1216002 - JavaScript error: resource://gre/modules/RequestSyncService.jsm, line 228: TypeError: 'continue' called on an object that does not implement interface IDBCursor.

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Fernando Jiménez Moreno [:ferjm]]

I've seen this error in the logcat while launching B2G desktop.

Comment 1

[Boris Zbarsky [:bzbarsky]]

[Tracking Requested - why for this release]: Broken code in initialization function, so I doubt this class behaves correctly.

So the relevant code looks like this:


217   addRegistration: function(aObj, aCb) {
218     debug('addRegistration');
219 
220     let key = this.principalToKey(aObj.principal);
221     if (!(key in this._registrations)) {
222       this._registrations[key] = {};
223     }
224 
225     this.scheduleTimer(aObj, function() {
226       this._registrations[key][aObj.data.task] = aObj;
227       if (aCb) {
228         aCb();
229       }
230     }.bind(this));
231   },

which means the real bug is in the caller: someone called addRegistration and passed IDBCursor.prototype.continue as the second arg without binding it to the relevant IDBCursor.

At first glance, the relevant caller is http://mxr.mozilla.org/mozilla-central/source/dom/requestsync/RequestSyncService.jsm?rev=f2ae60fdfa6d#104 added in bug 1194978.

Comment 2

[Andrea Marchesini [:baku]]

Attachment: rs.patch

I can also do cursor.continue.bind(this) but I use this approach elsewhere in this file.

Comment 3

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 8675900 [details] [diff] [review]
rs.patch

r=me.  Or you could do:

  this.addRegistration(cursor.value, cursor.continue.bind(cursor));

either way.

Comment 4

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/9519653b7f2e

Comment 5

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/9519653b7f2e

Comment 6

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

why is comment 5 tagged "backout"?

Comment 7

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

Sorry, accidentally used a not-quite-ready Bugherder branch and it got confused and marked every single thing from that merge as a backout.

Comment 8

[Liz Henry (:lizzard) (relman/hg->git project)]

Tracking for 43+ since this seems to have been a regression from bug 1194978. 
baku, do you want to request uplift to 43?

Comment 9

[Andrea Marchesini [:baku]]

Comment on attachment 8675900 [details] [diff] [review]
rs.patch

Approval Request Comment
[Feature/regressing bug #]: RequestSync
[User impact if declined]: a broken API
[Describe test coverage new/current, TreeHerder]: no tests.
[Risks and why]: no risk. The patch changes the scope in the use of a IDBCursor.
[String/UUID change made/needed]: none

Comment 10

[Liz Henry (:lizzard) (relman/hg->git project)]

Comment on attachment 8675900 [details] [diff] [review]
rs.patch

Minor fix to recent API regression, ok on m-c. Please uplift to aurora.

Comment 11

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/releases/mozilla-aurora/rev/e9f816591041