Fix the HL H5TB fill-info #5332 #5333
Conversation
byrnHDF
added
Priority - 1. High 🔼
byrnHDF
requested review from
lrknox,
derobins,
fortnern,
jhendersonHDF,
qkoziol,
vchoi-hdfgroup,
bmribler,
glennsong09,
mattjala and
brtnfld
as code owners
There was a problem hiding this comment.
https://my.cdash.org/viewTest.php?onlypassed&buildid=2833655
It would be nice if you include GH Actions for i386 as well in this PR.
i386 action already exists - just needed to remove special check for this failure |
| @@ -2676,6 +2686,13 @@ H5TBdelete_field(hid_t loc_id, const char *dset_name, const char *field_name) | |||
| if (H5TB_attach_attributes(table_title, loc_id, dset_name, nfields, tid_3) < 0) | |||
| goto out; | |||
|
|
|||
| if (NULL == (src_offset = (size_t *)malloc((size_t)nfields * sizeof(size_t)))) | |||
There was a problem hiding this comment.
It doesn't look like src_offset is used below here? The lines below are still assigning member_offset = H5Tget_member_offset(tid_3, (unsigned)i);
There was a problem hiding this comment.
Looks like I didn't save the change.
There was a problem hiding this comment.
It looks similar to the other cases where the H5TBget_field_info() call is needed, but it's hard to say. Would probably need to have a specific test for it to check.
Changed the test values to be unique to verify correct values are used.
Added a h5dump test to test_table so it could verify that fill values are correct.
Fixes #5332