2 * '\\' has a special meaning on phase 2 if and only if it is immediately
3 * followed by '\n'. In any other position it's left alone as any other
7 * Each instance of a backslash character (\) immediately followed by
8 * a new-line character is deleted, splicing physical source lines to
9 * form logical source lines. Only the last backslash on any physical
10 * source line shall be eligible for being part of such a splice.
11 * A source file that is not empty shall end in a new-line character,
12 * which shall not be immediately preceded by a backslash character
13 * before any such splicing takes place.
15 * Note that this happens on the phase 2, before we even think of any
16 * tokens. In other words, splicing is ignorant of and transparent for
17 * the rest of tokenizer.
22 /* This should result in "\a" */
23 /* XXX: currently sparse produces "a" */
24 /* Partially fixed: now it gives "\\a", which is a separate problem */
29 /* This should give 1 */
34 /* And this should give D, since '\n' is removed and we get no whitespace */
35 /* XXX: currently sparse produces 1 */
41 /* This should give '\a' - with no warnings issued */
42 /* XXX: currently sparse complains a lot and ends up producing a */
46 /* This should give nothing */
47 /* XXX: currently sparse produces more junk */
52 /* This should also give nothing */
53 /* XXX: currently sparse produces / * comment * / */
59 /* And this should complain since final newline should not be eaten by '\\' */
60 /* XXX: currently sparse does not notice */