公开(公告)号：US20140095519A1

公开(公告)日：2014-04-03

申请号：US13783141

申请日：2013-03-01

Applicant: ORACLE INTERNATIONAL CORPORATION

Inventor： Zhen Hua Liu ,  Ying Li ,  Hui Joe Chang

IPC: G06F17/30

CPC classification number: G06F16/8373

Abstract: Processes, machines, and stored instructions are provided for storing posting lists for tokens in XML documents and using the posting lists to process queries. For each occurrence of a token in the XML documents, a document processor adds an entry to a list for the token. The entry for the token maps the token to documents or nodes within the documents where the tokens can be found. The document processor may also detect tags in the XML documents and, for each occurrence of a tag, add an entry to a list for the tag. The entry for the tag specifies a range of locations covered by the tag. A query processor may then receive a full text query for evaluation against XML documents, and the query processor may determine a result set for the query using the lists for the tokens and/or the lists for the tags.

Abstract translation: 提供过程，机器和存储的指令用于存储XML文档中的令牌的发布列表，并使用发布列表来处理查询。 对于XML文档中每个出现的令牌，文档处理器将一个条目添加到令牌的列表中。 令牌的条目将令牌映射到可以找到令牌的文档中的文档或节点。 文档处理器还可以检测XML文档中的标签，并且对于标签的每次出现，向标签的列表中添加条目。 标签的条目指定标记覆盖的位置范围。 查询处理器然后可以接收用于针对XML文档的评估的全文查询，并且查询处理器可以使用标记的列表和/或标签的列表来确定查询的结果集。

公开(公告)号：US10915575B2

公开(公告)日：2021-02-09

申请号：US13783141

申请日：2013-03-01

Applicant: Oracle International Corporation

Inventor： Zhen Hua Liu ,  Ying Li ,  Hui Joe Chang

IPC: G06F16/835

Abstract: Methods and systems for improved query evaluation in computer systems are provided. A system receives a full text query for evaluation against a collection of XML documents. The query comprises specifies one or more tokens and one or more contexts. The contexts each represent a path within an XML expression. The query is used to make multiple determinations. First, determining, using token lists, first documents that contain the tokens included in the query. Second, determining, using tag lists, second documents that contain the contexts included in the query. Third, determining third documents by comparing the first documents and the second documents to determine which documents contain both the tokens and the contexts. The query is then evaluated in view of the third documents.