p86a-e

config/retriex/genre.yaml

@@ -105,6 +105,7 @@ parameters:
         - configuration_values.context_resolution.commercial_table_follow_up
         - configuration_values.context_resolution.referential_terms
         - configuration_values.context_resolution.history_anchor_enrichment
+        - configuration_values.context_resolution.product_list_followup
         - configuration_values.context_resolution.meta_query_guard
         - configuration_values.context_resolution.rag_anchor_enrichment
         review_path_groups:
@@ -1106,6 +1107,92 @@ parameters:
           - /\b(?:indikator(?:typ)?|indicator(?:\s+type)?|reagenz(?:satz|typ)?|reagent(?:\s+set|\s+type)?|typ|type)\s+[A-Za-zÄÖÜäöüß]{0,8}\s*\d{1,5}(?:\s*[A-ZÄÖÜ]{1,4})?(?:\s*%)?\b/iu
           template: '{anchor} {query}'
           max_query_terms: 2
+        product_list_followup:
+          origin: genre_native
+          enabled: true
+          # Handles referential follow-ups such as "links/preise zu den
+          # produkten aus dem shop". It only rewrites weak meta queries and
+          # uses product/model anchors from the latest assistant answer.
+          weak_query_max_terms: 4
+          weak_query_max_residual_terms: 0
+          max_anchors: 4
+          template: '{anchors}'
+          product_terms:
+          - produkt
+          - produkte
+          - produkten
+          - produkteintrag
+          - produkteinträge
+          - produkteintraege
+          - artikel
+          - gerät
+          - geraet
+          - geräte
+          - geraete
+          - modell
+          - modelle
+          shop_terms:
+          - shop
+          - shopdaten
+          - shop-daten
+          - link
+          - links
+          - url
+          - urls
+          - produktlink
+          - produktlinks
+          - preis
+          - preise
+          - kosten
+          - kostet
+          noise_terms:
+          - gebe
+          - gib
+          - zeige
+          - zeig
+          - nenne
+          - mir
+          - bitte
+          - link
+          - links
+          - url
+          - urls
+          - produktlink
+          - produktlinks
+          - shop
+          - shopdaten
+          - shop-daten
+          - preis
+          - preise
+          - kosten
+          - kostet
+          - produkt
+          - produkte
+          - produkten
+          - artikel
+          - gerät
+          - geraet
+          - geräte
+          - geraete
+          - modell
+          - modelle
+          - zu
+          - zum
+          - zur
+          - aus
+          - den
+          - der
+          - die
+          - das
+          - dem
+          - diese
+          - diesen
+          - dieser
+          - dazu
+          - davon
+          anchor_patterns:
+          - '/(?:^|\R)[^\S\r\n]*(?:\d+[.)][^\S\r\n]*)?(?P<anchor>Testomat(?:®)?[^\S\r\n]+\d{3,4}(?:[^\S\r\n]+[A-Za-z0-9-]{1,12}){0,2})\b/iu'
+          - '/\b(?P<anchor>Testomat(?:®)?[^\S\r\n]+(?:\d{3,4}(?:[^\S\r\n]+(?=[A-Z0-9-]*[A-Z])[A-Z0-9-]{2,12}){0,2}|EVO(?:[^\S\r\n]+[A-Z]{2,8})?|ECO(?:[- ]?(?:PLUS|C))?|DUO(?:[^\S\r\n]+\d{3,4})?|LAB(?:[^\S\r\n]+[A-Z-]{1,8}){1,2}))\b/iu'
         meta_query_guard:
           origin: genre_native
           meta_only_terms:
@@ -1336,6 +1423,8 @@ parameters:
           - evo
           - eco
           - plus
+          - self
+          - clean
           - c
           - duo
           adjacent_variant_patterns: