Bug 96444

Summary: IndexedDB: fire upgradeneeded even without an explicit integer version
Product: WebKit Reporter: David Grogan <dgrogan>
Component: New BugsAssignee: David Grogan <dgrogan>
Status: RESOLVED FIXED    
Severity: Normal CC: alecflett, jsbell, tony, webkit.review.bot
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Bug Depends on: 90411    
Bug Blocks: 96548, 96952    
Attachments:
Description Flags
Patch
none
Patch
none
Patch
none
Patch
none
Patch
none
Patch
none
Patch
none
ToT none

Description David Grogan 2012-09-11 16:06:03 PDT
IndexedDB: fire upgradeneeded even without an explicit integer version
Comment 1 David Grogan 2012-09-11 16:06:30 PDT
Created attachment 163468 [details]
Patch
Comment 2 David Grogan 2012-09-11 16:08:08 PDT
Created attachment 163469 [details]
Patch
Comment 3 David Grogan 2012-09-11 17:32:49 PDT
Created attachment 163488 [details]
Patch
Comment 4 David Grogan 2012-09-11 18:41:34 PDT
Created attachment 163492 [details]
Patch
Comment 5 David Grogan 2012-09-11 18:47:49 PDT
With this patch we do indeed pass the w3 tests Josh separated out.

Except one, http://w3c-test.org/webapps/IndexedDB/tests/submissions/Ms2ger/idbfactory_open9.htm .  It fails because of a different int-version issue.

This patch doesn't apply to ToT because it is based on 90411.
Comment 6 David Grogan 2012-09-12 15:59:26 PDT
Created attachment 163725 [details]
Patch
Comment 7 David Grogan 2012-09-12 16:03:56 PDT
jsbell@, could you take a look at this? As we predicted there's not much code change, mostly tests.
Comment 8 Joshua Bell 2012-09-13 11:29:27 PDT
Comment on attachment 163725 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=163725&action=review

LGTM!

> Source/WebCore/Modules/indexeddb/IDBLevelDBBackingStore.cpp:-342
> -    // FIXME: Change this to strictly greater than 0 once we throw TypeError for

This patch does not alter the (erroneous) behavior for open(name, -1), correct? It's still an alias for open(name), and that will be fixed (on the front end) in a separate bug?
Comment 9 David Grogan 2012-09-13 18:18:58 PDT
Created attachment 164020 [details]
Patch
Comment 10 WebKit Review Bot 2012-09-13 18:22:01 PDT
Attachment 164020 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'LayoutTests/ChangeLog', u'LayoutTests/http..." exit_code: 1
Source/WebCore/ChangeLog:9:  Need whitespace between colon and description  [changelog/filechangedescriptionwhitespace] [5]
Source/WebCore/ChangeLog:11:  Need whitespace between colon and description  [changelog/filechangedescriptionwhitespace] [5]
Total errors found: 2 in 30 files


If any of these errors are false positives, please file a bug against check-webkit-style.
Comment 11 David Grogan 2012-09-13 19:32:30 PDT
Created attachment 164030 [details]
Patch
Comment 12 David Grogan 2012-09-13 19:39:32 PDT
Comment on attachment 163725 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=163725&action=review

>> Source/WebCore/Modules/indexeddb/IDBLevelDBBackingStore.cpp:-342
>> -    // FIXME: Change this to strictly greater than 0 once we throw TypeError for
> 
> This patch does not alter the (erroneous) behavior for open(name, -1), correct? It's still an alias for open(name), and that will be fixed (on the front end) in a separate bug?

That's right.
Comment 13 David Grogan 2012-09-17 15:31:59 PDT
Created attachment 164462 [details]
ToT
Comment 14 David Grogan 2012-09-17 15:33:43 PDT
Tony, could you review this?  I'm not going to land it until tomorrow so it's not urgent at all.
Comment 15 David Grogan 2012-09-17 15:34:05 PDT
Tony, could you review this?  I'm not going to land it until tomorrow so it's not urgent at all.
Comment 16 WebKit Review Bot 2012-09-19 13:28:12 PDT
Comment on attachment 164462 [details]
ToT

Clearing flags on attachment: 164462

Committed r129037: <http://trac.webkit.org/changeset/129037>
Comment 17 WebKit Review Bot 2012-09-19 13:28:16 PDT
All reviewed patches have been landed.  Closing bug.