pear-link

pear-link

Parse Pear Links. This includes URLs with pear: & file: protocols.

npm install pear-link

Pear Link Format

The pear: protocol supports the following syntax:

pear://[<fork>.][<length>.]<keyOrAlias>[.<dhash>]<path>[?<search>][#<lochash>]

• fork is the fork id for the underlying hypercore.
• length is the length of the underlying hypercore.
• keyOrAlias is the z32 or hex encoded key or an alias for the key.
• path is zero or more path segments separated by a /.
• search is a query string of non-hierarchical data proceeded by a question mark (?).
• lochash is the fragment proceeded by a hash (#).

API

const plink = require('pear-link')

Exports a singleton with methods parse, serialize and normalize.

Supports aliases per pear-aliases.

const { protocol, pathname, search, hash, origin, drive } = plink.parse(url)

Parses the provided url string returning an object describing the URL's properties.

The returned object contains the following:

{
  protocol, // The url's protocol. Supported protocols include `pear:` & `file:`
  pathname, // The `some/path` in `pear://key/some/path`. Same as the standard URL pathname
  search,   // The query string in the URL, aka anything after a `?` including the `?`
  hash,     // The fragment part of the URL, eg `#fragment` in `pear://key/path#fragment`. Includes the `#`.
  origin: this.normalize(`${protocol}//${hostname}${pathname}`),
  drive: {
    key,    // The key for the hyperdrive. Usually the `key` part of a pear link.
    length, // The hyperdrive length defined in the URL.
    fork,   // The hyperdrive fork defined in the URL.
    hash,   // The dhash in the URL.
  }
}

const normalizedLink = plink.normalize(link)

Normalizes the link by removing trailing path separators (/).

const link = plink.serialize({ protocol, pathname, search, hash, drive })

Returns the serialized version of the object describing a URL as a string.

License

Apache-2.0