IBX-8897: Fixed incorrect extrapolation of parentheses in the name schema pattern #449
Conversation
|
| // Skip the group if it has no fields to parse | ||
| if (!preg_match('/<.*>/', $group)) { | ||
| continue; | ||
| } |
There was a problem hiding this comment.
After looking into this we could probably do more complex regex in L248:
// match groups "( )" containing fields "< >" - curly brackets without any fields are not a group, but a literal string
$foundGroups = preg_match_all('/\((.*<.+>.*)\)/U', $nameSchema, $groupArray);But as we discussed internally, I'm not sure which solution would be better for @ibexa/php-dev. The current one is maybe more readable.
POV ping @ibexa/php-dev
There was a problem hiding this comment.
If we want to proceed with the above, then tests related to this functionality should be expanded (which would be welcome by itself anyway, by the way).
This would ensure that any changes to the name schema processing, including regexp changes to improve performance, are not breaking anything - especially not any nested patterns.
There was a problem hiding this comment.
I've already modified a test to cover the "new" part of this functionality, is it sufficient of should i cover more cases like so?
alongosz
changed the title
Description:
Fixing the regression introduced in #242. Right now, for example, if we us this name schema
<title|(<field1> - <field2>)> (foobar)the foobar would be parsed to(EZMETAGROUP_0)regardless of the previously parsed fields. Grouping functionality with()has no use if the content inside this group has no<fields>.