Skip to content

Commit

Permalink
Merge pull request #5 from fkfest/auto/from-devel-repo
Browse files Browse the repository at this point in the history 
Merge changes from development repository
  • Loading branch information
@dnkats
dnkats authored Apr 12, 2024
2 parents 8b99180 + fb9b4b3 commit 4e1a5a5
Show file tree
Hide file tree
Showing 38 changed files with 1,815 additions and 843 deletions.
2 changes: 1 addition & 1 deletion .gitignore
View file
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
Manifest.toml
Manifest*.toml
docs/build
examples/*.out
examples/ElemCo.jl-devel
Expand Down
24 changes: 24 additions & 0 deletions CHANGELOG.md
View file
Original file line number Diff line number Diff line change
@@ -1,5 +1,29 @@
# Release notes

## Version [v0.11.0] - 2024.04.12

### Breaking

* `EC.ms` (previously of type `MSys`) in `ECInfo` is renamed to `EC.system` (of type `AbstractSystem`).
* `ECdriver` routine is moved to `CCDriver` module and renamed to `ccdriver`. The `fcidump` keyword-argument is now empty by default. It doesn't accept list of methods anymore, only one method at a time.
* The driver routines and macros return energies as `NamedTuple`.
* The SVD methods have to be called now as `SVD-<methodname>`, e.g., `svd-dcsd`.
* The `@svdcc` macro is renamed to `@dfcc` macro and calls the `dfccdriver` routine, which is intended as a driver routine for all DF-based correlation methods (i.e., methods which don't use the `EC.fd` integrals).

### Changed

* Renamed function `active_orbitals` to `oss_active_orbitals`.
* Renamed function `calc_ccsd_resid` to `calc_cc_resid`.
* `ECdriver` and `oss_active_orbitals` now return named tuples.
* Improved documentation of occupation strings syntax.
* Switched to `Atom` and `FlexibleSystem` from `AtomsBase` as the internal representation of the molecular system. The basis set is stored for each atom as `:basis` property (as `Dict{String,String}`, e.g., `system[1][:basis]["ao"]`). One can also set `:basis` property for the whole system.
* Renamed macro `@opt` to `@set`. `@opt` is now an alias of `@set`.

### Added

* The automatically generated `UCCSDT` and `UDC-CCSDT` methods have been added to the docs.
* SCS-MP2, SCS-CCSD and SCS-DCSD

## Version [v0.10.0] - 2024.02.21

### Breaking
Expand Down
5 changes: 4 additions & 1 deletion Project.toml
View file
Original file line number Diff line number Diff line change
@@ -1,10 +1,11 @@
name = "ElemCo"
uuid = "094d408e-8508-40f4-9646-a254980d91ac"
authors = ["Daniel Kats <[email protected]> and contributors"]
version = "0.10.0"
version = "0.11.0"

[deps]
ArgParse = "c7e460c6-2fb9-53a9-8c5b-16f535851c63"
AtomsBase = "a963bdd2-2df7-4f54-a1ee-49d51e6be12a"
Dates = "ade2ca70-3891-5945-98fb-dc099432e06a"
DocStringExtensions = "ffbed154-4ef7-542d-bbb7-c09d3a79fcae"
GaussianBasis = "9bb1a3dc-0d1c-467e-84f5-0c4ef701360a"
Expand All @@ -16,6 +17,8 @@ NPZ = "15e1cf62-19b3-5cfa-8e77-841668bca605"
Printf = "de0858da-6303-5e67-8744-51eddeeeb8d7"
TensorOperations = "6aa20fa7-93e2-5fca-9bc0-fbd0db3c71a2"
Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
Unitful = "1986cc42-f94f-5a68-af5c-568840ba703d"
UnitfulAtomic = "a7773ee8-282e-5fa2-be4e-bd808c38a91a"

[compat]
TensorOperations = ">=3.2.5"
Expand Down
2 changes: 0 additions & 2 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -38,8 +38,6 @@ The integrals are obtained from a FCIDUMP file or calculated using the `Gaussian

Requirements: julia (>1.8)

Packages: LinearAlgebra, NPZ, Mmap, TensorOperations, Printf, IterativeSolvers, GaussianBasis, DocStringExtensions, MKL(optional)

## Usage
For a development version of `ElemCo.jl`, clone the repository and create an alias to set the project to the `ElemCo.jl` directory,
```
Expand Down
Binary file modified docs/equations/equations.pdf
View file
Binary file not shown.
266 changes: 262 additions & 4 deletions docs/equations/equations.tex
View file
Original file line number Diff line number Diff line change
Expand Up @@ -37,23 +37,39 @@

\newcommand{\NL}{\nonumber \\}
\newcommand{\nl}{\nonumber \\ & }
\newcommand{\beq}{\begin{equation}\begin{aligned}}
\newcommand{\eeq}{\end{aligned}\end{equation}}
\newcommand{\half}{\frac{1}{2}}
\newcommand{\quart}{\frac{1}{4}}
\newcommand{\op}{\hat}
\newcommand{\mat}{\mathbf}
\newcommand{\vect}{\mathbf}
\newcommand{\tnsr}{}
\newcommand{\snam}[1]{{\rm #1}}
\DeclareMathOperator{\sgn}{sgn}
\DeclareMathOperator{\trace}{tr}
\newcommand{\tbra}{\texttt{bra}}
\newcommand{\tket}{\texttt{ket}}
\newcommand{\perm}[1]{{\cal P}\left(#1\right)}
\newcommand{\Perm}[2]{{\cal P}\left({#1}\veryshortarrow {#2}\right)}
\newcommand{\sop}[1]{{\cal S}\left(#1\right)}
\newcommand{\Sop}[2]{{\cal S}\left(#1,#2\right)}
\newcommand{\asop}[1]{{\cal A}\left(#1\right)}
\newcommand{\ASop}[2]{{\cal A}\left(#1;#2\right)}
\newcommand{\mat}{\mathbf}
\newcommand{\eq}[1]{Eq.~(\ref{#1})}
\newcommand{\eqs}[2]{Eqs.~(\ref{#1},~\ref{#2})}
\newcommand{\sect}[1]{Sec.~\ref{#1}}
\newcommand{\spa}[1]{{#1}}
\newcommand{\spb}[1]{\bar{#1}}
\newcommand{\half}{\frac{1}{2}}
\newcommand{\quart}{\frac{1}{4}}
\newcommand{\dg}{\dagger}
\newcommand{\lk}{\left}
\newcommand{\rk}{\right}
\newcommand{\ieq}{\stackrel{!}{=}}
\newcommand{\ElemCojl}{\textsf{ElemCo.jl} }
\newcommand{\quantwo}{\textsf{Quantwo} }
\newcommand{\cmd}[1]{\texttt{#1}}
% for presubscripts and presuperscripts (e.g., \pre{^x}U_a)
\newcommand{\pre}[1]{\,#1\!}
% Define \lt and \gt for `less than' and `greater than'
\mathchardef\lt="313C \mathchardef\gt="313E
% Set up `<' and `>' as angle brackets
Expand Down Expand Up @@ -82,11 +98,16 @@ \section{Notation}
The integrals are \textbf{not antisymmetrized} and denoted by $v_{pq}^{rs}$, where $p,q,r,s$ are indices of orbitals,
and the lower indices correspond to the creation and the upper indices to the annihilation operators in the Hamiltonian,
\begin{equation}
\op H = E_0 + h_p^q \op a^\dagger_p \op a_q +
\frac{1}{2} v_{pq}^{rs} \op a^\dagger_p \op a^\dagger_q \op a_s \op a_r,
\end{equation}
or for the normal-ordered Hamiltonian,
\begin{equation}
\op H_N = f_p^q \left\{\op a^\dagger_p \op a_q\right\}_N +
\frac{1}{2} v_{pq}^{rs} \left\{\op a^\dagger_p \op a^\dagger_q \op a_s \op a_r\right\}_N,
\end{equation}
i.e., $f_p^q = <p|\op f|q>$ and $v_{pq}^{rs} = <pq|rs>$.
i.e., $h_p^q = <p|\op h|q>$, $f_p^q = <p|\op f|q>$ and $v_{pq}^{rs} = <pq|rs>$.
Permutation operators:
\begin{equation}
Expand All @@ -112,6 +133,191 @@ \section{Notation}
\end{aligned}
\end{equation}
\chapter{Integrals}
\section{Density fitting and Cholesky decomposition}
\label{sec:df}
The electron-repulsion integrals in \ElemCojl are obtained either from an external program
through an FCIDUMP \cite{knowlesDeterminant1989} interface,
or are calculated using the density-fitting approximation using the \texttt{GaussianBasis.jl}
interface\cite{aroeiraFermi2022} to the \texttt{libcint}\cite{libcint} library.
In the density-fitting approximation, the electron-repulsion integrals are approximated by
\begin{equation}
v_{pq}^{rs} \approx v_{p}^{rP} \left[v^{-1}\right]_{PQ} v_{q}^{sQ},
\end{equation}
where $v_{p}^{rP}$ and $v_{q}^{sQ}$ are density-fitted 3-index integrals with auxiliary
basis functions $P,Q$,
\begin{equation}
v_{p}^{rP} = \int d\vect{r}_1 d\vect{r}_2 \frac{\phi^*_p(\vect{r}_1) \phi^r(\vect{r}_1) \phi^P(\vect{r}_2)}{|\vect{r}_1-\vect{r}_2|},
\end{equation}
and $v^{PQ}$ is the Coulomb metric matrix,
\begin{equation}
v^{PQ} = \int d\vect{r}_1 d\vect{r}_2 \frac{\phi^P(\vect{r}_1)\phi^Q(\vect{r}_2)}{|\vect{r}_1-\vect{r}_2|}.
\end{equation}
The Coulomb metric matrix is decomposed using the Cholesky factorization,
\begin{equation}
v^{PQ} = \sum_L L^{P}_{L} L^{Q}_{L},
\end{equation}
where $L^{P}_{L}$ is a lower triangular matrix.
Thereafter, a non-symmetric square root of the inverse, $C_P^L$, is
calculated by solving the equations
\begin{equation}
L^{P}_{L'} C_{P}^{L} = \delta^{L}_{L'},
\end{equation}
with $\delta^{K}_{L}$ being the Kronecker delta.
If $L^{P}_{L}$ is low-rank, the equation is solved using the QR decomposition,
otherwise it can be solved by simple back-substitution.
The transformed density-fitted integrals which are used throughout \ElemCojl
are then calculated by multiplying the density-fitted 3-index
integrals with the non-symmetric square root of the inverse,
\begin{equation}
v_{p}^{rL} = v_{p}^{rP} C_P^L,
\end{equation}
and the density-fitted 4-index integrals can be calculated by
\begin{equation}
v_{pq}^{rs} \approx \sum_L v_{p}^{rL} v_{q}^{sL}.
\end{equation}
If all integrals are calculated using the density-fitting approximation using \textit{mp2fit}
fitting basis in \ElemCojl,
the correction terms are added using the \textit{jkfit} fitting basis
to the one-body and zero-body terms of the Hamiltonian in order to ensure that the
reference energy and the Fock matrix from DF-HF is not changed by the density-fitting approximation,
\begin{equation}
\begin{aligned}
\tilde h_p^q &= \tilde f_p^q
- \sum_L \left(2v_{p}^{qL} v_{i}^{iL} - v_{p}^{iL} v_{i}^{qL}\right)\\
\tilde E_0 &= E_0 + h_i^i - \tilde h_i^i + \tilde f_I^I,
\end{aligned}
\end{equation}
where
\begin{equation}
\begin{aligned}
\tilde f_p^q &= h_p^q + \sum_{\tilde L} \left(2v_{p}^{q\tilde L} v_{\tilde i}^{\tilde i\tilde L}
- v_{p}^{\tilde i\tilde L} v_{\tilde i}^{q\tilde L}\right),
\end{aligned}
\end{equation}
and $I$ denotes the core orbitals (cf. \sect{sec:frozen-core}),
$\tilde i$ denote all occupied orbitals (including core)
and other indices do not include the core orbitals.
$\tilde L$ corresponds to the \textit{jkfit} density-fitting basis functions,
and $L$ corresponds to the \textit{mp2fit} density-fitting basis functions.
\section{Frozen-core approximation}
\label{sec:frozen-core}
The frozen-core approximation is used to reduce the number of orbitals in the correlated
calculation. The frozen-core approximation is implemented in \ElemCojl by
setting the corresponding integrals to zero and adding their contribution to the
one-particle and zero-particle part of the Hamiltonian,
\begin{equation}
\begin{aligned}
\tilde h_p^q &= h_p^q + 2 v_{pI}^{qI} - v_{pI}^{Iq}\\
\tilde E_0 &= E_0 + 2 h_I^I + 2 v_{IJ}^{IJ} - v_{IJ}^{JI},
\end{aligned}
\end{equation}
where $I, J$ denote the core orbitals, and other indices do not include the core orbitals.
For the UHF Hamiltonian, the frozen-core approximation is implemented as
\begin{equation}
\begin{aligned}
\tilde h_{\spa p}^{\spa q} &= h_{\spa p}^{\spa q} + v_{\spa p \spa I}^{\spa q\spa I} +
v_{\spa p \spb I}^{\spa q\spb I} - v_{\spa p\spa I}^{\spa I\spa q}\\
\tilde h_{\spb p}^{\spb q} &= h_{\spb p}^{\spb q} + v_{\spb p \spb I}^{\spb q\spb I} +
v_{\spb p \spa I}^{\spb q\spa I} - v_{\spb p\spb I}^{\spb I\spb q}\\
\tilde E_0 &= E_0 + h_{\spa I}^{\spa I} + h_{\spb I}^{\spb I} +
\half\left(v_{\spa I\spa J}^{\spa I\spa J} + v_{\spb I\spb J}^{\spb I\spb J}
+ 2 v_{\spa I\spb J}^{\spa I\spb J}
- v_{\spa I\spa J}^{\spa J\spa I} - v_{\spb I\spb J}^{\spb J\spb I}
\right).
\end{aligned}
\end{equation}
If the frozen-core approximation is used in combination with the density-fitting approximation,
\textit{jkfit} correction terms are added to the one-body and zero-body terms of the Hamiltonian in order to ensure that the
reference energy and the Fock matrix from DF-HF is not changed by the frozen-core approximation,
cf. \sect{sec:df}.
\chapter{Hartree-Fock}
\section{Density-fitted Hartree-Fock}
The density-fitted Hartree-Fock equations are given by
\begin{equation}
\begin{aligned}
f_\mu^\nu C_\nu^p &= S_\mu^\nu C_\nu^p \epsilon_p\\
f_\mu^\nu &= h_\mu^\nu + 2\sum_L\left(v_{\mu'}^{iL} C^{\dagger\mu'}_i\right) C_{P}^{L} v_{\mu}^{\nu P}
- v_{\mu}^{iL} v^{\dagger\nu}_{iL}\\
v_{\mu}^{iL} &= \left(v_{\mu}^{\nu P} C_{\nu}^i\right) C_P^L.
\end{aligned}
\end{equation}
Alternatively, the $v_{\mu}^{\nu L}$ integrals can be precomputed and the Fock matrix can be calculated as
\begin{equation}
\begin{aligned}
f_\mu^\nu &= h_\mu^\nu + 2\sum_L\left(v_{\mu'}^{iL} C^{\dagger\mu'}_i\right) v_{\mu}^{\nu L}
- v_{\mu}^{iL} v^{\dagger\nu}_{iL}\\
v_{\mu}^{iL} &= v_{\mu}^{\nu L} C_{\nu}^i.
\end{aligned}
\end{equation}
Note that our orbitals are real, and therefore $v^{\dagger\mu}_{iL} = v_{\mu}^{iL}$,
and $C^{\dagger\mu}_i = C_{\mu}^i$.
The unrestricted Hartree-Fock equations are given (for $\alpha$ spin) by
\begin{equation}
\begin{aligned}
\pre{^\alpha}f_{\mu}^{\nu} C_{\nu}^{\spa p} &= S_{\mu}^{\nu} C_{\nu}^{\spa p} \epsilon_{\spa p}\\
\pre{^\alpha}f_{\mu}^{\nu} &= h_{\mu}^{\nu} + \sum_L\left(v_{\mu'}^{\spa i L} C^{\dagger\mu'}_{\spa i}
+v_{\mu'}^{\spb i L} C^{\dagger\mu'}_{\spb i}\right) C_{P}^{L} v_{\mu}^{\nu P}
- v_{\mu}^{\spa i L} v^{\dagger\nu}_{\spa i L}\\
v_{\mu}^{\spa i L} &= \left(v_{\mu}^{\nu P} C_{\nu}^{\spa i}\right) C_{P}^L,
\end{aligned}
\end{equation}
and equations for $\beta$ spin can be obtained by swapping the spins.
The residual of the Hartree-Fock equations, which can be used in DIIS, is given by
\begin{equation}
\begin{aligned}
\label{eq:hf-residual}
\Delta f_{\mu}^{\nu} = S_{\mu}^{\nu'} D_{\nu'}^{\rho}f_{\rho}^{\nu} - f_{\mu}^{\nu'} D_{\nu'}^{\rho}S_{\rho}^{\nu},
\quad \mbox{ with } \quad D_{\mu}^{\nu} = C_{\mu}^i C_i^{\dagger\nu}.
\end{aligned}
\end{equation}
\section{(Bi-orthogonal) Hartree-Fock}
The closed-shell Hartree-Fock on top of the FCIDUMP integrals (including the case of similarity-transformed
Hamiltonians) is given by
\begin{equation}
\begin{aligned}
f_{\tilde p}^{\tilde q} C_{\tilde q}^p &= C_{\tilde p}^p \epsilon_p,\\
f_{\tilde p}^{\tilde q} &= h_{\tilde p}^{\tilde q} +
\gamma^{\tilde r}_{\tilde s} \left(V_{\tilde p\tilde r}^{\tilde q\tilde s}
- \half V_{\tilde p\tilde r}^{\tilde s\tilde q}\right),\\
\gamma^{\tilde r}_{\tilde s} &= 2\sum_{i \in \rm occ} \bar C^{\dg \tilde r}_{i} C_{\tilde s}^i,
\end{aligned}
\end{equation}
where tilde indices correspond to the original orbitals.
If the FCIDUMP is similarity-transformed, $\bar C^{\dg \tilde r}_{p} \neq C_{\tilde r}^p$,
and $\bar C^{\dg \tilde r}_{p}$ are obtained as an inverse of $C_{\tilde r}^p$ such that
$\bar C^{\dg \tilde r}_{p} C_{\tilde r}^r = \delta^{r}_{p}$.
The unrestricted Hartree-Fock equations are given (for $\alpha$ case) by
\begin{equation}
\begin{aligned}
\pre{^\alpha}f_{\tilde p}^{\tilde q} C_{\tilde q}^{\spa p} &= C_{\tilde p}^{\spa p} \epsilon_{\spa p},\\
\pre{^\alpha}f_{\tilde p}^{\tilde q} &= h_{\tilde p}^{\tilde q} +
\left(\pre{^\alpha}\gamma^{\tilde r}_{\tilde s} + \pre{^\beta}\gamma^{\tilde r}_{\tilde s}\right)
V_{\tilde p\tilde r}^{\tilde q\tilde s}
- \pre{^\alpha}\gamma^{\tilde r}_{\tilde s}V_{\tilde p\tilde r}^{\tilde s\tilde q},\\
\pre{^\alpha}\gamma^{\tilde r}_{\tilde s} &= \sum_{\spa i \in \rm occ} \bar C^{\dg \tilde r}_{\spa i} C_{\tilde s}^{\spa i},\\
\pre{^\beta}\gamma^{\tilde r}_{\tilde s} &= \sum_{\spb i \in \rm occ} \bar C^{\dg \tilde r}_{\spb i} C_{\tilde s}^{\spb i},
\end{aligned}
\end{equation}
and equations for $\beta$ spin can be obtained by swapping the spins.
Note that if the FCIDUMP is of UHF type, the original indices and integrals are spin-dependent,
which has to be taken into account in the equations.
The residual of the Hartree-Fock equations, which can be used in DIIS, is equivalent to
\eq{eq:hf-residual}, with the overlap matrix $S_{\mu}^{\nu}$ removed.
\chapter{CCSD and DCSD amplitude and $\Lambda$ equations}
\section{Closed-shell CCSD/DCSD Lagrangian} \label{sec:cs-ccsd}
The singles-dressed factorization of the closed-shell CCSD and DCSD amplitude equations
Expand Down Expand Up @@ -1273,6 +1479,58 @@ \chapter{Two determinant coupled cluster}
\begin{equation}
\brown{\Delta E_{\mathrm{IAS}}} = \brown{-W^{\spa{t}\spb{u}}_{\spa{u}\spb{t}} T^{\spa{t}}_{\spa{u}} T^{\spb{u}}_{\spb{t}}}.
\end{equation}
\chapter{Automatically generated UCCSDT and UDC-CCSDT}
Unrestricted implementations of CCSDT and DC-CCSDT\cite{kats19_dc,rishi19,schraivogel21_dc} were generated with version 1.0.1 of the \quantwo program\cite{quantwo}.
The \quantwo inputs are listed below.
\begin{itemize}
\item UCCSDT \quantwo input file:
\lstset{language=[LaTeX]Tex, frame=shadowbox, rulesepcolor=\color{gray}, basicstyle=\ttfamily\scriptsize, columns=fullflexible, breaklines=true}
\begin{lstlisting}
prog,spinintegr=0,nobrafac=1,explspin=1,algo=2
output,level=1,maxlenline=70
\beq
<\Phi^{a}_{i}| \op H (1 + \op T_2 + \op T_3) |0>_C
\eeq
\beq
<\Phi^{ab}_{ij}| \op H (1 + \op T_2 + \half \op T_2 \op T_2 + \op T_3) |0>_C
\eeq
\beq
<\Phi^{abc}_{ijk}| \op H (\op T_2 + \op T_3 + \half \op T_2 \op T_2 + \op T_2 \op T_3) |0>_C
\eeq
\end{lstlisting}
\item UDC-CCSDT \quantwo input file:
\begin{lstlisting}
%singles and doubles amplitude equations from UCCSDT
%we only modify the triples amplitude equation
prog,spinintegr=0,nobrafac=1,explspin=1,algo=2
output,level=1,maxlenline=70
\beq
<\Phi^{abc}_{ijk}| \op H (\op T_2 + \op T_3 + \frac{1}{2} \op T_2 \op T_2
+ \op T_2 \op T_3) |0>_C
+ (1 - \Perm{IJ}{JI} - \Perm{IK}{KI})(1 - \Perm{AB}{BA} - \Perm{AC}{CA})
(\sum_{LMDE} \tnsr \intg{LE}{MD} \tnsr T^{IL}_{AD} \tnsr T^{MJK}_{EBC})
- \frac{1}{2}(1 - \Perm{KI}{IK} - \Perm{KJ}{JK})
\sum_{LMDE} \tnsr \intg{LD}{ME} \tnsr T^{IJ}_{DE} \tnsr T^{LMK}_{ABC}
- \frac{1}{2}(1 - \Perm{CA}{AC} - \Perm{CB}{BC})
\sum_{LMDE} \tnsr \intg{LD}{ME} \tnsr T^{LM}_{AB} \tnsr T^{IJK}_{DEC}
+ \frac{1}{2}(1 - \Perm{IJ}{JI} - \Perm{IK}{KI})
\sum_{LMDE} \tnsr \intg{LD}{ME} \tnsr T^{LI}_{DE} \tnsr T^{MJK}_{ABC}
+ \frac{1}{2}(1 - \Perm{AB}{BA} - \Perm{AC}{CA})
\sum_{LMDE} \tnsr \intg{LD}{ME} \tnsr T^{LM}_{DA} \tnsr T^{IJK}_{EBC}
+ \frac{1}{2}(1 - \Perm{KI}{IK} - \Perm{KJ}{JK})(1 - \Perm{AB}{BA} - \Perm{AC}{CA})
\sum_{LMDE} \tnsr \intg{LD}{ME} \tnsr T^{IJ}_{AD} \tnsr T^{LMK}_{BEC}
+ \frac{1}{2}(1 - \Perm{IJ}{JI} - \Perm{IK}{KI})(1 - \Perm{CA}{AC} - \Perm{CB}{BC})
\sum_{LMDE} \tnsr \intg{LD}{ME} \tnsr T^{IL}_{AB} \tnsr T^{JMK}_{DEC}
\eeq
\end{lstlisting}
\end{itemize}
\noindent The program generates \textsf{TensorOperations} code.
The generated code used by \ElemCojl is located in the \texttt{src/algo} directory.
\addcontentsline{toc}{chapter}{Bibliography}
\bibliography{references.bib}
\bibliographystyle{naturemag}
Expand Down
Loading

2 comments on commit 4e1a5a5

@dnkats
Copy link
Member Author

@dnkats dnkats commented on 4e1a5a5 Apr 12, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@JuliaRegistrator register

@JuliaRegistrator
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Registration pull request created: JuliaRegistries/General/104752

Tip: Release Notes

Did you know you can add release notes too? Just add markdown formatted text underneath the comment after the text
"Release notes:" and it will be added to the registry PR, and if TagBot is installed it will also be added to the
release that TagBot creates. i.e.

@JuliaRegistrator register

Release notes:

## Breaking changes

- blah

To add them here just re-invoke and the PR will be updated.

Tagging

After the above pull request is merged, it is recommended that a tag is created on this repository for the registered package version.

This will be done automatically if the Julia TagBot GitHub Action is installed, or can be done manually through the github interface, or via:

git tag -a v0.11.0 -m "<description of version>" 4e1a5a50acec4c1dd28412c03dae06f5327d73e3
git push origin v0.11.0

Also, note the warning: This looks like a new registration that registers version 0.11.0.
Ideally, you should register an initial release with 0.0.1, 0.1.0 or 1.0.0 version numbers
This can be safely ignored. However, if you want to fix this you can do so. Call register() again after making the fix. This will update the Pull request.

Please sign in to comment.