Add more iterators on Group#157
Conversation
d9587cd to
dfd4c71
Compare
dfd4c71 to
b3e1d30
Compare
b3e1d30 to
d8ffaf3
Compare
d8ffaf3 to
ca10fab
Compare
|
@aldanor This should be good to go now |
|
@mulimoen Looks good overall. I started writing review but figured it would be easier to fix myself - I'll open a PR to your PR on your fork. |
|
One minor point - why the manual |
|
|
|
For reference: mulimoen#11 |
|
@aldanor Looks good to me now. I forgot how big this release is, I thought we already published attributes, but apparently not |
|
Hm, with times not being filled out, that's not good - but I'll open a separate PR to quickly fix that. This one - let's merge then. |
| } | ||
| } | ||
|
|
||
| pub enum IterationOrder { |
There was a problem hiding this comment.
#[derive(Clone, Copy, Debug, PartialEq, Eq)]?
| ) -> herr_t { | ||
| panic::catch_unwind(|| { | ||
| let other_data: &mut Vec<String> = unsafe { &mut *(op_data.cast::<Vec<String>>()) }; | ||
| pub enum TraversalOrder { |
There was a problem hiding this comment.
#[derive(Clone, Copy, Debug, PartialEq, Eq)]?
| let name = unsafe { std::ffi::CStr::from_ptr(name) }; | ||
| let info = unsafe { info.as_ref().unwrap() }; | ||
| let handle = Handle::try_new(id).unwrap(); | ||
| handle.incref(); |
There was a problem hiding this comment.
You could probably do Group::from_id(id).expect("invalid group id"), no?
The only difference would be it doesn't incref the handle there, wonder why it's increfed manually here?
| let vtable = op_data.cast::<Vtable<F, G>>(); | ||
| let vtable = unsafe { vtable.as_mut().expect("op_data is always non-null") }; | ||
| let name = unsafe { std::ffi::CStr::from_ptr(name) }; | ||
| let info = unsafe { info.as_ref().unwrap() }; |
There was a problem hiding this comment.
Replace all unwraps within ffi code with .expect()?
2 participants
This adds
groups,datasets, anddatatypestoGroup. This also adds the generaliter_visit, which can be adapted by users to fully use the capabilities ofH5Literate(sans iteration resumption).