add interface to build npy_array_list objects from C

[alcubierre-drive]

No description provided.

[oysteijo]

Wow! My first real external PR! I will look at this when I come home! fre. 20. sep. 2024 kl. 13:39 skrev Lennart Klebl ***@***.***>:

…

------------------------------ You can view, comment on, or merge this pull request online at: #30 Commit Summary - ac7933e <ac7933e> add interface to build npy_array_list objects from C File Changes (4 files <https://github.com/oysteijo/npy_array/pull/30/files>) - *M* .gitignore <https://github.com/oysteijo/npy_array/pull/30/files#diff-bc37d034bad564583790a46f19d807abfe519c5671395fd494d8cce506c42947> (4) - *M* README.md <https://github.com/oysteijo/npy_array/pull/30/files#diff-b335630551682c19a781afebcf4d07bf978fb1f8ac04c6bf87428ed5106870f5> (24) - *M* npy_array.c <https://github.com/oysteijo/npy_array/pull/30/files#diff-6d9506614727a2353da2185d8aa7322d1c0861321f053970c70395179ecf90f9> (14) - *M* npy_array.h <https://github.com/oysteijo/npy_array/pull/30/files#diff-f5e45fc1f6720efadd0846f92a30e422f331c3c1235064d3fc154da0531f6bf5> (3) Patch Links: - https://github.com/oysteijo/npy_array/pull/30.patch - https://github.com/oysteijo/npy_array/pull/30.diff — Reply to this email directly, view it on GitHub <#30>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABFBMLPEXC37TLG56EBCO6DZXQCNVAVCNFSM6AAAAABOR3KV6CVHI2DSMVQWIX3LMV43ASLTON2WKOZSGUZTQNJZGYZDGMI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[oysteijo]

It looks like npy_array_alloc() take a deep copy of an array. Maybe it is better to name it npy_array_deepcopy() ?

In my work I usually have huge arrays - there should be some check on the on the allocated memory.

[alcubierre-drive]

That makes a lot of sense. Changed it :)

[oysteijo]

I see nothing wrong! Merging!

[oysteijo]

Thanks a lot Lennart!

I think actually you are touching into the problem in #14. The user can have arrays where they do not "own" the items of the array_list. If you can think of somtheing better/simpler, I would love to hear ideas.

-Øystein

[alcubierre-drive]

How about a public and a private data pointer, where the private one is left as NULL in case the array doesn't own the data?

[oysteijo]

Hmmmm... Well... Sounds a bit like adding complexity. Then we should also have a function/macro that returns the right pointer when requested.

…

On Fri, 20 Sept 2024, 21:37 Lennart Klebl, ***@***.***> wrote: How about a public and a private data pointer, where the private one is left as NULL in case the array doesn't own the data? — Reply to this email directly, view it on GitHub <#30 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABFBMLJWDCJ2UFA2ENTCLVTZXR2QHAVCNFSM6AAAAABOR3KV6CVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRUGUYDANJUG4> . You are receiving this because you modified the open/close state.Message ID: ***@***.***>