I’ve made the ContentControllerTest whitespace agnostic, as it’s
possible that we change the whitespace semantics of includes in SS4
and I don’t want to make this test brittle to that.
The main benefit of this is so that authors who make use of
.editorconfig don't end up with whitespace changes in their PRs.
Spaces vs. tabs has been left alone, although that could do with a
tidy-up in SS4 after the switch to PSR-1/2.
The command used was this:
for match in '*.ss' '*.css' '*.scss' '*.html' '*.yml' '*.php' '*.js' '*.csv' '*.inc' '*.php5'; do
find . -path ./thirdparty -prune -o -type f -name "$match" -exec sed -i '' 's/[[:space:]]\+$//' {} \+
find . -path ./thirdparty -prune -o -type f -name "$match" | xargs perl -pi -e 's/ +$//'
done
Previously, we would be limited by the way allParams will return 3 parameters
at most. This way, we get the full URL.
Keep in mind, this code still needs a clean up, but at least it's not buggy now.
- Content filters included in SiteTree view
- View (tree/list) buttons included in SiteTree view
- Update view button styles for new layout
- Updated breadcrumbs for new layout
Remove "delete from live" duplicate action in favour of existing "unpublish" which is more consistent with current terminology
Add pop-up verification to destructive actions
Fix bug preventing side-by-side preview of archived pages
Fix bug in reporting publishing of error pages
Restoring a page without an available parent will restore to root
Historically, if you visit a foo action on Page, and Page_foo.ss doesn't exist, then it fails over
to Page.ss. The introduction of ContentController::getViewer() broke this, but this patch adds a
test for this case and fixes it.
It was identified by build failures on silverstripe/userforms when tested against the master branch.
Regression from f972466880.
Passes tests in 3.1, but fails in master due to Deprecation notices.
Tests needed adjustments because OldPageRedirector::find_old_page()
doesn't have the same method signature...
The problem is FulltextSearchable::enable() does two things:
It enables the extension, as well as sets the MySQL create table option
to MyISAM. Later tests run with the extension enabled, because it doesn't
get removed at the end of the test, but the table type is set back to
InnoDB when the test is reset and schema recreated.
This produces side-effects where later tests in a suite tries to run
ALTER TABLE on File and add fulltext indexes when the table type
is set to InnoDB, causing an error.
SiteTree versions that arent the live version shouldnt be accessed by
regular users, but the logic to check that was split off into canViewStage,
which wasnt checked by code that isnt specifically SiteTree aware
(like RestfulServer)
- Fix support for $allowed_children
- Added note when type selection is restricted
- Removed overly complex specs for "allowed children",
the data should be inferred from "disallowed children"
- Added support for SiteTree::$can_be_root
- Return raw JSON (not entity-encoded) from CMSMain->SiteTreeHints()
- Added tests for CMSMain->SiteTreeHints()
