Refactor database

[msricher]

@enjyashraf18 @FarnazH I rebased from atomdb/master and enjyashaf18/refacor-species so that they can be merged. atomdb/master can be rebased on top of this, or we can use this PR. We just need to fix the tests. @enjyashraf18 can you help with this? Otherwise I'll try later this week when I can type more again.

Tests are saying I can't import "sph_harm" from scipy.special, but I grepped every file and "sph_harm" doesn't even appear. I'm not sure what to do about this. Every test is failing due to some import issue or another.

[msricher]

My best attempt at merging it all together is now in branch master-all-prs. The data files from LFS are gone though.... @enjyashraf18 could you add them back in? I don't know which version is the most up-to-date.

[msricher]

This is ready for review. Failing tests are due to package qc-grid being broken. Can try to get that fixed tomorrow.

[msricher]

Issue #162 still exists, but aside from that every feature works now.

[PaulWAyers]

@marco-2023 can you look at the qc-grid issue? I think we already deal with the spherical harmonic issue in grid.

The sph_harm is because scipy changed the way it works and now sph_harm is not supported in the latest version. The new scipy uses
https://docs.scipy.org/doc/scipy/reference/generated/scipy.special.sph_harm_y.html
and another version (that computes all spherical harmonics up to some order)
https://docs.scipy.org/doc/scipy/reference/generated/scipy.special.sph_harm_y_all.html

One has to be careful because the arguments have changed; as I recall the $\theta$ and $\phi$ variables swapped (maybe also the indices specifying the degree/type).

[msricher]

I fixed the sph_harm issue for now by pinning the scipy version.

[msricher]

@enjyashraf18 The compile_species function isn't working for me:

Traceback (most recent call last):
  File "/home/michelle/Git/atomdb/run.py", line 5, in <module>
    compile_species(element_symbol(i), i, 1, dataset="gaussian")
    ~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/michelle/Git/atomdb/atomdb/species.py", line 789, in compile_species
    dump(DATASET_H5FILE, fields, dataset, mult)
    ~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/michelle/Git/atomdb/atomdb/species.py", line 804, in dump
    element_folder_creator.create_hdf5_file(DATASET_H5FILE, fields, dataset, mult)
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/michelle/Git/atomdb/atomdb/datasets/gaussian/h5file_creator.py", line 342, in create_hdf5_file
    DATASETS_H5FILE.create_group("/Datasets", dataset, f"{dataset} Data")
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/michelle/Git/atomdb/venv/lib/python3.14/site-packages/tables/file.py", line 975, in create_group
    parentnode = self._get_or_create_path(where, createparents)
  File "/home/michelle/Git/atomdb/venv/lib/python3.14/site-packages/tables/file.py", line 915, in _get_or_create_path
    return self.get_node(path)
           ~~~~~~~~~~~~~^^^^^^
  File "/home/michelle/Git/atomdb/venv/lib/python3.14/site-packages/tables/file.py", line 1798, in get_node
    node = self._get_node(nodepath)
  File "/home/michelle/Git/atomdb/venv/lib/python3.14/site-packages/tables/file.py", line 1741, in _get_node
    node = self._node_manager.get_node(nodepath)
  File "/home/michelle/Git/atomdb/venv/lib/python3.14/site-packages/tables/file.py", line 454, in get_node
    node = self.node_factory(key)
  File "/home/michelle/Git/atomdb/venv/lib/python3.14/site-packages/tables/group.py", line 1197, in _g_load_child
    node_type = self._g_check_has_child(childname)
  File "/home/michelle/Git/atomdb/venv/lib/python3.14/site-packages/tables/group.py", line 421, in _g_check_has_child
    raise NoSuchNodeError(
    ...<2 lines>...
    )
tables.exceptions.NoSuchNodeError: group ``/`` does not have a child named ``/Datasets``

I can't figure it out, do you know if you missed something while porting everything to the GLOBAL_DB?

[PaulWAyers]

@marco-2023 or @gabrielasd can you see if it's clear what's going on here?

[enjyashraf18]

Hi @msricher , as I mentioned, unfortunately I had to switch back to Windows (just temporarily), so I can’t really debug properly right now.

I checked the commits to see the latest changes, do you know when this issue started? Was it right after recreating datasets_data-v000.h5 or later?

[msricher]

I didn't recreate the database, I downloaded it from git LFS. It's the same file. I'll try figuring it out next week when I'm back.

…

On Thu, Apr 23, 2026, 06:30 Enjy Ashraf ***@***.***> wrote: *enjyashraf18* left a comment (theochem/AtomDB#166) <#166?email_source=notifications&email_token=ADLODOU7QDTSL3PK2K7PBVD4XHWFPA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIMZQGM3DKOJQGIY2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJNLQOJPWG33NNVSW45C7N5YGK3S7MNWGSY3L#issuecomment-4303659021> Hi @msricher <https://github.com/msricher> , as I mentioned, unfortunately I had to switch back to Windows (just temporarily), so I can’t really debug properly right now. I checked the commits to see the latest changes, do you know when this issue started? Was it right after recreating datasets_data-v000.h5 or later? — Reply to this email directly, view it on GitHub <#166?email_source=notifications&email_token=ADLODOU7QDTSL3PK2K7PBVD4XHWFPA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIMZQGM3DKOJQGIY2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJNLQOJPWG33NNVSW45C7N5YGK3S7MNWGSY3L#issuecomment-4303659021>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADLODOTOIP5HRXTG5T76CQ34XHWFPAVCNFSM6AAAAACX2ENGF6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DGMBTGY2TSMBSGE> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/ADLODOSYHG4J32RS2AVR7XD4XHWFPA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIMZQGM3DKOJQGIY2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJKTGN5XXIZLSL5UW64Y> and Android <https://github.com/notifications/mobile/android/ADLODOX6QTTHW6K7446TH6L4XHWFPA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIMZQGM3DKOJQGIY2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLTGN5XXIZLSL5QW4ZDSN5UWI>. Download it today! You are receiving this because you were mentioned.Message ID: ***@***.***>