web dev//crawling//parsing

Turning raw content (HTML, JSON, XML) into a structure a program can interpret. From "a string of characters" to "this is a title, this is a link, this is a paragraph".

Turning raw content (HTML, JSON, XML) into a structure a program can interpret. From "a string of characters" to "this is a title, this is a link, this is a paragraph".

The parser does not decide what matters; it only makes the structure available. What comes out is a tree (for HTML, the same kind of tree a browser builds as its DOM) that later stages walk.

It follows fetching and precedes both scraping, which picks specific data out of the tree, and indexing, which files the whole document. Links found while parsing feed discovery again.

Structured formats make parsing cheap and unambiguous, which is the whole argument for Structured Data and for APIs over pages.