[expand/normalize] the split_alpha_from_numeric option now applies to both e.g. A1 and 1A since we now strip out ordinal suffixes prior to normalization
This commit is contained in:
@@ -104,7 +104,7 @@ void add_normalized_strings_token(cstring_array *strings, char *str, token_t tok
|
||||
}
|
||||
}
|
||||
|
||||
if (is_numeric_token(token.type) && options.split_alpha_from_numeric && numeric_starts_with_alpha(str, token)) {
|
||||
if (is_numeric_token(token.type) && options.split_alpha_from_numeric) {
|
||||
normalize_token_options |= NORMALIZE_TOKEN_SPLIT_ALPHA_FROM_NUMERIC;
|
||||
normalize_token(strings, str, token, normalize_token_options);
|
||||
normalize_token_options ^= NORMALIZE_TOKEN_SPLIT_ALPHA_FROM_NUMERIC;
|
||||
|
||||
Reference in New Issue
Block a user