From 9478e157ec2cfe4de704b3bd78b07aee8824774f Mon Sep 17 00:00:00 2001
From: Kyle Gunger <kgunger12@gmail.com>
Date: Mon, 18 Apr 2022 01:19:02 -0400
Subject: Add tokenizer support for strings

---
 tnslc/dummy.tnsl           |  1 +
 tnslc/parse/token.tnsl     | 23 ++++++++++++++++++++++-
 tnslc/parse/tokenizer.tnsl |  5 ++---
 3 files changed, 25 insertions(+), 4 deletions(-)

(limited to 'tnslc')

diff --git a/tnslc/dummy.tnsl b/tnslc/dummy.tnsl
index edecab5..a2a8156 100644
--- a/tnslc/dummy.tnsl
+++ b/tnslc/dummy.tnsl
@@ -1,4 +1,5 @@
 /; main [float]
 	;int i = 0
+	;{}charp c = "abc"
 	;return a.b
 ;/
diff --git a/tnslc/parse/token.tnsl b/tnslc/parse/token.tnsl
index e18ade5..ce2bcdb 100644
--- a/tnslc/parse/token.tnsl
+++ b/tnslc/parse/token.tnsl
@@ -294,6 +294,27 @@
 	;return true
 ;/
 
+/; string_closed ({}charp dat, charp c) [bool]
+	/; if (len dat < 2)
+		;return false
+	;/
+
+	;charp closing = dat{0}
+	;bool escaping = false
+
+	/; loop (int i = 1; i < len dat) [i++]
+		/; if (dat{i} == closing && !escaping)
+			;return true
+		;; else if (dat{i} == '\\' && !escaping)
+			;escaping = true
+		;; else
+			;escaping = false
+		;/
+	;/
+
+	;return false
+;/
+
 /#
 	Get the token_type value for a given string of character points
 
@@ -301,7 +322,7 @@
 
 	/; if (len s` > 1)
 
-		/; if (is_numeric_literal(s))
+		/; if (is_numeric_literal(s) || s`{0} == '"' || s`{0} == '\'')
 			;return TOKEN_TYPE.LITERAL
 		;/
 
diff --git a/tnslc/parse/tokenizer.tnsl b/tnslc/parse/tokenizer.tnsl
index c6e1ab6..139877c 100644
--- a/tnslc/parse/tokenizer.tnsl
+++ b/tnslc/parse/tokenizer.tnsl
@@ -21,7 +21,8 @@
 /; break_token ({}charp dat, charp c) [bool]
 	/; if (len dat == 0)
 		;return false
-	
+	;; else if (dat{0} == '"' || dat{0} == '\'')
+		;return string_closed(dat, c)
 	;; else if (is_in_string(~RESERVED, dat{len dat - 1}))
 		/; if (is_in_string(~RESERVED, c))
 			;dat.append(c)
@@ -99,8 +100,6 @@
 		;out.append(ttk)
 	;/
 
-	;tnsl.io.println(tdat)
-
 	;out = strip_and_expand(~out)
 	;return ~out
 ;/
-- 
cgit v1.2.3