CSV Completeness Checker

A fast and easy-to-use tool that analyzes CSV files for data completeness by checking for missing/invalid values across different column combinations.

Features

• Checks for missing or invalid values across user defined column combinations.
• Saves and loads analysis configurations to and from JSON files for easy reuse.
• Interactive command line interface to guide you through the analysis setup.
• Full command line automation support for scripting and batch processing.

Build

Just let CMake do its magic.

Dependencies (automatically fetched):

• fast-cpp-csv-parser
• nlohmann_json
• cxxopts
• fmt

Usage

Interactive Mode

Run the application without arguments to use the interactive mode:

./csv-completeness-checker

The application will guide you through the following steps:

1. Provide a CSV or JSON file:

   • You can start with a .csv file to analyze.
   • Or, you can provide a .json file with a saved configuration from a previous session.

2. Select Fields to Analyze:

   • Choose the columns from your CSV file that you want to include in the analysis.

3. Define Invalid Values:

   • For each selected field, you can specify a comma-separated list of values that should be considered invalid or empty (e.g., N/A, Unspecified, -1).

4. Define Column Combinations:

   • Specify the combinations of columns you want to check for completeness. Use : to require multiple fields together and / inside a group to indicate alternatives. For example:
     • 1:3 checks rows where both column 1 and column 3 are valid.
     • 1:2:3/4 checks rows where columns 1 and 2 are valid and at least one of columns 3 or 4 is valid.
   • You can also provide multiple combinations separated by commas (e.g., 1:2, 1:2:3/4).

5. Save Configuration (Optional):

   • You can save your configuration (selected fields, invalid values, and combinations) to a JSON file for later use.

6. Process the CSV:

   • The tool will then process the CSV file and output the results for each column combination, showing both the raw counts (valid rows / total rows) and the completeness percentage.

Command Line Mode

Use command line arguments to automate the analysis without interactive prompts:

./csv-completeness-checker [OPTIONS]

Options:

Argument	Description
--csv, -c	Path to CSV file to analyze
--config, -C	Path to JSON configuration file (overrides --csv)
--output, -o	Path to save output JSON configuration
--fields, -f	Comma-separated field numbers to analyze
--invalid-values, -i	Invalid values mapping (format: field:value1,value2:field:value3...)
--combinations, -b	Column combinations to check (format: 1:2,1:3/4)
--format	Output format: text, json, csv, or keyvalue (will also enable quiet mode)
--silent, -q	Minimal output (only results)
--help, -h	Display help message