msys2_1.sh

#MSYS2 stuff, after it got installed in Windows or anywhere. Or try this with Conda:


echo Extra pip installer, version 4.1.7


# Update PATH to include necessary directories
export PATH="$PATH:$LOCALAPPDATA/miniconda3:$LOCALAPPDATA/miniconda3/Library/bin:$LOCALAPPDATA/Microsoft/WindowsApps:/c/ProgramData/chocolatey/bin:/c/ProgramData/chocoportable/bin:$LOCALAPPDATA/miniconda3/Scripts:$USERPROFILE/AppData/Local/Programs/Git/bin:$LOCALAPPDATA/Chocolatey:$LOCALAPPDATA/chocolatey/lib/make/tools/install/bin:/c/Windows/system32:/c/Windows:/c/Windows/System32/Wbem:/c/Windows/System32/WindowsPowerShell/v1.0:/c/Windows/System32/OpenSSH:$USERPROFILE/Chocolatey/bin:$LOCALAPPDATA/Microsoft/WindowsApps:$USERPROFILE/Documents/anaconda3/scripts:$PROGRAMDATA/chocolatey/bin:$USERPROFILE/Documents/Msys64/mingw64/bin"

# Add Python scripts directory to PATH
export PATH="$PATH:$USERPROFILE/AppData/Local/Packages/PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0/LocalCache/local-packages/Python312/Scripts"

pacman -S --noconfirm   mingw-w64-x86_64-python-pip

pip show pip | awk '/Location/ {print $2"/site-packages"}'

rm -rf "$(pip show pip | awk '/Location/ {print $2}')/EXTERNALLY-MANAGED"

pip config set global.break-system-packages true

#torch cpu:
pip config set global.index-url https://pypi.org/simple
pip config set global.extra-index-url https://download.pytorch.org/whl/cpu
pip config list

# Bad method, as manual: mkdir -p ~/.config/pip && echo -e "[global]\nbreak-system-packages = true" >> ~/.config/pip/pip.conf

#Needed for docling  : 

# General development tools

pacman -S --noconfirm base-devel

pacman -S --noconfirm --needed \
  libxml2-devel \
  libxslt-devel \
  git \
  mingw-w64-x86_64-cython \
  mingw-w64-x86_64-libxslt \
  mingw-w64-x86_64-libxml2  \
pacman -S --noconfirm neofetch  
#Install their general versions
: '
pacman -S --noconfirm --needed \

  gcc \
  clang \
  openssl \
  meson \
  ninja \

'

#Let us refresh just in case: 
pacman -Sy

# or the MSYS2-friendly Mingw-w64 versions
pacman -S --noconfirm --needed \
mingw-w64-x86_64-toolchain \
  mingw-w64-x86_64-python \
  mingw-w64-x86_64-meson \
  mingw-w64-x86_64-ninja \
  mingw-w64-x86_64-gcc \
  mingw-w64-x86_64-libopenmpt \
  mingw-w64-x86_64-cmake \
  mingw-w64-x86_64-make \
  mingw-w64-x86_64-ffmpeg \
  mingw-w64-x86_64-make \
  mingw-w64-x86_64-openssl \
  mingw-w64-x86_64-python3-pip \
  mingw-w64-x86_64-clang \
  mingw-w64-x86_64-openblas \
  mingw-w64-x86_64-vulkan-headers \
  mingw-w64-x86_64-vulkan-devel \
  mingw-w64-x86_64-cmake \ 
  mingw-w64-x86_64-ca-certificates \
  mingw-w64-x86_64-shaderc


  pacman -S --noconfirm --needed mingw-w64-x86_64-libc++    mingw64/mingw-w64-x86_64-lfortran    mingw-w64-x86_64-pkgconf

  pacman -S --noconfirm --needed mingw-w64-x86_64-libc++-dev
pacman -S --noconfirm --needed python-devel libatomic_ops-devel findutils

#!/bin/bash

# Define the profile file to write the environment setup
PROFILE_FILE="$HOME/.bashrc"

# Prepare the profile setup
{
    # 1. Set OpenSSL and PATH
    echo "export OPENSSL_ROOT_DIR=/mingw64"
    echo "export PATH=/mingw64/bin:\$PATH"

    # 2. Conda initialization
    echo "# Conda initialization"
    echo "MINICONDA_PATH=\"\${USERPROFILE}/AppData/Local/miniconda3\""
    echo "CONDA_INIT_SCRIPT=\"\${MINICONDA_PATH}/etc/profile.d/conda.sh\""
    echo "if [ -f \"\$CONDA_INIT_SCRIPT\" ]; then"
    echo "    source \"\$CONDA_INIT_SCRIPT\""
    echo "    #conda activate base"
    echo "fi"

    # 3. C Compiler setup
    echo "# C Compiler setup"
    echo "echo 'Checking for available C compilers...'"
    echo "if command -v clang &>/dev/null; then"
    echo "    export CC=\$(command -v clang)"
    echo "    export CXX=\$(command -v clang++)"
    echo "    echo 'Using Clang compiler: \$CC'"
    echo "    echo 'Using Clang++ compiler: \$CXX'"
    echo "    clang_version=\$(clang --version | head -n 1 | awk '{print \$3}')"
    echo "    export LD_LIBRARY_PATH=\"/mingw64/lib/clang/\$clang_version/lib:\$LD_LIBRARY_PATH\""
    echo "    export LDFLAGS=\"-L/mingw64/lib/clang/\$clang_version/lib \$LDFLAGS\""
    echo "    export LD_RUN_PATH=\"\$LD_LIBRARY_PATH\""
    echo "elif command -v gcc &>/dev/null; then"
    echo "    export CC=\$(command -v gcc)"
    echo "    export CXX=\$(command -v g++)"
    echo "    echo 'Using GCC compiler: \$CC'"
    echo "    echo 'Using G++ compiler: \$CXX'"
    echo "    gcc_version=\$(gcc --version | head -n 1 | awk '{print \$3}' | cut -d. -f1)"
    echo "    export LD_LIBRARY_PATH=\"/mingw64/lib/gcc/\$gcc_version:\$LD_LIBRARY_PATH\""
    echo "    export LDFLAGS=\"-L/mingw64/lib/gcc/\$gcc_version \$LDFLAGS\""
    echo "    export LD_RUN_PATH=\"\$LD_LIBRARY_PATH\""
    echo "else"
    echo "    echo 'Neither Clang nor GCC is installed. Please install one of them to proceed.'"
    echo "fi"

    # 4. Final LD* Variables Check
    echo "# Printing all LD* environment variables"
    echo "printenv | grep '^LD'"

    # 5. Checking which linker is being used
    echo "# Checking which linker is being used"
    echo "which ld"
    echo "which clang"
    echo "which clang++"
} >> "$PROFILE_FILE"

# Apply the changes immediately
source "$PROFILE_FILE"

# Final confirmation
echo "All configurations have been written to $PROFILE_FILE. Conda will activate automatically on future sessions."




echo "We check the include paths for Clang's libc++ version:"

# Step 3: Check if we found a valid version
if [ -z "$CXX_VERSION" ]; then
    echo "No Clang libc++ include directories found in /mingw64/include/c++/v1. Please check your installation."
else
    echo "Found Clang libc++ version: $CXX_VERSION. We set the paths thereto, but add them to your .bashrc etc later:"

    # Step 4: Set environment variables
    export CXXFLAGS="-I/mingw64/include/c++/v1"  # Specifies the Clang libc++ include path for C++ compilation.
    export CPLUS_INCLUDE_PATH="/mingw64/include/c++/v1"  # Specifically tells the C++ compiler where to find Clang's libc++ header files.
    export CPATH="$CPLUS_INCLUDE_PATH"  # General include path for both C and C++ compilers.

    # Step 5: Print what has been set
    echo "Set CXXFLAGS to: $CXXFLAGS"
    echo "Set CPLUS_INCLUDE_PATH to: $CPLUS_INCLUDE_PATH"
    echo "Set CPATH to: $CPATH"
fi
export PATH=/mingw64/bin:$PATH

# Step 6: Verify the paths
echo "Verify Clang libc++ include paths in /mingw64/include/c++/v1:"
ls /mingw64/include/c++/v1/

# Step 7: Print all C* related environment variables
echo "Current C* environment variables:"
printenv | grep '^C'


# "What is being installed?" "What dependencies am I pulling in?" and "What happens next?"

echo
echo "C Compiler Setup Script"
echo "----------------------------------------"
echo "This script ensures that you have a C compiler installed (either GCC or Clang)."
echo "If neither is found, it will guide you through installing one."

# Check if GCC or Clang are installed
echo "Checking for available C compilers..."

if ! command -v gcc >/dev/null && ! command -v clang >/dev/null; then
    echo "Error: No C compiler found (neither GCC nor Clang)."
    echo "Please install GCC or Clang before proceeding."
    exit 1
fi



# Set default compiler if not already set
echo "Setting up compilers..."



#gcc used: 
export CC=$(command -v gcc)
export CXX=$(command -v g++ )

#clang used: 
export CC=$(command -v clang )
export CXX=$(command -v clang++ )

# Show which compiler will be used
echo "Using compiler: $CC"
echo "Compiler setup completed."

#!/bin/bash

# Check if Clang is installed
if command -v clang &>/dev/null; then
    echo "Clang is installed. Setting up LD_* paths for Clang."

    # Set LD_* variables for Clang (assuming /usr/bin/clang as base directory for Clang)
    export LD_LIBRARY_PATH="/mingw64/lib/clang/$(clang --version | head -n 1 | awk '{print $3}')/lib:$LD_LIBRARY_PATH"
    export LDFLAGS="-L/mingw64/lib/clang/$(clang --version | head -n 1 | awk '{print $3}')/lib $LDFLAGS"
    export LD_RUN_PATH="$LD_LIBRARY_PATH"

    echo "Set LD_LIBRARY_PATH to: $LD_LIBRARY_PATH"
    echo "Set LDFLAGS to: $LDFLAGS"
    echo "Set LD_RUN_PATH to: $LD_RUN_PATH"

elif command -v gcc &>/dev/null; then
    echo "Clang not found. Falling back to GCC. Setting up LD_* paths for GCC."

    # Set LD_* variables for GCC (assuming /usr/bin/gcc as base directory for GCC)
    export LD_LIBRARY_PATH="/mingw64/lib/gcc/$(gcc --version | head -n 1 | awk '{print $3}' | cut -d. -f1)/$LD_LIBRARY_PATH"
    export LDFLAGS="-L/mingw64/lib/gcc/$(gcc --version | head -n 1 | awk '{print $3}' | cut -d. -f1) $LDFLAGS"
    export LD_RUN_PATH="$LD_LIBRARY_PATH"

    echo "Set LD_LIBRARY_PATH to: $LD_LIBRARY_PATH"
    echo "Set LDFLAGS to: $LDFLAGS"
    echo "Set LD_RUN_PATH to: $LD_RUN_PATH"

else
    echo "Neither Clang nor GCC is installed. Please install one of them to proceed."
fi

# Final step: List all LD* variables to confirm
echo "Printing all LD* environment variables:"
printenv | grep '^LD'

# Checking which linker is being used
echo "Checking which linker is being used:"
which ld
which clang
which clang++


mkdir -p Downloads
cd Downloads


# Python scientific and data tools
pacman -S --noconfirm --needed \
  mingw-w64-x86_64-python-numpy \
  mingw-w64-x86_64-python-numba \
  mingw-w64-x86_64-python-numexpr \
  mingw-w64-x86_64-python-numpydoc \
  mingw-w64-x86_64-python-opt_einsum \
  mingw-w64-x86_64-python-soundfile \
  mingw-w64-x86_64-python-pandas \
  mingw-w64-x86_64-python-scipy \
  mingw-w64-x86_64-python-matplotlib \
  mingw-w64-x86_64-python-openpyxl \
  mingw-w64-x86_64-libatomic_ops

# Set OpenSSL root directory
export OPENSSL_ROOT_DIR=/mingw64


pip install ninja
pip install ctypesgen
pip install setuptools
pip install -v clang
pip install --upgrade pip setuptools
pip install meson-python
pip install poetry
pip install lolcat
pip install pipwin




#Pytorch, you may try this, but requires admin install of Microsoft Visual C++:
: '
git clone https://github.com/smalltalkman/mingw-w64-python-pytorch.git
cd mingw-w64-python-pytorch
bash  ./utils/build.sh -64

cd ..


#You may try tricking it with pytorch Git clone and python setup.py build --cmake , etc. 
'


#Abseil full for ONNX install:
git clone https://github.com/abseil/abseil-cpp.git
cd abseil-cpp
cmake -DCMAKE_POSITION_INDEPENDENT_CODE=ON -DBUILD_SHARED_LIBS=ON ..
make -j$(nproc)
sudo make install
make install 
cd ..



#Or:
#curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
#python get-pip.py


pip install pandas -v
# you may get:   ../meson.build:5:13: ERROR: Command `/tmp/pip-install-szrpx4_7/pandas_18422d9ebc7a465b94960ab5fa3dfc92/generate_version.py --print` failed with status 1.
 
pip install numpy --no-binary numpy -v

pip install -v -U whisperx docling funasr openai-whisper open-interpreter tts
#This one is often missed:
pip install -U  onnxconverter_common
pip install -U  numpy --no-binary numpy -v

#For Whisperx : 
pip install -U  pyannote.audio  -v



git clone https://github.com/Dr-Noob/cpufetch
cd cpufetch
make
make install
./cpufetch  
cd ..


echo That peakperf does not work for MSYS2, but FYI: 
git clone https://github.com/Dr-Noob/peakperf
cd peakperf
./build.sh
make install
./peakperf



#This one is for Zorin, to avoid: WARNING: Error parsing dependencies of python-apt: Invalid version: '2.4.0-ubuntu4-zorin1'
sudo sed -i 's/^Version: \([0-9]\+\(\.[0-9]\+\)*\).*$/Version: \1/' /usr/lib/python3/dist-packages/python_apt-*.egg-info/PKG-INFO

: '
# See: https://github.com/abetlen/llama-cpp-python
Error: Can't find 'nmake' or 'CMAKE_C_COMPILER'
If you run into issues where it complains it can't find 'nmake' '?' or CMAKE_C_COMPILER, you can extract w64devkit as mentioned in llama.cpp repo and add those manually to CMAKE_ARGS before running pip install:

$env:CMAKE_GENERATOR = "MinGW Makefiles"
$env:CMAKE_ARGS = "-DGGML_OPENBLAS=on -DCMAKE_C_COMPILER=C:/w64devkit/bin/gcc.exe -DCMAKE_CXX_COMPILER=C:/w64devkit/bin/g++.exe"
'