From 5f6d17d3af8052da54afeb170c6986c8cec13d88 Mon Sep 17 00:00:00 2001
From: Case Duckworth
Date: Fri, 8 Jul 2022 16:38:49 -0500
Subject: Tokenize nesting?

---
 squ.awk | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 55 insertions(+), 3 deletions(-)

(limited to 'squ.awk')

diff --git a/squ.awk b/squ.awk
index d4753a0..2539786 100755
--- a/squ.awk
+++ b/squ.awk
@@ -10,6 +10,9 @@
 BEGIN {
 	nested = 0
 	buffer = ""
+	OFS = "\t"
+	STDERR = "/dev/stderr"
+	PRGN = "squawk"
 }
 
 {
@@ -17,15 +20,64 @@ BEGIN {
 }
 
 END {
-	ast = read(buffer)
+	read(buffer, ast)
 	eval(ast)
 }
 
 
-function eval(buf)
+function die(message, errcode)
 {
+	eprint(PRGN " ERROR" (message ? ": " message : ""))
+	exit (errcode ? errcode : 1)
 }
 
-function read(buf)
+function eprint(message)
 {
+	# Print MESSAGE to STDERR.
+	print(message) > STDERR
+}
+
+function eval(ast)
+{
+	# Evaluate multi-dimensional array AST.
+	for (w in ast) {
+		print ast[w]
+	}
+}
+
+function read(buf, ast)
+{
+	# Read string BUF into multi-dimensional array AST.
+	split(buf, b, "")
+	w = 1
+	word = ""
+	# Tokenize
+	for (c in b) {
+		# print c, b[c]
+		if (b[c] == "\\") {
+			word = word b[c++]
+		} else if (b[c] == "(") {
+			if (word) {
+				ast[w++] = word
+				word = ""
+			}
+			ast[w++] = "(" nested++
+		} else if (b[c] == ")") {
+			if (word) {
+				ast[w++] = word
+				word = ""
+			}
+			ast[w++] = ")" --nested
+		} else if (b[c] ~ /[ \t\r\n\f\v]/) {
+			if (word) {
+				ast[w++] = word
+				word = ""
+			}
+		} else {
+			word = word b[c]
+		}
+		if (nested < 0) {
+			die("Unmatched paren at " c)
+		}
+	}
 }
-- 
cgit 1.4.1-21-gabe81