Data Science with C++ exercise

Simple data science exercise demonstrating the use of these features:

• reading, writing, and parsing data
• header+source files
• OOP: inheritance, virtual methods, abstract base class, constructor+destructor, static variable+method (TODO), polymorphism (TODO), nested class (TODO)
• data structures: struct, class, unordered map, vector
• pointers, references (TODO)
• STL, Boost
• passing by reference/by value
• dynamic memory allocation (TODO)
• C++11, C++14 features: auto, alias
• templates (TODO)
• python like string formatting (with Boost Format)
• printing in color
• documentation
• testing (with Google Test)
• CLI arguments (TODO)
• make (TODO)
• include guards (via #pragma once)
• macros
• namespaces
• exceptions (TODO)
• regexes
• plotting (TODO)
• external library (TODO)
• declaration vs definition
• measing elapsed time
• profiling

Setup

# setup C++
sudo apt-get install g++

# install boost
sudo apt-get install libboost-all-dev

# install doxygen
sudo apt-get install doxygen
sudo apt-get install graphviz

# generate a project-specific doxygen configuration file
doxygen -g parsing-exercise.conf

# install astyle c++ prettyfier
download https://sourceforge.net/projects/astyle/files/latest/download?source=files
cd astyle/build/gcc
make

# install valgrind for measuring performance
sudo apt-get install libc6-dbg gdb valgrind
sudo apt-get install python graphviz
sudo pip install gprof2dot

# install code correctness recommender
sudo pip install --upgrade cppclean

# install google test
sudo apt-get install libgtest-dev
sudo apt-get install cmake # install cmake
cd /usr/src/gtest
sudo cmake CMakeLists.txt
sudo make
# copy or symlink libgtest.a and libgtest_main.a to your /usr/lib folder
sudo cp *.a /usr/lib

# make run.sh executable
chmod +x run.sh

# run
./run.sh

TODO

• better naming
• better optimization
• more comments
• use initializers whenever you can
• better project structure
• write blog post
• improve Boost formatting
• missing values