/**
* Name and address normalization utilities for cross-platform matching
*/
// Common abbreviations and their expansions
const ABBREVIATIONS: Record<string, string> = {
'st': 'street',
'ave': 'avenue',
'blvd': 'boulevard',
'rd': 'road',
'dr': 'drive',
'ln': 'lane',
'ct': 'court',
'pl': 'place',
'sq': 'square',
'pkwy': 'parkway',
'hwy': 'highway',
'apt': 'apartment',
'ste': 'suite',
'fl': 'floor',
'n': 'north',
's': 'south',
'e': 'east',
'w': 'west',
'ne': 'northeast',
'nw': 'northwest',
'se': 'southeast',
'sw': 'southwest',
'nyc': 'new york city',
'la': 'los angeles',
'sf': 'san francisco',
'dc': 'washington dc',
};
// Common restaurant name prefixes/suffixes to normalize
const NAME_PATTERNS: Array<{ pattern: RegExp; replacement: string }> = [
// Remove common suffixes
{ pattern: /\s+(restaurant|ristorante|bistro|cafe|café|bar|grill|kitchen|house|room)$/i, replacement: '' },
// Normalize "and" variations
{ pattern: /\s+(&|and)\s+/gi, replacement: ' & ' },
// Remove possessive
{ pattern: /'s\b/gi, replacement: '' },
// Normalize "the" at start
{ pattern: /^the\s+/i, replacement: '' },
];
/**
* Normalize a restaurant name for matching
*/
export function normalizeName(name: string): string {
let normalized = name.trim().toLowerCase();
// Apply name patterns
for (const { pattern, replacement } of NAME_PATTERNS) {
normalized = normalized.replace(pattern, replacement);
}
// Remove extra whitespace
normalized = normalized.replace(/\s+/g, ' ').trim();
// Remove accents
normalized = removeAccents(normalized);
return normalized;
}
/**
* Normalize an address for matching
*/
export function normalizeAddress(address: string): string {
let normalized = address.trim().toLowerCase();
// Expand abbreviations
const words = normalized.split(/\s+/);
const expandedWords = words.map(word => {
// Remove trailing periods
const clean = word.replace(/\.$/, '');
return ABBREVIATIONS[clean] || clean;
});
normalized = expandedWords.join(' ');
// Remove apartment/suite numbers (they vary too much)
normalized = normalized.replace(/\s+(apt|suite|ste|unit|#)\s*\w+/gi, '');
// Remove zip codes
normalized = normalized.replace(/\s+\d{5}(-\d{4})?$/, '');
// Remove extra whitespace
normalized = normalized.replace(/\s+/g, ' ').trim();
// Remove accents
normalized = removeAccents(normalized);
return normalized;
}
/**
* Normalize a city/neighborhood name
*/
export function normalizeLocation(location: string): string {
let normalized = location.trim().toLowerCase();
// Expand common city abbreviations
const words = normalized.split(/[\s,]+/);
const expandedWords = words.map(word => ABBREVIATIONS[word] || word);
normalized = expandedWords.join(' ');
// Remove state abbreviations at end
normalized = normalized.replace(/,?\s+[a-z]{2}$/i, '');
// Remove extra whitespace
normalized = normalized.replace(/\s+/g, ' ').trim();
// Remove accents
normalized = removeAccents(normalized);
return normalized;
}
/**
* Remove accents/diacritics from a string
*/
export function removeAccents(str: string): string {
return str.normalize('NFD').replace(/[\u0300-\u036f]/g, '');
}
/**
* Extract the core name from a restaurant name
* E.g., "Carbone NYC" -> "carbone"
*/
export function extractCoreName(name: string): string {
let core = normalizeName(name);
// Remove location suffixes (NYC, Downtown, etc.)
const locationSuffixes = /\s+(nyc|downtown|uptown|midtown|brooklyn|manhattan|chicago|la|sf|miami|boston|philly|dc|atl|houston|dallas|austin|seattle|portland|denver)$/i;
core = core.replace(locationSuffixes, '');
// Remove numbers at end (like "2.0" or "II")
core = core.replace(/\s+(\d+(\.\d+)?|i{1,3}|iv|v|vi{0,3})$/i, '');
return core.trim();
}
/**
* Parse a full address into components
*/
export interface AddressComponents {
street?: string;
city?: string;
state?: string;
zip?: string;
neighborhood?: string;
}
export function parseAddress(fullAddress: string): AddressComponents {
const components: AddressComponents = {};
// Try to extract zip code
const zipMatch = fullAddress.match(/\b(\d{5})(-\d{4})?\b/);
if (zipMatch) {
components.zip = zipMatch[1];
}
// Try to extract state (2-letter abbreviation)
const stateMatch = fullAddress.match(/,\s*([A-Z]{2})\s*(\d{5})?$/i);
if (stateMatch) {
components.state = stateMatch[1].toUpperCase();
}
// Split by commas
const parts = fullAddress.split(',').map(p => p.trim());
if (parts.length >= 1) {
components.street = parts[0];
}
if (parts.length >= 2) {
// Could be city or neighborhood
components.city = parts[1].replace(/\s+[A-Z]{2}\s*\d{5}.*$/, '').trim();
}
if (parts.length >= 3) {
// First part after street might be neighborhood
components.neighborhood = parts[1];
components.city = parts[2].replace(/\s+[A-Z]{2}\s*\d{5}.*$/, '').trim();
}
return components;
}
/**
* Create a normalized key for deduplication
*/
export function createDedupeKey(name: string, city: string): string {
const normName = extractCoreName(name);
const normCity = normalizeLocation(city);
return `${normName}|${normCity}`;
}
/**
* Normalize cuisine type
*/
export function normalizeCuisine(cuisine: string): string {
return cuisine
.toLowerCase()
.trim()
.replace(/\s+cuisine$/i, '')
.replace(/\s+food$/i, '')
.replace(/\s+/g, '-');
}
/**
* Parse comma-separated cuisines into array
*/
export function parseCuisines(cuisineString: string): string[] {
if (!cuisineString) return [];
return cuisineString
.split(/[,/]/)
.map(c => normalizeCuisine(c))
.filter(c => c.length > 0);
}
