Fix review 14/05/2020

lib/tasks/crawler.rake

@@ -14,14 +14,15 @@ namespace :crawler do
       html_jobs = Nokogiri::HTML.parse(URI.open("https://careerbuilder.vn/viec-lam/tat-ca-viec-lam-trang-#{page}-vi.html"))
       html_jobs = Nokogiri::HTML.parse(URI.open("https://careerbuilder.vn/viec-lam/tat-ca-viec-lam-trang-#{page}-vi.html"))
       # Loop item
       # Loop item
       html_jobs.css(".jobs-side-list .job-item").each do |item|
       html_jobs.css(".jobs-side-list .job-item").each do |item|
-        html_job_detail = Nokogiri::HTML.parse(URI.open(URI.encode(item.css(".figure .figcaption .title .job_link @href").text)))
+        url = item.css(".figure .figcaption .title .job_link @href").text
+        html_job_detail = Nokogiri::HTML.parse(URI.open(URI.encode(url)))
         if html_job_detail.at_css(".search-result-list-detail .tabs div#tab-1").nil?
         if html_job_detail.at_css(".search-result-list-detail .tabs div#tab-1").nil?
-          logger.warn "Another template #{item.css(".figure .figcaption .title .job_link @href").text}"
+          logger.warn "Another template #{url}"
           next
           next
         end
         end
         # Set salary, min-salary, max-salary
         # Set salary, min-salary, max-salary
         if item.at_css(".figure .figcaption .caption .salary").text.include? "USD"
         if item.at_css(".figure .figcaption .caption .salary").text.include? "USD"
-          logger.warn "Another template #{item.css(".figure .figcaption .title .job_link @href").text}"
+          logger.warn "Another template #{url}"
           next
           next
         end
         end
         salary = item.at_css(".figure .figcaption .caption .salary").text.gsub("$ ","")
         salary = item.at_css(".figure .figcaption .caption .salary").text.gsub("$ ","")
@@ -63,11 +64,12 @@ namespace :crawler do
         # Company attributes
         # Company attributes
         html_company_detail = Nokogiri::HTML.parse(open(URI.encode(item.css(".figure .image a @href").text)))
         html_company_detail = Nokogiri::HTML.parse(open(URI.encode(item.css(".figure .image a @href").text)))
         next if html_company_detail.at_css(".jobsby-company").nil?
         next if html_company_detail.at_css(".jobsby-company").nil?
+        company_css = ".jobsby-company .company-introduction .company-info .info "
         company_attributes = {
         company_attributes = {
-          title: html_company_detail.at_css(".jobsby-company .company-introduction .company-info .info .content .name").text,
-          address: html_company_detail.css(".jobsby-company .company-introduction .company-info .info .content p")[1].text,
-          logo: html_company_detail.at_css(".jobsby-company .company-introduction .company-info .info .img @src").text,
-          description: html_company_detail.at_css(".jobsby-company .company-introduction .company-info .info .content ul").inner_html.squish
+          title: html_company_detail.at_css(company_css + ".content .name").text,
+          address: html_company_detail.css(company_css + ".content p")[1].text,
+          logo: html_company_detail.at_css(company_css + ".img @src").text,
+          description: html_company_detail.at_css(company_css + ".content ul").inner_html.squish
         }
         }
         # Defind cities array
         # Defind cities array
         cities = item.css(".figure .figcaption .caption .location ul li").map do |city|
         cities = item.css(".figure .figcaption .caption .location ul li").map do |city|
@@ -80,7 +82,7 @@ namespace :crawler do
 
 
         result = CrawlerService.imports(job_attributes, company_attributes, cities, industries)
         result = CrawlerService.imports(job_attributes, company_attributes, cities, industries)
         
         
-        logger.info "Crawl success url : #{item.css(".figure .figcaption .title .job_link @href").text}"
+        logger.info "Crawl success url : #{url}"
         rescue Exception => e
         rescue Exception => e
           logger.error e
           logger.error e
           next
           next