* Update FCI example

* Add HFX on grids (fix pyscf#2011)

* Fix define_xc regarding the new eval_xc_eff in numint (fix pyscf#2092)

* syntax issue

* Fix tests

* Update FCI example

* Add HFX on grids (fix pyscf#2011)

* Fix define_xc regarding the new eval_xc_eff in numint (fix pyscf#2092)

* syntax issue

* Fix tests

* Fix the transformation matrix of Mole.decontracted_basis (fix pyscf#2348)

* Fix DHF dipole moment (pyscf#2362)

* fix dhf dipole moment

* dhf dipole sign

* Handle high angular momentum basis functions in pbc-GDF (fix pyscf#2169)

* Improve docs and examples (pyscf#2359)

* Improve the example of UHF Hamiltonian customization. (pyscf#2350)

* Add comments for tdscf modules (for issue pyscf#2246)

* Improve hf.dispersion imports

* Fix tdrhf

* Allow scientific notations in libxc.parse_xc (fix pyscf#1787)

* Fix define_xc example (pyscf#2364)

* Update FCI example

* Add HFX on grids (fix pyscf#2011)

* Fix define_xc regarding the new eval_xc_eff in numint (fix pyscf#2092)

* syntax issue

* Fix tests

* Improve PBC ECP accuracy (pyscf#2360)

* Fix dump_input in Cell

* Adjust ECP lattice sum (fix pyscf#2199)

* Missing import

* Some elements in pbc-nr_ecp are not zeroed out

* Restore old ecp implementation

* Restore incore.py

* Remove unused imports

* Improve smearning-ROHF (pyscf#2356)

* Fix smearning-ROHF

* Fix tests

* Disable Smearing-ROHF

* Enable spin-separated 3-RDMs (pyscf#2357)

* Add spin-separated rdm123

* Add test, fix lints, move example

* Add requested asserts to fci/test

* Wrong transition dipoles of triplet tddft states (fix pyscf#2370)

* Update tests

* Fix `examples/dft/31-xc_value_on_grid.py` (pyscf#2376)

* fix example

* fix minor

* comment

* unpack eval_rho correctly

* Update 31-xc_value_on_grid.py

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Bugs in ECP parser and integral code (fix pyscf#2363)

* Fix SGX initialization issue (fix pyscf#2065)

* Assert non-uniform mesh not being applied to FFT

* fix all ruff code C4 warnings

* Fix numpy.unique inverse indices due to NumPy 2.0 API changes

* Rewrite _break_dm_spin_symm (pyscf#1841)

* Rewrite _break_dm_spin_symm to form a doublet state density instead of zeroing out interatomic blocks of density matrix

* Control breaksym algorithm using UHF attributes

* Refactor UHF initial guess functions

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* deallocate memory in libxc interfaces

* add total_cycles in the mf class (pyscf#1764)

* add total_cycles in the mf class

Therefore, the total number of cycles can be extracted from the attribution of mf class. With this attribute, when replacing the randomized dm0 with another predicted dm0, the total number of cycles can be extracted to compare the reduced number of cycles.

* attribute name from total_cycles to scf_cycles 

 To minimize ambiguity of attribute name.

* Add docstring for the cycles attribute in SCF methods

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Save cycle count (pyscf#1351)

* Add a class to save iteration information and apply it to CCSD.

Signed-off-by: Kazuhito Matsuda <kazuhito.matuda@gmail.com>

* Reflect comments.

Signed-off-by: Kazuhito Matsuda <kazuhito.matuda@gmail.com>

* Simplify the storage of iteration info in ccsd

---------

Signed-off-by: Kazuhito Matsuda <kazuhito.matuda@gmail.com>
Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Improve docstring and other fixes (pyscf#2377)

* Add shls_slice kwarg for pyscf.df.incore (fix pyscf#1729)

* Fix example 31-xc_value_on_grid.py (fix pyscf#1725)

* Fix FCI wfn_symmetry example (fix pyscf#1726); simplify the fci direct_spin0_symm implementation

* Update GHF-X2C docstrings (fix pyscf#1707)

* Generating XC_CODES from libxc (fix pyscf#1692, pyscf#48)

* Lint issues

* Update copyright line (issue pyscf#1421)

* Add example for BCCD (fix pyscf#1591)

* Fix dump_input order in Cell.build (pyscf#2385)

* Fix dump_input order in Cell.build

* Restores the test

* traceless quadrupole moment implementation for SCF (pyscf#2255)

* traceless quadrupole moment implementation for SCF

* formatting and whitespace

* allow setting origin location in dip_moment

* pass with_origin through analyze; update dipole test to use explicit origin

* use (0,0,0) as default origin for dipole/quadrupole

* Add support for fractional coordinates in Cell (pyscf#2286)

* Add support for fractional coordinates in Cell

* Handle undefined attributes in Cell.to_mol()

* Optimize fft performance (pyscf#2276)

* Use scipy.fft module by default

* Optimize cache utilization of fft

* lint error

* fix flake F841 warning

* disable flake8

* Population analysis for KSCF (pyscf#2216)

* add mulliken_meta for kscf

* bug fix for j-build in get_jk

* bug fix for k-build in get_jk

* clean up mulliken_meta

* update k2gamma to_supercell_ao_integrals

* update pop test for k2gamma

* fix flake8

* update ref pop/chg value in kscf tests

* update comment

* Update the docstring of mulliken_meta

* Update docstring of mulliken_meta

* Update kuhf.py

---------

Co-authored-by: hongzhouye <>
Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Optimize nuclear hessian (pyscf#2241)

* Remove chkfile dependency in hessian

* Update krylov solver

* Update cphf

* Update hessian code

* Update eph hessian

* Fix a shape bug in ucphf

* Remove duplicated code

* clean up

* Adjust tests

* Update docstring of lib.linalg_helper.solve

* Fix PBC-TDDFT bugs and improve TDDFT eigenvalue solver (pyscf#2382)

* Add lr_eig solver for TDHF eigenvalues

* Fix complex conjugation bug in lr_eig

* Improve _lr_eig

* Update PBC-tddft using lr_eig

* Restore davidson_nosym solver in linalg_helper

* Simplify linalg_helper implementation

* Debug kshift for PBC-TDHF

* Fix kshift issue in KTDHF; Improve lr_eig

* Fix integration bugs in MGGA functional for pbc-dft

* tdhf code cleanup

* Add lr_eigh

* Add symmetry-adapted diagonalization in lr_eig

* Update tduhf and tduks

* Update tdghf, tdgks, tddhf, tddks

* Using lr_eigh in PBC-tdscf modules

* Update tests

* Update docstring and comments

* More efficient PBC-CDERI loading (pyscf#2392)

* More efficient PBC-CDERI loading

* correctness and readability

* workaround for h5py bug

* make requested changes

* Correct slicing for non-datasets

* Update docstring of _hstack_datasets

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Fix bug in GHF-X2C dipole function (pyscf#2396)

* Fix bug in GHF-X2C dipole function (fix pyscf#2393)

* typo

* Fix FCI wfn symmetry validation code for Dooh symmetry (fix pyscf#2383)

* fix SymAdaptedGHF.eig for complex orbitals

* add int1e_ovlp for fcidump to scf

* The missing quadrature adjustment in treutler_ahlrichs radial grids (pyscf#2186)

* The missing quadrature adjustment in treutler_ahlrichs radial grids.

* Upgrade tests wrt the change of Treutler-Ahlrichs grids

* Change default Treutler-Ahlrichs grids

* Updates for DF auxiliary basis generation (pyscf#2285)

* Auxbasis with keyword etb

* TODO: ORCA auto ABS

* Fix autoaux

* Update the default aug_etb settings

* Make autoaux match bse implementation

* Update autoaux tests

* Restore mdf test

* Update pyscf/df/addons.py

Co-authored-by: Susi Lehtola <susi.lehtola@alumni.helsinki.fi>

* Update comments

* lint warnings

* Improve df auxbasis output messages

---------

Co-authored-by: Susi Lehtola <susi.lehtola@alumni.helsinki.fi>

* Release 2.7 (pyscf#2407)

* Release 2.7

* add link to news page

* Fix an error in the native interface to the Basis Set Exchange
introduced in commit 430a9db and add
an example of its use.

* integral screening for gaunt and breit term (pyscf#2437)

* integral screening for gaunt and breit term

* Revert to use g_lssl integral it self to screen gaunt.
Since the q_cond for g_lssl and g_lsls are not symmetry, a asymmetry q_cond is added.

* mf object is reinitialized at line 241, if not, the previously initialized VHFOpt for gaunt will be used for breit as well and will thus cause wrong result.

* trim whitespace

* fix uccsd failing for empty active space

* add ccpy extension

* Improve error message for ao2mo module (fix pyscf#2473)

* Update 09-band_ase.py

In version 3.11.0 of ASE, the ase.lattice.bulk() function has been moved to ase.build.bulk.

* Fix compiling issues. Remove the non-standard ffsll function (pyscf#2449)

* Fix compiling issues. Remove the non-standard ffsll function

* add header for ffsll and use builtin if available

* add header for ffsll and use builtin if available

---------

Co-authored-by: chillenb <chillenbrand15@gmail.com>

* Update test tolerances (pyscf#2482)

* Update test tolerances

* Update tolerances

* Adds COSMO-RS functionality (pyscf#2480)

* Adds COSMO-RS functionality with tests and examples

* Fixes flake lint

* Fixes SCF energy test threshold

* Fixes SCF energy test threshold x2

* `SCFWithSolvent.get_fock` no longer requires specifying `dm` as keyword argument

* refactor dfmp2 & rpa (pyscf#2418)

* refactor dfmp2; add dfump2

* add parallel trisolve

* add except for loading libmp

* fix lint error

* refactor RPA code

* dtrtrs_ -> dtrsm

* dtrsm -> dtrsm_

* fix typo

* More error checks and cleanups

* Fix bugs introduced in the previous commit

---------

Co-authored-by: hongzhouye <>
Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Fixes calculating exchange integrals when only_dfj=True, issue pyscf#2456 (pyscf#2457)

* Fixes pyscf#2456

* Fixes bug in calculation of DF exchange integrals introduced in previous commit.

* Add tests

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Update 04-input_basis.py

The Link for dbasis set exchange is outdated (since more the one year) - the new URL is https://www.basissetexchange.org/

* Add minimally augmented basis sets

From Zheng, Jingjing, Xuefei Xu, and Donald G. Truhlar.
"Minimally augmented Karlsruhe basis sets."
Theoretical Chemistry Accounts 128 (2011): 295-305.
DOI: https://doi.org/10.1007/s00214-010-0846-z

Downloaded from http://sobereva.com/509 in NWChem format

Co-authored-by: Derk Kooi <derkkooi@gmail.com>

* Fix dump_chk in UCASSCF (fix pyscf#2432)

* Integer overflow due to the type casting issue in numpy-2.0 (fix pyscf#2481)

* Periodic stability analysis update to match with molecular version

* Fix: add missing conj() to t2 amps in init_amps of gccsd and uccsd routines (pyscf#2501)

* Fix: add missing conj() to t2 amps in init_amps

* Fix: remove .conj() from eris_oovv at the emp2 computation step

* fix mp2 compile errors (pyscf#2511)

* limit effects of NumPy 2 type promotion

* Add functions to read and write Cell geometry including lattice vectors (pyscf#2505)

* feat: Add functions to read and write Cell geometry

* examples: reading and writing of Cell geometry

* Allow scale different from 1 and fix import

* Fix lint/flake8 whitespace

* Allow lowercase Direct or Cartesian in VASP input

* Rebuild Cell after reading from string; Add tests

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Parallel in-place matrix transposition and scaling (pyscf#2507)

* in-place matrix scaling and transposition, imatcopy style

* manually collapse main loop nest

* use long long rather than ssize_t

* fix build with _OPENMP undefined

* make requested changes to simplify code

* remove unused defines

* reduce complex tilesize

* Update test_numpy_helper.py

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Add def2-mtzvp and def2-mtzvpp basis sets for 3c methods

* Fix iterative damping error (pyscf#2524)

* Fix iterative damping error

Fixes pyscf#2523

* Fix iterative damping for UCCSD and spin!=0

* Test UCCSD damping

* satisfy pylint

* Update the blacklist of dispersion correction

* some lightweight array helper functions (pyscf#2515)

* some lightweight array helper functions

* satisfy linter

* make requested changes

* Add out argument for entrywise_mul

* restore lib.zeros for testing

* Fix biased implementation for the becke radi method (pyscf#2499)

* Fix biased implementation for the becke radi method

This fixed the Becke radial integration scheme as described in pyscf#2490.

* Update test_grids.py

* Update radi.py

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Verify Examples (pyscf#2379)

* add verify_examples

* make executable

* minor fix relating to cwd

* add doc string and usage example

* Update to libxc 7.0.0 (pyscf#2458)

* Update to libxc 7.0.0

* Update dependency in CI jobs

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* a few rst fixes about RST213 in the doc, plus fixing one bad \[

* libqcschema: load qcschema json and build scf (pyscf#2296)

* libqcschema: load qcschema json and build scf

* Update test_libqcschema.py

* Update libqcschema.py

* Update libqcschema.py

* Move qcschem.py. Add nmo check.

* fix comment typo

* Handle lint warnings

* Update qcschema.py

* Fix aux file location in test_qcschema.py

* Update test_qcschema.py

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Adjust TDDFT amplitudes to follow the CIS convention (pyscf#2528)

* Align the strucure of TDDFT amplitudes with the CIS convention (fix pyscf#2246)

* Fix x2c.tdscf

* fix pair_mask bug in _dot_aow_ao_frac

* Multiple small fixes: bugs, comments, and tests (pyscf#2526)

* Handle MultiGridFFTDF2 in scf._response_functions

* Improve the output message when generating DF etb auxbasis

* NumPy objects in Mole.dumps

* Set hermi=0 as the default for df_jk.get_jk

* Fixes and updates for pbc.dft.numint

* Changing contraction order in eval_rho to reduce required FLOPs

* Update to_gpu interface for PBC modules

* Fix ASE interface

* Fix a TDDFT lr_eig convergence check bug

* Handling the leading '#' in basis string

* Exclude kpt, kpts and some attributes from to_gpu function

* Fix undefined variable in tdscf._lr_eig

* Add examples for DFT+U

* Update DFT-D3 example

* Add geometry optimization examples

* Exclude attribute mesh from to_gpu function

* Dyall basis missing from pip release

* max_cycle=0 bug in SOSCF (fix pyscf#2529)

* Adjust soscf tests

* Update the to_gpu conversion function

* Fix Coulomb imaginary part checker

* Adjust the initialization of some attributes

* dfmp2: fix invocation of mf.get_veff

* a few ao2mo type checks

* Accelerate DFT Density and XC Potential, Especially for MGGA (pyscf#2552)

* fix pair_mask bug in _dot_aow_ao_frac

* improve parallelization for small systems with MGGA; add criterion for MO-based vs DM-based density evaluation

* check with pair_mask will exist

* remove unnecessary transpose; adjust rho1/rho2 switch

* Update eval_rho1 in numint.py for non-hermitian dm

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Automatically apply SCF initial guess from existing wavefunction (pyscf#2541)

* Automatically apply SCF initial guess from existing wavefunction

* Update tests

* Add another _decompose_rdm1 for non-hermitian density matrices in df.grad.rhf (pyscf#2570)

* change _decompose_rdm1 using SVD for more general input dm matrix

* add if hermi for decompose dm in get_jk

* Improve TDDFT diagonalization program (pyscf#2561)

* Improve TDDFT eigen solver

* Bugfix and optimization for lr_eig

* Restore compatibility between lr_eig and real_eig

* Fix tdscf symmetric_orth function for complex vectors

* Reduce numerical errors in TDHF diagonalization

* Dimension check

* Fix tests

* Bugs in libxc.define_xc_ (fix pyscf#2551)

* Change pypi upload configurations in release pipeline

* Fix SR-ERI integral screening estimator (pyscf#2450)

* The SR-ERI estimator underestimated the upper bound when two basis functions were located at the same center. This fix adjusts the upper bound estimator.

* Prevent accident overwritten in ctypes.Structure

* Fix tests

* A few more small API updates and function improvments (pyscf#2573)

* Inconsistency between ROKS density matrix and its mo_coeff tag when reading RHF initial guess

* using a single stdout everywhere by default

* Mute unnecessary warning messages (fix pyscf#2558)

* Add kmesh attributes for PBC SCF class

* make the kpts attribute read only in PBC SCF class

* Improve kpts_to_kmesh transformation, supporting larger k-points mesh

* An incorrect warning for 'converged' attribute in CCSD (fix pyscf#2542)

* Update mcol example

* Fix GTH-PP integrals (fix pyscf#2575) (pyscf#2577)

* Fix GTH-PP integrals (fix pyscf#2575)

* Fix test

* lint

* Fix CCSD two-particle density matrix for complex orbitals (pyscf#2531)

* update hf make_rdm2 docs

* Fix the CCSD 2-rdm for complex orbitals

* Fix test

* Fix merging issues

* Fix TypeError when cc.frozen is nonzero numpy.integer (pyscf#2576)

* Fix TypeError when cc.frozen is nonzero numpy.integer

Previously, the frozen orbital check could attempt to call len() on a
numpy.int64, causing a TypeError. Add a hasattr check to properly handle
numpy integer types in the frozen orbital logic.

* Add same hasattr check to _response_dm1

* Add tests and fix frozen attributes in various modules

* Update test_ccsd.py

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Remove the misleading SCS-MP2 output message (fix pyscf#2537) (pyscf#2578)

* take cell.rcut into account when generating becke grids (pyscf#2460)

* take cell.rcut into account when generating becke grids

* 🐛 bugfix for lower dimensions

* revert gen_grids

* 🐛 fix issue with dimensions

* Add a test

* Fix test

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Update the mo_coeff argument treatments in CASCI and CASSCF scanner (pyscf#2581)

* Improve the RSH treatment in DFT (pyscf#2525)

* Compute LR or SR K matrices directly for certain XC functionals

* Fix libxc.rsh_coeff parser for custom RSH such as "hse06+HF"

* Better SR/LR treatments for RSH in TDDFT and scf/_response_fuctions.py

* Remove TDDFT code that were never used

* Fix various issues for previous commits and update tests

* Fix tests

* Drop invalid test

* Fix pbc.tdscf tests and update docstrings

* Improve TDDFT eigen solver

* Update _lr_eig

* error introduced in previous commit

* Bugfix and optimization for lr_eig

* Restore compatibility between lr_eig and real_eig

* Fix tdscf symmetric_orth function for complex vectors

* Reduce numerical errors in TDHF diagonalization

* Dimension check

* Release 2.8

* Update release pipeline; Disable Fortran extensions on macos-x86_64

* add documentation for libcint integral expressions (pyscf#2599)

* add documentation for libcint integral expressions

* Update moleintor.py

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* lint fix

* Add issue templates, encourage other questions in Discussions

As discussed at the second PySCF meeting Jan 13-14, 2025.

* update moleintor documentation (pyscf#2604)

* update moleintor documentation

* flake8 ignore line too long in moleintor docstring

* fix the long line the correct way

---------

Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Fix typos in doc examples (pyscf#2672)

* Fix typos in doc examples

* Fix doc atom_pure_symbol

* Update default auxiliary basis for def2-mtzvp and def2-mtzvpp

* Fix two links to the libxc website

* Fix dimension bug in spinor X2C code (pyscf#2611)

* Fix dimension bug in spinor X2C code

* syntax error in tests

* Fix dimension error

* Bug fix in C-PCM and SS(V)PE gradient

* Complex density matrices for DFHF (pyscf#2676)

* Complex dms for DFHF (fix pyscf#2670)

* Fix dimension issue

* Update dtype check

* Analytical PCM Hessian (pyscf#2691)

* Bug fix in C-PCM and SS(V)PE gradient

* Anaytical PCM Hessian

* Fix linter error

* Small fix to avoid crash in URPA for H atom and other systems with zero number of electrons in the beta spin channel

* twine==6.0.1

* Various small fixes and updates (pyscf#2677)

* Fix example pbc/33-soc_integrals.py (fix pyscf#2436)

* Compiling 4c1e integral

* Fix finite-size error in GDF casued by large k-mesh

* typo

* Guess mol.symmetry when reading from FCIDUMP (Fix pyscf#2586)

* Improve PBC GDF for dimension=0 systems (Fix pyscf#2608)

* Fix 21-nosymhf_then_symcasscf example (fix pyscf#2207)

* Skip integral unpack functions in RCCSD if applicable (fix pyscf#2346)

* Improve kpts_to_kmesh

* Fix a DF-UHF hessian dimension bug (fix pyscf#2674)

* lattice sum range issue for low-dimensional systems

* Adjust pbc GDF eigenvlaue decomposition accuracy

* Restore kpts_to_kmesh

* Fix AFTDF case for dimension=0 system

* Fix BPC 0d HF tests

* Adjust pbchf test threshold

* Improve mole.atom_coords() function

* Remove print statement

* UCCSD with density fitting (pyscf#2712)

* UCCSD with density fitting

* fix np.einsum options

* fix a few typo in mcscf

* Add example for ring CCD (fix pyscf#1149)

* Support "mol.symmetry = 'SO3'" in Mole.build (Fix pyscf#1992) (pyscf#2709)

* Support "mol.symmetry = 'SO3'" in Mole.build (Fix pyscf#1992)

* Update CI runner to ubuntu-latest

From ubuntu-20.04, which is about to be unavailable.

* Very small bug fix in nmr example

* Changed wrong < into <=

Enables slices for auxiliary basis in density fitting

* Fixes COSMO-RS functionality (pyscf#2733)

* Adds volume and level-of-theory parameters to write_cosmo_file

* Fixes git issues

* Changes COSMO-file format to Turbomole

* Adds note on the abscence of outlying charge correction in PCM computations

* Fixes COSMO-RS volume test

* PR review fixes

* Add linear dependency checks in tddft _lr_eig (pyscf#2735)

* Add linear dependency checks in tddft _lr_eig

* Symmetry detection bug due to np.round tolerance (pyscf#2715)

* Fix point-group symmetry verficication due to the np.round issue

* indexing bug

* Fix symmetry detection tests

* Fix argsort_coords

* Fix bug in sort_coords

* Update comments

* add dump_flags for NEVPT2 (pyscf#2747)

* add dump_flags for NEVPT2

* add dump_flags for NEVPT2

* Support FCIDUMP header ending with `/`

Currently only `&END` is supported but both are valid and some programs
use `/` exclusively.

* fix dfmp2

* Fix call to DFUMP2 in init

* fix docstring

* fix a bug in ecp parser

* Important NotImplementedError for sacasscf grad

The sacasscf gradients are not correct for unequal state weights!

* Fix DF auxiliary basis assignment for dict and ghost (pyscf#2753)

* Fix DF auxiliary basis assignment for dict and ghost

* Fix failing tests due to empty _basis

* Fix elif syntax

* Fix logic for handling default basis and add more tests

* Update tests that used DF with dict

* Better initial guess for smearing (pyscf#2728)

* better init guess for mu in smearing

* Update test_addons.py

* lint

---------

Co-authored-by: hongzhouye <>
Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>

* Fix libxc interface for passing omega (pyscf#2760)

* use xc_func_set_ext_params_name

* use xc_func_find_ext_params_name

* complete refactor

* fix kpts_band error

* Bump the cmake minimum requirements to 3.22 (aligned to ubuntu 22.04)

* fix near-symmetry issue with fromfile/fromstring and add warning

* Add a general driver for finite difference gradients and hessian (pyscf#2774)

* Add a finite difference driver

* Add geomopt interface

* restore pbc.dft.rks

* typo

* Make ``fci_dhf_slow.make_rdm1`` Faster (pyscf#2775)

* optimize ``fci_dhf_slow.make_rdm1``

* add tests for fci.fci_dhf_slow.make_rdm1

* Release 2.9 (pyscf#2765)

* Proposal: user-defined basis set aliases (pyscf#2776)

* More convenient handling of user-defined basis sets

* Add example for custom basis set aliases

* Renumber sub-examples in 04-input_basis.py

* cmake 4.0 compatibility issue (pyscf#2777)

* cmake 4.0 compatibility issue

* Add python 3.13 in conda build

* fix error message for functional prefix

* Fix potential bug of 6th order derivative in `dft.xc_deriv.transform_xc` (pyscf#2784)

* fix dft.xc_deriv._diagonal_indices for (possibly) even higher derivatives

* change how diagonal_indices tested

* update conda build in actions

* correct SUBGROUP of C2h

* ECP and auxiliary basis set auto assignment (pyscf#2796)

* ecp and auxbasis auto assignment

* BSE predfined auxbasis

* typo

* bugfix

* Assignment of the predefined auxbasis

* Fix absent auxbasis

* Improve warning message when auxbasis is not found

* Update examples (pyscf#2795)

* Update examples and documents

* Fix libxc format output

* Update DF example

* Authors list

* Fix autoabs example

* Remove misleading comments (fix pyscf#2787)

* Some comments to explain the code

* Fix density fitting method conversion issue (fix pyscf#2788) (pyscf#2803)

* Fix to_hf and to_ks for GDF SCF methods (fix pyscf#2788)

* Fix a class name composition bug

* Handle PBC DF conversion

* import error

---------

Signed-off-by: Kazuhito Matsuda <kazuhito.matuda@gmail.com>
Co-authored-by: Qiming Sun <osirpt.sun@gmail.com>
Co-authored-by: Xubo Wang <34044625+xubwa@users.noreply.github.com>
Co-authored-by: Brian Zhao <zdj519@gmail.com>
Co-authored-by: matthew-hennefarth <matthew.hennefarth@gmail.com>
Co-authored-by: Frédéric Chapoton <chapoton@unistra.fr>
Co-authored-by: Susi Lehtola <susi.lehtola@gmail.com>
Co-authored-by: Xiaojie Wu <wxj6000@gmail.com>
Co-authored-by: Haiyang Yu <1161283769@qq.com>
Co-authored-by: Kazuhito MATSUDA <kazuhito.matuda@gmail.com>
Co-authored-by: Christopher Hillenbrand <chillenbrand15@gmail.com>
Co-authored-by: Hong-Zhou Ye <hzyechem@gmail.com>
Co-authored-by: Huanchen Zhai <hczhai@ucla.edu>
Co-authored-by: Susi Lehtola <susi.lehtola@alumni.helsinki.fi>
Co-authored-by: Xing Zhang <zhangxing.nju@gmail.com>
Co-authored-by: gkc1000 <gkc1000@gmail.com>
Co-authored-by: juanjoaucar <93452131+juanjoaucar@users.noreply.github.com>
Co-authored-by: Sebastian Ehlert <28669218+awvwgk@users.noreply.github.com>
Co-authored-by: Ivan Chernyshov <ivan-chernyshoff@yandex.ru>
Co-authored-by: kvkarandashev <kvkarandashev@gmail.com>
Co-authored-by: jonas-greiner <107192561+jonas-greiner@users.noreply.github.com>
Co-authored-by: steto123 <steffen@sthomas.de>
Co-authored-by: Sebastian Ehlert <sehlert@microsoft.com>
Co-authored-by: Derk Kooi <derkkooi@gmail.com>
Co-authored-by: Ethan Vo <ethan.vo@hotmail.com>
Co-authored-by: rhkim79 <50161497+rhkim79@users.noreply.github.com>
Co-authored-by: Timothy Berkelbach <tberkel@users.noreply.github.com>
Co-authored-by: Michal Krompiec <16683215+mkrompiec@users.noreply.github.com>
Co-authored-by: Thijs Vogels <t.vogels@me.com>
Co-authored-by: Andrew Jenkins <63872995+ajjenkins1@users.noreply.github.com>
Co-authored-by: Kyle Bystrom <kylebystrom@gmail.com>
Co-authored-by: Chaoqun ZHANG <cq_zhang@outlook.com>
Co-authored-by: Al Nejati <alireza.n.j@gmail.com>
Co-authored-by: simonganne01 <93279056+simonganne01@users.noreply.github.com>
Co-authored-by: Verena Neufeld <veri7neuf@gmail.com>
Co-authored-by: Yann Damour <77277447+Ydrnan@users.noreply.github.com>
Co-authored-by: Henry Wang <henrywang0142857@gmail.com>
Co-authored-by: henryw7 <60555480+henryw7@users.noreply.github.com>
Co-authored-by: Egor Trushin <e.v.trushin@gmail.com>
Co-authored-by: Huanchen Zhai <44282896+hczhai@users.noreply.github.com>
Co-authored-by: jeanwsr <srwang20@fudan.edu.cn>
Co-authored-by: Matthew R Hermes <mrhermes@uchicago.edu>
Co-authored-by: Oskar Weser <oskar.weser@gmail.com>
Co-authored-by: Victor Yu <victor.wz.yu.1012@gmail.com>
Co-authored-by: hongzhouye <>
Co-authored-by: Zhenyu Zhu ajz34 <ajz34@outlook.com>
Co-authored-by: Dayou Zhang <zhan6350@umn.edu>