Version 1.5.0
Public BerryPath release notes for version 1.5.0.
Version 1.5.0
New
-
Matching rules now use a friendlier rule builder. Pick a field from a searchable list with type icons (text, number, date, yes/no). Operators show in plain words (is, is not, is one of, has any value). Values come from your own product feed with click-to-pick suggestions and counts.
-
"Is one of" and "is none of" now use a chip picker so you can quickly add many values at once.
-
Yes/No fields show the actual values from your feed (for example Ja / Nee) instead of a generic true/false toggle.
-
The same searchable field picker is now used on the dataset Mapping tab, so finding the right feed field is easier with large feeds.
Improved
-
Combine options renamed from ALL rules (AND) / ANY rule (OR) to Match all rules / Match any rule.
-
Dropdowns now flip up when there is no space below, follow the page when you scroll and keep loading long field and value lists while you scroll.
-
Search in field and value pickers now receives focus as soon as you open the picker.
-
For dataset answers, you can now choose how multiple choices in one product field are split. / remains the default.
-
The builder now shows dataset answer previews in a modal, including example feed values and the choices they create.
-
Matching now shows questions by level from the start question. Questions reached through different answers at the same level share the same route number; disconnected questions stay marked at the bottom.
-
Dataset attribute questions now show their product match count directly in Matching, with an expandable value list and product preview per value.
-
Numeric fields with units like 12 mm or 3mm are now accepted as values.
Fixed
-
Higher Order Value no longer shows a negative percentage in Reports or flow analytics; lower or missing assisted order value now falls back to 0%.
-
Product feed values with unicode codes, such as Q²M, now display correctly in dropdowns and value pickers.
-
Value pickers now keep complete field values with a slash, such as 500g/m2, instead of splitting them.
-
The product match counter now saves the visible filter rules before recounting.
-
Automatic USP labels based on question and answer now follow the selected language in preview and the Flow widget.
-
The product match panel for dataset attribute questions now opens without an Alpine error.
-
Dataset attribute questions now show the number of values as values, so it is no longer confused with product matches.
-
The product preview for dataset attribute questions no longer stays stuck on loading matches after products are already visible.
-
Chip input for "is one of" no longer grows tall when many values are selected; the panel stays compact and the chips live inside the dropdown.
-
Saved matching operators (like is not) stay selected reliably after reload.