Skip to main content

Company Parser Implementation Guide

Overview​

The InpiAPI::Parsers::CompanyParser class handles parsing and processing of company data from external APIs. This guide documents the current implementation structure.

Class Structure​

Constants​

The parser uses several constants to define field processing rules:

  • ROOT_FIELDS_TO_EXTRACT - Fields extracted from root data
  • ROOT_FIELDS_TO_DELETE - Fields removed after processing
  • FORMALITY_FIELDS_TO_EXTRACT - Formality-specific fields
  • FORMALITY_FIELDS_TO_DELETE - Formality fields to clean up
  • ZERO_VALUE_FIELDS - Fields with zero values to remove
  • ENTERPRISE_CONTENT_FIELDS - Enterprise content fields
  • ENTERPRISE_CONTENT_FIELDS_TO_DELETE - Enterprise fields to clean up

Key Methods​

Data Extraction Methods​

  • extract_root_data - Main orchestrator for root data extraction
  • extract_root_fields - Extracts specific fields from raw data
  • remove_processed_root_fields - Cleans up processed fields
  • merge_formality_fields - Handles formality data merging
  • remove_processed_formality_fields - Cleans up formality fields

Identity Processing Methods​

  • identite - Main identity processing method
  • person_has_identite? - Validates identity data presence
  • prepare_identite_data - Prepares data for processing
  • build_enterprise_data - Builds enterprise data structure
  • merge_description - Handles description merging
  • merge_content_fields - Processes content field operations

Notes about updated fields​

There is 3 fields related to update time

  • updated in companies table is storing the fierld value from the api
  • api_updated_at. is for the last time the companyParser ran successfully for that record
  • updated_at is the rails fields storing the last change on comapnies/companies_infos for that record

Data Normalization​

  • normalize_diffusion_insee - Processes INSEE diffusion flags
  • remove_zero_value_fields - Removes fields with zero values

Usage Guidelines​

Basic Usage​

parser = InpiAPI::Parsers::CompanyParser.new(raw_data, company)
parser.update
if parser.validate_extracted_data
  # Processing successful
else
  # Handle errors: parser.errors
end

Data Processing Flow​

  1. Extract root data using extract_root_data
  2. Process identity information via identite
  3. Normalize and clean data
  4. Validate extracted data
  5. Update company record

Error Handling​

The parser includes comprehensive error handling:

  • Validation of required data fields
  • Graceful handling of missing or malformed data
  • Detailed error reporting via parser.errors
  • Rollback capabilities for failed operations

Best Practices​

  1. Always validate data before processing
  2. Check parser.errors after processing
  3. Use appropriate error handling for your use case
  4. Test with various data formats to ensure robustness

Testing​

When testing the parser:

RSpec.describe InpiAPI::Parsers::CompanyParser do
  let(:parser) { described_class.new(valid_data, company) }
  
  it 'processes valid data successfully' do
    parser.update
    expect(parser.validate_extracted_data).to be true
  end
  
  it 'handles invalid data gracefully' do
    invalid_parser = described_class.new(invalid_data, company)
    invalid_parser.update
    expect(invalid_parser.validate_extracted_data).to be false
    expect(invalid_parser.errors).to be_present
  end
end

Integration​

The parser integrates with:

  • UpdateCompanyJob for background processing
  • ParserJob system for job management
  • Company models for data persistence
  • Various association models (facilities, rights, etc.)