From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 28 Mar 2011 03:58:22 +0000 (+1030)
Subject: failtest: continue (without failing) if we run off end of --failpath=
X-Git-Url: https://git.ozlabs.org/?p=ccan;a=commitdiff_plain;h=3ad57bc902f31ba347df59de04168f00c626d924;hp=024fbb5f9682d3187b65849948c372c3879ed9bd

failtest: continue (without failing) if we run off end of --failpath=

This is important because we tell people to use --failpath to reproduce a
failure, but the fail path we list only goes up to the last failure injection
if the child dies, for example.

Thanks to David Gibson for prodding me to fix this...
---

diff --git a/ccan/failtest/failtest.c b/ccan/failtest/failtest.c
index 7a6f5220..b25b049e 100644
--- a/ccan/failtest/failtest.c
+++ b/ccan/failtest/failtest.c
@@ -364,7 +364,10 @@ static bool should_fail(struct failtest_call *call)
 		/* + means continue after end, like normal. */
 		if (*failpath == '+')
 			failpath = NULL;
-		else {
+		else if (*failpath == '\0') {
+			/* Continue, but don't inject errors. */
+			return call->fail = false;
+		} else {
 			if (tolower((unsigned char)*failpath)
 			    != info_to_arg[call->type])
 				errx(1, "Failpath expected '%c' got '%c'\n",