Building a Parser with ProGrammar
ProGrammar divides the overall task of building a parser into two distinct phases. The first phase is defining the grammar, which is done in the visual IDE. The second phase involves calling the parse engine from your application, in order to parse data and process the results. This separation of tasks is important because it allows you to focus entirely on the process of defining a grammar-until you get it right-before writing application-specific code. In fact, by the time your application calls the parser, most of the work will already have been accomplished in the visual environment. Typically, your program needs only to pull the data that it requires from a parse tree.
The following steps are involved in building and using a parser:
| 1 | Write a grammar that describes the syntax of the data that will be parsed. The ProGrammar IDE allows you to write grammars, test them by parsing sample data, and debug them interactively. A binary grammar file is generated by the IDE whenever a project is built. This file contains instructions for parsing a particular type of data, and is used directly by the parse engine at runtime. |
| 2 | Call the parser from your application. Client applications interact with parse engine by calling the ProGrammar API. The API provides methods to load grammars, parse data, process errors, and retrieve data elements from the parse tree. The API can be called as an ActiveX control or a native C++ library. |
| 3 | Retrieve data from the parse tree. After each successful parse, the parse engine generates a parse tree, which is a data structure that maps grammar symbols to elements in the input stream. The client application calls API methods to navigate, search, and retrieve data elements from the parse tree. |
For comments or questions about this site, please contact
webmaster@programmar.com
Copyright © 1998-2008 NorKen Technologies, Inc. All rights reserved.