Files
baffle-hub/app/services/geolite_country_importer.rb

288 lines
9.1 KiB
Ruby

require 'csv'
class GeoliteCountryImporter
BATCH_SIZE = 1000
def initialize(file_path, data_import:)
@file_path = file_path
@data_import = data_import
@total_records = 0
@processed_records = 0
@failed_records = 0
@errors = []
@locations_cache = {}
end
def import
Rails.logger.info "Starting import for file: #{@file_path}"
Rails.logger.info "File exists: #{File.exist?(@file_path)}"
Rails.logger.info "File size: #{File.size(@file_path)} bytes" if File.exist?(@file_path)
# Check if file is actually a zip by reading the magic bytes
is_zip_file = check_if_zip_file
Rails.logger.info "File is zip: #{is_zip_file}"
if is_zip_file
Rails.logger.info "Calling import_from_zip"
import_from_zip
else
Rails.logger.info "Calling regular import (not zip)"
load_locations_data
import_csv_file(@file_path)
end
{
total_records: @total_records,
processed_records: @processed_records,
failed_records: @failed_records,
errors: @errors
}
end
private
def check_if_zip_file
# Check if the file starts with ZIP magic bytes (PK\x03\x04)
File.open(@file_path, 'rb') do |file|
header = file.read(4)
return header == "PK\x03\x04"
end
rescue => e
Rails.logger.error "Error checking if file is zip: #{e.message}"
false
end
def import_from_zip
require 'zip'
require 'stringio'
Rails.logger.info "Processing zip file directly: #{@file_path}"
# Read the entire ZIP file content into memory first
zip_content = File.binread(@file_path)
Zip::File.open_buffer(StringIO.new(zip_content)) do |zip_file|
# First, see what's in the zip
Rails.logger.info "Files in zip:"
zip_file.each do |entry|
Rails.logger.info " - #{entry.name} (#{entry.size} bytes)"
end
# First, load location data from zip
load_locations_data_from_zip(zip_file)
# Then process block files from zip
zip_file.each do |entry|
if entry.name.include?('Blocks') && entry.name.end_with?('.csv')
Rails.logger.info "Processing block file from zip: #{entry.name}"
process_csv_from_zip(zip_file, entry)
end
end
end
rescue => e
Rails.logger.error "Error processing ZIP file: #{e.message}"
Rails.logger.error e.backtrace.join("\n")
raise
end
def process_csv_from_zip(zip_file, entry)
zip_file.get_input_stream(entry) do |io|
# Read the entire content from the stream
content = io.read
# Try different encodings if UTF-8 fails
encodings = ['UTF-8', 'ISO-8859-1', 'Windows-1252']
encoding_used = nil
encodings.each do |encoding|
begin
# Parse the CSV content from the string
CSV.parse(content, headers: true, header_converters: :symbol, encoding: encoding) do |row|
@total_records += 1
begin
import_record(row)
@processed_records += 1
rescue => e
@failed_records += 1
@errors << "Row #{@total_records}: #{e.message} - Data: #{row.to_h}"
end
update_progress_if_needed
end
encoding_used = encoding
Rails.logger.info "Successfully processed #{entry.name} with #{encoding} encoding"
break
rescue CSV::InvalidEncodingError => e
Rails.logger.warn "Failed to process #{entry.name} with #{encoding} encoding: #{e.message}"
next if encoding != encodings.last
raise e if encoding == encodings.last
end
end
unless encoding_used
@errors << "Failed to process #{entry.name} with any supported encoding"
end
end
end
def load_locations_data_from_zip(zip_file)
require 'zip'
# Find all location files and prioritize English
location_entries = zip_file.select { |entry| entry.name.include?('Locations') && entry.name.end_with?('.csv') }
# Sort to prioritize English locations file
location_entries.sort_by! { |entry| entry.name.include?('Locations-en') ? 0 : 1 }
location_entries.each do |entry|
Rails.logger.info "Loading locations from: #{entry.name}"
zip_file.get_input_stream(entry) do |io|
# Read the entire content from the stream
content = io.read
# Try different encodings if UTF-8 fails
encodings = ['UTF-8', 'ISO-8859-1', 'Windows-1252']
encodings.each do |encoding|
begin
# Parse the CSV content from the string
CSV.parse(content, headers: true, header_converters: :symbol, encoding: encoding) do |row|
geoname_id = row[:geoname_id]
next unless geoname_id
@locations_cache[geoname_id] = {
country_iso_code: row[:country_iso_code],
country_name: row[:country_name],
continent_code: row[:continent_code],
continent_name: row[:continent_name],
is_in_european_union: row[:is_in_european_union]
}
end
Rails.logger.info "Loaded locations from #{entry.name} with #{encoding} encoding"
break
rescue CSV::InvalidEncodingError => e
Rails.logger.warn "Failed to load locations from #{entry.name} with #{encoding} encoding: #{e.message}"
next if encoding != encodings.last
raise e if encoding == encodings.last
end
end
end
end
Rails.logger.info "Loaded #{@locations_cache.size} location records"
end
def import_csv_file(csv_file)
CSV.foreach(csv_file, headers: true, header_converters: :symbol, encoding: 'UTF-8') do |row|
@total_records += 1
begin
import_record(row)
@processed_records += 1
rescue => e
@failed_records += 1
@errors << "Row #{@total_records}: #{e.message} - Data: #{row.to_h}"
# Update progress every 100 records or on error
update_progress_if_needed
end
update_progress_if_needed
end
end
def import_record(row)
network = row[:network]
geoname_id = row[:geoname_id]
registered_country_geoname_id = row[:registered_country_geoname_id]
is_anonymous_proxy = row[:is_anonymous_proxy] == '1'
is_satellite_provider = row[:is_satellite_provider] == '1'
is_anycast = row[:is_anycast] == '1'
unless network
raise "Missing required field: network"
end
# Validate network format
IPAddr.new(network) # This will raise if invalid
# Get location data - prefer geoname_id, then registered_country_geoname_id
location_data = @locations_cache[geoname_id] || @locations_cache[registered_country_geoname_id] || {}
additional_data = {
geoname_id: geoname_id,
registered_country_geoname_id: registered_country_geoname_id,
represented_country_geoname_id: row[:represented_country_geoname_id],
continent_code: location_data[:continent_code],
continent_name: location_data[:continent_name],
country_name: location_data[:country_name],
is_in_european_union: location_data[:is_in_european_union],
is_satellite_provider: is_satellite_provider,
is_anycast: is_anycast
}.compact
NetworkRange.upsert(
{
network: network,
country: location_data[:country_iso_code],
is_proxy: is_anonymous_proxy,
source: 'geolite_country',
additional_data: additional_data,
updated_at: Time.current
},
unique_by: :index_network_ranges_on_network_unique
)
end
def update_progress_if_needed
if (@processed_records + @failed_records) % 100 == 0
@data_import.update_progress(
processed: @processed_records,
failed: @failed_records,
total_records: @total_records,
stats: {
total_records: @total_records,
current_file: File.basename(@file_path),
locations_loaded: @locations_cache.size,
recent_errors: @errors.last(5)
}
)
end
end
def load_locations_data
locations_files = find_locations_files
locations_files.each do |locations_file|
CSV.foreach(locations_file, headers: true, header_converters: :symbol, encoding: 'UTF-8') do |row|
geoname_id = row[:geoname_id]
next unless geoname_id
@locations_cache[geoname_id] = {
country_iso_code: row[:country_iso_code],
country_name: row[:country_name],
continent_code: row[:continent_code],
continent_name: row[:continent_name],
is_in_european_union: row[:is_in_european_union]
}
end
end
end
def find_locations_files
if @file_path.end_with?('.zip')
base_dir = File.dirname(@file_path)
base_name = File.basename(@file_path, '.zip')
# Look for English locations file first, then any locations file
[
File.join(base_dir, "#{base_name}-Locations-en.csv"),
Dir[File.join(base_dir, "#{base_name}-Locations-*.csv")].first
].compact.select { |file| File.exist?(file) }
else
base_dir = File.dirname(@file_path)
Dir[File.join(base_dir, "*Locations*.csv")].select { |file| File.exist?(file) }
end
end
end