RESOLVED FIXED 102267
IndexedDB: Test conversions batch 4
https://bugs.webkit.org/show_bug.cgi?id=102267
Summary IndexedDB: Test conversions batch 4
David Grogan
Reported 2012-11-14 12:15:58 PST
IndexedDB: Test conversions batch 4
Attachments
Patch (63.56 KB, patch)
2012-11-14 12:34 PST, David Grogan
no flags
Patch (65.96 KB, patch)
2012-11-14 16:47 PST, David Grogan
no flags
David Grogan
Comment 1 2012-11-14 12:34:42 PST
David Grogan
Comment 2 2012-11-14 12:36:42 PST
Alec or Josh, could you take a look at this?
Joshua Bell
Comment 3 2012-11-14 15:15:11 PST
Comment on attachment 174225 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=174225&action=review lgtm > LayoutTests/storage/indexeddb/lazy-index-population-expected.txt:47 > +FIXME: Script shouldn't have to explicitly close connection after it failed to open. http://crbug.com/161066 Does this repro in DRT? If so, a WebKit bug reference would be more appropriate I think. > LayoutTests/storage/indexeddb/tutorial.html:-1 > -<html><title>IndexedDB Tutorial</title> Being paranoid, I'm going to point out places where this test may have been doing unique things... > LayoutTests/storage/indexeddb/tutorial.html:-122 > - while (db.objectStoreNames.length) Deleting all the existing object stores by referencing db.objectStoreNames.length / db.objectStoreNames[0] appears to only be done here and in the deleteAllObjectStores() function in shared.js, and this overall refactor to use indexedDBTest is removing most calls to it. We should probably keep one around. > LayoutTests/storage/indexeddb/tutorial.html:-307 > - // Note that multiple objectStore (or index) method calls will return the same object, within This should be validated by mozilla/object-identity.html but we may want to add setting a property on the object.
Joshua Bell
Comment 4 2012-11-14 15:16:42 PST
Comment on attachment 174225 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=174225&action=review > LayoutTests/ChangeLog:11 > + I think tutorial.html has outlived its usefulness now that there are ... and I agree with removing tutorial.html
David Grogan
Comment 5 2012-11-14 16:43:10 PST
Comment on attachment 174225 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=174225&action=review >> LayoutTests/storage/indexeddb/lazy-index-population-expected.txt:47 >> +FIXME: Script shouldn't have to explicitly close connection after it failed to open. http://crbug.com/161066 > > Does this repro in DRT? If so, a WebKit bug reference would be more appropriate I think. Done. >> LayoutTests/storage/indexeddb/tutorial.html:-122 >> - while (db.objectStoreNames.length) > > Deleting all the existing object stores by referencing db.objectStoreNames.length / db.objectStoreNames[0] appears to only be done here and in the deleteAllObjectStores() function in shared.js, and this overall refactor to use indexedDBTest is removing most calls to it. We should probably keep one around. Good point. I'd bet deleteAllObjectStores doesn't go away anytime soon. Note added to shared.js though. >> LayoutTests/storage/indexeddb/tutorial.html:-307 >> - // Note that multiple objectStore (or index) method calls will return the same object, within > > This should be validated by mozilla/object-identity.html but we may want to add setting a property on the object. Added.
David Grogan
Comment 6 2012-11-14 16:47:03 PST
David Grogan
Comment 7 2012-11-14 16:48:14 PST
Tony, could you review this batch of tests?
WebKit Review Bot
Comment 8 2012-11-14 17:19:34 PST
Comment on attachment 174285 [details] Patch Clearing flags on attachment: 174285 Committed r134700: <http://trac.webkit.org/changeset/134700>
WebKit Review Bot
Comment 9 2012-11-14 17:19:38 PST
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.