From 47a08b42df1255227f54a2846404649ee4de6e98 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 3 Jun 2014 21:17:57 +1000 Subject: [PATCH] Scanner: minor text updates Signed-off-by: NeilBrown --- csrc/scanner.mdc | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/csrc/scanner.mdc b/csrc/scanner.mdc index 05df51d..d15b44b 100644 --- a/csrc/scanner.mdc +++ b/csrc/scanner.mdc @@ -90,7 +90,7 @@ The different tokens are numbers, words, marks, strings, comments, newlines, EOF, and indents, each of which is examined in detail below. There are various cases where no token can be found in part of the -input. All of these will be reported as an `TK_error` token. +input. All of these will be reported as a `TK_error` token. It is possible to declare a number of strings which form distinct tokens (rather than being grouped as e.g. 'word'). These are given @@ -260,6 +260,9 @@ and the length of the list must be given (`known_count`). Tokens matching these known words are reported as the index of the list added to `TK_reserved`. +If identifiers are ignored, then any work which is not listed as a +known word results in an error. + ###### token config parameters const char **words_marks; int known_count; -- 2.43.0