Cross-reference between ANSI Standard and Test Programs (cont.) Section 7: Expressions 7.2: Syntax 42 43 61 143 144 145 41 7.4: Semantics 43 61 143 144 145 146 35 167 168 170 173 174 176 177 180 182 183 43 117 119 120 121 124 127 128 169 175 178 Section 8: Implementation-Supplied Functions 8.2: Syntax 130 131 143 144 145 146 147 148 149 150 164 8.4: Semantics 114 115 116 117 119 120 121 123 124 127 128 130 131 132 133 Cross-reference between ANSI Standard and Test Programs (cont.) 12.2: Syntax 1 6 165 172 173 174 175 192 193 203 204 Section 14: The Data-, Read, and Restore-Statements Cross-reference between ANSI Standard and Test Programs (cont.) 16.4: Semantics 151 153 154 155 156 157 158 159 160 161 162 163 164 167 Section 17: The Randomize Statement 17.2: Syntax 131 17.4: Semantics 131 Section 18: The Remark-Statement 18.2: Syntax 15 18.4: Semantics 15 Appendix A Differences between Versions 1 and 2 of the Minimal BASIC Test Programs of In the development Version variety of changes in the test system. some stylistic. Below is a list of differences. we introduced a wide Some were substantive, the more significant do Perhaps the most extensive change has to with the more complete treatment of the errors and exceptions which must be detected and reported by a conforming processor. We've tried to make clear the distinction between the two and just what conformance entails in each case. Also, Version 2 tests a wider variety of anomalous conditions for the processor to handle. It is in this area of helpful recovery from programmer mistakes that the Minimal BASIC standard imposes stricter requirements than other language standards and the tests reflect this emphasis. 2. Version 2 differs significantly from Version 1 in its treatment of accuracy requirements. We abandoned any attempt to compute internal accuracy for the purpose of judging conformance as being too vulnerable to the problems of circularity. Rather we formulated a criterion of accuracy, and computed the required results outside the program itself. The programs therefore generally contain only simple IF statements comparing constants or variables (no lengthy expressions). Those test sections where we did attempt some internal computation of accuracy, e.g., the error measure and computation of accuracy of constants and variables, are informative only. 3. There are a number of new informative tests for the RND function. These are to help users whose applications are strongly dependent on a nearly patternless RND sequence. The overall structure of the test system is more explicit. The group numbering should help to explain why testing of certain sections of the ANSI standard had to precede others. Also, it should be easier to isolate the programs relevant to the testing of a given section by referring to the group structure. We tried to be especially careful to keep the printed output of the various tests as consistent as their subject matter would allow. In particular, we always made sure that the programs stated as explicitly as possible what was necessary for the test to pass or fail and that this message was surrounded by triple asterisks. References 1. American National Standard for Minimal BASIC, X3.60-1978, 2. 3. 4. J. A. Lee, A Candidate Standard for Fundamental BASIC, July 1973 BASIC, Software T. R. Hopkins, PBASIC - A Verifier for 2, Knuth, The Art of Computer Programming, Vol. D. E. (1969) Publishing |
0 Reviews Reviews aren't verified, but Google checks for and removes fake content when it's identified Write review |