Skip to content
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
665 lines (479 sloc) 18.4 KB

devsetup for macOS

This tutorial explains in detail the setup process for macOS.

To make your life easier, install devsetup for macOS.

Installing GCC

The first thing we need to install is the GNU Compiler Collection (GCC).

  1. Open the terminal and run the following command:

    # Install the Apple Command Line Tools package
    xcode-select --install
  2. Install the additional SDK headers:

    sudo installer -pkg /Library/Developer/CommandLineTools/Packages/macOS_SDK_headers_for_macOS_10.14.pkg -target /

Installing Homebrew

Homebrew is the missing package manager for macOS.

  1. Open the terminal and run the following script:

    /usr/bin/ruby -e "$(curl -fsSL"

Installing Git

Git is a free and open source distributed version control system.

  1. Install Git:

    brew install git
  2. Configure Git:

    # User name and email
    git config --global "Your name"
    git config --global ""
    # Tells git that we want color in our UI
    git config --global color.ui true
    git config --global color.status.changed "blue normal"
    git config --global color.status.untracked "red normal"
    git config --global color.status.added "magenta normal"
    git config --global color.status.updated "green normal"
    git config --global color.status.branch "yellow normal bold"
    git config --global color.status.header "white normal bold"

Updating bash

  1. Update bash:

    brew install bash
  2. Add the new shell to the list of allowed shells:

    sudo bash -c 'echo /usr/local/bin/bash >> /etc/shells'
  3. Change to the new shell:

    chsh -s /usr/local/bin/bash
  4. Check the bash version:

    bash --version
    # GNU bash, version 5.0.7(1)-release
  5. Install bash TAB completion 2:

    brew install bash-completion@2
  6. Install Informative git-prompt:

    brew install bash-git-prompt
  7. Install Docker bash TAB complete:

    • Docker-machine tab complete:
    sudo curl -L \
    -o /usr/local/etc/bash_completion.d/docker-machine
    • Docker-compose tab complete:
    sudo curl -L \
    -o /usr/local/etc/bash_completion.d/docker-compose

Update the .bash_profile file and terminal theme

This is the terminal using my personal theme and .bash_profile configuration. Feel free to copy and use it.

Gomes-Borges macOS terminal

  1. Download and open the macOS.terminal file:

    cd ${HOME}
    curl -L ${THEME_URL} -o macOS.terminal
    open macOS.terminal
  2. Once the terminal is open, set the theme as default:

    • In the top menu bar, select Terminal then Preferences
    • Select the Profiles Tab
    • Click and highlight the macOS themes
    • Press the Default button near the bottom of the window
  3. Update the .bash_profile file using the following commands:

    • Backup the current .bash_profile

      if [[ -f "${HOME}/.bash_profile" ]]; then
          cp "${HOME}/.bash_profile" "${HOME}/.bash_profile.bkp"
          echo "" > "${HOME}/.bash_profile"
    • Open the .bash_profile file

      open -e "${HOME}/.bash_profile"
    • Update the .bash_profile copying and pasting the text below. Then save and close the .bash_profile file.

      # Bash Profile
      # Copyright (c) Marcos Gomes-Borges
      # LS Colors
      # CLICOLOR use ANSI color sequences to distinguish file types
      export CLICOLOR=true
      export LSCOLORS=gxegbxdxcxahadabafacge
      alias ls='ls -GFh'
      # COLORS \e[<prefix>;<color>m;<decoration>
      # Prevent Mac OS ._ in in tar.gz files
      export COPYFILE_DISABLE=true
      # Homebrew
      export PATH="/usr/local/bin:${PATH}"
      # Homebrew completion
      if [ -f "$(brew --prefix)/etc/bash_completion.d/brew" ]; then
          . "$(brew --prefix)/etc/bash_completion.d/brew"
      # Bash completion@2
      if [ -f /usr/local/share/bash-completion/bash_completion ]; then
          . /usr/local/share/bash-completion/bash_completion
      # Bash-Git-prompt
      if [ -f "$(brew --prefix)/opt/bash-git-prompt/share/" ]; then
          __GIT_PROMPT_DIR="$(brew --prefix)/opt/bash-git-prompt/share"
          source "$(brew --prefix)/opt/bash-git-prompt/share/"
      # Python virtual environment
      # Allow pip only for active virtual environment
      # Use 'gpip' or 'gpip3' for global environment
      export PIP_REQUIRE_VIRTUALENV=true
          PIP_REQUIRE_VIRTUALENV="" pip "${@}"
          PIP_REQUIRE_VIRTUALENV="" pip3 "${@}"
      # Pyenv Python version management
      if command -v pyenv 1>/dev/null 2>&1; then
          eval "$(pyenv init -)"
          pyenv virtualenvwrapper
                      "\n${MAGENTA}[\d \t] ${YELLOW}$(python --version 2>&1)${NONE}\
                      "\n${VIRTUAL_ENV:+($(basename ${VIRTUAL_ENV}))}\\$ "\
                      " (%s)"'

Improve command line history search

On the the terminal application, copy and paste the following commands:

read -r -d '' INPUTRC <<"EOF"

set colored-stats on
set mark-symlinked-directories on
set show-all-if-ambiguous on
set show-all-if-unmodified on
set visible-stats on
set completion-ignore-case on
TAB: menu-complete

echo "${INPUTRC}" > ${HOME}/.inputrc

Installing multiple Python versions

Pyenv lets you easily switch between multiple versions of Python. It's simple.

  1. Install pyenv:

    brew install pyenv
  2. Install pyenv-virtualenvwrapper :

    brew install pyenv-virtualenvwrapper
  3. Install a few different versions of Python:

    # Init Pyenv
    eval "$(pyenv init -)"
    # BUGFIX pyenv 1.2.x on macOS 10.14.x
    if [ "$(sw_vers -productVersion | cut -c -5)" == "10.14" ]; then
        export SDKROOT=/Applications/
        export MACOSX_DEPLOYMENT_TARGET=10.14
    # Install Python
    PYTHON_CONFIGURE_OPTS="--enable-framework" pyenv install 2.7.15
    PYTHON_CONFIGURE_OPTS="--enable-framework" pyenv install 3.6.5
    PYTHON_CONFIGURE_OPTS="--enable-framework" pyenv install 3.7.3
  4. Set the Python version you want as default:

    pyenv global 3.6.5
  5. Setup the virtualenvwrapper:

    pyenv virtualenvwrapper

Installing Python libraries

  1. Install dependencies:

    # Leave Python virtual environment
    # Install Graphviz - Keras dependency
    brew install Graphviz
  2. Create a Python virtual environment:

     # Set Python 3.6.5
     pyenv shell 3.6.5
     mkvirtualenv py365
     pip install --upgrade pip

    Remember to activate Python environment using workon py365

  3. Installing TensorFlow:

    Check the TensorFlow documentation for more details

    pip install tensorflow
  4. Install Keras:

    pip install h5py \
                pydot \
  5. Install scientific libraires:

    pip install numpy \
                pandas \
                scipy \
                pillow \
                scikit-learn \
                scikit-image \
  6. Install basic libraires:

    pip install progressbar2 \
                requests \
                Cython \
                pylint \
                autopep8 \
                pytest \
  7. Install Matplotlib:

    pip install ipympl matplotlib
    # Fix Matplotlib backend
    mkdir -p "${HOME}/.config/matplotlib"
    echo "backend : TkAgg" > "${HOME}/.config/matplotlib/matplotlibrc"
  8. Install Jupyter lab:

    # Leave Python virtual environment
    # Install Node
    brew install node
    # Activate Python environment
    workon py365
    # Install Jupyterlab
    pip install jupyterlab jupyterlab-git
    jupyter labextension install @jupyterlab/toc
    jupyter labextension install @jupyterlab/git
    jupyter serverextension enable --py jupyterlab_git
    jupyter labextension install @jupyter-widgets/jupyterlab-manager
    jupyter labextension install jupyter-matplotlib

Visual Studio Code - VS Code

VS Code: Code editing. Redefined. For Windows, Linux and macOS.

  1. Launching VS Code from terminal:

    You can also run VS Code from terminal by typing code after adding it to the path:

    1. Launch VS Code.
    2. Open the Command Palette (⇧⌘P) and type shell command.
    3. Select the option Install 'code' command in PATH.
    4. Restart the terminal application to update the PATH.

    You'll be able to type code . in any folder to start editing files in that folder.

  2. User Settings:

    Inside script directory you find my personal VS Code settings, vs-code-settings.json. Feel free to copy and add it to your VS Code user settings.

  3. VS Code as default git editor and diff tool:

    You can set Visual Studio Code as default git editor and diff tool.

    On the terminal application, copy and paste the following commands:

    # Set Code as default git editor
    git config --global core.editor 'code --wait'
    # Set Code as diff tool
    git config --global diff.tool vscode
    git config --global difftool.vscode.cmd 'code --wait --diff $LOCAL $REMOTE'

LaTeX with VS Code

Requeriments (LaTeX):

  • LaTeX for Linux and Windows: I recommend installing TeX Live.
  • LaTeX for macOS: I recommend installing MacTeX, which contains all of TeX Live and a few Mac-specific additions.

To edit and compile LaTeX projects using VS Code, I recommend LaTeX Workshop, which is an extension aiming to provide all-in-one features and utilities for LaTeX typesetting.

Check the LaTeX Workshop website to discover more about the extension.

To use LaTeX Workshop correctly, be sure to create a directory called build in the same place of your LaTeX file. To configure LaTeX Workshop, take a look on my recipe file vs-code-settings.json.

LaTeX with VSCode

Installing FFmpeg

FFmpeg is the leading multimedia framework, able to decode, encode, transcode, mux, demux, stream, filter and play pretty much anything that humans and machines have created.

# Leave Python virtual environment

# Install FFmpeg with all modules
brew install ffmpeg $(brew options ffmpeg | grep -E '^--with-' - | tr '\n' ' ')

Installing OpenCV 4

Copy and past the following commands:

  1. Download, configure and install Opencv

    # Leave Python virtual environment
    # Installing Opencv dependencies
    brew install cmake pkg-config jpeg webp gphoto2 libpng libtiff openexr eigen tbb
    # Activate Python environment
    workon py365
    pip install pip numpy
  2. DOWNLOADING OpenCV files

    cd ${HOME}
    curl -L ${OPENCV_URL} -o "opencv.tar.gz"
    curl -L ${OPENCV_CONTRIB_URL} -o "opencv_contrib.tar.gz"
    tar -xzf opencv.tar.gz
    tar -xzf opencv_contrib.tar.gz
    mkdir -p opencv-${OPENCV_VERSION}/build
    cd opencv-${OPENCV_VERSION}/build
  3. CONFIGURING OpenCV installation

    # Automatically ENABLE/DISABLE Qt
    if ls ${HOME}/Qt*/*/clang_64/bin/qmake 1>/dev/null 2>&1; then
    QT_LIB=$(echo ${HOME}/Qt*/*/clang_64/lib/cmake)
    QT_QMAKE=$(echo ${HOME}/Qt*/*/clang_64/bin/qmake)
    # Get PATH for Python libraries
    py3_exec=$(which python3)
    py3_config=$(python3 -c "from distutils.sysconfig import get_config_var as s; print(s('LIBDIR'))")
    py3_include=$(python3 -c "import distutils.sysconfig as s; print(s.get_python_inc())")
    py3_packages=$(python3 -c "import distutils.sysconfig as s; print(s.get_python_lib())")
    py3_numpy=$(python3 -c "import numpy; print(numpy.get_include())")
    # Configure OpenCV via CMake
          -D CMAKE_INSTALL_PREFIX=/usr/local \
          -D OPENCV_EXTRA_MODULES_PATH="../../opencv_contrib-${OPENCV_VERSION}/modules" \
          -D WITH_CUDA=OFF \
          -D WITH_QT="${QT_ON_OFF}" \
          -D CMAKE_PREFIX_PATH="${QT_LIB}" \
          -D QT5Core_DIR="${QT_LIB}/Qt5Core" \
          -D QT5Gui_DIR="${QT_LIB}/Qt5Gui" \
          -D QT5Test_DIR="${QT_LIB}/Qt5Test" \
          -D QT5Widgets_DIR="${QT_LIB}/Qt5Widgets" \
          -D Qt5Concurrent_DIR="${QT_LIB}/Qt5Concurrent" \
          -D WITH_GTK=OFF \
          -D WITH_OPENGL=ON \
          -D WITH_OPENVX=ON \
          -D WITH_OPENCL=ON \
          -D WITH_TBB=ON \
          -D WITH_EIGEN=ON \
          -D WITH_GDAL=ON \
          -D WITH_FFMPEG=ON \
          -D BUILD_TIFF=ON \
          -D BUILD_opencv_python2=OFF \
          -D BUILD_opencv_python3=ON \
          -D PYTHON_DEFAULT_EXECUTABLE="${py3_exec}" \
          -D PYTHON3_EXECUTABLE="${py3_exec}" \
          -D PYTHON3_INCLUDE_DIR="${py3_include}" \
          -D PYTHON3_PACKAGES_PATH="${py3_packages}" \
          -D PYTHON3_LIBRARY="${py3_config}" \
          -D PYTHON3_NUMPY_INCLUDE_DIRS="${py3_numpy}" \
          -D BUILD_EXAMPLES=ON \
          -D CMAKE_CXX_STANDARD=14 ..
  4. Make & make install

    make -j"$(nproc)"
    sudo make -j"$(nproc)" install
  5. Test installation with Python

    # Activate Python environment
    workon py365
    # Check if the OpenCV library is well configured
    python -c "import cv2;  print(cv2.__version__)"
    # The output should be 4.1.0
  6. Test installation with C++

    We can run some sample code located at opencv/samples/cpp:

    # Access CPP examples
    cd ${HOME}/opencv-${OPENCV_VERSION}/samples/cpp
    # Active OpenCV virtual environment
    workon opencv
    # Compile and run without CUDA support
    g++ -ggdb $(pkg-config --cflags --libs opencv) facedetect.cpp -o /tmp/test && /tmp/test
  7. Test installation with Qt

    • The very first thing to do is instruct Qt where to find pkg-config

      which pkg-config
      # returns /usr/local/bin/pkg-config
    • We need to add /usr/local/bin to PATH

      Go to Project, expand Build Environment and add /usr/local/bin to PATH.

      Don’t forget to add a colon ":" before appending /usr/local/bin (see the screenshot below).

      Qt build settings for OpenCV

    • You also need to add a new variable called PKG_CONFIG_PATH and set it to the directory that contains opencv.pc for your OpenCV 3 installation. You can find it using the following command:

      find /usr/local -name "opencv.pc"
      # /usr/local/lib/pkgconfig/opencv.pc
    • Now that the paths are correctly set up, we need to add a few lines to our project file .pro to tell qmake to use pkg-config for OpenCV:

      # Tells Qmake to use pkg-config for OpenCV
      QT_CONFIG -= no-pkg-config
      CONFIG += link_pkgconfig
      PKGCONFIG += opencv
      # Add OpenCV Libraries
      LIBS += $(pkg-config --libs opencv)
    • You may receive a runtime error on macOS High Sierra. Note the actual error message may vary sometimes:

      dyld: Symbol not found: __cg_jpeg_resync_to_restart
      Referenced from: /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO
      Expected in: /usr/local/lib/libJPEG.dylib
      in /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO
      The program has unexpectedly finished.

      To fix it go to Project -> Run -> Run Environment -> Unset DYLD_LIBRARY_PATH. See image below:

      DYLD problem

  8. Sym-link OpenCV to your virtual environment (if necessary)

    Copy and paste the following commands on your terminal:

    # Get the OpenCV library path
    # Activate TARGET Python environment
    workon TARGET_ENV
    # Get the site-packages path of the virtual environment
    site_packages=$(python3 -c "import distutils.sysconfig as s; print(s.get_python_lib())")
    # Create a link to OpenCV library
    ln -s ${opencv_lib} ${site_packages}
You can’t perform that action at this time.