Add initial code, just to get started

1 files changed, 65 insertions, 0 deletions

diff --git a/lexer.c b/lexer.c
new file mode 100644
index 0000000..68e925b
--- /dev/null
+++ b/lexer.c
@@ -0,0 +1,65 @@
+#include <u.h>
+#include <libc.h>
+#include <bio.h>
+
+#include "apl9.h"
+
+Rune primfuncnames[] = L"+-×÷*⍟⌹○!?|⌈⌊⊥⊤⊣⊢=≠≤<>≥≡≢∨∧⍲⍱↑↓⊂⊃⊆⌷⍋⍒⍳⍸∊⍷∪∩~,⍪⍴⌽⊖⍉⍎⍕";
+Rune primmonopnames[] = L"¨⍨⌸⌶&";
+Rune primdyadopnames[] = L"⍣.∘⍤⍥@⍠⌺";
+Rune primhybridnames[] = L"/\⌿⍀";
+
+Datum *
+lexline(Rune *line, int *ntoks)
+{
+	int offset = 0;
+	int len = runestrlen(line);
+	Datum *tokens = mallocz(sizeof(Datum) * MAX_LINE_TOKENS, 1);
+	*ntoks = 0;
+	while(offset < len){
+		if(isspacerune(line[offset])){
+			offset++;
+			continue;
+		}else if(runestrchr(L"(){}[]", line[offset])){
+			switch(line[offset]){
+			case '(': tokens[*ntoks].tag = LParTag; break;
+			case ')': tokens[*ntoks].tag = RParTag; break;
+			case '{': tokens[*ntoks].tag = LCurlTag; break;
+			case '}': tokens[*ntoks].tag = RCurlTag; break;
+			case '[': tokens[*ntoks].tag = LBracketTag; break;
+			case ']': tokens[*ntoks].tag = RBracketTag; break;
+			}
+			tokens[*ntoks].strrep = runesmprint("%C", line[offset]);
+			offset++;
+		}else if(runestrchr(primfuncnames, line[offset])){
+			tokens[*ntoks].tag = FunctionTag;
+			tokens[*ntoks].strrep = runesmprint("%C", line[offset]);
+			offset++;
+		}else if(runestrchr(primmonopnames, line[offset])){
+			tokens[*ntoks].tag = MonadicOpTag;
+			tokens[*ntoks].strrep = runesmprint("%C", line[offset]);
+			offset++;
+		}else if(runestrchr(primdyadopnames, line[offset])){
+			tokens[*ntoks].tag = DyadicOpTag;
+			tokens[*ntoks].strrep = runesmprint("%C", line[offset]);
+			offset++;
+		}else if(isdigitrune(line[offset])){
+			char buf[64];
+			char *p = buf;
+			while(isdigitrune(line[offset])){
+				p += runetochar(p, &line[offset]);
+				offset++;
+			}
+			*p = 0;
+			tokens[*ntoks].tag = ArrayTag;
+			tokens[*ntoks].array = mkscalarint(atoll(buf));
+			tokens[*ntoks].strrep = runesmprint("%s", buf);
+		}else{
+			print("Can't lex: %S\n", &line[offset]);
+			*ntoks = 0;
+			break;
+		}
+		(*ntoks)++;
+	}
+	return tokens;
+}
\ No newline at end of file