1
1
mirror of https://github.com/kanaka/mal.git synced 2024-11-04 05:26:00 +03:00

basic: Fix handling of unterminated strings by rewriting string reader.

I've essentially translated the BBC BASIC FNunquote_string into these
more primitive dialects, including writing an implementation of INSTR in CBM BASIC because it's not natively present.  The result is ugly but functional.

Fun fact: QBasic thinks INSTR uses INSTR([start, ]needle, haystack);
BBC BASIC uses INSTR(needle, haystack[, start]).
This commit is contained in:
Ben Harris 2019-06-22 17:27:46 +01:00
parent eacb46df65
commit b69cc9068e

View File

@ -167,12 +167,27 @@ SUB READ_FORM
REM PRINT "READ_STRING"
C=ASC(MID$(T$,LEN(T$),1))
IF C<>34 THEN R=-1:ER=-1:E$="expected '"+CHR$(34)+"', got EOF":GOTO READ_FORM_RETURN
R$=MID$(T$,2,LEN(T$)-2)
S1$=CHR$(92)+CHR$(92):S2$=CHR$(127):GOSUB REPLACE: REM protect backslashes
S1$=CHR$(92)+CHR$(34):S2$=CHR$(34):GOSUB REPLACE: REM unescape quotes
#cbm S1$=CHR$(92)+"n":S2$=CHR$(13):GOSUB REPLACE: REM unescape newlines
#qbasic S1$=CHR$(92)+"n":S2$=CHR$(10):GOSUB REPLACE: REM unescape newlines
S1$=CHR$(127):S2$=CHR$(92):GOSUB REPLACE: REM unescape backslashes
J=2:R$=""
READ_STRING_LOOP:
#qbasic I=INSTR(J,T$,CHR$(92))
#cbm I=J
#cbm INSTR_LOOP:
#cbm IF I>LEN(T$) THEN I=0:GOTO INSTR_DONE
#cbm IF MID$(T$,I,1)=CHR$(92) THEN GOTO INSTR_DONE
#cbm I=I+1
#cbm GOTO INSTR_LOOP
#cbm INSTR_DONE:
IF I=0 THEN GOTO READ_STRING_DONE
R$=R$+MID$(T$,J,I-J)
C$=MID$(T$,I+1,1)
#qbasic IF C$="n" THEN R$=R$+CHR$(10) ELSE R$=R$+C$
#cbm IF C$="n" THEN R$=R$+CHR$(13)
#cbm IF C$<>"n" THEN R$=R$+C$
J=I+2
GOTO READ_STRING_LOOP
READ_STRING_DONE:
IF J=LEN(T$)+1 THEN R=-1:ER=-1:E$="expected '"+CHR$(34)+"', got EOF":GOTO READ_FORM_RETURN
R$=R$+MID$(T$,J,LEN(T$)-J)
REM intern string value
B$=R$:T=4:GOSUB STRING
GOTO READ_FORM_RETURN