[fsp] Let extensions decide about the file system id.

[fsp] Let extensions decide about the file system id.

Previously, providing extensions were mounting virtual file systems by passing
the display name only. As a result, there was no easy way to detect, if the
virtual file system is already mounted.

This is a big issue for event pages, and extensions providing a remote file
system (not archive via file handlers). In such situation, they have to mount
the virtual file system as soon as the providing extension is launched.
However, the extension may be suspended (event page), and then launched again
once a request is emited to it. In such situation, the providing extension
would mount the file system again, and cause duplicated entries in Files app.

To prevent that, providing extensions need to be able to detect if they already
mounted the virtual file system.

The current solution let's extensions mount after each startup, but it would
fail quietly if a file system with the same id is already mounted.

Alternative would be creating a method which would enumerate already registered
file systems by the extension, but that would be an overkill.

Note, that this strategy is same as for context menus api:
https://developer.chrome.com/extensions/contextMenus#method-create

TEST=unit_tests: *FileSystemProvider*, browser_tests: *FileSystemProvider*
BUG=248427
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=272232

[mtomasz, 2014-05-21 08:59:29]

@kinaba: PTAL at C++.
@hirono: PTAL at JS.
@benwells: PAT at IDL.

This CL is mostly mechanical:

1. FileSystemId is now determined by providing extensions, not by file system
provider api.

2. FileSystemId is now a string, not a number.

3. For mount paths, an auto incremented number (index) is used, instead of a
file system id, to avoid exploitation (since fileSystemId is a user provided
string).

Thanks.

[hirono, 2014-05-21 10:09:53]

https://codereview.chromium.org/294073007/diff/1/chrome/test/data/extensions/...
File chrome/test/data/extensions/api_test/file_system_provider/read_file/test.js
(right):

https://codereview.chromium.org/294073007/diff/1/chrome/test/data/extensions/...
chrome/test/data/extensions/api_test/file_system_provider/read_file/test.js:93:
* @param {number} inFileSystemId ID of the file system.
nit: string
Same for followings.

[mtomasz, 2014-05-22 01:18:36]

I've simplified the code by removing:

> 3. For mount paths, an auto incremented number (index) is used, instead of a
> file system id, to avoid exploitation (since fileSystemId is a user provided
> string).

The file system id is not escaped and used for creating the mount path safely.
This makes the code simpler, and URLs persistent, which may be critical eg.
for creating shortcuts in Files app.

PTAL.

[mtomasz, 2014-05-22 02:16:54]

https://codereview.chromium.org/294073007/diff/1/chrome/test/data/extensions/...
File chrome/test/data/extensions/api_test/file_system_provider/read_file/test.js
(right):

https://codereview.chromium.org/294073007/diff/1/chrome/test/data/extensions/...
chrome/test/data/extensions/api_test/file_system_provider/read_file/test.js:93:
* @param {number} inFileSystemId ID of the file system.
On 2014/05/21 10:09:54, hirono wrote:
> nit: string
> Same for followings.

Done.

[hirono, 2014-05-22 03:46:12]

lgtm with nits.

https://codereview.chromium.org/294073007/diff/40001/chrome/common/extensions...
File chrome/common/extensions/api/file_browser_private.idl (right):

https://codereview.chromium.org/294073007/diff/40001/chrome/common/extensions...
chrome/common/extensions/api/file_browser_private.idl:237: // Id the provided
file system (for proviided file systems).
typo: proviided
nit: Id the -> Id of the? I'm not sure.

[benwells, 2014-05-22 03:58:58]

lgtm

[kinaba, 2014-05-22 05:23:32]

lgtm

[mtomasz, 2014-05-22 07:04:22]

The CQ bit was checked by mtomasz@chromium.org

[commit-bot: I haz the power, 2014-05-22 07:06:31]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mtomasz@chromium.org/294073007/60001

[mtomasz, 2014-05-22 12:53:04]

The CQ bit was unchecked by mtomasz@chromium.org

[mtomasz, 2014-05-22 12:53:10]

The CQ bit was checked by mtomasz@chromium.org

[commit-bot: I haz the power, 2014-05-22 12:54:50]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mtomasz@chromium.org/294073007/60001

[commit-bot: I haz the power, 2014-05-22 18:06:41]

Change committed as 272232