[PATCH] test/T360-symbol-hiding.sh: trim extra output from readelf

[PATCH] test/T360-symbol-hiding.sh: trim extra output from readelf

[David Bremner]

readelf on (at least) ppc64le generates symbol versioning (sometimes)
at the end of the line inside '()' and some extension to the Ndx name
inside '[]'. Remove this output to allow our simple column based
parsing to work.
---
 test/T360-symbol-hiding.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

I dislike this feeling of playing whack-a-mole parsing ill-defined
output, but here we are. I looked into alternatives a bit, but nothing
grabbed me. abi-dumper is a perl script wrapping readelf, so if we
wanted we could delegate the hacks to a third party tool.

I also thought about combining the sed into the awk call, but it
seemed like too much work to me, probably because I am not an awk
expert.

We could also just drop the test. In Debian we have dpkg-gensymbols
that manages more or less the same task; I'm not sure if other
environments have something similar.

diff --git a/test/T360-symbol-hiding.sh b/test/T360-symbol-hiding.sh
index 3f00dd38..b2a5694a 100755
--- a/test/T360-symbol-hiding.sh
+++ b/test/T360-symbol-hiding.sh
@@ -26,7 +26,7 @@ test_begin_subtest 'checking output'
 test_expect_equal "$result" "$output"
 
 test_begin_subtest 'comparing existing to exported symbols'
-readelf -Ws $NOTMUCH_BUILDDIR/lib/libnotmuch.so | \
+readelf -Ws $NOTMUCH_BUILDDIR/lib/libnotmuch.so | sed -e 's/\[[^]]*\]//' -e 's/([^)]*)$//' |\
     awk '$4 == "FUNC" && $5 == "GLOBAL" && $7 != "UND" {print $8}' | sort -u > ACTUAL
 sed -n 's/^\(notmuch_[a-zA-Z0-9_]*\)[[:blank:]]*(.*/\1/p' $NOTMUCH_SRCDIR/lib/notmuch.h | sort -u > EXPORTED
 test_expect_equal_file EXPORTED ACTUAL
-- 
2.29.2

Re: [PATCH] test/T360-symbol-hiding.sh: trim extra output from readelf

[David Bremner]

David Bremner <david@tethera.net> writes:

> readelf on (at least) ppc64le generates symbol versioning (sometimes)
> at the end of the line inside '()' and some extension to the Ndx name
> inside '[]'. Remove this output to allow our simple column based
> parsing to work.

Simplified version (thanks Tomi) applied.

d