Closed Bug 595573 Opened 16 years ago Closed 15 years ago

Extensions uninstalled after browser restart, due to bug 533038

Categories

(Toolkit :: Add-ons Manager, defect)

defect
Not set
major

Tracking

()

VERIFIED FIXED
mozilla2.0b7
Tracking Status
blocking2.0 --- beta7+

People

(Reporter: fehe, Assigned: mwu)

References

Details

(Keywords: regression)

Attachments

(1 file)

User-Agent: Mozilla/5.0 (Windows NT 5.1; rv:2.0b6pre) Gecko/20100911 Firefox/4.0b6pre Build Identifier: Mozilla/5.0 (Windows NT 5.1; rv:2.0b6pre) Gecko/20100911 Firefox/4.0b6pre ID:20100911154345 Certain extension are automatically uninstalled by Minefield, on browser restart, due to the landing of bug 533038 Reproducible: Always Steps to Reproduce: 1. Obtain the Sept 11 nightly build: http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/2010-09-11-04-mozilla-central/ 2. Start Minefield with a new profile 3. Disable extension compatibility checking: extensions.checkCompatibility.4.0b and set to false 4. Install Forecastfox: https://addons.mozilla.org/en-US/firefox/addon/398/ 5. Install the latest Greasemonkey nightly build: https://arantius.com/misc/gm-nightly/ 6. Install QuickRestart: https://addons.mozilla.org/en-US/firefox/addon/3559/ 7. Restart 8. Notice that all three extensions are installed 9. Restart 10. Notice that all three extensions have been uninstalled! 11. Go to Help --> Troubleshooting Information --> Open Containing Folder --> extensions 12. Notice the extensions are there as XPI files and not installed.
Blocks: packedxpi
blocking2.0: --- → ?
Keywords: regression
Version: unspecified → Trunk
Assignee: nobody → mwu
blocking2.0: ? → beta6+
The regex at https://mxr.mozilla.org/mozilla-central/source/toolkit/mozapps/extensions/XPIProvider.jsm#151 is failing for GUID em:ids leading to https://mxr.mozilla.org/mozilla-central/source/toolkit/mozapps/extensions/XPIProvider.jsm#5913 & https://mxr.mozilla.org/mozilla-central/source/toolkit/mozapps/extensions/XPIProvider.jsm#1651 being logged. LOG addons.xpi: Ignoring file entry whose name is not a valid add-on ID: C:\Documents and Settings\<user>\Application Data\Mozilla\Firefox\Profiles\<profile>\extensions\{1280606b-2510-4fe0-97ef-9b5a22eafe41}.xpi LOG addons.xpi: Add-on {1280606b-2510-4fe0-97ef-9b5a22eafe41} removed from app-profile A workaround is to change the em:id in install.rdf to an email address instead of a GUID.
Status: UNCONFIRMED → NEW
Ever confirmed: true
That's my fault for forgetting about GUID form IDs in the review,however I can't see how this is passing tests so I suspect that tests aren't running for the packed case right now.
oh wait, we may not have any tests of GUID IDs,that seems like a bit of an oversight.
Attached patch FixSplinter Review
Thanks for catching this. The {UUID}.xpi file doesn't pass the extension id test so we need to make sure we strip the .xpi before testing the extension id.
Attachment #474452 - Flags: review?(dtownsend)
Comment on attachment 474452 [details] [diff] [review] Fix ># HG changeset patch ># Parent 0f98f549bc00128957e99ff4bc7a03d9d486846d > >diff --git a/toolkit/mozapps/extensions/XPIProvider.jsm b/toolkit/mozapps/extensions/XPIProvider.jsm >--- a/toolkit/mozapps/extensions/XPIProvider.jsm >+++ b/toolkit/mozapps/extensions/XPIProvider.jsm >@@ -5905,26 +5905,27 @@ DirectoryInstallLocation.prototype = { > continue; > > let id = entry.leafName; >- Don't remove this line
Attachment #474452 - Flags: review?(dtownsend) → review+
Status: NEW → RESOLVED
Closed: 15 years ago
Flags: in-testsuite+
Resolution: --- → FIXED
Target Milestone: --- → mozilla2.0b6
Fixed in: Mozilla/5.0 (Windows NT 5.1; rv:2.0b6pre) Gecko/20100912 Firefox/4.0b6pre ID:20100912041924 http://hg.mozilla.org/mozilla-central/rev/cd3c926a7413 Thanks
Status: RESOLVED → VERIFIED
Was this pushed to all builds? I think I was seeing this bug in Linux.
(In reply to comment #8) > Was this pushed to all builds? I think I was seeing this bug in Linux. The code in question is used on all platforms.
Anything we should care of for manual tests?
Flags: in-litmus?
OS: Windows XP → All
Hardware: x86 → All
(In reply to comment #10) > Anything we should care of for manual tests? I think we're covered here now.
Flags: in-litmus? → in-litmus-
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: