move h5 ref page from dft_tools and update - #20
Merged
Merged
Conversation
The VASP converter wrote the number of spin channels 'ns' into dft_symmcorr_input/SP instead of the SP flag, so a paramagnetic calculation ended up with SP = 1 there while dft_input/SP = 0. Shift it by one, as is already done in convert_dft_input(). This has no effect on any calculation: triqs_dft_tools.symmetry.Symmetry reads SP but only uses it in commented-out code, and the VASP symmetry group is a stub with n_symm = 0. The four VASP converter reference archives are updated accordingly (dft_symmcorr_input/SP: 1 -> 0). Also fix the documented shape of the wannier centres returned by read_all_wannier90_data(), which are [n_spin_blocks, n_wannier, 3] and not [n_spin_blocks, 3, 3].
The description of the hdf5 structure written by the converters lived in the DFTTools documentation, but since the converters themselves now live here, this is where it belongs. Move it over and rework it against the archives produced by the test suite: - state for every group which converter writes it, which converter and SumkDFT argument names it, and what reads it back, plus an overview table of all groups - document the storage conventions (lists/dicts as subgroups, complex arrays with a trailing real/imag axis, n_spin_blocs, 1-based band indices, energy units) - document the groups that were missing entirely: dft_misc_input, dft_symmcorr_input/dft_symmpar_input, dft_parproj_input, dft_bands_input, dft_transp_input, dft_contours_input, dft_update - correct dft_input: 'irep' not 'irrep', n_reps/dim_reps are per-shell lists, T has dimension (2l+1)*(SO+1), proj_mat_csc has four indices over all shells, bz_weights is a deprecated duplicate of kpt_weights, and kpts/kpt_weights/kpt_basis are not VASP specific - note where converters deviate: extra keys in shells/corr_shells, the W90 band_window layout, the Elk band_window spin count, dummy entries for T/dim_reps/n_parproj, and the VASP symmetry stub - drop the stale reference to a Wien2k section in the H(k) part and point at triqs_dftkit.hk.Converter The DFTTools side still has to be updated to drop its copy and link here.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR brings the https://triqs.github.io/dft_tools/latest/h5structure.html documentation / reference card over to the converters where it belongs. I also let Claude verify and update all sections with the actual converter data generated.