Bug 151752

Summary:

Modern IDB: IDBTransaction::error is not exposed

Product:

WebKit

Reporter:

Brady Eidson <beidson>

Component:

WebCore Misc.

Assignee:

Brady Eidson <beidson>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

achristensen, alecflett, commit-queue, jsbell

Priority:

Version:

Safari 9

Hardware:

Unspecified

OS:

Unspecified

Bug Depends on:

151753

Bug Blocks:

149117, 150882

Attachments:

Description	Flags
Patch v1	none

Brady Eidson

Reported 2015-12-02 10:12:43 PST

Modern IDB: IDBTransaction::error is not exposed There's at least 3 layout tests that hit this, so I'll get all three of those running.

Attachments
Patch v1 (12.44 KB, patch) 2015-12-02 11:40 PST, Brady Eidson	no flags	Details Formatted Diff Diff
View All Add attachment proposed patch, testcase, etc.

Brady Eidson

Comment 1 2015-12-02 11:40:14 PST

Created attachment 266458 [details] Patch v1

WebKit Commit Bot

Comment 2 2015-12-02 12:32:53 PST

Comment on attachment 266458 [details] Patch v1 Clearing flags on attachment: 266458 Committed r192956: <http://trac.webkit.org/changeset/192956>

WebKit Commit Bot

Comment 3 2015-12-02 12:32:55 PST

All reviewed patches have been landed. Closing bug.

Darin Adler

Comment 4 2015-12-02 17:01:37 PST

Comment on attachment 266458 [details] Patch v1 View in context: https://bugs.webkit.org/attachment.cgi?id=266458&action=review > Source/WebCore/Modules/indexeddb/client/IDBTransactionImpl.cpp:119 > RefPtr<DOMError> IDBTransaction::error() const Why does this return a RefPtr? Should just return a DOMError*, not a RefPtr. Unless it manufactures a new error object.

Brady Eidson

Comment 5 2015-12-02 20:44:15 PST

(In reply to comment #4) > Comment on attachment 266458 [details] > Patch v1 > > View in context: > https://bugs.webkit.org/attachment.cgi?id=266458&action=review > > > Source/WebCore/Modules/indexeddb/client/IDBTransactionImpl.cpp:119 > > RefPtr<DOMError> IDBTransaction::error() const > > Why does this return a RefPtr? Should just return a DOMError*, not a RefPtr. > Unless it manufactures a new error object. All of the abstract bindings left behind from Legacy IDB inappropriately return RefPtrs like this. And, sadly, sometimes LegacyIDB does (incorrectly) manufacture a new object each call instead of retaining the same one for repeated calls. Instead of trying to clean up that mess while both implementations live side by side, I consider fixing it to be part of "Remove LegacyIDB"

Note You need to log in before you can comment on or make changes to this bug.