Amiga GFA BASIC 3.5 — inherited statement and operator evidence
Amiga GFA BASIC 3.5 — inherited statement and operator evidence
Scope and source boundary
This pass checks selected statement and operator candidates against the version-labelled material in Chapters 18–19 of Das große GFA-BASIC-Buch zum Amiga (AM-BOOK-DE-1990). Chapter 18 opens on printed p. 451 by identifying its matrix/vector commands as new in Version 3.5 and stating that old commands have basically not changed. The examples in these chapters therefore provide direct 3.5 use evidence for the exact inherited forms shown. They do not prove every overload, grammar alternative, edge case, or full compatibility of the corresponding 3.0 chapter.
Page references below are printed book pages. The vector example on p. 471 and the DATA-pointer example on p. 480 were checked against page images, in addition to the extracted text.
Directly evidenced inherited operators
| Candidate | 3.5 evidence | Demonstrated form | Boundary |
|---|---|---|---|
Numeric + and - |
pp. 466–467 | MAT ADD c()=b()+b() and MAT SUB a()=b()-c() |
Confirms numeric arithmetic use only; not precedence, comparison, or string concatenation |
* and / |
pp. 469, 473 | MAT MUL a()=b()*c() and MAT MUL v(),1/SQR(l) |
Confirms the shown numeric forms only; not precedence or all operand types |
The arithmetic evidence does not establish the operator hierarchy, associativity, unary plus/minus, string concatenation, exponentiation, integer division, modulo, comparison operators, or logical operators. Although the code example contains negative DATA values on p. 471, that does not by itself establish general unary-operator parsing in expressions; both prefix forms remain candidates. The assignment sign in the examples is not treated as evidence for equality comparison.
Directly evidenced inherited statements
| Candidate | 3.5 evidence | Demonstrated form | Boundary |
|---|---|---|---|
PRINT |
pp. 458, 471, 480 | Ordinary screen output in matrix and DATA-pointer examples | Does not confirm PRINT AT or WRITE |
LOCATE |
p. 471 | LOCATE 1,12 in the vector-animation example |
Confirms this cursor-positioning form only |
FOR / NEXT |
pp. 463–464, 471, 480 | Ascending FOR ... TO ... NEXT and descending FOR j=i TO 1 STEP -1 ... NEXT |
Does not confirm the DOWNTO spelling or editor ENDFOR replacement |
WHILE / WEND |
p. 480 | WHILE _DATA ... WEND; the example exits when the DATA pointer reaches zero |
Confirms the demonstrated condition form, not every expression or loop edge case |
COLOR |
p. 471 | COLOR 3 and COLOR 1 |
Does not confirm the distinct SETCOLOR palette-component form |
DRAW TO |
p. 471 | A procedure draws calculated endpoints using DRAW TO |
Does not confirm PLOT, LINE, or other DRAW forms |
The candidate manifest splits grouped records where the example demonstrates only one member or form: PRINT and LOCATE are separated from PRINT AT and WRITE; COLOR from SETCOLOR; DRAW TO from PLOT, LINE, and other DRAW forms. Prefix unary plus/minus remain grouped and unconfirmed.
Similar-looking forms deliberately not conflated
MAT PRINT #1andMAT INPUT #1are matrix-specific commands; they do not confirm ordinaryPRINT #,INPUT #, orLINE INPUT #file statements.MAT READis not used as evidence for ordinaryREAD.MAT ADD,MAT SUB, andMAT MULare matrix statements; their names do not confirm same-named Chapter 4 functions.COLORis not treated as interchangeable withSETCOLOR.- The
DRAW TOexample does not confirm the fullDRAWcommand family.
Result
Eight inherited statement/operator candidate records receive narrow direct 3.5 evidence in this pass: numeric +/-, *//, PRINT, LOCATE, FOR/NEXT, WHILE/WEND, COLOR, and DRAW TO. Their original source origin remains the 3.0 base manual; confirmation source and printed pages are recorded separately.
No API registry declaration, implementation state, runtime-conformance claim, or language-core count is changed. Platform-specific graphics forms remain candidates for the Amiga API profile, not language-core declarations. Unsupported siblings and operators remain documented in the 3.0 base only; their absence from these examples is not evidence of removal or incompatibility.