Links

📦 CPAN Distribution: https://metacpan.org/pod/OMOP::CSV::Validator

OMOP CSV Validator

The OMOP CSV Validator is a CLI tool (and module) that validates OMOP CDM CSV files against their expected data types. Rather than relying solely on Types::Standard or similar libraries, it converts SQL schemas derived from the OMOP Common Data Model (CDM) PostgreSQL DDL files into JSON schemas. It then utilizes JSON::Validator, which scales efficiently with large datasets and provides meaningful error messages.

Features

• DDL Parsing: Automatically converts PostgreSQL OMOP CDM DDL into JSON schemas.
• Version Independent Works with any DDL (e.g., 5.3, 5.4).
• CSV Validation: Validates CSV files using JSON::Validator.
• Modular Design: Separate CLI and module for easy testing and integration.

Installation

This project uses cpanm along with a cpanfile to manage dependencies. It is recommended to install dependencies locally using local::lib.

Step 1: Install cpanminus

If you don't have cpanm installed, run:

sudo apt-get install cpanminus

If you don't have the gcc compiler and other default Linux utils installed please do:

sudo apt-get install gcc make libperl-dev

Step 2: Set Up local::lib

Configure a local library in your home directory. For example:

cpanm --local-lib=~/perl5 local::lib && eval $(perl -I ~/perl5/lib/perl5/ -Mlocal::lib)

Then, add this settings to your shell profile (e.g. ~/.bashrc or ~/.zshrc) so that your shell knows about your local library.

echo 'eval $(perl -I ~/perl5/lib/perl5/ -Mlocal::lib)' >> ~/.bashrc

Step 3: Download and installation:

From CPAN

From any directory:

cpanm -n OMOP::CSV::Validator

From Github

1. Clone the repository:

git clone https://github.com/mrueda/omop-csv-validator.git
cd omop-csv-validator

2. Install Dependencies from:

cpanm -n --installdeps .

This command reads the included cpanfile and installs all required dependencies into your local library directory.

Usage

Command-Line Interface

Once dependencies are installed, you can run the CLI tool as follows:

(If you installed fron CPAN then you can simply run omop-csv-validator).

bin/omop-csv-validator --ddl path/to/OMOPCDM_ddl.sql --input path/to/data.csv --sep ","

With the included example data:

bin/omop-csv-validator --ddl ddl/OMOPCDM_postgresql_5.4_ddl.sql -i example/DRUG_EXPOSURE.csv -sep $'\t'

Example of an error in table person:

Options

Usage:
      omop_csv_validator --ddl DDL.sql --input DATA.csv [--sep $'\t'] [--table person] [--save-schemas schemas.json]

Options:
    --ddl
        (required) Path to the PostgreSQL DDL file defining OMOP CDM table
        structures.

    --input
        (required) Path to the input CSV file to validate.

    --sep
        CSV field separator (default: comma). For tab, use: --sep $'\t'

    --table, -t
        (optional) Table name to validate against. If not provided, the
        script will attempt to derive the table name from the CSV filename.

    --save-schemas
        (optional) Path to a file where the DDL-derived schemas should be
        saved (in JSON format).

    --no-color, -nc
        (Optional) Turn off STDOUT color

    --help, -h
        Display this help message.

    --version, -V
        Show the script's version (which corresponds to
        "OMOP::CSV::Validator::VERSION").

Running Tests

To run the test suite, execute:

prove -l t/

Utilities

• reorder-csv.pl

See directory utils.

Notes

TIMESTAMP format

The TIMESTAMP field from the DDL is translated to the date-time format as defined in the JSON Schema specification (ISO 8601):

YYYY-MM-DDTHH:MM:SSZ

If your TIMESTAMP fields (ex: visit_end_datetime) follow the date format YYYY-MM-DD, you can ignore any validation errors related to the missing time component.

Integers Length

Depending on how you generate integers for person_id, you might end up with values longer than 20 characters. When this occurs, the validator will report an error like /person_id: Expected integer - got number. In such cases, you can safely ignore or discard the message.

Author

Written by Manuel Rueda, PhD. Info about CNAG can be found at https://www.cnag.eu.

Contributing

Contributions, issues, and feature requests are welcome. Please check the issues page for details.

License

This project is released under the Artistic License 2.0.