%global debug_package %{nil}

%global pypi_name pytorch-tokenizers

Name:          python-%{pypi_name}
Version:       1.2.0
Release:       0%{?dist}
Summary:       C++ implementations for various tokenizers
License:       BSD-3-Clause
URL:           https://github.com/meta-pytorch/tokenizers
Source0:       %{url}/archive/%{version}.tar.gz#/tokenizers-%{version}.tar.gz
# Use system libraries instead of bundled submodules. Upstream default is OFF
# so existing users are unaffected; distro builds set TOKENIZERS_USE_SYSTEM_LIBS=ON.
Patch0:        tokenizers-use-system-libs.patch
Patch1:        tokenizers-Fix-include-of-absl.patch

ExclusiveArch: aarch64 x86_64
BuildRequires: gcc gcc-c++
BuildRequires: abseil-cpp-devel
BuildRequires: json-devel
BuildRequires: pcre2-devel
BuildRequires: re2-devel
BuildRequires: sentencepiece-devel
BuildRequires: pybind11-devel
BuildRequires: python3-cmake
BuildRequires: python3-setuptools
BuildRequires: python3-pybind11
BuildRequires: python3-sentencepiece
BuildRequires: python3-tiktoken
BuildRequires: python3-torch
BuildRequires: python3-tokenizers
BuildRequires: python3-pytest

%global _description %{expand:
C++ implementations for various tokenizers (sentencepiece, tiktoken etc).
Useful for other PyTorch repos such as torchchat, ExecuTorch to build LLM
runners using ExecuTorch stack or AOT Inductor stack.
}

%description %_description

%package -n     python3-%{pypi_name}
Summary:        %{summary}

%description -n python3-%{pypi_name} %_description


%prep
%autosetup -p1 -n tokenizers-%{version}

%generate_buildrequires
%pyproject_buildrequires

%build
export CMAKE_ARGS="-DTOKENIZERS_USE_SYSTEM_LIBS=ON -DSUPPORT_REGEX_LOOKAHEAD=ON"
%pyproject_wheel

%install
%pyproject_install
%pyproject_save_files pytorch_tokenizers

%files -n python3-%{pypi_name} -f %{pyproject_files}
%license LICENSE
%doc README.md

%changelog
* Sun Oct 19 2025 Peter Robinson <pbrobinson@fedoraproject.org> - 1.2.0-1
- Initial package