At present, most original classic manuscripts of Dongba script have been collected by wellknown institutions from more than ten countries. As academic researchersare decentralized, it is very inconvenient for them to communicate with each other. The construction of a sharing platform for ancient books of Dongba script is beneficial for emergency treatment and inheritance of classic culture. In allusion to digitalization and data storage of ancient book resources of Dongba script, a digitalization method is presented in this paper for printing books known as Annotated General Catalog of Ancient Books of Ethnic Minorities in China (Naxi Volume) based on the analysis of existing information extraction approaches and data storage modes. Moreover, metadata is also adopted to refer to the bibliography of ancient books of Dongba script, which are extracted from printing books. And ultimately, XML database is employed to manage the digitalized contents. According to the experimental results, the information extraction approach proposed in this paper is able to extract contents accurately direct at the elaborate structure of the bibliography for ancient books of Dongba script on one hand and provides structured retrieval means on the other hand. As a result, both feasibility and validity of such an approach areverified. This research has important reference meanings for the digitalization and semistructured data management of ancient books of ethnic minorities.