README

Overview

This program provides a CLI tool to parse CSV files, filtering out columns & rows based on criteria specified in a configuration file (.\config\config.json). Users can optionally override settings using command-line arguments, ensuring flexibility and adaptability to various use cases.

Configuration

In the follow example(s) $ROOT is the directory of the binary file or .exe.

By default, the program will search for a config directory & config file in the following locations: $ROOT/config & $ROOT/config/config.json respectively.

If the config folder and/or the config file is not found, the program will generate a 'dummy' config file with similar structure & values to the snippet below. You can find the newly created file at $ROOT/config/config.json.

The configuration file ($ROOT/config/config.json) should be formatted as follows:

{
  "source": "\\windows\\path\\to\\source.csv",
  "output_type": "csv",
  "output_path": "linux_style/path/to/output.csv",
  "has_headers": true,
  "fields": [
    "Field1",
    "Field2",
    "Field3"
  ],
  "unique_fields": [
    "unique_fields_to_include"
  ],
  "include_cols_with": {
    "Field1": [
      "FilterCriteria1",
      "FilterCriteria2"
    ],
    "Field2": [
      "FilterCriteria3",
      "FilterCriteria4"
    ]
  }
}

• Pro-Tip!: The code handles both Windows and Linux-style paths. That being said - Filesystem themselves may not play nicely if you're mixing OS paths.

Fields:

• source: Path to the input CSV file.
• output_type: Desired output format (e.g., csv).
• output_path: Path for the output CSV file.
• has_headers: Boolean value indicating whether the CSV file has headers.
• fields: An array of fields to always include in the output.
• unique_fields: An array of fields to include in the output only if they are unique. (Optional - Leave list blank if not needed)
• include_cols_with: A dictionary defining filtering criteria the keys are the columns, and the list of values are values that should be included in the output.

Command Line Interface

Most commands are also implemented as CLI arguments. You can view the help message using the following command:

.\csv_parser_rs --help
# or
.\csv_parser_rs -h

You can run the parser using the following command:

.\csv_parser_rs [source] [-c config_file] [-t output_type] [-o output_path]

Arguments:

• source: (Optional) First argument - Path to the source CSV file; overrides the source in config.json.
• -c, --config: (Optional) Path to an alternative configuration file; overrides the default.
• -t, --output_type: (Optional) Specify the output type (stdout, csv); defaults to the value in config.json.
• -o, --output_path: (Optional) Specify the output file path; overrides the output_path in config.json.

Output Types

The tool supports two output types:

• stdout: Print the results to the standard output.
• csv: Save the results to a specified CSV file.

Usage Example

To run the parser with a custom configuration file (ie: One that is not in the assumed location):

.\csv_parser_rs -c path\to\config.json

To override the configuration using CLI arguments:

.\csv_parser_rs path\to\input.csv -t stdout -o path\to\output.csv

FAQ's

What happens to duplicates exactly?

The program will only include the first occurrence of a row with a unique field in the output. That's to say - the first row with a unique field will be included when reading from row/line 1 to the last row/line that is populated.

Weird caveat:

Technically - a blank cell is a unique value (Regardless of if it actually has content or not, or a space etc.) - because of this it's important to understand that if you're filtering (via unique_fields ) on a column that has blank cells in it, the results will follow the same logic - ie: only a single blank cell will be included in the output.

What happens if the unique column contains blank values?

The program will only include the first occurrence of a row with a unique field in the output.

What happens if has_headers is set to false?

If has_headers is set to false, the program will treat the first row as a data row and include it in the output. This includes the ability to consider it a filterable row.

What happens if the config is not provided/missing a field/malformed?

If the directory & file are missing entirely, the program will generate a dummy config directory and file in the assumed location. If the file is missing a field, the program will generate an error message and exit, no file operations will be performed at all.

Why Rust?

Rust is a systems programming language that provides memory safety, zero-cost abstractions, and concurrency.

Author

Blake B.

License

This project is licensed under the MIT License.