Databases Reference
In-Depth Information
wasn't possible to review each one, so we ranked projects by their size and level of
activity.
Packaging up the results included getting statistics out of the source code project
(see the source code processing section), determining the location of the source on
the code filer, and building a searchable index.
6.3.9 Project Searching
By the time we added support for projects as separate entities in the system, CNet
had open-sourced their enterprise search application (“Solr”). Solr provided a nice
layer on top of Lucene, which simplified the work we had to do to add project-
centric search support.
We created a Solr index schema that had the following fields, among others
(Table
6.1
):
Table 6.1: Excerpt from solr index schema for projects
6.4 Public Site Architecture
The above describes a mixture of ad hoc back end systems, and the components
used to provide search services. The actual architecture used to handle both end
user (browser-based) queries and partner/enterprise API requests is a combination
of a few additional systems (Fig.
6.2
).
Search WWH ::
Custom Search