Shrink space required by history log file

Shrink space required by history log file

1) Use seconds-since-epoch in log record keys, instead of full date strings.
2) Add support for inflating/deflating urls.
3) Deflate URLS prior to storaing in log file.
4) Remove obsolete code for reloading history after remote changes.

BUG=461176
TEST=browser_test: FileManagerJsTest.*

Committed: https://crrev.com/1f0ca0da1b05c4ac1781063fcf7cfbb195e2a606
Cr-Commit-Position: refs/heads/master@{#317965}

[Steve McKay, 2015-02-24 02:38:46]

Log entries now look like:

COPY RECORD:
[0,"1400539064_464768","google-drive","$/removable/USB%20Drive/DCIM/robbieshade2.jpg","$/drive-ebbe9ae388f1daaf9b4260a9ff64b62252e473f1/root/Chrome%20OS%20Cloud%20backup/2015-02-23/robbieshade2.jpg"],

IMPORT RECORD:
[1,"1400539060_690154","google-drive"],

[Steve McKay, 2015-02-25 00:07:57]

smckay@chromium.org changed reviewers:
+ mtomasz@chromium.org

[Steve McKay, 2015-02-25 00:07:58]

Dude. I totally thought I mailed this last night. Today, I was like "What's with
Tomasz?". Now I'm like, "what's with me?"

[mtomasz, 2015-02-25 00:40:17]

lgtm with nits

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
File ui/file_manager/file_manager/background/js/import_history.js (right):

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/background/js/import_history.js:429: var url =
importer.inflateAppUrl(
We're losing backward compatibility. What will happen to pictures imported by
users on dev channel? Are we OK with it?

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
File ui/file_manager/file_manager/common/js/importer_common.js (right):

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/common/js/importer_common.js:394: * Reconstitutes a
url previous deflated by {@code deflateAppUrl}.
nit: Can you add a comment what it returns if it wasn't deflated previously by
deflateAppUrl?

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/common/js/importer_common.js:452: return
String(Date.parse(date)).substring(0, 10);
nit: This will not work as intended for pictures created in 1999. I still have
old sd cards from that period. I had my first digital camera in 1997. Can we
trim last 3 digits instead of taking first 10?

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
File ui/file_manager/file_manager/common/js/importer_common_unittest.js (right):

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/common/js/importer_common_unittest.js:232:
assertTrue(deflated.substring(0, 1) === '$');
nit: assertEquals can be used for better logging.

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/common/js/importer_common_unittest.js:238:
assertEquals(importer.inflateAppUrl(deflated), url);
nit: Opposite order of arguments. Should be assertEquals(expected, actual)

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
File ui/file_manager/file_manager/foreground/js/file_manager.js (right):

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/foreground/js/file_manager.js:810: if
(!event.entry) {
nit: Can we add a comment when can it happen that entry is not passed?

[Steve McKay, 2015-02-25 01:01:25]

Respond to review comments. Also remove a little more unused sync related code.

[Steve McKay, 2015-02-25 01:31:51]

Respond to review comments.

[Steve McKay, 2015-02-25 01:32:12]

Patchset #3 (id:40001) has been deleted

[Steve McKay, 2015-02-25 01:49:39]

Simplify some of the history initialization complexity added in earlier patch.

[Steve McKay, 2015-02-25 01:51:57]

Dones. Made a few more small improvements. Please take another look.

If you're happy, please do hit the commit checkbox for me.

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
File ui/file_manager/file_manager/background/js/import_history.js (right):

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/background/js/import_history.js:429: var url =
importer.inflateAppUrl(
On 2015/02/25 00:40:17, mtomasz wrote:
> We're losing backward compatibility. What will happen to pictures imported by
> users on dev channel? Are we OK with it?

They'll still have the pictures, but they'll have to be deduped as part of
content dedupe.

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
File ui/file_manager/file_manager/common/js/importer_common.js (right):

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/common/js/importer_common.js:394: * Reconstitutes a
url previous deflated by {@code deflateAppUrl}.
On 2015/02/25 00:40:17, mtomasz wrote:
> nit: Can you add a comment what it returns if it wasn't deflated previously by
> deflateAppUrl?

Done.

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/common/js/importer_common.js:452: return
String(Date.parse(date)).substring(0, 10);
On 2015/02/25 00:40:17, mtomasz wrote:
> nit: This will not work as intended for pictures created in 1999. I still have
> old sd cards from that period. I had my first digital camera in 1997. Can we
> trim last 3 digits instead of taking first 10?

Oh, really good catch. Fixed.

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
File ui/file_manager/file_manager/common/js/importer_common_unittest.js (right):

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/common/js/importer_common_unittest.js:232:
assertTrue(deflated.substring(0, 1) === '$');
On 2015/02/25 00:40:17, mtomasz wrote:
> nit: assertEquals can be used for better logging.

Done.

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/common/js/importer_common_unittest.js:238:
assertEquals(importer.inflateAppUrl(deflated), url);
On 2015/02/25 00:40:17, mtomasz wrote:
> nit: Opposite order of arguments. Should be assertEquals(expected, actual)

Done.

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
File ui/file_manager/file_manager/foreground/js/file_manager.js (right):

https://codereview.chromium.org/954583004/diff/20001/ui/file_manager/file_man...
ui/file_manager/file_manager/foreground/js/file_manager.js:810: if
(!event.entry) {
On 2015/02/25 00:40:17, mtomasz wrote:
> nit: Can we add a comment when can it happen that entry is not passed?

Actually it should never be null. I might have added this when investigating
other issues. If we see problems in the future, we'll revisit adding
protections. For now, reverted.

[mtomasz, 2015-02-25 02:02:40]

lgtm with nit

https://codereview.chromium.org/954583004/diff/80001/ui/file_manager/file_man...
File ui/file_manager/file_manager/background/js/import_history.js (right):

https://codereview.chromium.org/954583004/diff/80001/ui/file_manager/file_man...
ui/file_manager/file_manager/background/js/import_history.js:256: return /**
@type {!Promise.<!importer.ImportHistory>} */ (this.whenReady_);
nit: Is this cast necessary?

[Steve McKay, 2015-02-25 02:08:01]

https://codereview.chromium.org/954583004/diff/80001/ui/file_manager/file_man...
File ui/file_manager/file_manager/background/js/import_history.js (right):

https://codereview.chromium.org/954583004/diff/80001/ui/file_manager/file_man...
ui/file_manager/file_manager/background/js/import_history.js:256: return /**
@type {!Promise.<!importer.ImportHistory>} */ (this.whenReady_);
On 2015/02/25 02:02:39, mtomasz wrote:
> nit: Is this cast necessary?

Yeah. The compiler complains if parameterized types don't match exactly...even
when the runtime type implements/extends the declared type.

[Steve McKay, 2015-02-25 02:08:13]

The CQ bit was checked by smckay@chromium.org

[commit-bot: I haz the power, 2015-02-25 02:08:49]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/954583004/80001

[commit-bot: I haz the power, 2015-02-25 03:21:09]

Committed patchset #4 (id:80001)

[commit-bot: I haz the power, 2015-02-25 03:21:53]

Patchset 4 (id:??) landed as
https://crrev.com/1f0ca0da1b05c4ac1781063fcf7cfbb195e2a606
Cr-Commit-Position: refs/heads/master@{#317965}