Structured Document Search via Parse Tree Rewriting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current structured document management systems face inefficiencies in searching XML data due to limitations in applying refining conditions early in the search process, particularly with nested queries, and struggle with hierarchical structures and element identifiers, leading to suboptimal search performance.
Innovation Solution
A structured document management device and method that includes a parser to decompose query data into a static single assignment form, a constraint acquiring unit to relax hierarchical conditions, and a parse tree rewriting unit to rewrite queries for early refinement, allowing for efficient search by tracing paths and using element IDs to eliminate redundant evaluations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If SAX parsing process is used to check structural conditions, then the search process can access lower layer from higher layer, but it is not possible to apply refining condition unless the lower layer is traced, requiring tracing from top layers to lower layers
Solution Approach 1:
The patent applies preliminary action by performing indexing on structured documents before the actual search query is executed. The constraint acquiring unit pre-processes the document structure to identify and store structural conditions and element identifiers at various levels. When a query is received, these pre-computed constraints can be applied immediately without tracing from the top layer, thus resolving the contradiction between easy access direction and search time.
2Productivity
If refining condition is applied early to reduce intermediate data, then search speed is accelerated, but this technique is not compatible with nested query consisting of plurality of subqueries
Solution Approach 1:
The patent segments the nested query into multiple subqueries and processes each subquery independently. The constraint acquiring unit divides the complex nested query structure into separate constraint sets for each subquery level. This allows early refinement to be applied at each segment level, maintaining search speed while achieving compatibility with nested queries through systematic decomposition of the query structure.
Solution Approach 2:
The patent implements a nested processing structure where outer query constraints are applied first, then inner query constraints are applied to the refined result set. This nested application of constraints mirrors the nested structure of the query itself, allowing early refinement at each nesting level while maintaining compatibility with the overall nested query structure.
3Loss of time
If conditions are copied to all possible locations at the stage of creating the view, then the need to unnecessarily trace elements of the view is eliminated, but the same conditions are evaluated at a plurality of locations and the effect of refinement is not sufficient
Solution Approach 1:
The patent applies local quality by making constraint application location-specific rather than universal. The constraint acquiring unit analyzes the query structure and applies constraints only at the specific locations where they are most effective for refinement. This prevents redundant evaluation at all possible locations while still eliminating unnecessary tracing, achieving both time savings and effective refinement by tailoring constraint application to local query characteristics.
Data Source
AI summary
According to an embodiment, a structured document management device includes a parser; a constraint acquiring unit; and a parse tree rewriting unit. The structured document receiving unit is configured to receive an input of a structured document having a layered logical structure. The parser is configured to parse the syntax of received query data to acquire a parse tree in which the query data is decomposed so as to meet a static single assignment form which is defined such that each variable is assigned exactly once. The constraint acquiring unit is configured to acquire a hierarchical condition that the variables of the parse tree need to meet and constraints that include conditions for values of the respective variables. The parse tree rewriting unit is configured to rewrite the query data in such a form that the structured document is refinable by the constraints.