1. Introduction

SPRUT is a translator of a compiler internal representation description (IRD) into Standard Procedural Interface (SPI). The most convenient form of the internal representation is a directed graph. IRD defines structure of the graph. SPI provides general graph manipulating functions. The defined graph nodes can be decorated with attributes of arbitrary types.

IRD declares types of nodes of the graph. Nodes contains fields, part of them represents links between nodes, and another part of them stores attributes of arbitrary types. To make easy describing internal representation the IRD supports explicitly single inheritance in node types and also can model multiple inheritance. There can be several levels of internal representation description in separate files. The nodes of one level refer to the nodes of previous levels. Therefore each next level enriches source program internal representation. For example, the zero level representation may be internal representation for scanner, the first level may be internal representation for parser, and so on.

SPI can contains functions to construct and destroy graphs and graph nodes, to copy graphs or graph nodes, to read and write graphs or graph nodes from (to) files, to print graphs or graph nodes, to check up constraints on graph, to traverse graphs, and to transform acyclic graphs in some commonly used manners. SPI can also check up the most important constraints on internal representation during work with node fields. SPI can automatically maintain back links between internal representation nodes.

Using SPRUT has the following advantages:

1. brief and concise notation for internal representation
2. improving maintainability and as consequence reliability of the compiler
3. user is freed from the task of writing large amounts of relatively simple code

The following sections define the internal representation description language, explain SPI, and present some examples.