624230 - Error: 0x80070057 (NS_ERROR_ILLEGAL_VALUE) [nsINavHistoryService.executeQueries]

Status: RESOLVED FIXED

(Firefox :: Bookmarks & History, defect)

Description

[ithinc]

User-Agent:       Mozilla/5.0 (Windows NT 5.1; rv:2.0b9pre) Gecko/20110108 Firefox/4.0b9pre
Build Identifier: Mozilla/5.0 (Windows NT 5.1; rv:2.0b9pre) Gecko/20110108 Firefox/4.0b9pre

Error: Component returned failure code: 0x80070057 (NS_ERROR_ILLEGAL_VALUE) [nsINavHistoryService.executeQueries]
Source file: chrome://browser/content/places/browserPlacesViews.js
Line: 83

Please try ... catch the executeQueries for invalid query, like what is done in Firefox 3.6.


Reproducible: Always

Steps to Reproduce:
Execute the following in Javascript shell:

if (PlacesToolbarHelper._viewElt._placesView)
  PlacesToolbarHelper._viewElt.controllers
  .removeController(PlacesToolbarHelper._viewElt._placesView._controller);

PlacesToolbarHelper.customizeStart();
PlacesToolbarHelper._place = "place:folder=-1";
PlacesToolbarHelper.customizeDone();

Comment 1

[Shawn Wilsher :sdwilsh]

I'm not sure we actually care about random queries that are not generated by us being forced to display here.  Normal code paths that a user would hit are unit tested.

Comment 2

[Rickfan]

it's ok on ff 4.0 beta 11,but it's come out on ff 4.0 beta 12

Error: uncaught exception: addToLocal: [Exception... "Component returned failure code: 0x80070057 (NS_ERROR_ILLEGAL_VALUE) [nsINavHistoryService.executeQuery]"  nsresult: "0x80070057 (NS_ERROR_ILLEGAL_VALUE)"  location: "JS frame :: resource://gre/modules/PlacesUtils.jsm :: PU_getFolderContents :: line 790"  data: no]

the code is:
var historyService = PlacesUtils.history; 
var options = historyService.getNewQueryOptions();
var query = historyService.getNewQuery();
query.setFolders([folder], 1);
historyService.executeQuery(query, options);

Comment 3

[Marco Bonardo [:mak]]

sounds correct to me, if folder doesn't exist it should throw

Comment 4

[Rickfan]

(In reply to comment #3)
> sounds correct to me, if folder doesn't exist it should throw

Sorry, it's my fault, thanks

Comment 5

[ithinc]

This seems resolved in the nightly build, maybe by bug 641074?

Comment 6

[Marco Bonardo [:mak]]

Yes, that uniformed the behavior across invalid queries and invalid folder shortcuts. Indeed this case was throwing only for an invalid shortcut, but not for any invalid query, as I initially thought.