Abstract This is a technical note that assumes reader familiarity with Pagan's paper on converting scaled down Pascal p‐code programming language interpreters into simple compilers.1 This note discusses the extension of Pagan's methods to a full‐scale Pascal‐to‐C translator. In order to make procedure calls and typed function calls work properly for such a translator, it was found necessary to add type information to one of the p‐code calling instructions (and alter the Pascal‐to‐p‐code translator accordingly). A table of execution times for Pagan's ‘50 primes’ bench‐mark program shows the improvements obtained as a result of successive refinements in the Pascal‐to‐C translator, until the present version that uses C ‘register variables’ for integer arithmetic.2 Programs compiled using this system can also be linked into C programs as functions or procedures and can be debugged using standard C debuggers.

Halstead's Software Science formulas and the Air Force's Rome Air Development Center software effort data base are used to derive a software project effort estimator. This estimator is: T = 94S/sup 2/ and T =381/sup 2/ where T =total programmer and software engineer months charged to the project S = thousands of statements of source Fortran, excluding comments I = thousands of assembly language source statements, excluding comments Overall project effort is obtained by applying the estimator on a subroutine by subroutine basis and adding the results. On a data base of 21 projects, ranging in size from 5, 000 to 164, 000 delivered machine instructions, a correlation coefficient of 0. 64 was obtained between data and estin-,Lator. Grouping projects together in twos and threes raised the corre3aAion coefficient to 0. 91 .

A method is presented for directly transforming an arbitrary LR( k ) grammar to an equivalent LR(1) grammar. It is further shown that the method transforms an arbitrary prefix-free LR( k ) grammar to an equivalent LR(0) grammar. It is argued that the method is efficient and offers some advantages over traditional “look-ahead” parsing methods. Finally, it is demonstrated that the method can be used to transform an LR(1) grammar to an equivalent SLR(1) grammar, which in turn can be easily transformed to an equivalent (1, 1) bounded right-context grammar.

Abstract This paper describes our experience in using a common set of code generating subroutines as the basis for compilers of ALGOL W, ALGOL 60 and subsets of PL/I and COBOL. The advantages of using a common set of primitives for implementing more than one compiler include: (1) greatly reduced effort to implement additional languages; (2) simplified maintenance of the generated compilers; (3) standardized diagnostic messages for all compilers implemented using this system; and (4) simplified documentation of the compilers.