基於BODE擷取系統的自動化網頁再生技術

Abstract

為了整合各個應用領域的多樣化的網路電子文件，在擴充性、網路傳輸、資訊交換、及資料整合管理等各方面都具有優越特性之XML標準應運而生。目前XML文件的轉換大多以XSLT語法來表示，例如，XSLT能將XML文件轉換成HTML網頁。開發XSLT是一項與輸出端有高度關聯性的程序，因此不易撰寫。

本論文利用一個稱為BODE系統的視覺化工具來製作自動化的XSLT產生器。藉由紀錄資料擷取的程序，BODE系統會學習來源網頁文件與目的XML資料文件之間的映對關係，並從這個映對關係計算出XSLT程式碼。由於BODE系統的資料擷取步驟是以視覺化的方式進行，然後再自動地計算出XSLT，因此，綜合這兩個步驟就能形成視覺化的XSLT產生器。

XML is a standard way of representing data in a structured format, and being applied for various application domains, such as data exchange, data integration, electronic publishing, and e-Business. XSLT is a language commonly used to transform XML documents, e.g., to transform a XML document to a web page (i.e. HTML file). Since XSLT is tightly related with the representation on the display screen, cooperation between experienced programmer and art designer invokes high cost of developing the XSLT code.

In this thesis, an automatic XSLT generator, based on a visualized web extraction tool, called BODE system, is developed. By extracting the web pages, the BODE system records the mapping from the HTML document of the web page to the XML document which stores the extracted data. From the recorded mapping, the reverse direction transformation, i.e. from the XML document to the HTML document, is derived and outputted as the XSLT code. Since the extraction step is performed by using the visualized tool, and the step of deriving XSLT is automatic, these two steps comprise a visualized XSLT generation system.