I am attaching a new test case avro-1034-test-2.c. This test case
results in a segmentation fault when I apply Doug's patch
does not result in a segmentation fault when I apply my patch
I don't know if the problem is with Doug's patch or if I am operating
incorrectly on the resolved value in my test case. Specifically, I am
calling avro_value_reset() on the resolved value.
In this test case I populate a source value (nested) with a nested
array and read it using a resolved reader value (resolved_record). I
then call avro_value_reset on both the source value and the resolved
reader value. I then re-populate the source value with a larger nested
array, and proceed to read it using the same resolved reader.
On this second read attempt, the code seg-faults, when I am using
Doug's patch. However, if I do not call avro_value_reset() on the
resolved reader value, then the code does not seg-fault and appears to
operate correctly, again using Doug's patch.
Is it permissible to call avro_value_reset() on the resolved reader