Julianus Pfeuffer
Julianus Pfeuffer
While your PR is a great addition that we were discussing some time ago already for autowrap (to have a choice between views and copies), I am not sure if...
We are actually happy about some feedback here. We restrained from implementing too many "views", yet, since beginner python programmers might not be very familiar with such concepts and this...
Hmm I think the problem might be, that your PyFloatDataArray does not own shared_ptr\ but shared_ptr\.data. This does not increase the reference count to the parent object. If the original...
But doesn't your buffer only keep data_array.data "alive"? You would need to keep a shared pointer to the whole FloatDataArray object to increase its reference count and prevent it from...
Ah so you want to change how pyopenms.FloatDataArray works internally, not what pyopenms.FloatDataArray.get_data returns? Yes should work.
Hmm I think we meant a different "metavalue" storage. Don't we have a DataProcessing field somewhere or was this only for MS experiment. Or SearchEngineSettings? Although it is not directly...
@copilot What do you think about the comments?
@copilot you don't need special static variables to store the names. They should always be Tool:param_name.
I already found one problem. We use inline conversions for Strings! We do: ```python def load(self, filename , MSExperiment in_1 ): """ load(self, filename: Union[bytes, str, String] , in_1: MSExperiment...
I give up. Conclusion: autowrap comment parsing is still trash and needs to be fixed first.