CSV Processing Library

A lightweight, dependency-free Rust library for processing CSV data with column selection and row filtering capabilities. The library provides a C-compatible interface, making it suitable for integration with other programming languages and systems.

Features

• Process CSV data from both strings and files
• Select specific columns for output
• Apply multiple filter conditions to rows
• Support for various comparison operators (=, !=, <, >, <=, >=)
• Zero external dependencies
• C-compatible interface
• Comprehensive test coverage

Installation

Since this is a Rust library that compiles to a C-compatible library, you'll need to have Rust installed on your system. The library can be built using Cargo, Rust's package manager.

1. Add the library to your Cargo.toml:

[lib]
name = "csv"
crate-type = ["cdylib"]

2. Build the library:

cargo build --release

This will create a shared library in the target/release directory.

Usage

C Interface

The library provides two main functions:

void processCsv(const char[] csv_data, const char[] selected_columns, const char[] row_filter_definitions);
void processCsvFile(const char[] csv_file_path, const char[] selected_columns, const char[] row_filter_definitions);

Parameters

• csv_data/csv_file_path: The CSV data as a string or file path
• selected_columns: Comma-separated list of column headers to include in the output
• row_filter_definitions: Newline-separated list of filter conditions

Filter syntax

Filters support the following operators:

• = (equals)
• != (not equals)
• < (less than)
• > (greater than)
• <= (less than or equal)
• >= (greater than or equal)

Multiple filters for the same column are treated as OR conditions, while filters across different columns are treated as AND conditions.

Example

// Processing CSV data directly
const char* csv_data = "name,age,salary\nJohn,30,50000\nJane,25,60000";
const char* selected_columns = "name,salary";
const char* filters = "age>20\nsalary>=55000";

processCsv(csv_data, selected_columns, filters);

// Processing CSV from a file
const char* file_path = "employees.csv";
processCsvFile(file_path, selected_columns, filters);

Output Format

The output is printed to stdout in CSV format, with the selected columns in the order specified in the selected_columns parameter. For example:

name,salary
Jane,60000

Error Handling

The library handles various error conditions and provides appropriate error messages:

• Invalid CSV format
• Missing headers
• Invalid filter syntax
• File access errors
• Character encoding errors

Errors are printed to stderr, and the program exits with status code 1 when an error occurs.

Testing

The library includes a comprehensive test suite covering various scenarios:

• Basic column selection
• Different filter combinations
• Edge cases
• Error conditions
• Double quote handling in headers
• Multiple filter conditions per column

Run the tests using:

cargo test -- --test-threads=1

Important Note: The flag -- --test-threads=1 is required when running tests. This is because some tests intentionally trigger process::exit and panic conditions to verify error handling. Running tests in a single thread prevents these termination conditions from interfering with other test executions.

Limitations

• Assumes CSV data is comma-delimited
• Doesn't support quoted fields containing commas
• All comparisons are performed lexicographically
• Filter values must be exact matches (no partial matches or regular expressions)
• Headers must be unique

Performance Considerations

• The library processes CSV data line by line
• No additional memory allocation for intermediate results
• Filters are evaluated lazily
• Column selection is performed during output generation

Thread Safety

The library is not explicitly thread-safe. When using in a multi-threaded environment, ensure that calls to the processing functions are properly synchronized.