Improve pgbench error reporting.

This would have been worth doing on general principle anyway, but the
recent addition of an expression syntax to pgbench makes it an even
better idea than it would have been otherwise.

Fabien Coelho

contrib/pgbench/exprscan.l

@@ -17,6 +17,13 @@ static int	yyline = 0, yycol = 0;
 static YY_BUFFER_STATE scanbufhandle;
 static char *scanbuf;
 static int	scanbuflen;
+
+/* context information for error reporting */
+static char *expr_source = NULL;
+static int expr_lineno = 0;
+static char *expr_full_line = NULL;
+static char *expr_command = NULL;
+static int expr_col = 0;
 %}
 
 %option 8bit
@@ -56,7 +63,9 @@ space			[ \t\r\f]
 
 .				{
 					yycol += yyleng;
-					fprintf(stderr, "unexpected character \"%s\"\n", yytext);
+					syntax_error(expr_source, expr_lineno, expr_full_line, expr_command,
+								 "unexpected character", yytext, expr_col + yycol);
+					/* dead code, exit is called from syntax_error */
 					return CHAR_ERROR;
 				}
 %%
@@ -64,20 +73,27 @@ space			[ \t\r\f]
 void
 yyerror(const char *message)
 {
-	/* yyline is always 1 as pgbench calls the parser for each line...
-	 * so the interesting location information is the column number */
-	fprintf(stderr, "%s at column %d\n", message, yycol);
-	/* go on to raise the error from pgbench with more information */
+	syntax_error(expr_source, expr_lineno, expr_full_line, expr_command,
+				 message, NULL, expr_col + yycol);
 }
 
 /*
  * Called before any actual parsing is done
  */
 void
-expr_scanner_init(const char *str)
+expr_scanner_init(const char *str, const char *source,
+				  const int lineno, const char *line,
+				  const char *cmd, const int ecol)
 {
 	Size	slen = strlen(str);
 
+	/* save context informations for error messages */
+	expr_source = (char *) source;
+	expr_lineno = (int) lineno;
+	expr_full_line = (char *) line;
+	expr_command = (char *) cmd;
+	expr_col = (int) ecol;
+
 	/*
 	 * Might be left over after error
 	 */
@@ -105,4 +121,9 @@ expr_scanner_finish(void)
 {
 	yy_delete_buffer(scanbufhandle);
 	pg_free(scanbuf);
+	expr_source = NULL;
+	expr_lineno = 0;
+	expr_full_line = NULL;
+	expr_command = NULL;
+	expr_col = 0;
 }

contrib/pgbench/pgbench.c

@@ -287,6 +287,7 @@ typedef struct
 	int			type;			/* command type (SQL_COMMAND or META_COMMAND) */
 	int			argc;			/* number of command words */
 	char	   *argv[MAX_ARGS]; /* command word list */
+	int			cols[MAX_ARGS]; /* corresponding column starting from 1 */
 	PgBenchExpr *expr;			/* parsed expression */
 } Command;
 
@@ -2185,6 +2186,32 @@ parseQuery(Command *cmd, const char *raw_sql)
 	return true;
 }
 
+void
+syntax_error(const char *source, const int lineno,
+			 const char *line, const char *command,
+			 const char *msg, const char *more, const int column)
+{
+	fprintf(stderr, "%s:%d: %s", source, lineno, msg);
+	if (more != NULL)
+		fprintf(stderr, " (%s)", more);
+	if (column != -1)
+		fprintf(stderr, " at column %d", column);
+	fprintf(stderr, " in command \"%s\"\n", command);
+	if (line != NULL)
+	{
+		fprintf(stderr, "%s\n", line);
+		if (column != -1)
+		{
+			int i;
+
+			for (i = 0; i < column - 1; i++)
+				fprintf(stderr, " ");
+			fprintf(stderr, "^ error found here\n");
+		}
+	}
+	exit(1);
+}
+
 /* Parse a command; return a Command struct, or NULL if it's a comment */
 static Command *
 process_commands(char *buf, const char *source, const int lineno)
@@ -2229,6 +2256,7 @@ process_commands(char *buf, const char *source, const int lineno)
 
 		while (tok != NULL)
 		{
+			my_commands->cols[j] = tok - buf + 1;
 			my_commands->argv[j++] = pg_strdup(tok);
 			my_commands->argc++;
 			if (max_args >= 0 && my_commands->argc >= max_args)
@@ -2246,9 +2274,10 @@ process_commands(char *buf, const char *source, const int lineno)
 
 			if (my_commands->argc < 4)
 			{
-				fprintf(stderr, "%s: missing argument\n", my_commands->argv[0]);
-				exit(1);
+				syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+							 "missing arguments", NULL, -1);
 			}
+
 			/* argc >= 4 */
 
 			if (my_commands->argc == 4 || /* uniform without/with "uniform" keyword */
@@ -2263,41 +2292,38 @@ process_commands(char *buf, const char *source, const int lineno)
 			{
 				if (my_commands->argc < 6)
 				{
-					fprintf(stderr, "%s(%s): missing threshold argument\n", my_commands->argv[0], my_commands->argv[4]);
-					exit(1);
+					syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+								 "missing threshold argument", my_commands->argv[4], -1);
 				}
 				else if (my_commands->argc > 6)
 				{
-					fprintf(stderr, "%s(%s): too many arguments (extra:",
-							my_commands->argv[0], my_commands->argv[4]);
-					for (j = 6; j < my_commands->argc; j++)
-						fprintf(stderr, " %s", my_commands->argv[j]);
-					fprintf(stderr, ")\n");
-					exit(1);
+					syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+								 "too many arguments", my_commands->argv[4],
+								 my_commands->cols[6]);
 				}
 			}
 			else /* cannot parse, unexpected arguments */
 			{
-				fprintf(stderr, "%s: unexpected arguments (bad:", my_commands->argv[0]);
-				for (j = 4; j < my_commands->argc; j++)
-					fprintf(stderr, " %s", my_commands->argv[j]);
-				fprintf(stderr, ")\n");
-				exit(1);
+				syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+							 "unexpected argument", my_commands->argv[4],
+							 my_commands->cols[4]);
 			}
 		}
 		else if (pg_strcasecmp(my_commands->argv[0], "set") == 0)
 		{
 			if (my_commands->argc < 3)
 			{
-				fprintf(stderr, "%s: missing argument\n", my_commands->argv[0]);
-				exit(1);
+				syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+							 "missing argument", NULL, -1);
 			}
 
-			expr_scanner_init(my_commands->argv[2]);
+			expr_scanner_init(my_commands->argv[2], source, lineno,
+							  my_commands->line, my_commands->argv[0],
+							  my_commands->cols[2] - 1);
 
 			if (expr_yyparse() != 0)
 			{
-				fprintf(stderr, "%s: parse error\n", my_commands->argv[0]);
+				/* dead code: exit done from syntax_error called by yyerror */
 				exit(1);
 			}
 
@@ -2309,8 +2335,8 @@ process_commands(char *buf, const char *source, const int lineno)
 		{
 			if (my_commands->argc < 2)
 			{
-				fprintf(stderr, "%s: missing argument\n", my_commands->argv[0]);
-				exit(1);
+				syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+							 "missing argument", NULL, -1);
 			}
 
 			/*
@@ -2339,12 +2365,13 @@ process_commands(char *buf, const char *source, const int lineno)
 					pg_strcasecmp(my_commands->argv[2], "ms") != 0 &&
 					pg_strcasecmp(my_commands->argv[2], "s") != 0)
 				{
-					fprintf(stderr, "%s: unknown time unit '%s' - must be us, ms or s\n",
-							my_commands->argv[0], my_commands->argv[2]);
-					exit(1);
+					syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+								 "unknown time unit, must be us, ms or s",
+								 my_commands->argv[2], my_commands->cols[2]);
 				}
 			}
 
+			/* this should be an error?! */
 			for (j = 3; j < my_commands->argc; j++)
 				fprintf(stderr, "%s: extra argument \"%s\" ignored\n",
 						my_commands->argv[0], my_commands->argv[j]);
@@ -2353,22 +2380,22 @@ process_commands(char *buf, const char *source, const int lineno)
 		{
 			if (my_commands->argc < 3)
 			{
-				fprintf(stderr, "%s: missing argument\n", my_commands->argv[0]);
-				exit(1);
+				syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+							 "missing argument", NULL, -1);
 			}
 		}
 		else if (pg_strcasecmp(my_commands->argv[0], "shell") == 0)
 		{
 			if (my_commands->argc < 1)
 			{
-				fprintf(stderr, "%s: missing command\n", my_commands->argv[0]);
-				exit(1);
+				syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+							 "missing command", NULL, -1);
 			}
 		}
 		else
 		{
-			fprintf(stderr, "Invalid command %s\n", my_commands->argv[0]);
-			exit(1);
+			syntax_error(source, lineno, my_commands->line, my_commands->argv[0],
+						 "invalid command", NULL, -1);
 		}
 	}
 	else

contrib/pgbench/pgbench.h

@@ -47,7 +47,12 @@ extern PgBenchExpr *expr_parse_result;
 extern int      expr_yyparse(void);
 extern int      expr_yylex(void);
 extern void expr_yyerror(const char *str);
-extern void expr_scanner_init(const char *str);
+extern void expr_scanner_init(const char *str, const char *source,
+							  const int lineno, const char *line,
+							  const char *cmd, const int ecol);
+extern void syntax_error(const char* source, const int lineno, const char* line,
+						 const char* cmd, const char* msg, const char* more,
+						 const int col);
 extern void expr_scanner_finish(void);
 
 extern int64 strtoint64(const char *str);