{ "edges": [ { "to": "f1f3e2d0-48a9-4439-978c-271b61fb0995", "from": "b803208f-0943-450b-b971-3efaaa0a09a9", "toPort": "0", "fromPort": "0" }, { "to": "2b922787-6e81-4c99-8a4c-51b8cfd17fe4", "from": "c7f93646-8a4c-4afe-a613-4495eee1f218", "toPort": "0", "fromPort": "0" }, { "to": "53119f38-dd63-42e3-9062-1b4529956cce", "from": "56d2d3bf-3398-42b5-bd85-f8c07ac2240f", "toPort": "0", "fromPort": "0" }, { "to": "b803208f-0943-450b-b971-3efaaa0a09a9", "from": "2b922787-6e81-4c99-8a4c-51b8cfd17fe4", "toPort": "0", "fromPort": "0" }, { "to": "9c09bd5e-dceb-45b4-9986-3e5aac436713", "from": "802b7a74-7695-49d9-91db-f03ec621414d", "toPort": "0", "fromPort": "0" }, { "to": "1cdf0632-9041-4417-a0b8-852647c21eda", "from": "7329e920-df75-4a7b-bc5f-e9102c78c6f5", "toPort": "0", "fromPort": "0" }, { "to": "c7f93646-8a4c-4afe-a613-4495eee1f218", "from": "218438d3-6f88-49b7-b11d-9b2d92f74454", "toPort": "0", "fromPort": "0" }, { "to": "56d2d3bf-3398-42b5-bd85-f8c07ac2240f", "from": "676cff78-9de6-4d99-9dcb-424325a590f1", "toPort": "0", "fromPort": "0" }, { "to": "676cff78-9de6-4d99-9dcb-424325a590f1", "from": "1cdf0632-9041-4417-a0b8-852647c21eda", "toPort": "0", "fromPort": "0" }, { "to": "218438d3-6f88-49b7-b11d-9b2d92f74454", "from": "9c09bd5e-dceb-45b4-9986-3e5aac436713", "toPort": "0", "fromPort": "0" }, { "to": "802b7a74-7695-49d9-91db-f03ec621414d", "from": "53119f38-dd63-42e3-9062-1b4529956cce", "toPort": "0", "fromPort": "0" } ], "nodes": [ { "id": "1cdf0632-9041-4417-a0b8-852647c21eda", "name": "Set Google Maps Configuration", "type": "set", "params": { "fields": { "values": [ { "name": "city", "stringValue": "={{ $json.city || $json['City Name'] || '' }}" }, { "name": "keyword", "stringValue": "real estate agent" }, { "name": "baseUrl", "stringValue": "https://www.google.com/maps/search/" }, { "name": "searchUrl", "stringValue": "={{ 'https://www.google.com/maps/search/real+estate+agent+in+' + String($json.city || $json['City Name'] || '').replace(/\\s+/g, '+') }}" } ] }, "options": {} }, "position": [ 976, 320 ] }, { "id": "802b7a74-7695-49d9-91db-f03ec621414d", "name": "Parse Full Business Info", "type": "code", "params": { "jsCode": "// Process ALL input items - n8n processes items individually, so we need to handle all\nconst inputItems = $input.all() || [$input.item];\nconst output = [];\n\n// Process each item individually\nfor (const inputItem of inputItems) {\n let originalData = {};\n let currentMapUrl = '';\n \n // Get HTML from ScrapeOps detail page first\n let htmlContent = '';\n if (inputItem.json.response && typeof inputItem.json.response === 'string') {\n htmlContent = inputItem.json.response;\n } else if (typeof inputItem.json === 'string' && inputItem.json.length > 1000) {\n htmlContent = inputItem.json;\n } else if (inputItem.json.body && typeof inputItem.json.body === 'string') {\n htmlContent = inputItem.json.body;\n } else if (inputItem.json.data && typeof inputItem.json.data === 'string') {\n htmlContent = inputItem.json.data;\n }\n \n // Extract mapUrl from HTML\n let extractedMapUrl = '';\n if (htmlContent && typeof htmlContent === 'string' && htmlContent.length > 100) {\n const mapUrlPatterns = [\n /https?:\\/\\/[^\"'<>\\s]*maps\\.google\\.com[^\"'<>\\s]*place[^\"'<>\\s]*/i,\n /maps\\.google\\.com[^\"'<>\\s]*place[^\"'<>\\s]*/i,\n /\\/maps\\/place\\/[^\"'<>\\s\\?]+/i\n ];\n for (const pattern of mapUrlPatterns) {\n const match = htmlContent.match(pattern);\n if (match && match[0]) {\n extractedMapUrl = match[0];\n if (!extractedMapUrl.startsWith('http')) {\n extractedMapUrl = 'https://' + extractedMapUrl.split(/[\"'<>\\s\\?]/)[0];\n } else {\n extractedMapUrl = extractedMapUrl.split(/[\"'<>\\s\\?]/)[0];\n }\n break;\n }\n }\n if (!extractedMapUrl) {\n const placeIdPatterns = [/1s([^:!\\/\"'<>\\s]+)/, /place\\/([^\\/\\?\"'<>\\s]+)/];\n for (const pattern of placeIdPatterns) {\n const match = htmlContent.match(pattern);\n if (match && match[1]) {\n const nameMatch = htmlContent.match(/]*>([^<]+)<\\/h1>/i) || \n htmlContent.match(/]*class=[\"'][^\"']*qBF1Pd[^\"']*[\"'][^>]*>([^<]+)<\\/div>/i);\n if (nameMatch && nameMatch[1]) {\n const bn = nameMatch[1].trim().replace(/\\s+/g, '+');\n extractedMapUrl = `https://www.google.com/maps/place/${bn}/data=!4m7!3m6!1s${match[1]}`;\n break;\n }\n }\n }\n }\n }\n \n // Match with Parse Business Details\n try {\n const parseNode = $('Parse Business Listings');\n if (parseNode && parseNode.all) {\n const parseData = parseNode.all() || [];\n if (extractedMapUrl) {\n const getPlaceId = (url) => {\n if (!url) return '';\n const m1 = url.match(/1s([^:!\\/]+)/);\n const m2 = url.match(/place\\/([^\\/\\?]+)/);\n return m1 ? m1[1] : (m2 ? m2[1] : '');\n };\n const extractedPlaceId = getPlaceId(extractedMapUrl);\n for (let i = 0; i < parseData.length; i++) {\n const itemData = parseData[i].json || parseData[i];\n const itemMapUrl = itemData.mapUrl || '';\n if (itemMapUrl) {\n const itemPlaceId = getPlaceId(itemMapUrl);\n if (itemPlaceId && extractedPlaceId && itemPlaceId === extractedPlaceId) {\n originalData = itemData; currentMapUrl = itemMapUrl; break;\n }\n if (itemPlaceId && extractedPlaceId && \n (itemPlaceId.includes(extractedPlaceId) || extractedPlaceId.includes(itemPlaceId))) {\n originalData = itemData; currentMapUrl = itemMapUrl; break;\n }\n if (itemMapUrl.includes(extractedPlaceId) || extractedMapUrl.includes(itemPlaceId)) {\n originalData = itemData; currentMapUrl = itemMapUrl; break;\n }\n }\n }\n }\n if (!originalData || Object.keys(originalData).length === 0) {\n const scrapeOpsNode = $('ScrapeOps: Fetch Business Details');\n if (scrapeOpsNode && scrapeOpsNode.all) {\n const scrapeOpsData = scrapeOpsNode.all() || [];\n let currentIndex = -1;\n for (let i = 0; i < scrapeOpsData.length; i++) {\n const itemHtml = scrapeOpsData[i].json.response || scrapeOpsData[i].json.body || scrapeOpsData[i].json.data || '';\n if (itemHtml === htmlContent || \n (itemHtml && htmlContent && itemHtml.substring(0, 500) === htmlContent.substring(0, 500))) {\n currentIndex = i; break;\n }\n }\n if (currentIndex >= 0 && currentIndex < parseData.length) {\n originalData = parseData[currentIndex].json || parseData[currentIndex];\n currentMapUrl = originalData.mapUrl || '';\n } else if (inputItem.index !== undefined && inputItem.index !== null && inputItem.index < parseData.length) {\n originalData = parseData[inputItem.index].json || parseData[inputItem.index];\n currentMapUrl = originalData.mapUrl || '';\n }\n } else if (inputItem.index !== undefined && inputItem.index !== null && inputItem.index < parseData.length) {\n originalData = parseData[inputItem.index].json || parseData[inputItem.index];\n currentMapUrl = originalData.mapUrl || '';\n }\n }\n if (!originalData || Object.keys(originalData).length === 0) {\n output.push({\n json: {\n businessName: 'MATCHING_ERROR', phone: '', website: '', rating: '',\n totalReviews: '', address: '', city: '', mapUrl: extractedMapUrl || '',\n category: 'real estate agent', lgbtqFriendly: false,\n checkedAt: new Date().toISOString(),\n error: 'Could not match HTML to original business data.'\n }\n });\n continue;\n }\n }\n } catch (error) {\n output.push({\n json: {\n businessName: 'MATCHING_ERROR', phone: '', website: '', rating: '',\n totalReviews: '', address: '', city: '', mapUrl: extractedMapUrl || '',\n category: 'real estate agent', lgbtqFriendly: false,\n checkedAt: new Date().toISOString(),\n error: 'Error matching data: ' + String(error.message || 'Unknown error')\n }\n });\n continue;\n }\n \n if (!originalData.mapUrl && currentMapUrl) originalData.mapUrl = currentMapUrl;\n else if (!originalData.mapUrl && extractedMapUrl) originalData.mapUrl = extractedMapUrl;\n if (htmlContent.length > 2000000) htmlContent = htmlContent.substring(0, 2000000);\n \n function extractText(html, pattern) {\n const match = html.match(pattern);\n if (match && match[1]) return match[1].replace(/<[^>]*>/g, '').replace(/\\s+/g, ' ').trim();\n return '';\n }\n function extractAllMatches(html, pattern) {\n const matches = [];\n let match;\n while ((match = pattern.exec(html)) !== null) { if (match[1]) matches.push(match[1].trim()); }\n return matches;\n }\n \n let businessName = originalData.businessName || '';\n let phone = originalData.phone || '';\n let website = originalData.website || '';\n let rating = '';\n let totalReviews = '';\n let address = originalData.address || '';\n let city = originalData.city || '';\n let category = originalData.category || 'real estate agent';\n let lgbtqFriendly = false;\n let review1 = '', review2 = '', review3 = '';\n \n if (htmlContent && htmlContent.length > 100) {\n \n // EXTRACT PHONE\n if (!phone || phone.length < 7) {\n const m = htmlContent.match(/data-item-id=[\"']phone:tel:([^\"']+)[\"']/i);\n if (m) { phone = m[1].replace(/[^0-9+]/g, ''); if (!phone.startsWith('+')) phone = '+' + phone; }\n }\n if (!phone || phone.length < 7) {\n const telLinks = extractAllMatches(htmlContent, /href=[\"']tel:([^\"']+)[\"']/gi);\n if (telLinks.length > 0) { phone = telLinks[0].replace(/[^0-9+]/g, ''); if (!phone.startsWith('+') && phone.length >= 10) phone = '+' + phone; }\n }\n if (!phone || phone.length < 7) {\n const m = htmlContent.match(/<[^>]*data-item-id=[\"'][^\"']*phone[^\"']*[\"'][^>]*>([\\s\\S]{0,500})<\\/button>/i) ||\n htmlContent.match(/<[^>]*data-item-id=[\"'][^\"']*phone[^\"']*[\"'][^>]*>([\\s\\S]{0,500})<\\/a>/i);\n if (m) {\n const t = m[1].match(/<[^>]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/[^>]*>/i);\n if (t) { phone = t[1].replace(/[^0-9+()-]/g, '').trim(); if (phone.length < 7 || phone.length > 15) phone = originalData.phone || ''; else if (!phone.startsWith('+') && phone.length >= 10) phone = '+' + phone; }\n }\n }\n \n // EXTRACT WEBSITE\n if (!website || website.length < 5) {\n const m = htmlContent.match(/<[^>]*data-item-id=[\"']authority[\"'][^>]*href=[\"']([^\"']+)[\"'][^>]*>/i);\n if (m) { website = m[1]; if (website.startsWith('www.')) website = 'http://' + website; website = website.split('#')[0].trim(); if (website.includes('google.com') || website.includes('googleapis.com') || website.includes('gstatic.com')) website = originalData.website || ''; }\n }\n if (!website || website.length < 5) {\n const m = htmlContent.match(/aria-label=[\"']Website:[^\"']*[\"'][^>]*href=[\"']([^\"']+)[\"']/i);\n if (m) { website = m[1]; if (website.startsWith('www.')) website = 'http://' + website; website = website.split('#')[0].trim(); if (website.includes('google.com') || website.includes('googleapis.com') || website.includes('gstatic.com')) website = originalData.website || ''; }\n }\n \n // EXTRACT ADDRESS\n if (!address || address.length < 5) {\n const m = htmlContent.match(/<[^>]*data-item-id=[\"']address[\"'][^>]*>([\\s\\S]{0,500})<\\/button>/i) ||\n htmlContent.match(/<[^>]*data-item-id=[\"']address[\"'][^>]*>([\\s\\S]{0,500})<\\/div>/i);\n if (m) { const t = m[1].match(/<[^>]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/[^>]*>/i); if (t) { address = t[1].replace(/<[^>]*>/g, '').replace(/\\s+/g, ' ').trim(); if (address.length < 10 || address.length > 300) address = originalData.address || ''; } }\n }\n if (!address || address.length < 5) {\n const m = htmlContent.match(/aria-label=[\"']Address:[^\"']*[\"'][^>]*>([\\s\\S]{0,500})<\\/button>/i);\n if (m) { const t = m[1].match(/<[^>]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/[^>]*>/i); if (t) { address = t[1].replace(/<[^>]*>/g, '').replace(/\\s+/g, ' ').trim(); if (address.length < 10 || address.length > 300) address = originalData.address || ''; } }\n }\n \n // ============================================================\n // EXTRACT LGBTQ+ FRIENDLY\n // ============================================================\n const lgbtqPatterns = [\n /lgbtq[^\"'<>]{0,50}friendly/i,\n /lgbt[^\"'<>]{0,50}friendly/i,\n /identifies as lgbtq\\+?\\s*friendly/i,\n /lgbtq\\+?\\s*owned/i,\n /gay.friendly/i,\n /transgender.friendly/i,\n /pride[^\"'<>]{0,30}friendly/i\n ];\n for (const pattern of lgbtqPatterns) {\n if (pattern.test(htmlContent)) {\n lgbtqFriendly = true;\n break;\n }\n }\n \n // ============================================================\n // EXTRACT RATING\n // ============================================================\n const f7niceBlock = htmlContent.match(/F7nice[^>]*>([\\s\\S]{0,500}?)<\\/div>/i);\n if (f7niceBlock && f7niceBlock[1]) {\n const ariaHidden = f7niceBlock[1].match(/aria-hidden=(?:\\\\?\"|\\\")true(?:\\\\?\\\"|\\\")\\s*>(\\d+\\.?\\d*)<\\/span>/i);\n if (ariaHidden && ariaHidden[1]) {\n const n = parseFloat(ariaHidden[1]);\n if (n >= 1.0 && n <= 5.0) rating = ariaHidden[1];\n }\n }\n if (!rating) {\n const m = htmlContent.match(/aria-label=(?:\\\\?\"|\\\")(\\d+\\.?\\d*)\\s+stars?\\s*(?:\\\\?\"|\\\")/i);\n if (m && m[1]) { const n = parseFloat(m[1]); if (n >= 1.0 && n <= 5.0) rating = m[1]; }\n }\n if (!rating) {\n const raw = extractText(htmlContent, /]*class=[\"'][^\"']*MW4etd[^\"']*[\"'][^>]*>([^<]+)<\\/span>/i).replace(/[^0-9.]/g, '');\n const n = parseFloat(raw);\n if (raw && !isNaN(n) && n >= 1.0 && n <= 5.0) rating = raw;\n }\n const ratingNum = parseFloat(rating);\n if (!rating || isNaN(ratingNum) || ratingNum < 1.0 || ratingNum > 5.0) {\n rating = originalData.rating || '';\n }\n \n // ============================================================\n // EXTRACT TOTAL REVIEWS\n // ============================================================\n if (f7niceBlock && f7niceBlock[1]) {\n const reviewsInBlock = f7niceBlock[1].match(/aria-label=(?:\\\\?\"|\\\")([0-9,]+)\\s+reviews(?:\\\\?\"|\\\")/i);\n if (reviewsInBlock && reviewsInBlock[1]) totalReviews = reviewsInBlock[1].replace(/,/g, '');\n if (!totalReviews) {\n const parenInBlock = f7niceBlock[1].match(/\$([0-9,]+)\$/);\n if (parenInBlock && parenInBlock[1]) totalReviews = parenInBlock[1].replace(/,/g, '');\n }\n }\n if (!totalReviews) {\n const roleImg = htmlContent.match(/role=(?:\\\\?\"|\\\"|')img(?:\\\\?\"|\\\"|')[^>]*aria-label=(?:\\\\?\"|\\\")([0-9,]+)\\s+reviews(?:\\\\?\"|\\\")[\\s>]/i);\n if (roleImg && roleImg[1]) totalReviews = roleImg[1].replace(/,/g, '');\n }\n if (!totalReviews) {\n const strictMatch = htmlContent.match(/aria-label=[\"\\\\]+([0-9,]+)\\s+reviews[\"\\\\]+/i);\n if (strictMatch && strictMatch[1]) totalReviews = strictMatch[1].replace(/,/g, '');\n }\n if (!totalReviews) {\n const raw = extractText(htmlContent, /]*class=[\"'][^\"']*UY7F9[^\"']*[\"'][^>]*>([^<]+)<\\/span>/i).replace(/[^0-9]/g, '');\n if (raw && raw.length >= 1 && raw.length <= 6) totalReviews = raw;\n }\n const totalNum = parseInt(totalReviews);\n if (!totalReviews || isNaN(totalNum) || totalNum < 1 || totalNum > 999999) {\n totalReviews = '';\n }\n \n // EXTRACT BUSINESS NAME\n if (!businessName || businessName.length < 3) {\n for (const p of [\n /]*class=[\"'][^\"']*qBF1Pd[^\"']*[\"'][^>]*>([^<]+)<\\/div>/i,\n /]*>([^<]+)<\\/h1>/i,\n /]*>([^<]+)<\\/h2>/i\n ]) {\n businessName = extractText(htmlContent, p);\n if (businessName && businessName.length >= 3) break;\n }\n businessName = businessName.replace(/\\s+/g, ' ').trim();\n if (!businessName || businessName.length < 3) businessName = originalData.businessName || '';\n }\n \n // ============================================================\n // EXTRACT REVIEWS - FULL TEXT\n // ============================================================\n \n function cleanReviewText(text) {\n if (!text) return '';\n text = text.replace(/<[^>]*>/g, ' ');\n text = text.replace(/&/g, '&').replace(/</g, '<').replace(/>/g, '>').replace(/"/g, '\"').replace(/'/g, \"'\").replace(/ /g, ' ');\n text = text.replace(/\\\\u0026/g, '&').replace(/\\\\u003c/g, '<').replace(/\\\\u003e/g, '>').replace(/\\\\n/g, ' ').replace(/\\\\\"/g, '\"');\n text = text.replace(/\\bMore\\b/g, '').replace(/\\bLess\\b/g, '').replace(/\\bTranslate\\b/g, '');\n text = text.replace(/\\bLike\\b/g, '').replace(/\\bShare\\b/g, '').replace(/\\bReport\\b/g, '');\n text = text.replace(/Response from the owner[\\s\\S]{0,500}?(?=\\s{2,}|$)/gi, '');\n text = text.replace(/Owner[''s]*\\s*response[\\s\\S]{0,500}?(?=\\s{2,}|$)/gi, '');\n text = text.replace(/^\\s*[Nn][Ee][Ww]\\s+/g, '');\n text = text.replace(/[★⭐☆✩✫✬✭✮✯✰]/g, '');\n text = text.replace(/\\d+\\s*stars?/gi, '');\n text = text.replace(/\\d+\\s+(days?|weeks?|months?|years?)\\s+ago/gi, '');\n text = text.replace(/\\b(a|an)\\s+(day|week|month|year)\\s+ago\\b/gi, '');\n text = text.replace(/\\d{1,2}\\/\\d{1,2}\\/\\d{2,4}/g, '');\n text = text.replace(/\\b(January|February|March|April|May|June|July|August|September|October|November|December)\\s+\\d{4}\\b/gi, '');\n text = text.replace(/Local\\s+Guide\\s*[·•]\\s*\\d+\\s+reviews?\\s*[·•]?\\s*(\\d+\\s+photos?)?/gi, '');\n text = text.replace(/\\d+\\s+reviews?\\s*[·•]\\s*\\d+\\s+photos?/gi, '');\n text = text.replace(/[\\uE000-\\uF8FF\\uFFF0-\\uFFFF]/g, '');\n text = text.replace(/[\\r\\n\\t]+/g, ' ').replace(/\\s{2,}/g, ' ').trim();\n return text;\n }\n \n function areSimilar(a, b) {\n if (!a || !b) return false;\n const shorter = a.length < b.length ? a : b;\n const longer = a.length < b.length ? b : a;\n const checkLen = Math.min(shorter.length, 120);\n if (longer.includes(shorter.substring(0, checkLen))) return true;\n if (shorter.includes(longer.substring(0, checkLen))) return true;\n if (a.substring(0, 80) === b.substring(0, 80)) return true;\n return false;\n }\n \n let allCandidates = [];\n function addCandidate(text, source) {\n if (!text) return;\n const cleaned = cleanReviewText(text);\n if (!cleaned || cleaned.length < 40) return;\n if (cleaned.match(/^[\\d\\s\\.,]+$/)) return;\n allCandidates.push({ text: cleaned, length: cleaned.length, source });\n }\n \n // SOURCE 1: data-expanded-text (char by char)\n const expandedAttrPattern = /data-expanded-text=/gi;\n let attrMatch;\n while ((attrMatch = expandedAttrPattern.exec(htmlContent)) !== null) {\n const startPos = attrMatch.index + 'data-expanded-text='.length;\n const quoteChar = htmlContent[startPos];\n if (quoteChar !== '\"' && quoteChar !== \"'\") continue;\n let endPos = startPos + 1, fullText = '';\n while (endPos < htmlContent.length && endPos < startPos + 10000) {\n if (htmlContent[endPos] === quoteChar && htmlContent[endPos - 1] !== '\\\\') { fullText = htmlContent.substring(startPos + 1, endPos); break; }\n endPos++;\n }\n if (fullText) addCandidate(fullText, 'expanded');\n }\n \n // SOURCE 2: data-original-text (char by char)\n const originalTextPattern = /data-original-text=/gi;\n let origMatch;\n while ((origMatch = originalTextPattern.exec(htmlContent)) !== null) {\n const startPos = origMatch.index + 'data-original-text='.length;\n const quoteChar = htmlContent[startPos];\n if (quoteChar !== '\"' && quoteChar !== \"'\") continue;\n let endPos = startPos + 1, fullText = '';\n while (endPos < htmlContent.length && endPos < startPos + 10000) {\n if (htmlContent[endPos] === quoteChar && htmlContent[endPos - 1] !== '\\\\') { fullText = htmlContent.substring(startPos + 1, endPos); break; }\n endPos++;\n }\n if (fullText) addCandidate(fullText, 'original');\n }\n \n // SOURCE 3: data-review-id blocks\n const reviewBlockPositions = [];\n const reviewIdFinder = /data-review-id=\"([^\"]+)\"/gi;\n let ridMatch;\n while ((ridMatch = reviewIdFinder.exec(htmlContent)) !== null) reviewBlockPositions.push({ pos: ridMatch.index, id: ridMatch[1] });\n for (let bi = 0; bi < reviewBlockPositions.length; bi++) {\n const startPos = reviewBlockPositions[bi].pos;\n const endPos = bi + 1 < reviewBlockPositions.length ? Math.min(reviewBlockPositions[bi + 1].pos, startPos + 8000) : startPos + 8000;\n const chunk = htmlContent.substring(startPos, endPos);\n const spanCandidates = [];\n const spanPat = /]*>([\\s\\S]{40,3000}?)<\\/span>/gi;\n let spanMatch;\n while ((spanMatch = spanPat.exec(chunk)) !== null) {\n const raw = spanMatch[1];\n if ((raw.match(/<[^>]+>/g) || []).length > 12) continue;\n const cleaned = cleanReviewText(raw);\n if (cleaned && cleaned.length >= 40) spanCandidates.push(cleaned);\n }\n spanCandidates.sort((a, b) => b.length - a.length);\n if (spanCandidates.length > 0) addCandidate(spanCandidates[0], 'review_block_' + bi);\n }\n \n // SOURCE 4: wiI7pd class\n const wiPat = /]*class=\"[^\"]*wiI7pd[^\"]*\"[^>]*>([\\s\\S]{40,8000}?)<\\/span>/gi;\n let wiMatch;\n while ((wiMatch = wiPat.exec(htmlContent)) !== null) addCandidate(wiMatch[1], 'wiI7pd');\n \n // SOURCE 5: MyEned class\n const myPat = /]*class=\"[^\"]*MyEned[^\"]*\"[^>]*>([\\s\\S]{40,8000}?)<\\/span>/gi;\n let myMatch;\n while ((myMatch = myPat.exec(htmlContent)) !== null) addCandidate(myMatch[1], 'MyEned');\n \n // SOURCE 6: rsqaWe class\n const rsqaPat = /]*class=\"[^\"]*rsqaWe[^\"]*\"[^>]*>([\\s\\S]{40,8000}?)<\\/span>/gi;\n let rsqaMatch;\n while ((rsqaMatch = rsqaPat.exec(htmlContent)) !== null) addCandidate(rsqaMatch[1], 'rsqaWe');\n \n // SOURCE 7: JSON-LD\n const jsonPat = /]*type=\"application\\/ld\\+json\"[^>]*>([\\s\\S]*?)<\\/script>/gi;\n let jsonMatch;\n while ((jsonMatch = jsonPat.exec(htmlContent)) !== null) {\n try {\n const obj = JSON.parse(jsonMatch[1]);\n const walkJson = (o) => {\n if (!o || typeof o !== 'object') return;\n if (o['@type'] === 'Review' && o.reviewBody) addCandidate(o.reviewBody, 'jsonld');\n if (Array.isArray(o)) o.forEach(walkJson);\n else Object.values(o).forEach(v => { if (typeof v === 'object') walkJson(v); });\n };\n walkJson(obj);\n } catch(e) {}\n }\n \n // SOURCE 8: meta reviewBody\n const metaPat = /]*itemprop=\"reviewBody\"[^>]*content=\"([^\"]{40,})\"[^>]*>/gi;\n let metaMatch;\n while ((metaMatch = metaPat.exec(htmlContent)) !== null) addCandidate(metaMatch[1], 'meta');\n \n // DEDUPLICATE AND SELECT BEST 3\n allCandidates.sort((a, b) => b.length - a.length);\n let finalReviews = [];\n for (const candidate of allCandidates) {\n if (finalReviews.length >= 3) break;\n if (!finalReviews.some(r => areSimilar(r, candidate.text))) finalReviews.push(candidate.text);\n }\n \n // Fallback\n if (finalReviews.length === 0) {\n const broadCandidates = [];\n const broadPat = /<(?:span|div|p)[^>]*>([\\s\\S]{150,3000}?)<\\/(?:span|div|p)>/gi;\n let broadMatch;\n while ((broadMatch = broadPat.exec(htmlContent)) !== null) {\n const raw = broadMatch[1];\n if ((raw.match(/<[^>]+>/g) || []).length > 8) continue;\n const cleaned = cleanReviewText(raw);\n const reviewWords = /\\b(agent|realtor|property|home|house|listing|sale|purchase|buyer|seller|market|negotiation|closing|mortgage|neighborhood|recommend|professional|excellent|responsive|knowledgeable|helpful|trust|deal|offer|contract)\\b/i;\n if (cleaned && cleaned.length >= 150 && reviewWords.test(cleaned)) broadCandidates.push(cleaned);\n }\n broadCandidates.sort((a, b) => b.length - a.length);\n for (const text of broadCandidates) {\n if (finalReviews.length >= 3) break;\n if (!finalReviews.some(r => areSimilar(r, text))) finalReviews.push(text);\n }\n }\n \n review1 = finalReviews[0] || '';\n review2 = finalReviews[1] || '';\n review3 = finalReviews[2] || '';\n }\n \n output.push({\n json: {\n businessName: businessName || originalData.businessName || '',\n phone: phone || originalData.phone || '',\n website: website || originalData.website || '',\n rating: rating || '',\n totalReviews: totalReviews || '',\n address: address || originalData.address || '',\n city: city || originalData.city || '',\n mapUrl: originalData.mapUrl || '',\n category: category || originalData.category || 'real estate agent',\n lgbtqFriendly: lgbtqFriendly,\n checkedAt: originalData.checkedAt || new Date().toISOString(),\n review1: review1,\n review2: review2,\n review3: review3\n }\n });\n}\n\nreturn output;" }, "position": [ 1872, 320 ] }, { "id": "9c09bd5e-dceb-45b4-9986-3e5aac436713", "name": "Read Previous Entries from Sheet", "type": "integration", "params": { "options": {}, "service": "googleSheets", "operation": "read", "sheetName": { "__rl": true, "mode": "list", "value": 1703015671, "cachedResultUrl": "https://docs.google.com/spreadsheets/d/1C7OAR6d6bngkrCw-On7zoIYY0QjobLVdDaP_d7u-hKU/edit#gid=1703015671", "cachedResultName": "Real Estate Agent Finder" }, "documentId": { "__rl": true, "mode": "url", "value": "https://docs.google.com/spreadsheets/d/1C7OAR6d6bngkrCw-On7zoIYY0QjobLVdDaP_d7u-hKU/edit?gid=0#gid=0" } }, "position": [ 2096, 320 ] }, { "id": "7329e920-df75-4a7b-bc5f-e9102c78c6f5", "name": "Form: Enter City to Search", "type": "trigger", "params": { "kind": "n8n-nodes-base.formTrigger", "path": "3cc8189c-deb7-496c-9962-7ba89dd92a94", "options": {}, "formTitle": "Tell Which City", "formFields": { "values": [ { "fieldLabel": "City Name", "requiredField": true } ] } }, "position": [ 752, 320 ] }, { "id": "676cff78-9de6-4d99-9dcb-424325a590f1", "name": "ScrapeOps: Search Google Maps", "type": "integration", "params": { "url": "={{ $json.searchUrl }}", "service": "scrapeops", "operation": "run", "returnType": "htmlResponse", "__originalType": "@scrapeops/n8n-nodes-scrapeops.ScrapeOps", "advancedOptions": { "wait": "=12000", "render_js": true, "residential_proxy": false } }, "position": [ 1200, 320 ] }, { "id": "56d2d3bf-3398-42b5-bd85-f8c07ac2240f", "name": "Parse Business Listings", "type": "code", "params": { "jsCode": "const item = $input.item;\nconst config = $('Set Google Maps Configuration').item.json;\n\n// City validation: Check if city is valid\nconst cityName = config.city || config['City Name'] || '';\nif (!cityName || cityName.trim().length === 0) {\n return [{\n json: {\n error: true,\n message: 'Invalid city name or no businesses found. Please enter a valid Google Maps city.'\n }\n }];\n}\n\n// Get HTML from ScrapeOps htmlResponse - it's in $json.response\nlet htmlContent = '';\nif (item.json.response && typeof item.json.response === 'string') {\n htmlContent = item.json.response;\n} else if (typeof item.json === 'string' && item.json.length > 1000) {\n htmlContent = item.json;\n} else if (item.json.body && typeof item.json.body === 'string') {\n htmlContent = item.json.body;\n} else if (item.json.data && typeof item.json.data === 'string') {\n htmlContent = item.json.data;\n}\n\n// Limit HTML size to prevent memory issues\nif (htmlContent.length > 1000000) {\n htmlContent = htmlContent.substring(0, 1000000);\n}\n\nif (!htmlContent || htmlContent.length < 100) {\n return [{\n json: {\n businessName: 'No HTML content',\n phone: '',\n website: '',\n rating: '',\n totalReviews: '',\n address: '',\n city: config.city || '',\n mapUrl: '',\n category: config.keyword || 'real estate agent',\n checkedAt: new Date().toISOString(),\n error: 'No HTML content received'\n }\n }];\n}\n\n// Helper function to extract text from HTML tag\nfunction extractText(html, pattern) {\n const match = html.match(pattern);\n if (match && match[1]) {\n return match[1].replace(/<[^>]*>/g, '').replace(/\\s+/g, ' ').trim();\n }\n return '';\n}\n\n// Helper function to extract attribute value\nfunction extractAttribute(html, pattern) {\n const match = html.match(pattern);\n if (match && match[1]) {\n return match[1].trim();\n }\n return '';\n}\n\n// Helper function to extract all matches\nfunction extractAllMatches(html, pattern) {\n const matches = [];\n let match;\n while ((match = pattern.exec(html)) !== null) {\n if (match[1]) {\n matches.push(match[1].trim());\n }\n }\n return matches;\n}\n\nconst results = [];\nconst seenPlaceIds = new Set();\n\n// NEW APPROACH: Extract by mapUrl first (most reliable identifier)\n// Step 1: Find all unique mapUrls with their place IDs - use flexible pattern\nconst mapUrlPatterns = [\n /href=[\"']([^\"']*maps\\.google\\.com[^\"']*place[^\"']*\\/data=!4m[^\"']*)[\"']/gi,\n /href=[\"']([^\"']*maps\\.google\\.com[^\"']*place[^\"']*)[\"']/gi,\n /href=[\"']([^\"']*google\\.com\\/maps\\/place[^\"']*)[\"']/gi\n];\n\nconst mapUrls = [];\nfor (const mapUrlPattern of mapUrlPatterns) {\n let mapUrlMatch;\n while ((mapUrlMatch = mapUrlPattern.exec(htmlContent)) !== null && mapUrls.length < 100) {\n let mapUrl = mapUrlMatch[1];\n if (!mapUrl.startsWith('http')) {\n mapUrl = 'https://' + mapUrl;\n }\n \n // Extract place ID to ensure uniqueness\n const placeIdMatch = mapUrl.match(/1s([^:!\\/]+)/) || mapUrl.match(/place\\/([^\\/\\?]+)/);\n const placeId = placeIdMatch ? placeIdMatch[1] : mapUrl;\n \n if (!seenPlaceIds.has(placeId) && mapUrl.includes('place')) {\n seenPlaceIds.add(placeId);\n mapUrls.push({ mapUrl: mapUrl, placeId: placeId });\n }\n }\n if (mapUrls.length > 0) break; // Use first pattern that finds URLs\n}\n\n// If no mapUrls found, try to extract business cards from HTML structure\nif (mapUrls.length === 0) {\n // Fallback: Extract business cards using HTML structure\n const cardPatterns = [\n /]*class=[\"'][^\"']*Nv2PK[^\"']*[\"'][^>]*>([\\s\\S]{300,8000})<\\/div>/gi,\n /]*href=[\"'][^\"']*\\/maps\\/place\\/[^\"']*[\"'][^>]*>([\\s\\S]{300,8000})<\\/a>/gi,\n /]*data-result-index[^>]*>([\\s\\S]{300,8000})<\\/div>/gi\n ];\n \n const allCards = [];\n for (const pattern of cardPatterns) {\n let match;\n while ((match = pattern.exec(htmlContent)) !== null && allCards.length < 100) {\n if (match[1] && match[1].length > 200) {\n allCards.push(match[1]);\n }\n }\n if (allCards.length > 0) break;\n }\n \n // Process cards and extract mapUrls from them\n for (const cardHtml of allCards) {\n const cardMapUrlMatch = cardHtml.match(/href=[\"']([^\"']*maps\\.google\\.com[^\"']*place[^\"']*)[\"']/i) ||\n cardHtml.match(/href=[\"']([^\"']*google\\.com\\/maps\\/place[^\"']*)[\"']/i);\n if (cardMapUrlMatch && cardMapUrlMatch[1]) {\n let mapUrl = cardMapUrlMatch[1];\n if (!mapUrl.startsWith('http')) {\n mapUrl = 'https://' + mapUrl;\n }\n const placeIdMatch = mapUrl.match(/1s([^:!\\/]+)/) || mapUrl.match(/place\\/([^\\/\\?]+)/);\n const placeId = placeIdMatch ? placeIdMatch[1] : mapUrl;\n if (!seenPlaceIds.has(placeId)) {\n seenPlaceIds.add(placeId);\n mapUrls.push({ mapUrl: mapUrl, placeId: placeId, cardHtml: cardHtml });\n }\n } else {\n // Even without mapUrl, try to extract business info from card\n const businessName = extractText(cardHtml, /]*class=[\"'][^\"']*qBF1Pd[^\"']*[\"'][^>]*>([^<]+)<\\/div>/i) ||\n extractText(cardHtml, /]*>([^<]+)<\\/h3>/i) ||\n extractAttribute(cardHtml, /aria-label=[\"']([^\"']+)[\"']/i) ||\n '';\n if (businessName && businessName.trim().length > 2) {\n const cardId = 'card_' + allCards.indexOf(cardHtml);\n if (!seenPlaceIds.has(cardId)) {\n seenPlaceIds.add(cardId);\n mapUrls.push({ mapUrl: '', placeId: cardId, cardHtml: cardHtml });\n }\n }\n }\n }\n}\n\n// Step 2: For each mapUrl, find its card and extract all data\nfor (const { mapUrl: currentMapUrl, placeId, cardHtml: preExtractedCard } of mapUrls) {\n // Use pre-extracted card if available, otherwise find the card\n let cardHtml = preExtractedCard || '';\n let mapUrlIndex = -1;\n \n if (!cardHtml) {\n // Find the card that contains this mapUrl\n mapUrlIndex = htmlContent.indexOf(currentMapUrl);\n if (mapUrlIndex === -1) continue;\n \n // Extract a card context around this mapUrl (3000 chars before and after)\n const cardStart = Math.max(0, mapUrlIndex - 3000);\n const cardEnd = Math.min(htmlContent.length, mapUrlIndex + currentMapUrl.length + 3000);\n cardHtml = htmlContent.substring(cardStart, cardEnd);\n } else {\n mapUrlIndex = cardHtml.indexOf(currentMapUrl);\n if (mapUrlIndex === -1) mapUrlIndex = 0; // Use start if not found\n }\n \n // Try to find the actual card boundaries (only if not pre-extracted)\n if (!preExtractedCard && mapUrlIndex >= 0) {\n const cardStartPattern = /]*(?:class=[\"'][^\"']*(?:Nv2PK|qBF1Pd|result)[^\"']*[\"']|data-result-index)[^>]*>/i;\n \n // Find the start of the card\n const cardStart = Math.max(0, mapUrlIndex - 3000);\n let actualStart = cardStart;\n for (let i = mapUrlIndex; i >= Math.max(0, mapUrlIndex - 3000); i -= 100) {\n const checkHtml = htmlContent.substring(Math.max(0, i - 500), mapUrlIndex);\n const startMatch = checkHtml.match(cardStartPattern);\n if (startMatch) {\n actualStart = Math.max(0, i - 500) + checkHtml.lastIndexOf(startMatch[0]);\n break;\n }\n }\n \n // Find the end of the card\n const cardEnd = Math.min(htmlContent.length, mapUrlIndex + currentMapUrl.length + 3000);\n let actualEnd = cardEnd;\n let divCount = 0;\n let foundStart = false;\n for (let i = actualStart; i < Math.min(htmlContent.length, actualStart + 5000); i++) {\n if (htmlContent.substring(i, i + 4) === '') {\n divCount--;\n if (foundStart && divCount === 0) {\n actualEnd = i + 6;\n break;\n }\n }\n }\n \n cardHtml = htmlContent.substring(actualStart, actualEnd);\n mapUrlIndex = cardHtml.indexOf(currentMapUrl);\n if (mapUrlIndex === -1) mapUrlIndex = 0;\n }\n \n if (cardHtml.length < 200) continue;\n \n // STEP 1: Extract business name and address from mapUrl (most reliable)\n let businessName = '';\n let address = '';\n \n // Try to extract mapUrl from card if not provided\n if (!currentMapUrl || currentMapUrl.length === 0) {\n const cardMapUrlMatch = cardHtml.match(/href=[\"']([^\"']*maps\\.google\\.com[^\"']*place[^\"']*)[\"']/i) ||\n cardHtml.match(/href=[\"']([^\"']*google\\.com\\/maps\\/place[^\"']*)[\"']/i);\n if (cardMapUrlMatch && cardMapUrlMatch[1]) {\n currentMapUrl = cardMapUrlMatch[1];\n if (!currentMapUrl.startsWith('http')) {\n currentMapUrl = 'https://' + currentMapUrl;\n }\n mapUrlIndex = cardHtml.indexOf(currentMapUrl);\n if (mapUrlIndex === -1) mapUrlIndex = 0;\n }\n }\n \n const placeMatch = currentMapUrl ? currentMapUrl.match(/\\/place\\/([^\\/]+)/) : null;\n if (placeMatch && placeMatch[1]) {\n let placePath = decodeURIComponent(placeMatch[1]);\n placePath = placePath.replace(/\\+/g, ' ').trim();\n \n // Split by ' - ' or ' | ' - business name is usually first, address is after\n const parts = placePath.split(/\\s*[-|]\\s*/);\n if (parts.length >= 2) {\n businessName = parts[0].trim();\n // Address is everything after the first part\n address = parts.slice(1).join(', ').trim();\n } else if (parts.length === 1) {\n businessName = parts[0].trim();\n }\n \n // Clean business name - remove pipe separators and trailing location info\n businessName = businessName\n .replace(/\\s*\\|.*$/i, '')\n .trim();\n \n // Clean address - remove business name words if they appear\n if (address && businessName) {\n const nameWords = businessName.split(/\\s+/).filter(w => w.length > 4);\n let cleanedAddress = address;\n for (const word of nameWords) {\n const wordRegex = new RegExp(word.replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\$&'), 'gi');\n cleanedAddress = cleanedAddress.replace(wordRegex, '');\n }\n address = cleanedAddress.replace(/^[-|,\\s]+|[-|,\\s]+$/g, '').trim();\n \n // If address is too short or looks like business name, clear it\n if (address.length < 5 || address.toLowerCase().includes(businessName.toLowerCase().substring(0, 15))) {\n address = '';\n }\n }\n }\n \n // If no business name from mapUrl, try HTML extraction with multiple patterns\n if (!businessName || businessName.length < 3) {\n const namePatterns = [\n /]*class=[\"'][^\"']*qBF1Pd[^\"']*[\"'][^>]*>([^<]+)<\\/div>/i,\n /]*>([^<]+)<\\/h3>/i,\n /]*href=[\"'][^\"']*\\/maps\\/place[^\"']*[\"'][^>]*>([^<]+)<\\/a>/i,\n /]*class=[\"'][^\"']*qBF1Pd[^\"']*[\"'][^>]*>([^<]+)<\\/span>/i,\n /]*class=[\"'][^\"']*fontHeadlineSmall[^\"']*[\"'][^>]*>([^<]+)<\\/div>/i,\n /]*role=[\"']heading[\"'][^>]*>([^<]+)<\\/div>/i,\n /aria-label=[\"']([^\"']+)[\"']/i\n ];\n \n for (const pattern of namePatterns) {\n businessName = extractText(cardHtml, pattern) || extractAttribute(cardHtml, pattern) || '';\n if (businessName && businessName.length >= 3) break;\n }\n \n businessName = businessName.replace(/\\s+/g, ' ').trim();\n }\n \n // Filter invalid names\n if (!businessName || businessName.length < 3 || businessName.length > 150 ||\n /^(directions|save|share|more|view|see|click|here|search|map)$/i.test(businessName)) {\n continue;\n }\n \n // STEP 2: Extract rating and reviews from card\n let rating = '';\n rating = extractText(cardHtml, /]*class=[\"'][^\"']*MW4etd[^\"']*[\"'][^>]*>([^<]+)<\\/span>/i) ||\n extractText(cardHtml, /(\\d+\\.?\\d*)\\s*(?:star)/i) ||\n '';\n rating = rating.replace(/[^0-9.]/g, '').trim();\n \n let totalReviews = '';\n totalReviews = extractText(cardHtml, /]*class=[\"'][^\"']*UY7F9[^\"']*[\"'][^>]*>([^<]+)<\\/span>/i) ||\n extractText(cardHtml, /(\\d+(?:,\\d+)?)\\s*(?:review)/i) ||\n extractText(cardHtml, /\$(\\d+(?:,\\d+)?)\$/i) ||\n '';\n totalReviews = totalReviews.replace(/[^0-9,]/g, '').trim();\n \n // STEP 3: Extract address from Google Maps HTML\n // Note: mapUrl path usually only contains business name, not full address\n // So we must extract address from HTML elements\n if (!address || address.length < 5) {\n // Method 1: Extract from data-item-id=\"address\" (MOST RELIABLE)\n const addressSectionMatch = cardHtml.match(/<[^>]*data-item-id=[\"']address[\"'][^>]*>([\\s\\S]{0,500})<\\/button>/i) ||\n cardHtml.match(/<[^>]*data-item-id=[\"']address[\"'][^>]*>([\\s\\S]{0,500})<\\/div>/i);\n if (addressSectionMatch && addressSectionMatch[1]) {\n const addressTextMatch = addressSectionMatch[1].match(/<[^>]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/[^>]*>/i);\n if (addressTextMatch && addressTextMatch[1]) {\n address = addressTextMatch[1].replace(/<[^>]*>/g, '').replace(/\\s+/g, ' ').trim();\n if (address.length > 10 && address.length < 300) {\n // Found address from data-item-id\n } else {\n address = '';\n }\n }\n }\n \n // Method 2: Extract from aria-label=\"Address:\"\n if (!address || address.length < 5) {\n const addressAriaMatch = cardHtml.match(/aria-label=[\"']Address:[^\"']*[\"'][^>]*>([\\s\\S]{0,500})<\\/button>/i);\n if (addressAriaMatch && addressAriaMatch[1]) {\n const addressTextMatch = addressAriaMatch[1].match(/<[^>]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/[^>]*>/i);\n if (addressTextMatch && addressTextMatch[1]) {\n address = addressTextMatch[1].replace(/<[^>]*>/g, '').replace(/\\s+/g, ' ').trim();\n if (address.length > 10 && address.length < 300) {\n // Found address from aria-label\n } else {\n address = '';\n }\n }\n }\n }\n \n // Method 3: Fallback - extract from Io6YTe elements and filter\n if (!address || address.length < 5) {\n const addressPatterns = [\n /]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/div>/gi,\n /]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/span>/gi\n ];\n \n const allAddressCandidates = [];\n for (const pattern of addressPatterns) {\n const matches = extractAllMatches(cardHtml, pattern);\n for (const match of matches) {\n const cleanMatch = match.replace(/<[^>]*>/g, '').replace(/\\s+/g, ' ').trim();\n if (cleanMatch.length > 10 && cleanMatch.length < 300) {\n allAddressCandidates.push(cleanMatch);\n }\n }\n }\n \n // Filter and select the best address candidate\n for (const candidate of allAddressCandidates) {\n // Skip if it's a phone number\n if (candidate.match(/^\\+?[\\d\\s\\-\$\$]{10,20}$/) && \n (candidate.match(/\\+/) || candidate.match(/\$?\\d{3}\$?[\\s.-]?\\d{3}[\\s.-]?\\d{4}/))) {\n continue;\n }\n // Skip common non-address text\n if (candidate.match(/^(Open|Closed|Call|Website|Directions|Save|Share|Hours|Monday|Tuesday|Wednesday|Thursday|Friday|Saturday|Sunday|nexhealth|souldentalnyc|P2Q2|Your Maps activity|LGBTQ|women-owned)$/i)) {\n continue;\n }\n // Skip if it's just the business name\n if (businessName && candidate.toLowerCase().includes(businessName.toLowerCase().substring(0, 20))) {\n continue;\n }\n // Prefer candidates that look like addresses (contain comma, numbers, or street indicators)\n if (candidate.includes(',') || candidate.match(/\\d/) || \n candidate.match(/\\b(street|st|avenue|ave|road|rd|boulevard|blvd|drive|dr|lane|ln|way|circle|ct|court|plaza|pl|suite|ste|unit|apt|apartment|floor|fl|w|e|n|s|north|south|east|west)\\b/i)) {\n address = candidate;\n break;\n }\n }\n }\n }\n \n // STEP 4: Extract phone from Google Maps HTML\n let phone = '';\n \n // Method 1: Extract from data-item-id=\"phone:tel:\" attribute (MOST RELIABLE)\n const phoneDataIdMatch = cardHtml.match(/data-item-id=[\"']phone:tel:([^\"']+)[\"']/i);\n if (phoneDataIdMatch && phoneDataIdMatch[1]) {\n phone = phoneDataIdMatch[1].replace(/[^0-9+]/g, '');\n if (!phone.startsWith('+')) {\n phone = '+' + phone;\n }\n }\n \n // Method 2: Extract from tel: href links\n if (!phone || phone.length < 7) {\n const telLinks = extractAllMatches(cardHtml, /href=[\"']tel:([^\"']+)[\"']/gi);\n if (telLinks.length > 0) {\n let rawPhone = telLinks[0];\n phone = rawPhone.replace(/[^0-9+]/g, '');\n if (!phone.startsWith('+') && phone.length >= 10) {\n phone = '+' + phone;\n }\n }\n }\n \n // Method 3: Extract from Io6YTe element with phone pattern (near data-item-id=\"phone\")\n if (!phone || phone.length < 7) {\n // Find the phone section by looking for data-item-id containing \"phone\"\n const phoneSectionMatch = cardHtml.match(/<[^>]*data-item-id=[\"'][^\"']*phone[^\"']*[\"'][^>]*>([\\s\\S]{0,500})<\\/button>/i) ||\n cardHtml.match(/<[^>]*data-item-id=[\"'][^\"']*phone[^\"']*[\"'][^>]*>([\\s\\S]{0,500})<\\/a>/i);\n if (phoneSectionMatch && phoneSectionMatch[1]) {\n const phoneTextMatch = phoneSectionMatch[1].match(/<[^>]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/[^>]*>/i);\n if (phoneTextMatch && phoneTextMatch[1]) {\n phone = phoneTextMatch[1].replace(/[^0-9+()-]/g, '').trim();\n if (phone.length >= 7 && phone.length <= 15) {\n // Format phone - ensure + prefix if it looks international\n if (!phone.startsWith('+') && phone.length >= 10) {\n phone = '+' + phone;\n }\n } else {\n phone = '';\n }\n }\n }\n }\n \n // Method 4: Extract from visible text patterns in Io6YTe elements (fallback)\n if (!phone || phone.length < 7) {\n const phoneElements = extractAllMatches(cardHtml, /<[^>]*class=[\"'][^\"']*Io6YTe[^\"']*[\"'][^>]*>([^<]+)<\\/[^>]*>/gi);\n for (const element of phoneElements) {\n const cleanElement = element.replace(/<[^>]*>/g, '').replace(/\\s+/g, ' ').trim();\n // Check if this looks like a phone number\n if (cleanElement.match(/^\\+?[\\d\\s\\-\$\$]{10,20}$/) && \n (cleanElement.match(/\\+/) || cleanElement.match(/\$?\\d{3}\$?[\\s.-]?\\d{3}[\\s.-]?\\d{4}/))) {\n phone = cleanElement.replace(/[^0-9+()-]/g, '').trim();\n if (phone.length >= 7 && phone.length <= 15) {\n if (!phone.startsWith('+') && phone.length >= 10) {\n phone = '+' + phone;\n }\n break;\n }\n }\n }\n }\n \n // STEP 6: Extract website from Google Maps HTML\n let website = '';\n \n // Method 1: Extract from data-item-id=\"authority\" (MOST RELIABLE - Google Maps uses this)\n const websiteSectionMatch = cardHtml.match(/<[^>]*data-item-id=[\"']authority[\"'][^>]*href=[\"']([^\"']+)[\"'][^>]*>/i);\n if (websiteSectionMatch && websiteSectionMatch[1]) {\n website = websiteSectionMatch[1];\n if (website.startsWith('www.')) {\n website = 'http://' + website;\n }\n // Keep query parameters (like utm_campaign) but clean up\n website = website.split('#')[0].trim();\n }\n \n // Method 2: Extract from aria-label=\"Website:\" attribute\n if (!website) {\n const websiteAriaMatch = cardHtml.match(/aria-label=[\"']Website:[^\"']*[\"'][^>]*href=[\"']([^\"']+)[\"']/i);\n if (websiteAriaMatch && websiteAriaMatch[1]) {\n website = websiteAriaMatch[1];\n if (website.startsWith('www.')) {\n website = 'http://' + website;\n }\n website = website.split('#')[0].trim();\n }\n }\n \n // Method 3: Extract from href links near data-item-id=\"authority\"\n if (!website) {\n const authoritySectionMatch = cardHtml.match(/<[^>]*data-item-id=[\"']authority[\"'][^>]*>([\\s\\S]{0,500})<\\/a>/i);\n if (authoritySectionMatch && authoritySectionMatch[1]) {\n const hrefMatch = authoritySectionMatch[1].match(/href=[\"']([^\"']+)[\"']/i);\n if (hrefMatch && hrefMatch[1]) {\n website = hrefMatch[1];\n if (website.startsWith('www.')) {\n website = 'http://' + website;\n }\n if (!website.includes('google.com') && !website.includes('maps.google.com') &&\n !website.includes('googleapis.com') && !website.includes('gstatic.com') &&\n (website.startsWith('http://') || website.startsWith('https://'))) {\n website = website.split('#')[0].trim();\n } else {\n website = '';\n }\n }\n }\n }\n \n // Method 4: Fallback - find website links in the card (excluding Google domains)\n if (!website) {\n const cardLinks = extractAllMatches(cardHtml, /href=[\"']([^\"']+)[\"']/gi);\n for (const link of cardLinks) {\n if (link && !link.includes('google.com/maps') && !link.includes('maps.google.com') && \n !link.includes('googleapis.com') && !link.includes('gstatic.com') &&\n !link.includes('nexhealth.com') && !link.includes('zocdoc.com') &&\n !link.startsWith('tel:') && !link.startsWith('mailto:') && !link.startsWith('#') &&\n (link.startsWith('http://') || link.startsWith('https://') || link.startsWith('www.'))) {\n website = link;\n if (website.startsWith('www.')) {\n website = 'http://' + website;\n }\n website = website.split('#')[0].trim();\n break;\n }\n }\n }\n \n website = website.trim();\n \n // Only add if we have at least a business name\n if (businessName && businessName.length > 2) {\n results.push({\n json: {\n businessName: businessName,\n phone: phone || '',\n website: website || '',\n rating: rating || '',\n totalReviews: totalReviews || '',\n address: address || '',\n city: config.city || '',\n mapUrl: currentMapUrl || '',\n category: config.keyword || 'real estate agent',\n checkedAt: new Date().toISOString()\n }\n });\n }\n}\n\n// City validation: Return error if no businesses found\nif (results.length === 0) {\n return [{\n json: {\n error: true,\n message: 'Invalid city name or no businesses found. Please enter a valid Google Maps city.'\n }\n }];\n}\n\nreturn results;" }, "position": [ 1424, 320 ] }, { "id": "53119f38-dd63-42e3-9062-1b4529956cce", "name": "ScrapeOps: Fetch Business Details", "type": "integration", "params": { "url": "={{ $json.mapUrl }}", "service": "scrapeops", "operation": "run", "returnType": "htmlResponse", "__originalType": "@scrapeops/n8n-nodes-scrapeops.ScrapeOps", "advancedOptions": { "wait": "=12000", "render_js": true, "residential_proxy": false } }, "position": [ 1648, 320 ] }, { "id": "218438d3-6f88-49b7-b11d-9b2d92f74454", "name": "Compare & Deduplicate Leads", "type": "code", "params": { "jsCode": "// Helper function to normalize phone numbers (remove all non-digit characters)\nfunction normalizePhone(phone) {\n if (!phone) return '';\n let normalized = String(phone).replace(/[^0-9+]/g, '');\n if (normalized.startsWith('+')) {\n normalized = '+' + normalized.substring(1).replace(/[^0-9]/g, '');\n } else {\n normalized = normalized.replace(/[^0-9]/g, '');\n }\n return normalized;\n}\n\n// Helper function to normalize business names (trim, lowercase, remove extra spaces)\nfunction normalizeBusinessName(name) {\n if (!name) return '';\n return String(name).trim().toLowerCase().replace(/\\s+/g, ' ');\n}\n\n// Get current items from Parse Full Business Info node\nlet currentItems = [];\ntry {\n const parseNode = $('Parse Full Business Info');\n if (parseNode && parseNode.all) {\n const parseData = parseNode.all() || [];\n currentItems = parseData.map(i => i.json || i).filter(item => \n item && item.businessName && \n String(item.businessName) !== 'No businesses found' && \n String(item.businessName) !== 'No HTML content' &&\n String(item.businessName) !== 'MATCHING_ERROR'\n );\n }\n} catch (error) {\n return [{\n json: {\n businessName: 'Error accessing Parse node',\n phone: '',\n website: '',\n rating: '',\n totalReviews: '',\n address: '',\n city: '',\n mapUrl: '',\n category: 'real estate agent',\n lgbtqFriendly: false,\n status: 'old',\n checkedAt: new Date().toISOString(),\n error: String(error.message || 'Unknown error')\n }\n }];\n}\n\nif (currentItems.length === 0) {\n return [{\n json: {\n businessName: 'No businesses found',\n phone: '',\n website: '',\n rating: '',\n totalReviews: '',\n address: '',\n city: '',\n mapUrl: '',\n category: 'real estate agent',\n lgbtqFriendly: false,\n status: 'old',\n checkedAt: new Date().toISOString()\n }\n }];\n}\n\n// Get previous entries from Google Sheets\nlet previousItems = [];\ntry {\n const readNode = $('Read Previous Entries from Sheet');\n if (readNode && readNode.all) {\n const prevData = readNode.all() || [];\n previousItems = prevData.map(item => {\n const json = item.json || item;\n return {\n businessName: String(json.businessName || json.BusinessName || json['Business Name'] || ''),\n phone: String(json.phone || json.Phone || json['Phone'] || ''),\n website: String(json.website || json.Website || json['Website'] || ''),\n rating: String(json.rating || json.Rating || json['Rating'] || ''),\n totalReviews: String(json.totalReviews || json.TotalReviews || json['Total Reviews'] || ''),\n address: String(json.address || json.Address || json['Address'] || ''),\n city: String(json.city || json.City || json['City'] || ''),\n mapUrl: String(json.mapUrl || json.MapUrl || json['Map URL'] || ''),\n category: String(json.category || json.Category || json['Category'] || ''),\n lgbtqFriendly: json.lgbtqFriendly || json.LgbtqFriendly || json['Lgbtq Friendly'] || false,\n status: String(json.status || json.Status || json['Status'] || '')\n };\n }).filter(prev => prev.businessName && String(prev.businessName).length > 0);\n }\n} catch (error) {\n previousItems = [];\n}\n\n// Create a set of normalized previous business+phone combinations\nconst prevKeys = new Set();\npreviousItems.forEach(prev => {\n const normalizedName = normalizeBusinessName(prev.businessName);\n const normalizedPhone = normalizePhone(prev.phone);\n if (normalizedName && normalizedName.length > 0) {\n const key = `${normalizedName}||${normalizedPhone}`;\n prevKeys.add(key);\n if (normalizedPhone && normalizedPhone.length > 0) {\n prevKeys.add(normalizedName);\n }\n }\n});\n\n// Process current items and mark as new or old\nconst results = [];\nconst seen = new Set();\n\ncurrentItems.forEach(item => {\n const originalBusinessName = String(item.businessName || '').trim();\n const originalPhone = String(item.phone || '').trim();\n \n const normalizedName = normalizeBusinessName(originalBusinessName);\n const normalizedPhone = normalizePhone(originalPhone);\n \n const currentKey = `${normalizedName}||${normalizedPhone}`;\n \n if (seen.has(currentKey)) {\n return;\n }\n seen.add(currentKey);\n \n let isExisting = false;\n if (normalizedPhone && normalizedPhone.length > 0) {\n const exactKey = `${normalizedName}||${normalizedPhone}`;\n isExisting = prevKeys.has(exactKey);\n }\n \n if (!isExisting && normalizedName && normalizedName.length > 0) {\n isExisting = prevKeys.has(normalizedName);\n }\n \n if (!isExisting) {\n for (const prev of previousItems) {\n const prevNormalizedName = normalizeBusinessName(prev.businessName);\n const prevNormalizedPhone = normalizePhone(prev.phone);\n \n if (normalizedName === prevNormalizedName) {\n if (normalizedPhone && normalizedPhone.length > 0 && prevNormalizedPhone && prevNormalizedPhone.length > 0) {\n if (normalizedPhone === prevNormalizedPhone) {\n isExisting = true;\n break;\n }\n } else {\n isExisting = true;\n break;\n }\n }\n }\n }\n \n const status = isExisting ? 'old' : 'new';\n \n results.push({\n json: {\n businessName: originalBusinessName,\n phone: originalPhone,\n website: String(item.website || ''),\n rating: String(item.rating || ''),\n totalReviews: String(item.totalReviews || ''),\n address: String(item.address || ''),\n city: String(item.city || ''),\n mapUrl: String(item.mapUrl || ''),\n category: String(item.category || 'real estate agent'),\n lgbtqFriendly: item.lgbtqFriendly || false,\n status: status,\n checkedAt: String(item.checkedAt || new Date().toISOString()),\n review1: String(item.review1 || ''),\n review2: String(item.review2 || ''),\n review3: String(item.review3 || '')\n }\n });\n});\n\nreturn results.length > 0 ? results : [{\n json: {\n businessName: 'No data',\n phone: '',\n website: '',\n rating: '',\n totalReviews: '',\n address: '',\n city: '',\n mapUrl: '',\n category: 'real estate agent',\n lgbtqFriendly: false,\n status: 'old',\n checkedAt: new Date().toISOString()\n }\n}];" }, "position": [ 2320, 320 ] }, { "id": "c7f93646-8a4c-4afe-a613-4495eee1f218", "name": "Filter New Leads Only", "type": "if", "params": { "conditions": { "string": [ { "value1": "={{ $json.status }}", "value2": "new" } ] } }, "position": [ 2544, 320 ] }, { "id": "2b922787-6e81-4c99-8a4c-51b8cfd17fe4", "name": "Save New Leads to Sheet", "type": "integration", "params": { "columns": { "value": { "city": "={{ $json.city }}", "phone": "={{ $json.phone }}", "mapUrl": "={{ $json.mapUrl }}", "rating": "={{ $json.rating }}", "status": "={{ $json.status }}", "address": "={{ $json.address }}", "review1": "={{ $json.review1 }}", "review2": "={{ $json.review2 }}", "review3": "={{ $json.review3 }}", "website": "={{ $json.website }}", "category": "={{ $json.category }}", "checkedAt": "={{ $json.checkedAt }}", "businessName": "={{ $json.businessName }}", "totalReviews": "={{ $json.totalReviews }}", "lgbtqFriendly": "={{ $json.lgbtqFriendly }}" }, "schema": [ { "id": "businessName", "type": "string", "display": true, "required": false, "displayName": "businessName", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "phone", "type": "string", "display": true, "required": false, "displayName": "phone", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "website", "type": "string", "display": true, "required": false, "displayName": "website", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "rating", "type": "string", "display": true, "required": false, "displayName": "rating", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "totalReviews", "type": "string", "display": true, "required": false, "displayName": "totalReviews", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "address", "type": "string", "display": true, "required": false, "displayName": "address", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "city", "type": "string", "display": true, "required": false, "displayName": "city", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "category", "type": "string", "display": true, "required": false, "displayName": "category", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "mapUrl", "type": "string", "display": true, "required": false, "displayName": "mapUrl", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "checkedAt", "type": "string", "display": true, "required": false, "displayName": "checkedAt", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "lgbtqFriendly", "type": "string", "display": true, "removed": false, "required": false, "displayName": "lgbtqFriendly", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "status", "type": "string", "display": true, "required": false, "displayName": "status", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "review1", "type": "string", "display": true, "required": false, "displayName": "review1", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "review2", "type": "string", "display": true, "required": false, "displayName": "review2", "defaultMatch": false, "canBeUsedToMatch": true }, { "id": "review3", "type": "string", "display": true, "required": false, "displayName": "review3", "defaultMatch": false, "canBeUsedToMatch": true } ], "mappingMode": "defineBelow", "matchingColumns": [], "attemptToConvertTypes": false, "convertFieldsToString": false }, "options": {}, "service": "googleSheets", "operation": "append", "sheetName": { "__rl": true, "mode": "list", "value": 1703015671, "cachedResultUrl": "https://docs.google.com/spreadsheets/d/1C7OAR6d6bngkrCw-On7zoIYY0QjobLVdDaP_d7u-hKU/edit?gid=1703015671#gid=1703015671", "cachedResultName": "Real Estate Agent Finder" }, "documentId": { "__rl": true, "mode": "url", "value": "https://docs.google.com/spreadsheets/d/1C7OAR6d6bngkrCw-On7zoIYY0QjobLVdDaP_d7u-hKU/edit?gid=0#gid=0" } }, "position": [ 2768, 320 ] }, { "id": "b803208f-0943-450b-b971-3efaaa0a09a9", "name": " Send Gmail Alert", "type": "integration", "params": { "sendTo": "example@.com", "message": "={{ 'New Real Estate Agent Found\\n\\n' + 'Sheet URL: ' + 'https://docs.google.com/spreadsheets/d/1C7OAR6d6bngkrCw-On7zoIYY0QjobLVdDaP_d7u-hKU/edit?gid=0#gid=0' }}", "options": { "appendAttribution": false }, "service": "gmail", "subject": "={{ '📍 New Businesses Found in ' + $('Set Google Maps Configuration').first().json.city + ' (Real Estate Agent)' }}", "emailType": "text", "operation": "read" }, "position": [ 2992, 320 ] }, { "id": "f1f3e2d0-48a9-4439-978c-271b61fb0995", "name": "Send Slack Alert", "type": "integration", "params": { "text": "New Real Estate Agent Found: https://docs.google.com/spreadsheets/d/1C7OAR6d6bngkrCw-On7zoIYY0QjobLVdDaP_d7u-hKU/edit?gid=0#gid=0", "user": { "__rl": true, "mode": "id", "value": "U08P301T9HA" }, "select": "user", "service": "slack", "operation": "read", "otherOptions": { "includeLinkToWorkflow": true } }, "position": [ 3216, 320 ] } ] }

Scrape Google Maps realtor leads with ScrapeOps, Google Sheets, Gmail and Slack