

I have plans to expand the test coverage in
provider_async_file_util_unittest.cc, and then we will have to replace that
blind success call, with checking if the entry already exists, then
CreateEntry().

[mtomasz, 2014-07-16 09:36:20]

The CQ bit was checked by mtomasz@chromium.org

[commit-bot: I haz the power, 2014-07-16 09:38:36]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mtomasz@chromium.org/393233002/20001

[commit-bot: I haz the power, 2014-07-16 22:08:37]

Change committed as 283538