About

Data Explorer is a tool for performing various operations on datasets. It supports operations like aggregation (average, minimum, and maximum) and grouping, providing a simple interface for querying data.

Building:

Clone and use CMake with GCC/Clang/MSVC to compile the project and tests from an IDE or command line. CMake should:

• configure everything automatically,
• download and build dependency,
• compile and create binaries.

Used tools and libs

Tool	Windows	Ubuntu
OS version	10 22H2	24.04
GCC	13.1.0	13.2.0
CMake	3.30.2	3.28.3
Git	2.46.0	2.43.0
cpputils	1.0.0	1.0.0
GoogleTest	1.15.2	1.15.2

Usage:

Launching:

Launch the binary and pass a single parameter, which is the file name containing data:

$ data-explorer sample.txt

Interface

Application expects a series of commands as text lines. Each line should have the following structure:

<operation> <aggregation column> <grouping column>  

where:

• operation is an aggregation function of one of the following types:
  • avg: calculates the average of a set of values.
  • min: finds the minimum value in a set of values.
  • max: finds the maximum value in a set of values.
• aggregation column is a numerical data column using which aggregation will be done.
• grouping column is a column used for grouping results.

Usage Example:

$ avg score movie_name

Example output:

Execute: AVG score GROUPED BY movie_name
Execution time: 12us
Results:
ender's_game 8
pulp_fiction 6
inception 8

License

This project is licensed under the MIT License. See the LICENSE file for details.

The project uses the following open-source software:

Name	License	Home	Description
cpputils	MIT	https:/przemek83/cpputils	collection of C++ utility classes
GoogleTest	BSD-3-Clause	https:/google/googletest	testing framework

Testing

For testing purposes, gtest framework is used. Build the project first. Make sure that the data-explorer-test target is built. Modern IDEs supporting CMake also support running tests with monitoring of failures. But in case you would like to run it manually, go to the build/test directory, where the⁣ binary data-explorer-test should be available. Launching it should produce the following output on Linux:

$ ./data-explorer-test 
Running main() from <path>/data-explorer/build/_deps/googletest-src/googletest/src/gtest_main.cc
[==========] Running 44 tests from 8 test suites.
[----------] Global test environment set-up.
[----------] 1 test from DataExplorer
[ RUN      ] DataExplorer.executeQuery
[       OK ] DataExplorer.executeQuery (0 ms)

(...)

[ RUN      ] UserInterfaceTest.GetQueryOperationWrongAggregatingColumn
[       OK ] UserInterfaceTest.GetQueryOperationWrongAggregatingColumn (0 ms)
[----------] 7 tests from UserInterfaceTest (0 ms total)

[----------] Global test environment tear-down
[==========] 44 tests from 8 test suites ran. (1 ms total)
[  PASSED  ] 44 tests.

As an alternative, CTest can be used to run tests from build directory:

$ ctest
Test project <path>/data-explorer/build
    Start  1: DataExplorer.executeQuery
1/44 Test  #1: DataExplorer.executeQuery ...................................   Passed    0.00 sec

(...)

    Start 44: UserInterfaceTest.GetQueryOperationWrongAggregatingColumn
44/44 Test #44: UserInterfaceTest.GetQueryOperationWrongAggregatingColumn ...   Passed    0.00 sec

100% tests passed, 0 tests failed out of 44

Total Test time (real) =   0.10 sec

Additional info

As speed is the most important expectation from the task, there was some optimization was performed. Ones with the biggest impact:

• Used std::unordered_map instead of std::map.
• Used std::vectors to store data and passed by const reference.
• Storing strings as mapped values (std::string <-> unsigned int) and usage of indexes for operations (performance and significant memory optimization).
• Minimized copying.

Potential further improvements

For bigger datasets and more sophisticated operations, the following enhancements might be viable:

• Usage for multithreading in calculations using std::async + std::future.
• Usage of MPI (make sense with more sophisticated calculations).
• GPU calculations (in case of more complex calculations).