Structured array highlevel accessors #3098
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #3098 +/- ##
=======================================
Coverage 89.92% 89.93%
=======================================
Files 29 29
Lines 32362 32389 +27
Branches 5802 5802
=======================================
+ Hits 29101 29128 +27
Misses 1859 1859
Partials 1402 1402
Flags with carried forward coverage won't be shown. Click here to find out more.
🚀 New features to boost your workflow:
|
benjeffery
force-pushed
the
structured-array-highlevel
branch
from
941ba92 to
2e25635
Compare
| @@ -10760,7 +10760,7 @@ TreeSequence_get_individuals_metadata_offset(TreeSequence *self, void *closure) | |||
| } | |||
| individuals = self->tree_sequence->tables->individuals; | |||
| ret = TreeSequence_make_array( | |||
| self, individuals.num_rows + 1, NPY_UINT32, individuals.metadata_offset); | |||
| self, individuals.num_rows + 1, NPY_UINT64, individuals.metadata_offset); | |||
There was a problem hiding this comment.
Hmm, did the tests pass before because we didn't look at the contents of the offset arrays?
There was a problem hiding this comment.
Yeah, the low-level tests don't check the contents, just the semantics and lifetime/memory concerns.
python/tskit/trees.py
Outdated
| ) | ||
|
|
||
| @property | ||
| def individuals_metadata_offset(self): |
There was a problem hiding this comment.
I'm not sure we want to offer this array in the high-level API. It may just be confusing. I would pull this out of the current PR, as it's orthogonal
There was a problem hiding this comment.
Offset accessors removed in b74b824
|
@jeromekelleher one more thought here - we could return the raw metadata bytes if the codec doesn't support structuring? |
|
Let's not - maybe we'll think of a better way of doing things in future and having it error out now leaves those options open. |
No description provided.