panFMP is a generic and flexible framework for building geoscientific data portals independent of metadata formats and protocols. Data providers can be harvested with commonly used protocols (e.g. Open Archives Initiative Protocol for Metadata Harvesting or Open Geospatial Consortium Catalogue Service) and metadata description standards like DIF or ISO 19115. The new Java based portal software supports any XML format and makes metadata searchable through Apache Lucene using XPath. Software administrators are free to define which fields are searchable independent of its type. In addition, by extending Lucene algorithms, we have significantly improved queries for numerical and date/time ranges. The harvested metadata are stored in separate indices which makes it possible to combine these in several different portals. The new portal specific Java API and web service interface is highly customizable and supports custom front-ends for users, provides automatic query completion, and additional visualization in maps.