shithub: rgbds

Download patch

ref: 6655e04ef01bd78acd01f5d9efca4d50340b0a0f
parent: 40c6b840f8078f6a4ac494d30c638696a115f3e2
author: Rangi <[email protected]>
date: Tue Mar 2 14:35:24 EST 2021

Restore the "EOF-newline" lexer hack

This was removed in b3c0db218d564bbc9a801d8635591fd969c36e82
(along with two unrelated changes).

Removing this hack introduced issue #742, whereby INCLUDing
a file without a trailing newline can cause a syntax error.

A more proper fix would involve Bison's tracking locations,
but for now the EOF-newline hack fixes the issue while only
affecting some reported errors (expecting "newline"
instead of "end of file").

Fixes #742

--- a/src/asm/lexer.c
+++ b/src/asm/lexer.c
@@ -349,6 +349,7 @@
 	bool atLineStart;
 	uint32_t lineNo;
 	uint32_t colNo;
+	int lastToken;
 
 	struct IfStack *ifStack;
 
@@ -372,6 +373,7 @@
 {
 	state->mode = LEXER_NORMAL;
 	state->atLineStart = true; /* yylex() will init colNo due to this */
+	state->lastToken = T_EOF;
 
 	state->ifStack = NULL;
 
@@ -2235,17 +2237,15 @@
 		return T_STRING;
 	lexer_SetMode(LEXER_NORMAL);
 
-	// If a macro is invoked on the last line of a file, with no blank
-	// line afterwards, returning EOF afterwards will cause Bison to
-	// stop parsing, despite the lexer being ready to output more.
-	// To avoid this, return T_NEWLINE for EOF as well.
 	if (c == '\r' || c == '\n') {
 		shiftChars(1);
 		/* Handle CRLF */
 		if (c == '\r' && peek(0) == '\n')
 			shiftChars(1);
+		return T_NEWLINE;
 	}
-	return T_NEWLINE;
+
+	return T_EOF;
 }
 
 #undef append_yylval_tzString
@@ -2456,16 +2456,22 @@
 	int token = lexerModeFuncs[lexerState->mode]();
 
 	if (token == T_EOF) {
-		/* Try to switch to new buffer; if it succeeds, scan again */
-		dbgPrint("Reached EOB!\n");
-		/* Captures end at their buffer's boundary no matter what */
-		if (!lexerState->capturing) {
-			if (!yywrap())
-				goto restart;
-			dbgPrint("Reached end of input.\n");
-			return T_EOF;
+		if (lexerState->lastToken != T_NEWLINE) {
+			dbgPrint("Forcing EOL at EOF\n");
+			token = T_NEWLINE;
+		} else {
+			/* Try to switch to new buffer; if it succeeds, scan again */
+			dbgPrint("Reached EOF!\n");
+			/* Captures end at their buffer's boundary no matter what */
+			if (!lexerState->capturing) {
+				if (!yywrap())
+					goto restart;
+				dbgPrint("Reached end of input.\n");
+				return T_EOF;
+			}
 		}
 	}
+	lexerState->lastToken = token;
 
 	lexerState->atLineStart = token == T_NEWLINE;
 
@@ -2524,6 +2530,7 @@
 					 * We know we have read exactly "ENDR", not e.g. an EQUS
 					 */
 					lexerState->captureSize -= strlen("ENDR");
+					lexerState->lastToken = T_POP_ENDR; // Force EOL at EOF
 					goto finish;
 				}
 				level--;
@@ -2586,6 +2593,7 @@
 				 * We know we have read exactly "ENDM", not e.g. an EQUS
 				 */
 				lexerState->captureSize -= strlen("ENDM");
+				lexerState->lastToken = T_POP_ENDM; // Force EOL at EOF
 				goto finish;
 			}
 		}
--- a/src/asm/parser.y
+++ b/src/asm/parser.y
@@ -603,20 +603,19 @@
 
 %%
 
-asmfile		: lines last_line
+asmfile		: lines
 ;
 
+/* Note: The lexer adds T_NEWLINE at the end of the input */
 lines		: %empty
 		| lines line
 ;
 
-last_line	: label
-		| label cpu_command
-		| label macro
-		| label directive
-		| assignment_directive
-;
-line		: last_line T_NEWLINE
+line		: label T_NEWLINE
+		| label cpu_command T_NEWLINE
+		| label macro T_NEWLINE
+		| label directive T_NEWLINE
+		| assignment_directive T_NEWLINE
 		| line_directive /* Directives that manage newlines themselves */
 		| error T_NEWLINE { /* Continue parsing the next line on a syntax error */
 			fstk_StopRept();
--- a/test/asm/block-comment-termination-error.err
+++ b/test/asm/block-comment-termination-error.err
@@ -1,5 +1,5 @@
 ERROR: block-comment-termination-error.asm(1):
     Unterminated block comment
 ERROR: block-comment-termination-error.asm(1):
-    syntax error, unexpected end of file
+    syntax error, unexpected newline
 error: Assembly aborted (2 errors)!
--- a/test/asm/code-after-endm-endr-endc.err
+++ b/test/asm/code-after-endm-endr-endc.err
@@ -7,9 +7,9 @@
 ERROR: code-after-endm-endr-endc.asm(17):
     syntax error, unexpected PRINTLN, expecting newline
 ERROR: code-after-endm-endr-endc.asm(19):
-    syntax error, unexpected PRINTLN, expecting end of file or newline
+    syntax error, unexpected PRINTLN, expecting newline
 ERROR: code-after-endm-endr-endc.asm(23):
     syntax error, unexpected PRINTLN, expecting newline
 ERROR: code-after-endm-endr-endc.asm(25):
-    syntax error, unexpected PRINTLN, expecting end of file or newline
+    syntax error, unexpected PRINTLN, expecting newline
 error: Assembly aborted (7 errors)!