Use this field to create different extractors for different areas of your source content.įixed - to add a fixed value to all documents with this attribute.Įxpressions - to enter a JSONPath expression in Selectors. In some sources, you can also use JavaScript. For most sources, you can use a regular expression or a glob expression. The pattern that defines the URLs to which this extractor and its rules apply. JSONPath is a query language for JSON that is similar to XPath. This is a common use case when you use an API crawler.
JSONPath - use this document extractor when you need to crawl JSON data to extract attribute values. For example, if you want to replace parts of a URL that you get from the page metadata before storing it as an attribute, you must use a JS extractor. JS - use this document extractor when you want to use a JavaScript function that returns attribute values. For example, titles and images are sometimes within CSS tags. These tags appear in the page source.ĬSS - use this document extractor when you want to use CSS queries to extract data from CSS tags. These tags do not appear in the page source. XPath - use this document extractor when you want to use an XPath query to extract data from HTML tags. One way to handle this is to create one document extractor for /personal, one for /loans, and one for /commercial. For example, you have a banking website with three sections - personal, loans, and commercial, each with a different URL and attribute pattern. Usually, you would create one document extractor for each content section that needs different URL matching rules and attribute tagging rules. You can create more than one document extractor for a source. This is because you did not configure the crawler to extract the average_rating attribute. In the previous example, you won't be able to show users an average_rating for each content piece even if your original content has field this and you have configured an average_rating attribute. Unless you extract an attribute, you can't use that attribute to create a search experience. Show a results page with a title, description, and image for each item.Īllow users to filter by content_type equals, say, news, or blog.Īllow users to arrange results in ascending or descending of the title.
0 kommentar(er)
