Addon testing suite #5058
Addon testing suite #5058
Conversation
bakulf
force-pushed
the
addon
branch
8 times, most recently
from
46881f9 to
a7eb906
Compare
Codecov ReportBase: 25.70% // Head: 28.26% // Increases project coverage by
Additional details and impacted files@@ Coverage Diff @@
## main #5058 +/- ##
==========================================
+ Coverage 25.70% 28.26% +2.56%
==========================================
Files 12 255 +243
Lines 607 14970 +14363
Branches 325 8567 +8242
==========================================
+ Hits 156 4231 +4075
- Misses 262 4251 +3989
- Partials 189 6488 +6299
Flags with carried forward coverage won't be shown. Click here to find out more.
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
| assert(await vpn.isFeatureFlippedOff('addonSignature')); | ||
| } | ||
| it('Empty addon index', async () => { | ||
| await vpn.resetAddons('01_empty_manifest'); |
There was a problem hiding this comment.
nit: I think it would be better if we reset to the empty manifest on the beforeEach block everytime and then in the test cases we defaulted to fetchAddons. IIUC right now it is necessary to call reset as the first thing in each test case in order not to leak state from one test to the next. Is that correct? WDYT?
There was a problem hiding this comment.
Maybe we even do that for ALL tests so that addon changes are never leaked no matter what people do? Not sure there a global beforeEach block, if not just ignore.
There was a problem hiding this comment.
At the end of any test we call hardReset() that deletes all the settings.
tests/functional/helper.js
Outdated
| json.type === 'reset_addons' && !('error' in json), | ||
| `Command failed: ${json.error}`); | ||
|
|
||
| await this.waitForCondition(async () => this.lastAddonLoadingCompleted()); |
There was a problem hiding this comment.
I'm confused. Does the _writeCommand('reset_addons') end up having connect called again?
There was a problem hiding this comment.
no. _writeCommand() writes something on a established connected channel (websocket for functional tests, wasm JS for wasm).
This is a way to test addons in functional tests. It works this way:
tests/functional/addonswe create a folder for each "scenarios" we want to expose. We can expose a manifest.json directly, or a few folders, one for each addon.I migrated all the tests and I wrote a few more. But we can do more! Here is the list to start: