### Install Desbordante from Source

Source: https://github.com/desbordante/desbordante-core/blob/main/README_PYPI.md

Install Desbordante from source after cloning the repository. This involves setting up a virtual environment and installing dependencies.

```bash
python3 -m venv .venv
source .venv/bin/activate
python3 -m pip install .
```

--------------------------------

### Install Desbordante and Dependencies

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_1_typo_miner.ipynb

Installs the desbordante library, downloads a sample dataset, and installs colorama and jellyfish for text processing and visualization.

```python
!pip install desbordante==2.3.2
!wget https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/Workshop.csv
!pip install colorama jellyfish
```

--------------------------------

### Install Desbordante and Download Dataset

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_2_deduplication.ipynb

Installs the Desbordante library and downloads a sample CSV file for deduplication. Ensure you have pip and wget installed.

```python
!pip install desbordante==2.3.2
!wget https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/duplicates.csv
```

--------------------------------

### Install Desbordante from PyPI

Source: https://github.com/desbordante/desbordante-core/blob/main/README_PYPI.md

Install the Desbordante package using pip. This is the recommended method for most users.

```bash
pip install desbordante
```

--------------------------------

### Compile and Install Boost with Clang

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Compile and install the Boost library using Clang. This ensures ABI compatibility when using Clang.

```sh
./bootstrap.sh
sudo ./b2 install -a toolset=clang cxxflags="-stdlib=libc++" linkflags="-stdlib=libc++"
```

--------------------------------

### Install Levenshtein Library

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Matching_Dependencies.ipynb

Installs the Levenshtein library, which is used for calculating string distances and similarities.

```python
!pip install Levenshtein
```

--------------------------------

### Install Desbordante and Pandas

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Approximate_Functional_Dependencies_Mining.ipynb

Installs the necessary libraries for running AFD discovery. Ensure you have a stable internet connection for the download.

```python
!pip install desbordante==2.3.2
!pip install pandas
```

--------------------------------

### Download Example Datasets

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Approximate_Inclusion_Dependencies Mining.ipynb

Downloads the 'employee.csv' and 'project_assignments.csv' datasets required for the demonstration. These files are hosted on GitHub.

```python
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/employee.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/project_assignments.csv
```

--------------------------------

### Install Desbordante and Download Datasets

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_3_anomaly_detection.ipynb

Installs the Desbordante library and downloads necessary CSV datasets for anomaly detection. Ensure you have internet access and the necessary permissions to install packages and save files.

```python
!pip install desbordante==2.3.2
!wget https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/cargo_data_1.csv
!wget https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/cargo_data_2.csv
!wget https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/cargo_data_3.csv
```

--------------------------------

### Install CMake and LLVM/Clang

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Install CMake and the latest LLVM/Clang compiler suite. Sets Clang as the default compiler for the current session.

```sh
sudo apt install cmake
bash -c "$(wget -O - https://apt.llvm.org/llvm.sh)"
export CXX=clang++ 
export CXXFLAGS="-stdlib=libc++"
# libc++ is fully compatible with GCC's ABI, so you can omit the next line if you want to use libstdc++ ABI:
export LDFLAGS="-lc++abi"
```

--------------------------------

### Run Basic Mining FD Example

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/README.md

Execute the basic mining FD example using Python 3. Ensure you are in the 'desbordante' folder when running this command, as it references datasets relative to the project root.

```bash
python3 examples/basic/mining_fd.py
```

--------------------------------

### Install Desbordante Library

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Association_Rules.ipynb

Install the Desbordante library using pip. Ensure you have version 2.3.2 or compatible.

```python
!pip install desbordante==2.3.2
```

--------------------------------

### Install LLVM Clang 17 with Homebrew

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Install LLVM Clang version 17 using Homebrew. Replace 'llvm@17' with the desired version if needed.

```shell
brew install llvm@17
```

--------------------------------

### Install Ubuntu Dependencies

Source: https://github.com/desbordante/desbordante-core/blob/main/README.md

Installs necessary packages for building on Ubuntu, including GCC, CMake, Ninja, Boost, and Python. Sets GCC as the default C++ compiler.

```sh
sudo apt update && sudo apt upgrade
sudo apt install g++ cmake ninja-build libboost-all-dev python3 python3-venv
export CXX=g++
```

--------------------------------

### Install Desbordante Core via Pip

Source: https://github.com/desbordante/desbordante-core/blob/main/README.md

Use this command to install the Desbordante library from the Python Package Index. Ensure you have Python 3.8 or higher installed.

```sh
$
pip install desbordante

```

--------------------------------

### Install Boost on Older Ubuntu Versions

Source: https://github.com/desbordante/desbordante-core/blob/main/README.md

Downloads, extracts, and installs a specific version of the Boost library from source on older Ubuntu systems where APT repositories may not have compatible versions.

```sh
wget https://archives.boost.io/release/1.89.0/source/boost_1_89_0.tar.gz
tar xzvf boost_1_89_0.tar.gz
cd boost_1_89_0 && ./bootstrap.sh
sudo ./b2 install --prefix=/usr/
```

--------------------------------

### Install Xcode Command Line Tools

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Install Xcode Command Line Tools if they are not already present on your system. This is a prerequisite for many development tasks on macOS.

```shell
xcode-select --install
```

--------------------------------

### Install Python Bindings

Source: https://github.com/desbordante/desbordante-core/blob/main/src/python_bindings/CMakeLists.txt

Installs the compiled Python bindings module to the root installation directory when DESBORDANTE_BINDINGS is set to 'INSTALL'.

```cmake
if(DESBORDANTE_BINDINGS STREQUAL "INSTALL")
    # Install Python bindings module to system
    # Places the compiled library in the root installation directory
    install(TARGETS ${MODULE_NAME} LIBRARY DESTINATION .)
endif()
```

--------------------------------

### Execute SPLIT Algorithm and Get DDs

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Differential_Dependencies.ipynb

This snippet shows how to initialize the SPLIT algorithm, load data, execute the dependency finding process, and retrieve the differential dependencies.

```python
algo = desbordante.dd.algorithms.Split()
algo.load_data(table=dataset)
algo.execute(difference_table=dif_table)
dds = algo.get_dds()
for dd in dds:
  print(dd)
```

--------------------------------

### Install CMake and Python with Homebrew

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Use Homebrew to install CMake and Python 3. After installation, verify CMake version and ensure Homebrew packages are in your PATH.

```shell
brew install cmake python3
```

--------------------------------

### Install GCC 14 with Homebrew

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Install the GCC 14 compiler using the Homebrew package manager. This is required for building with GCC.

```shell
brew install gcc@14
```

--------------------------------

### Install CMake and Boost on macOS using Homebrew

Source: https://github.com/desbordante/desbordante-core/blob/main/README.md

Installs CMake and the Boost library on macOS using the Homebrew package manager. Includes instructions for adding Homebrew to the environment path if CMake is not found.

```sh
brew install cmake boost
```

--------------------------------

### Load and Run AC Mining Example

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Algebraic_Constraints.ipynb

Loads the dataset using pandas and runs the AC mining algorithm with default parameters. Ensure the TABLE variable is set correctly.

```python
table_for_example = pandas.read_csv(TABLE, sep=SEPARATOR, header=HEADER)
run_ac_mining(table_for_example, ['Delivery date', 'Dispatch date'])
```

--------------------------------

### Install Desbordante and Dependencies

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Approximate_Inclusion_Dependencies Mining.ipynb

Installs the desbordante library along with pandas and tabulate for data manipulation and display. Ensure you have a compatible Python environment.

```python
!pip install desbordante==2.3.2
!pip install pandas
!pip install tabulate
```

--------------------------------

### Install Desbordante Type Stubs

Source: https://github.com/desbordante/desbordante-core/blob/main/README_PYPI.md

Install type stubs for Desbordante if type hints are not working in your IDE. Note that stubs may not always be up-to-date with the latest package version.

```sh
pip install desbordante-stubs
```

--------------------------------

### Executing Algebraic Constraint Algorithm

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Algebraic_Constraints.ipynb

Runs the Default algorithm from desbordante.ac with specified parameters on Strength and Agility columns using addition. Requires the desbordante library to be installed.

```python
operation, operation_name = OPERATIONS['+']

algo = desbordante.ac.algorithms.Default()

df_without_id = table_for_example[['Strength', 'Agility']]

algo.load_data(table=df_without_id)

algo.execute(p_fuzz=P_FUZZ, fuzziness=FUZZINESS, bumps_limit=BUMPS_LIMIT, weight=WEIGHT,
        bin_operation='+', ac_seed=AC_SEED, iterations_limit=ITERATIONS_LIMIT)
```

--------------------------------

### Create Small DataFrame

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Denial_Constraints.ipynb

Creates a small Pandas DataFrame to illustrate the concept of denial constraints with a concrete example.

```python
data = {'Name': ['Alice', 'Bob', 'Carol'],
        'Grade': [3, 4, 4],
        'Salary': [3000, 4000, 4000]}

small_dataset = pd.DataFrame(data)
small_dataset
```

--------------------------------

### Download Dataset

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Algebraic_Constraints.ipynb

Downloads the cargo_march.csv dataset using wget. This is a prerequisite for running the AC mining example.

```bash
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/cargo_march.csv
TABLE = 'cargo_march.csv'
```

--------------------------------

### Ar/Apriori Library Configuration

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/ar/apriori/CMakeLists.txt

Configures the Ar/Apriori library by setting its name, adding sources, and linking necessary libraries. This setup is crucial for building the Apriori algorithm module.

```cmake
set(NAME ar.apriori)
desbordante_add_lib(NAME)
target_sources(${NAME} PRIVATE apriori.cpp candidate_hash_tree.cpp)
target_link_libraries(
    ${NAME} PRIVATE ${DESBORDANTE_PREFIX}::ar ${DESBORDANTE_PREFIX}::algos
                    spdlog::spdlog_header_only magic_enum::magic_enum Boost::headers
)
```

--------------------------------

### Setup Pandas Display Options

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_2_deduplication.ipynb

Configures pandas to display all columns, full column width, and without truncating row representations. This is useful for inspecting DataFrames.

```python
from collections import defaultdict, deque

import desbordante
import pandas


def setup_pandas_print():
    pandas.set_option('display.max_columns', None)
    pandas.set_option('display.width', None)
    pandas.set_option('display.max_colwidth', None)
    pandas.set_option('display.expand_frame_repr', False)

setup_pandas_print()
```

--------------------------------

### Execute DES Algorithm and Get NARs

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Numerical_Association_Rules.ipynb

Initializes the DES algorithm, loads data, executes the algorithm with specified parameters, and retrieves the found NARs. Ensure 'dataset' is loaded before execution.

```python
algo = desbordante.nar.algorithms.DES()
algo.load_data(table=dataset)
algo.execute(minconf=0.7, minsup=0.1, population_size=500, seed=5854,
             max_fitness_evaluations=700)
nars = algo.get_nars()
for nar in nars:
  print(nar)
```

--------------------------------

### Initialize and Run Deduplication

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_2_deduplication.ipynb

Initializes the deduplication process by getting functional dependencies (FDs) from the dataset and then prompts the user for input to select columns for deduplication. It displays the original record count and FD information before user interaction.

```python
print(f'Original records: {len(df)}')
print()

fds = get_1lhs_fds(df, ALGORITHM_TYPE, CONFIG)
print_fd_info(df, fds)
lhs_column = int(input('LHS column index: '))
fds = list(filter(lambda fd: fd[0] == lhs_column, fds))
if not fds:
    print('No FDs with this LHS!')
else:
    print('RHS columns:')
    print('
'.join(f'{rhs}: {df.columns[rhs]}' for _, rhs in fds))
    chosen_cols = sorted(set(map(int, input('RHS columns to use (indices): ').split())))
    matches_required = int(input('Equal columns to consider duplicates: '))

    new_rows = get_deduped_rows(df, chosen_cols, matches_required, fds)
    print()

    print(f'Resulting records: {len(new_rows)}. Duplicates found: {len(df) - len(new_rows)}')
    new_df = pandas.DataFrame(new_rows, columns=df.columns)

    print(new_df)
    new_df.to_csv(OUTPUT_FILE, index=False)
```

--------------------------------

### Build Benchmarks

Source: https://github.com/desbordante/desbordante-core/blob/main/src/tests/benchmark/README.md

Command to build the benchmark executable. Use the --benchmark flag to include benchmarks in the build process.

```bash
./build.sh --benchmark
```

--------------------------------

### Import Libraries and Load Dataset

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Approximate_Functional_Dependencies_Mining.ipynb

Imports the Desbordante and Pandas libraries and downloads a sample CSV dataset for demonstration. The dataset contains product information including ID, name, and price.

```python
import desbordante as db
import pandas as pd
```

```python
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/inventory_afd.csv
```

```python
df = pd.read_csv("inventory_afd.csv")
df
```

--------------------------------

### Discover Exact Denial Constraints

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Denial_Constraints.ipynb

Initializes the Desbordante Default algorithm, loads the dataset, and executes the discovery of exact denial constraints. The discovered constraints are then printed.

```python
algo = desbordante.dc.algorithms.Default()
algo.load_data(table=dataset)
algo.execute(evidence_threshold=0, shard_length=0)
exact_dcs = algo.get_dcs()

print("Discovered DCs:")
for dc in exact_dcs:
    print(dc)
```

--------------------------------

### Discovering Matching Dependencies with Increased Minimum Support

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Matching_Dependencies.ipynb

This snippet demonstrates how to increase the minimum support to find more reliable dependencies. It calculates the minimum support based on the dataset size and then executes the algorithm.

```python
algo.execute(column_matches=column_matches, min_support=round(len(second_dataset) * 1.5))
mds = algo.get_mds()
print('Found MDs:')
print(*(f'{i + 1} {md}' for i, md in enumerate(mds)), sep='\n')
```

--------------------------------

### Numerical Association Rules Examples

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Numerical_Association_Rules.ipynb

Examples of numerical association rules found using the DES algorithm, illustrating relationships between various attributes and their associated support and confidence.

```plaintext
NAR 1:
Intelligence[4 - 10]
Shedding[Moderate]
      |
      |
      V
Friendliness[6 - 10]
Life Span[9 - 16]
   support = 0.5660377358490566
   confidence = 0.9574468085106383

NAR 2:
Health Issues Risk[Moderate]
Life Span[8 - 14]
      |
      |
      V
Friendliness[5 - 8]
   support = 0.33962264150943394
   confidence = 0.7714285714285715

NAR 3:
Size[1 - 2]
Intelligence[5 - 8]
Grooming Needs[Moderate]
Weight[15.246273 - 68.261820]
      |
      |
      V
Shedding[Moderate]
   support = 0.05660377358490566
   confidence = 0.9

NAR 4:
Friendliness[5 - 10]
Exercise Requirements[1.708423 - 2.261994]
Type[Working]
      |
      |
      V
Life Span[10 - 16]
Training Difficulty[4 - 9]
   support = 0.08176100628930817
   confidence = 0.7222222222222222
```

--------------------------------

### Configure and Build Boost with LLVM Clang 17

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Configure and build the Boost library using LLVM Clang 17. This involves creating a user-config.jam file specifying the clang++ path and running the b2 build script with specific flags.

```shell
./bootstrap.sh
echo "using darwin : : $(brew --prefix llvm@17)/bin/clang++ ;" > user-config.jam
sudo ./b2 install -a --prefix=/usr/local --user-config=user-config.jam \
 cxxflags=\"-std=c++11 -I$(brew --prefix llvm@17)/include\" \
 linkflags=\"-L$(brew --prefix llvm@17)/lib/c++\"
export BOOST_ROOT=/usr/local/
```

--------------------------------

### Configure and Build Boost with GCC 14

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Configure and build the Boost library using GCC 14. This involves creating a user-config.jam file to specify the compiler and then running the b2 build script.

```shell
./bootstrap.sh 
echo "using darwin : : g++-14 ;" > user-config.jam
sudo ./b2 install --user-config=user-config.jam --layout=versioned
export BOOST_ROOT=/usr/local/
```

--------------------------------

### Get All Statistics as a String

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Simple_Statistics.ipynb

This snippet retrieves all computed statistics for the dataset and formats them as a plain text string, suitable for direct printing. This is a convenient way to get a quick overview of the dataset's statistics.

```python
print(data_stats.get_all_statistics_as_string())
```

--------------------------------

### CMakeLists.txt for Order Algorithm Library

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/od/order/CMakeLists.txt

Defines the library name, adds it using desbordante_add_lib, specifies source files, and links necessary libraries including magic_enum, desbordante components, spdlog, and Boost.

```cmake
set(NAME od.order)
desbordante_add_lib(NAME)
target_sources(
    ${NAME} PRIVATE dependency_checker.cpp list_lattice.cpp order.cpp order_utility.cpp
                    sorted_partitions.cpp
)
target_link_libraries(${NAME} PUBLIC magic_enum::magic_enum)
target_link_libraries(
    ${NAME} PRIVATE ${DESBORDANTE_PREFIX}::config ${DESBORDANTE_PREFIX}::model::table
                    spdlog::spdlog_header_only ${DESBORDANTE_PREFIX}::algos Boost::headers
)
```

--------------------------------

### Configure ASan Build Options

Source: https://github.com/desbordante/desbordante-core/blob/main/CMakeLists.txt

Sets up AddressSanitizer (ASan) options, including '-fsanitize=address'. Includes a specific workaround for Ubuntu with libc++ that may cause false positives.

```cmake
if(ASAN)
    set(ASAN_OPTS "-fsanitize=address")

    execute_process(COMMAND grep -q "Ubuntu" /etc/os-release RESULT_VARIABLE IS_UBUNTU)
    if(IS_UBUNTU EQUAL 0 AND CMAKE_CXX_FLAGS MATCHES "-stdlib=libc\+\+")
        message(
            WARNING
                "Running on Ubuntu with libc++. ASAN is broken in Ubuntu package with libc++15-19. If you are using \"these versions of libc++, set ASAN_OPTIONS=alloc_dealloc_mismatch=0 to suppress alloc-dealloc-mismatch "
                "check or consider using another distro or higher libc++ version for full ASAN coverage"
        )
    endif()

    string(
        JOIN
        ";"
        DEBUG_BUILD_OPTS
        "${DEBUG_BUILD_OPTS}"
        "-Wno-error" 
        "${ASAN_OPTS}"
    )
    set(DEBUG_LINK_OPTS "${ASAN_OPTS}")
elseif(UBSAN)
    string(
        JOIN
        ";"
        UBSAN_OPTS
        "-fsanitize=undefined"
        "-fsanitize=float-divide-by-zero"
        "-Wno-error"
        "-fno-sanitize=signed-integer-overflow"
        "-fno-sanitize=shift"
        "-fno-sanitize-recover=all"
    )
    if(CMAKE_CXX_COMPILER_ID STREQUAL "Clang" AND CMAKE_HOST_APPLE)
        string(JOIN ";" UBSAN_OPTS ${UBSAN_OPTS}
               "-fsanitize-ignorelist=${CMAKE_SOURCE_DIR}/ub_sanitizer_ignore_list.txt"
        )
    endif()
    string(JOIN ";" DEBUG_BUILD_OPTS "${DEBUG_BUILD_OPTS}" "${UBSAN_OPTS}")
    set(DEBUG_LINK_OPTS "${UBSAN_OPTS}")
endif()
```

--------------------------------

### Executing Inclusion Dependencies Discovery

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Inclusion_Dependencies_Mining.ipynb

Initializes the Default algorithm, loads the specified tables, executes the discovery process, and retrieves the discovered INDs. The `tables` variable should be a list of tuples, where each tuple contains the file path, delimiter, and a boolean indicating if the first row is a header.

```python
TABLES = [(f'{table_name}.csv', ',', True) for table_name in ['course', 'department', 'instructor', 'student', 'teaches']]

algo = desbordante.ind.algorithms.Default()
algo.load_data(tables=TABLES)
algo.execute()
inds = algo.get_inds()
```

--------------------------------

### Download and Extract Boost

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Download and extract the Boost source code. This is a prerequisite for building with Clang on Ubuntu.

```sh
curl https://archives.boost.io/release/1.86.0/source/boost_1_86_0.tar.bz2 --output "boost_1_86_0.tar.bz2"
tar xvjf boost_1_86_0.tar.bz2 && rm boost_1_86_0.tar.bz2
cd boost_1_86_0
```

--------------------------------

### Calculating Levenshtein Similarity

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Matching_Dependencies.ipynb

Demonstrates the usage of the LevenshteinSimilarity function to compare strings. It shows examples of identical strings and strings with a small difference.

```python
print(LevenshteinSimilarity("lion","lion"))
print(LevenshteinSimilarity("bear","beer"))
```

--------------------------------

### Configure Utility Bindings Library

Source: https://github.com/desbordante/desbordante-core/blob/main/src/python_bindings/CMakeLists.txt

Defines and configures the 'bindlib.util' library, listing its source files and linking required libraries.

```cmake
set(NAME bindlib.util)
desbordante_add_lib(NAME OBJECT)
target_sources(
    ${NAME}
    PRIVATE py_util/create_dataframe_reader.cpp
            py_util/dataframe_reader.cpp
            py_util/get_py_type.cpp
            py_util/iterable_sequence_stream.cpp
            py_util/logging.cpp
            py_util/opt_to_py.cpp
            py_util/py_to_any.cpp
            py_util/table_serialization.cpp
            py_util/value_to_py.cpp
)
target_link_libraries(
    ${NAME} PRIVATE magic_enum::magic_enum spdlog::spdlog_header_only Boost::headers
                    pybind11::pybind11
)
set_target_properties(${NAME} PROPERTIES CXX_VISIBILITY_PRESET "hidden")
```

--------------------------------

### Import Required Libraries

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Algebraic_Constraints.ipynb

Imports the desbordante library along with pandas, operator, and tabulate for data manipulation and display. Ensure these libraries are installed before running.

```python
import desbordante
import pandas
import operator
from tabulate import tabulate
```

--------------------------------

### Configure sd Binding

Source: https://github.com/desbordante/desbordante-core/blob/main/src/python_bindings/CMakeLists.txt

Adds the sd binding, focused on verification, and lists its required libraries.

```cmake
desbordante_add_bind(
    sd
    SRCS
    sd/bind_sd_verification.cpp
    LIBS
    ${DESBORDANTE_PREFIX}::sd::verifier
    magic_enum::magic_enum
    Boost::headers
)
```

--------------------------------

### Create Benchmark Executable

Source: https://github.com/desbordante/desbordante-core/blob/main/src/tests/benchmark/CMakeLists.txt

Defines the benchmark executable, its source files, include directories, and links necessary libraries and dependencies.

```cmake
set(NAME ${DESBORDANTE_PREFIX}.benchmark)
add_executable(${NAME})
target_sources(${NAME} PRIVATE main.cpp benchmark_cli.cpp benchmark_comparer.cpp)
target_include_directories(${NAME} PRIVATE ${DESBORDANTE_COMMON_INCLUDE_DIRS})
target_link_libraries(
    ${NAME}
    PRIVATE ${DESBORDANTE_PREFIX}::testlib::common
            ${DESBORDANTE_PREFIX}::dc::fastadc
            ${DESBORDANTE_PREFIX}::dd::split
            ${DESBORDANTE_PREFIX}::ind
            ${DESBORDANTE_PREFIX}::ind::mind
            ${DESBORDANTE_PREFIX}::fd::hy
            ${DESBORDANTE_PREFIX}::fd::pyro
            ${DESBORDANTE_PREFIX}::fd::tane
            ${DESBORDANTE_PREFIX}::fd::aid
            ${DESBORDANTE_PREFIX}::md::hy
            ${DESBORDANTE_PREFIX}::md::hy::preprocessing
            ${DESBORDANTE_PREFIX}::nar::des
            Boost::program_options
            magic_enum::magic_enum
            spdlog::spdlog_header_only
)
add_dependencies(
    ${NAME} Desbordante.zip.datasets Desbordante.zip.highway-neighbors
    Desbordante.zip.performance-tests
)
```

--------------------------------

### Download Datasets

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Functional_Dependencies_Mining.ipynb

Download the sample datasets required for functional dependency discovery. These files contain the data to be analyzed.

```python
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/duplicates_short.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/university_fd.csv
```

--------------------------------

### Configure Util Library

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/util/CMakeLists.txt

Defines the 'util' library, adds its source files, and links necessary external libraries.

```cmake
set(NAME util)
desbordante_add_lib(NAME OBJECT)
target_sources(
    ${NAME} PRIVATE convex_hull.cpp create_dd.cpp levenshtein_distance.cpp qgram_vector.cpp
                    worker_thread_pool.cpp
)
target_link_libraries(${NAME} PUBLIC magic_enum::magic_enum)
target_link_libraries(${NAME} PRIVATE spdlog::spdlog_header_only Boost::headers)
```

--------------------------------

### Executing Dependency Analysis with Custom Minimum Support

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Matching_Dependencies.ipynb

Executes the dependency analysis algorithm with a specified minimum support value. This allows for the discovery of dependencies that might otherwise be pruned due to low support. The discovered dependencies are then retrieved and printed.

```python
algo.execute(column_matches=column_matches, min_support=6)
mds = algo.get_mds()
print('Found MDs:')
print(*(f'{i + 1} {md}' for i, md in enumerate(mds)), sep='
')
```

--------------------------------

### Setup Pandas Display Options

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_3_anomaly_detection.ipynb

Configures pandas display options to ensure that all columns, rows, and content are visible when printing dataframes. This is useful for inspecting large or complex datasets.

```python
import desbordante
import pandas


def setup_pandas_print():
    pandas.set_option('display.max_columns', None)
    pandas.set_option("display.max_rows", None)
    pandas.set_option('display.width', None)
    pandas.set_option('display.max_colwidth', None)
    pandas.set_option('display.expand_frame_repr', False)

setup_pandas_print()
```

--------------------------------

### Downloading Dataset

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Algebraic_Constraints.ipynb

Downloads the 'player_stats.csv' dataset using wget. Ensure you have internet connectivity.

```bash
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/86085bbe5fb310795a8da70c63978f07b5d96a3b/examples/datasets/player_stats.csv
```

--------------------------------

### Import Libraries and Setup Pandas

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_1_typo_miner.ipynb

Imports necessary libraries including functools, itertools, operator, colorama, jellyfish, desbordante, and pandas. It also configures pandas display options for better readability of DataFrames.

```python
from functools import reduce
from itertools import groupby, islice
import operator

from colorama import Style, Fore
from jellyfish import levenshtein_distance
import desbordante
import pandas

def setup_pandas_print():
    pandas.set_option('display.max_columns', None)
    pandas.set_option('display.width', None)
    pandas.set_option('display.max_colwidth', None)
    pandas.set_option('display.expand_frame_repr', False)

setup_pandas_print()
```

--------------------------------

### Download Datasets

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Association_Rules.ipynb

Download the transactional datasets required for association rule mining using wget.

```bash
!wget https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/rules_book_rows.csv
!wget https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/rules_book.csv
```

--------------------------------

### FastADC Main Library Configuration

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/dc/FastADC/CMakeLists.txt

Defines the main 'dc.fastadc' library, adds its source file, and links it with project-specific and external libraries.

```cmake
set(NAME dc.fastadc)
desbordante_add_lib(NAME)
target_sources(${NAME} PRIVATE fastadc.cpp)
target_link_libraries(
    ${NAME}
    PUBLIC ${DESBORDANTE_PREFIX}::config ${DESBORDANTE_PREFIX}::model::table
    PRIVATE ${DESBORDANTE_PREFIX}::dc::fastadc::structs spdlog::spdlog_header_only
            magic_enum::magic_enum Boost::headers
)
```

--------------------------------

### Find Matching Dependencies using HyMD

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Matching_Dependencies.ipynb

Initializes the HyMD algorithm, loads data, defines column matches using Levenshtein similarity against self, executes the algorithm, and prints the found matching dependencies (MDs). Requires the 'desbordante' library and pandas.

```python
import desbordante.md.algorithms
import desbordante.md.column_matches

Levenshtein = desbordante.md.column_matches.Levenshtein

algo = desbordante.md.algorithms.HyMD()
algo.load_data(left_table=first_dataset)
column_matches = [Levenshtein(i, i) for i in range(len(first_dataset.columns))]
algo.execute(column_matches=column_matches)
mds = algo.get_mds()
print('Found MDs:')
print(*(f'{i + 1} {md}' for i, md in enumerate(mds)), sep='\n')
```

--------------------------------

### Import and Use Desbordante Python Module

Source: https://github.com/desbordante/desbordante-core/blob/main/README.md

Demonstrates how to import the Desbordante Python module after building it and how to control its logging level.

```python
import logging
import desbordante

# Get the logger provided by the library
log = logging.getLogger("desbordante")

# Set the desired log level using standard logging constants
log.setLevel(logging.INFO) # Or logging.DEBUG, logging.TRACE, etc.
```

--------------------------------

### Download Dataset Files

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Inclusion_Dependencies_Mining.ipynb

Downloads several CSV files containing relational data necessary for demonstrating inclusion dependency mining. These files include course, department, instructor, student, and teaches information.

```bash
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/course.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/department.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/instructor.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/student.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/teaches.csv
```

--------------------------------

### Download Sample Datasets

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Approximate_Unique_Column_Combinations_Mining.ipynb

Downloads the necessary CSV files for AUCC mining demonstration.

```python
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ucc_datasets/aucc.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ucc_datasets/aucc_correct.csv
```

--------------------------------

### Verify [id] -> [name] FD

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Approximate_and_Exact_Functional_Dependencies_Verification.ipynb

Initializes the FD verification algorithm, loads the dataset, and executes the verification for the FD [id] -> [name]. This snippet demonstrates the basic workflow for checking a specific FD.

```python
algo = db.afd_verification.algorithms.Default()
algo.load_data(table=("duplicates_short.csv", ",", True))
algo.execute(lhs_indices=[0], rhs_indices=[2])
print_results_for_fd(algo, data, 0, 2)
```

--------------------------------

### CMakeLists.txt for Pyro Library

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/fd/pyro/CMakeLists.txt

Configures the build for the Pyro library, including setting the library name, adding source files, and linking necessary dependencies.

```cmake
set(NAME fd.pyro)
desbordante_add_lib(NAME)
target_sources(${NAME} PRIVATE pyro.cpp)
target_link_libraries(
    ${NAME} PRIVATE spdlog::spdlog_header_only ${DESBORDANTE_PREFIX}::fd::pyro::common
                    ${DESBORDANTE_PREFIX}::fd::pli magic_enum::magic_enum Boost::headers
)
```

--------------------------------

### Configure Euler FD Library

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/fd/eulerfd/CMakeLists.txt

This snippet sets up the Euler FD library using CMake. It defines the library name, adds it as a library, specifies source files, and links necessary libraries.

```cmake
set(NAME fd.euler)
desbordante_add_lib(NAME)
target_sources(${NAME} PRIVATE cluster.cpp eulerfd.cpp mlfq.cpp search_tree.cpp)
target_link_libraries(
    ${NAME}
    PUBLIC ${DESBORDANTE_PREFIX}::config
    PRIVATE ${DESBORDANTE_PREFIX}::fd Boost::headers
)
```

--------------------------------

### Set Release Build Options

Source: https://github.com/desbordante/desbordante-core/blob/main/CMakeLists.txt

Initializes release build options based on general build options.

```cmake
set(RELEASE_BUILD_OPTS ${BUILD_OPTS})
```

--------------------------------

### Download Dataset

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Denial_Constraints.ipynb

Downloads the 'taxes.csv' dataset from a GitHub repository. This dataset will be used for demonstrating denial constraints.

```python
!wget https://raw.githubusercontent.com/Desbordante/desbordante-core/refs/heads/main/examples/datasets/taxes.csv
```

--------------------------------

### Configure fd.fun Library with CMake

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/fd/fun/CMakeLists.txt

This snippet sets up the fd.fun library using CMake. It defines the library name, adds its source files, and links necessary external libraries.

```cmake
set(NAME fd.fun)
desbordante_add_lib(NAME)
target_sources(${NAME} PRIVATE fun.cpp)
target_link_libraries(
    ${NAME} PRIVATE ${DESBORDANTE_PREFIX}::fd::pli spdlog::spdlog_header_only Boost::headers
)
```

--------------------------------

### CMakeLists.txt for cind.verifier

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/cind/cind_verifier/CMakeLists.txt

Configures the cind.verifier library, specifying its sources and linking necessary dependencies.

```cmake
set(NAME cind.verifier)
desbordante_add_lib(NAME)
target_sources(${NAME} PRIVATE cind_verifier.cpp)
target_link_libraries(
    ${NAME}
    PUBLIC ${DESBORDANTE_PREFIX}::config
    PRIVATE ${DESBORDANTE_PREFIX}::util magic_enum::magic_enum spdlog::spdlog_header_only
            Boost::headers
)
```

--------------------------------

### Download Dataset

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Simple_Statistics.ipynb

Download the 'Workshop.csv' dataset from a specified URL using wget. This dataset will be used for statistical analysis.

```bash
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/Workshop.csv
```

--------------------------------

### Download Dataset Files

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Approximate_and_Exact_Inclusion_Dependencies_Verification.ipynb

Downloads the necessary CSV files (orders.csv, products.csv, customers.csv) from a GitHub repository. These datasets are used in the demonstrations of IND and AIND verification.

```bash
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/orders.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/products.csv
!wget -q https://raw.githubusercontent.com/Desbordante/desbordante-core/main/examples/datasets/ind_datasets/customers.csv
```

--------------------------------

### Add 'model.bitset' Test Suite

Source: https://github.com/desbordante/desbordante-core/blob/main/src/tests/unit/CMakeLists.txt

Configures the 'model.bitset' unit test suite with its source file and libraries.

```cmake
desbordante_add_test(
    model.bitset SRCS test_bitset.cpp LIBS spdlog::spdlog_header_only Boost::headers
)
```

--------------------------------

### Configure PFD Verifier Library

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/fd/pfd_verifier/CMakeLists.txt

Sets the library name, adds source files, and links necessary libraries for the PFD Verifier. Includes configuration, model table, and magic enum dependencies.

```cmake
set(NAME fd.pfd_verifier)
desbordante_add_lib(NAME)
target_sources(${NAME} PRIVATE pfd_verifier.cpp)
target_link_libraries(
    ${NAME}
    PUBLIC ${DESBORDANTE_PREFIX}::config
    PRIVATE ${DESBORDANTE_PREFIX}::model::table magic_enum::magic_enum Boost::headers
)
```

--------------------------------

### Tane Library Configuration

Source: https://github.com/desbordante/desbordante-core/blob/main/src/core/algorithms/fd/tane/CMakeLists.txt

Configures the main Tane library, including its sources and linking against TaneCommon, fd::pli, fd::afd_metric, and Boost.

```cmake
set(NAME fd.tane)
desbordante_add_lib(NAME)
target_sources(${NAME} PRIVATE afd_measures.cpp tane.cpp)
target_link_libraries(${NAME} PUBLIC magic_enum::magic_enum)
target_link_libraries(
    ${NAME} PRIVATE ${DESBORDANTE_PREFIX}::fd::tane::common ${DESBORDANTE_PREFIX}::fd::pli
                    ${DESBORDANTE_PREFIX}::fd::afd_metric Boost::headers
)
```

--------------------------------

### Run Benchmarks Locally

Source: https://github.com/desbordante/desbordante-core/blob/main/src/tests/benchmark/README.md

Executes the compiled benchmark suite. Optionally, specify files for comparing previous results and saving new results.

```bash
cd build/target
./Desbordante_benchmark
```

```bash
./Desbordante_benchmark old-results.json new-results.json
```

--------------------------------

### Configure Algorithm Parameters for Anomaly Detection

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_3_anomaly_detection.ipynb

Sets up parameters for pandas.read_csv, exact and approximate FD algorithms, and a metric verifier. Includes configuration for metrics and their algorithms.

```python
# Parameters for pandas.read_csv(...).
HEADER = 0
SEPARATOR = ","

# Algorithm that finds exact FDs and its config.
EXACT_ALGORITHM_TYPE = desbordante.fd.algorithms.Default
EXACT_ALGORITHM_CONFIG = {}

# Algorithm that finds approximate FDs and its config.
APPROXIMATE_ALGORITHM_TYPE = desbordante.afd.algorithms.Default
ERROR = [0.01, 0.03, 0.05]
APPROXIMATE_ALGORITHM_CONFIG = {'error': ERROR}

METRIC_VERIFIER = "MetricVerifier"
METRIC_VERIFIER_CONFIG = {
    "lhs_indices": [1],
    "rhs_indices": [3],
    "metric": "euclidean",
    "metric_algorithm": "brute",
    "parameter": 4,
}

# available metrics: euclidean, levenshtein, cosine
# available metric algorithms: brute, approx, calipers
# parameter: desired distance related to metric. e.g., for euclidean metric parameter=4 means that euclidean distance has to be no larger than 4

# Variables to simplify the configuration string construction below.
EXACT_ALGORITHM = EXACT_ALGORITHM_TYPE.__name__
APPROXIMATE_ALGORITHM = APPROXIMATE_ALGORITHM_TYPE.__name__

CONFIG_STRING = f"""Starting anomaly detection scenario with parameters:
{ERROR=}
{HEADER=}
{SEPARATOR=}
{EXACT_ALGORITHM=}
{APPROXIMATE_ALGORITHM=}
{METRIC_VERIFIER=} """
```

--------------------------------

### Download and Extract Boost Source

Source: https://github.com/desbordante/desbordante-core/wiki/Building

Download the latest Boost library source code from the official website and extract it. This is a prerequisite for building Boost with GCC or LLVM Clang.

```shell
cd ~/Downloads
curl https://archives.boost.io/release/1.86.0/source/boost_1_86_0.tar.bz2 --output "boost_1_86_0.tar.bz2"
tar xvjf boost_1_86_0.tar.bz2 && rm boost_1_86_0.tar.bz2
cd boost_1_86_0
```

--------------------------------

### Simple Benchmark Implementation

Source: https://github.com/desbordante/desbordante-core/blob/main/src/tests/benchmark/README.md

Registers a straightforward benchmark that executes an algorithm with specified parameters and an optional name suffix. Allows setting a custom time threshold.

```C++
void XXBenchmark(BenchmarkRunner& runner, BenchmarkComparer& comparer) {
    algos::StdParamsMap params = {/* algo parameters */};
    // Arguments are: CSVConfig, other algo parameters and test name suffix
    // Returns generated benchmark name
    auto bench_name = runner.RegisterSimpleBenchmark(kHugeDataset, std::move(params), "simple");
    // Next line is optional, default threshold is 10%
    // SetThreshold takes benchmark name and desired threshold
    comparer.SetThreshold(bench_name, 20);
}
```

--------------------------------

### Create Input Data Directory

Source: https://github.com/desbordante/desbordante-core/blob/main/datasets/CMakeLists.txt

Creates the input data directory within the runtime output. This is safe for incremental builds.

```cmake
set(INPUT_DATA_DIR "${CMAKE_RUNTIME_OUTPUT_DIRECTORY}/input_data")
file(MAKE_DIRECTORY ${INPUT_DATA_DIR})
```

--------------------------------

### Configure pfd Binding

Source: https://github.com/desbordante/desbordante-core/blob/main/src/python_bindings/CMakeLists.txt

Adds the pfd binding, focused on verification, and lists its required libraries.

```cmake
desbordante_add_bind(
    pfd
    SRCS
    pfd/bind_pfd_verification.cpp
    LIBS
    ${DESBORDANTE_PREFIX}::fd::pfd_verifier
    magic_enum::magic_enum
    Boost::headers
)
```

--------------------------------

### Initialize and Execute Statistics Algorithm

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Simple_Statistics.ipynb

Initialize the Default statistics algorithm from Desbordante and load the pandas DataFrame into it. Then, execute the algorithm to compute statistics.

```python
data_stats = db.statistics.algorithms.Default()
data_stats.load_data(table=dataset)
data_stats.execute()
```

--------------------------------

### Build Desbordante Project

Source: https://github.com/desbordante/desbordante-core/blob/main/README.md

Compiles the Desbordante project, including tests. The `--pybind` flag can be used to also build the Python module.

```sh
./build.sh
```

```sh
./build.sh --pybind
```

--------------------------------

### Discover Approximate Inclusion Dependencies

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Approximate_Inclusion_Dependencies Mining.ipynb

Initializes the Mind algorithm for IND discovery and loads the specified tables. It then executes the discovery process with an error threshold of 0.3 and prints all found INDs.

```python
algo = desbordante.ind.algorithms.Mind()

TABLES = [
    ("employee.csv", ',', True),
    ("project_assignments.csv", ",", True),
]

algo.load_data(tables=TABLES)
algo.execute(error=0.3)

for ind in algo.get_inds():
    print("IND:", ind)
```

--------------------------------

### Fetch and Unpack Datasets

Source: https://github.com/desbordante/desbordante-core/blob/main/datasets/CMakeLists.txt

Iterates through declared datasets, fetching and unpacking them into the input data directory if enabled. Includes error handling for missing dataset files.

```cmake
foreach(name IN LISTS DATASETS)
    if(DESBORDANTE_FETCH_DATASETS)
        desbordante_fetch_datasets(NAME ${name} DOWNLOAD_DIR ${CMAKE_CURRENT_LIST_DIR})
    else()
        message(WARNING "There could be updates in dataset ${name} (Fetch disabled)")
    endif()
    set(FILE_PATH "${CMAKE_CURRENT_LIST_DIR}/${name}.zip")
    if(NOT EXISTS ${FILE_PATH})
        message(FATAL_ERROR "Cannot access ${FILE_PATH}: no such file. Datasets are required for \n        unit tests or benchmarks."
        )
    endif()
    desbordante_unpack_datasets(STAMP_FILE PATH ${FILE_PATH} EXTRACT_DIR ${INPUT_DATA_DIR})
    add_custom_target(Desbordante.zip.${name} DEPENDS ${STAMP_FILE})
endforeach()
```

--------------------------------

### Import Libraries

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Association_Rules.ipynb

Import the necessary libraries, Desbordante for association rule mining and Pandas for data manipulation.

```python
import desbordante
import pandas as pd
```

--------------------------------

### Configure Data Bindings Library

Source: https://github.com/desbordante/desbordante-core/blob/main/src/python_bindings/CMakeLists.txt

Defines and configures the 'bindlib.data' library, specifying its source files and linking necessary dependencies.

```cmake
set(NAME bindlib.data)
desbordante_add_lib(NAME OBJECT)
target_sources(${NAME} PRIVATE data/bind_data_types.cpp)
target_link_libraries(${NAME} PRIVATE pybind11::pybind11)
set_target_properties(${NAME} PROPERTIES CXX_VISIBILITY_PRESET "hidden")
```

--------------------------------

### Typo Miner Configuration Parameters

Source: https://github.com/desbordante/desbordante-core/blob/main/examples/notebooks/Desbordante_demo_scenario_1_typo_miner.ipynb

Sets up various parameters for value cluster filtering, FD algorithms (exact and approximate), and pandas read_csv configuration. Includes assertions to validate error values for different algorithms.

```python
# Value cluster filtering parameters.
RADIUS = 3
RATIO = 0.1

# Algorithm that finds exact FDs and its config.
EXACT_ALGORITHM_TYPE = desbordante.fd.algorithms.Default
EXACT_ALGO_CONFIG = {}

# Algorithm that finds approximate FDs and its config.
APPROXIMATE_ALGORITHM_TYPE = desbordante.afd.algorithms.Default
ERROR = 0.005 # Highest error for almost holding FDs.
APPROXIMATE_ALGO_CONFIG = {'error': ERROR}

# Parameters for pandas.read_csv(...).
DATASET_PATH = 'Workshop.csv'
HEADER = 0
SEPARATOR = ','

# Index of the almost holding FD. Chosen in advance purely for
# demonstration purposes. In a real usage scenario this should be a
# choice for the user.
FD_INDEX = 2


assert APPROXIMATE_ALGO_CONFIG['error'] > 0.0, 'Typo mining relies on non-zero error'
assert EXACT_ALGO_CONFIG.get('error', 0.0) == 0.0, 'Error must be 0 for precise algorithm'

# Variables to simplify the configuration string construction below.
EXACT_ALGORITHM = EXACT_ALGORITHM_TYPE.__name__
APPROXIMATE_ALGORITHM = APPROXIMATE_ALGORITHM_TYPE.__name__

# A message containing all variables used by this usage scenario, to be
# displayed to the user.
CONFIG_STRING = f"""
Starting typo discovery scenario with parameters:
{RADIUS=}
{RATIO=}
{ERROR=}
{DATASET_PATH=}
{EXACT_ALGORITHM=}
{APPROXIMATE_ALGORITHM=}
{HEADER=}
{SEPARATOR=} """

```

--------------------------------

### Add 'ind.faida' Test Suite

Source: https://github.com/desbordante/desbordante-core/blob/main/src/tests/unit/CMakeLists.txt

Configures the 'ind.faida' unit test suite, specifying its source file and required libraries.

```cmake
desbordante_add_test(
    ind.faida
    SRCS
    test_faida.cpp
    LIBS
    ${DESBORDANTE_PREFIX}::testlib::common
    ${DESBORDANTE_PREFIX}::ind
    ${DESBORDANTE_PREFIX}::ind::faida
    ${DESBORDANTE_PREFIX}::testlib::ind
    spdlog::spdlog_header_only
    magic_enum::magic_enum
    Boost::headers
)
```