JSON Web Scraper
Declarative extraction of structured JSON from HTML using Cheerio.
- Node.js
- Cheerio
- JSON
Turn HTML into structured records
The scraper uses a declarative schema to describe which elements and attributes to extract from a document.
Schema-driven extraction across directories
I built tooling around HTML retrieval, Cheerio selection and structured JSON output. Targeted extraction scenarios include business directory listings and geographic catalog pages such as Google Maps listings and Just Dial vendor profiles.
Parse the document with Cheerio
Cheerio provides DOM querying without executing the page in a full browser. This fits content already present in HTML; browser-rendered content needs a separate strategy.
Keep extraction rules explicit
Selectors and transforms belong in a schema that can be reviewed alongside the expected result. The portfolio does not claim a throughput benchmark.
Extraction targets and schema validation
Declarative extraction schemas are tested against directory structures like Google Maps and Just Dial, parsing tabular phone, address, and rating records directly into validated JSON without launching headless browsers.