翻译文献The Extension C API.docVIP

大连交通大学2016届本科生毕业设计（论文）外文翻译  PAGE - 17 - 外文原文 The Extension C API This chapter is about teaching SQLite new tricks. The previous chapter dealt with generic database work; this chapter is about being creative. The latter half of the API—the Extension API—offers three basic ways to extend (or customize) SQLite, through the creation of user-defined functions, aggregates, and collation sequences. User-defined functions are SQL functions that map to some implementation that you write. They are callable from within SQL. For example, you could create a function hello_newman() that returns the string Hello Jerry and, once it is registered, call it from SQL as follows: sqlite select hello_newman() as reply; reply Hello Jerry This is a special version of the SQLite command-line program I customized myself that includes hello_newman(). When I get bored, I call it occasionally. Aggregates are a special form of function and work in much the same way except that they operate on sets of records and return aggregated values or expressions computed over a partic-ular column in the set. Or they may be computed from multiple columns. Standard aggregates that are built into SQLite include SUM(), COUNT(), and AVG(), for example. Collations are methods of comparing and sorting text. That is, given two strings, a Collation would decide which one is greater, or whether the two are equal. The default Collation in SQLite is BINARY—it compares strings byte by byte using memcmp(). While this collation happens to work well for words in the English language (and other languages that use UTF-8 encoding), it doesn’t necessarily work so well for other languages that use different encodings. This is why SQLite version 3 introduced user-defined collations. This chapter covers each of these three user-defined extension facilities and their associ-ated API functions. As you will see, when combined with other facilities such as triggers and conflict resolution, the user-defined extensions API can pro