发明名称 |
Multistage pipeline for feeding joined tables to a search system |
摘要 |
Flattening a relational database. A method includes accessing a first database table. A first entry value is identified in a first entry from the first database table. The first entry value represents a plurality of entry values of corresponding entries in a second database table. An array is created that includes one or more representations of the plurality of entry values. At least a portion of the first database table is converted to one or more hierarchical record documents, with columns of the first database table represented at a first hierarchical level and values in entries in the columns represented at a second hierarchical level below the first hierarchical level for a corresponding column. Converting the first database table to one or more hierarchical record documents includes substituting the array in a hierarchical record document for each entry in the plurality of entries having the first entry value. |
申请公布号 |
US8190597(B1) |
申请公布日期 |
2012.05.29 |
申请号 |
US201113029977 |
申请日期 |
2011.02.17 |
申请人 |
MILLETT RONALD P.;KING MORAY B.;PERFECT SEARCH CORPORATION |
发明人 |
MILLETT RONALD P.;KING MORAY B. |
分类号 |
G06F7/00;G06F17/30 |
主分类号 |
G06F7/00 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|