On 07/05/2013 02:20 AM, Nathan Scott wrote:
I've also added qa/875 which I think is exposing some issues (no .out
file yet as a result) - could you take a look at that for me while I'm
away next week? Thanks!!! Test 720 would be a good comparison point.
We may have a bit of work to do in __pmUnparseHostAttrsSpec still, and
there's lots of nasty looking errors from valgrind coming out the test
program (parsehostattrspec.c). cheers. -- Nathan
Hi Nathan,
I've attached two patches for review:
__pmUnparseHostAttrsSpec.patch: Adds the missing code for supporting
local:// and unix://. This fixes test 720 and produces reasonable output
for test 875 (also attached).
memoryleak.patch: fixes the memory leak in the parsehostattrs.c. I based
freeing the hash array in the caller on similar code in logutil.c. Would
it may make more sense to free it in __pmFreeHostAttrsSpec()?
Dave
__pmUnparseHostAttrsSpec.patch
Description: Text Data
memleak.patch
Description: Text Data
875.out
Description: Text document
|