Using PHP to build a custom search engine _php Tutorial

Build a custom search engine with PHP

While Google and its family of products are almost omnipotent, powerful Web search engines do not work well for every site. If the site content has been highly specialized or clearly categorized, then you need to use Sphinx and PHP to create an optimized local search system.

In the Internet age, people want information to be packaged like fast food: it can be quickly and easily accessible, and is divided into small units (or byte-size units). In fact, even the most common Web sites now require a variety of menus with a quick browse style in order to meet the needs of impatient and eager users of information:

· RSS is a pizza courier who will bring the freshly baked pizza to the door.

• The blog is a local Chinese restaurant, offering you a favorite flavor dish.

• The forum is a commonplace (or perhaps more appropriately, a scene of looting of food in the "Animal House").

• Search is like eating a buffet dinner at a local restaurant: Keep the food you want to eat, as long as your gut--and your chair--can hold.

Fortunately, PHP developers can find a variety of RSS, blog and forum software to create or improve the site. And while Google and other search sites are almost omnipotent and do filtering communications, search engines don't necessarily adapt well to each site.

For example, if the Web site offers hundreds of new and refurbished Porsche parts, Google may find your site through a generalized search such as "Carrera parts", but for more specific "used 1991 Porsche 911 Targa Headlig HT Bezel "Query, it may not get accurate results.

If the site content is highly specialized, or if the visitor expects the search function to be similar to a real workflow, it is best to add a local search system tailored to your site based on the global search engine for the Web (for more examples of specialized searches, see "a needle in a billion Haystacks ").

Learn how to add a fast, efficient, open source, and free search engine to your PHP site with this article. This article does not develop a visible Web site. Instead, focus on the components needed to deliver effective search results: Databases, indexes, search engines, and PHP application programming interfaces (APIs).

Access to excellent Sphinx

To provide a custom search feature for your site, you must have a data source and the ability to search for that data source. For a WEB application, the data source is typically a relational database with some search capabilities built in (equality is a simple search operator, like the SQL operator). However, some searches may be more specific than searches that the database can perform, or the search may be overly complex, causing the inherent SQL JOIN to be unresponsive.

Haidilaozhen

Many sites offer content specific to a particular industry, profession, or entertainment, such as pharmaceuticals, legal, music, and car repair. In-depth study of these content may require the use of special tools or training, or the use of only one lead to generate relevant practical results.

Here are some common search scenarios that require a custom search system:

• Find all the articles on Stanley Cup written by Joe Hockey.

• Find the latest drivers for the HP LaserJet 3015 All-in-one printer.

• Find Dinosaur Jr. Take part in a TV episode of David's late-night talk show.

To speed up the search, you can rearrange the tables and thus simplify the underlying queries (table and SQL query optimizations are highly dependent on patterns and engines.) You can find various articles and books on database performance through online search. In addition, you can add a specialized search engine. Which form of search engine to apply is also dependent on the form (and quantity) and budget of the data. There are a number of options available: You can connect a Google tool to your network, buy Endeca or other large business search products, or try Lucene. But in many cases, the use of commercial products has been a bit of a fuss, or a waste of operational budgets, and Lucene did not provide PHP APIs when it was written in July 2007.

As an alternative, consider Sphinx, an open source and free search engine that can search for text very quickly. For example, in an almost 300,000 row and five index columns

http://www.bkjia.com/PHPjc/1044162.html www.bkjia.com true http://www.bkjia.com/PHPjc/1044162.html techarticle Build a custom search engine with PHP while Google and its family of products are almost omnipotent, the powerful Web search engine does not work well for every site. If the site is inside ...

