[src/trunk]: src/tests/usr.bin/xlint/lint1 lint: reproduce the crash from ckg...

[rillig <rillig%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/d8fe8ae34e56
branches:  trunk
changeset: 959607:d8fe8ae34e56
user:      rillig <rillig%NetBSD.org@localhost>
date:      Sat Feb 20 09:40:42 2021 +0000

description:
lint: reproduce the crash from ckgetopt.c 1.2

The intention of the getopt check was to analyze only those while loops
whose condition consists of the usual getopt call.  For all other while
loops, ck.while_level was intended to stay at 0.

This was not the case in ckgetopt.c 1.2 and has been fixed in ckgetopt.c
1.3.  The code did not document the intended invariants though, which it
should have done.  This will be done in a follow-up commit.

diffstat:

 tests/usr.bin/xlint/lint1/msg_338.c |  10 +++++++++-
 1 files changed, 9 insertions(+), 1 deletions(-)

diffs (23 lines):

diff -r 7570e1aa08e9 -r d8fe8ae34e56 tests/usr.bin/xlint/lint1/msg_338.c
--- a/tests/usr.bin/xlint/lint1/msg_338.c       Sat Feb 20 09:36:30 2021 +0000
+++ b/tests/usr.bin/xlint/lint1/msg_338.c       Sat Feb 20 09:40:42 2021 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: msg_338.c,v 1.1 2021/02/19 12:28:56 rillig Exp $       */
+/*     $NetBSD: msg_338.c,v 1.2 2021/02/20 09:40:42 rillig Exp $       */
 # 3 "msg_338.c"
 
 // Test for message: option '%c' should be handled in the switch [338]
@@ -41,5 +41,13 @@
                }
        }
 
+       /* A while loop that is not related to getopt is simply skipped. */
+       while (o != 0) {
+               switch (o) {
+               case '?':
+                       o = ':';
+               }
+       }
+
        return 0;
 }