MNT Fix test that didn't account for i18n spaces #611
Conversation
|
lint failure is unrelated and will be fixed by #610 |
| @@ -1434,8 +1434,8 @@ public function testDBFieldArgs() | |||
| $node = $result['data']['readOneDataObjectFake'] ?? null; | |||
| $this->assertEquals('This is a varchar field', $node['myField']); | |||
| $this->assertEquals('Saturday', $node['date1']); | |||
| $this->assertEquals('2/29/20, 5:00 PM', $node['date2']); | |||
| $this->assertEquals('5:00:00 PM', $node['date3']); | |||
| $this->assertMatchesRegularExpression('#2/29/20,\h5:00\hPM#u', $node['date2']); | |||
There was a problem hiding this comment.
Actually it might be html5 overriding it - https://github.com/silverstripe/silverstripe-framework/blob/5/src/Forms/DateField.php#L450
However there's a comment that this should force a particular standard of date formats, so even then we still shouldn't get thin spaces?
There was a problem hiding this comment.
This one is draft for a reason.
There was a problem hiding this comment.
Based on php/php-src#11600 it seems that there was an intentional change in "ICU" which seems to be what's used under-the-hood for the i18n date formatting. This change includes the narrow non-breaking spaces which seem like they're here to stay.
We'll need to update a bunch of tests to account for it, including this one.
|
Fixes https://github.com/silverstripe/silverstripe-graphql/actions/runs/11061549418/job/30734504257?pr=610
The date formatter used is
IntlDateFormatterwhich sometimes uses narrow non-breaking spaces. These are valid, given the intention of usingIntlDateFormatteris to give a format that is correct for the locale used on the front-end, and in some locales the narrow non-breaking space is correct.This test doesn't have to care what specific spaces are used - just that the correct date is there in the correct format.
Issue