libnpmsearch
v9.0.0
Published
Programmatic API for searching in npm and compatible registries.
Downloads
2,808,200
Readme
libnpmsearch
libnpmsearch
is a Node.js library for
programmatically accessing the npm search endpoint. It does not support
legacy search through /-/all
.
Table of Contents
Example
const search = require('libnpmsearch')
console.log(await search('libnpm'))
=>
[
{
name: 'libnpm',
description: 'programmatic npm API',
...etc
},
{
name: 'libnpmsearch',
description: 'Programmatic API for searching in npm and compatible registries',
...etc
},
...more
]
Install
$ npm install libnpmsearch
API
opts
for libnpmsearch
commands
The following opts are used directly by libnpmsearch
itself:
opts.limit
- Number of results to limit the query to. Default: 20opts.from
- Offset number for results. Used withopts.limit
for pagination. Default: 0opts.detailed
- If true, returns an object withpackage
,score
, andsearchScore
fields, withpackage
being what would usually be returned, and the other two containing details about how that package scored. Useful for UIs. Default: falseopts.sortBy
- Used as a shorthand to setopts.quality
,opts.maintenance
, andopts.popularity
with values that prioritize each one. Should be one of'optimal'
,'quality'
,'maintenance'
, or'popularity'
. Default:'optimal'
opts.maintenance
- Decimal number between0
and1
that defines the weight ofmaintenance
metrics when scoring and sorting packages. Default:0.65
(same asopts.sortBy: 'optimal'
)opts.popularity
- Decimal number between0
and1
that defines the weight ofpopularity
metrics when scoring and sorting packages. Default:0.98
(same asopts.sortBy: 'optimal'
)opts.quality
- Decimal number between0
and1
that defines the weight ofquality
metrics when scoring and sorting packages. Default:0.5
(same asopts.sortBy: 'optimal'
)
libnpmsearch
uses npm-registry-fetch
.
Most options are passed through directly to that library, so please refer to
its own opts
documentation
for options that can be passed in.
A couple of options of note for those in a hurry:
opts.token
- can be passed in and will be used as the authentication token for the registry. For other ways to pass in auth details, see the n-r-f docs.
> search(query, [opts]) -> Promise
query
must be either a String or an Array of search terms.
If opts.limit
is provided, it will be sent to the API to constrain the number
of returned results. You may receive more, or fewer results, at the endpoint's
discretion.
The returned Promise resolved to an Array of search results with the following format:
{
name: String,
version: SemverString,
description: String || null,
maintainers: [
{
username: String,
email: String
},
...etc
] || null,
keywords: [String] || null,
date: Date || null
}
If opts.limit
is provided, it will be sent to the API to constrain the number
of returned results. You may receive more, or fewer results, at the endpoint's
discretion.
For streamed results, see search.stream
.
Example
await search('libnpm')
=>
[
{
name: 'libnpm',
description: 'programmatic npm API',
...etc
},
{
name: 'libnpmsearch',
description: 'Programmatic API for searching in npm and compatible registries',
...etc
},
...more
]
> search.stream(query, [opts]) -> Stream
query
must be either a String or an Array of search terms.
If opts.limit
is provided, it will be sent to the API to constrain the number
of returned results. You may receive more, or fewer results, at the endpoint's
discretion.
The returned Stream emits one entry per search result, with each entry having the following format:
{
name: String,
version: SemverString,
description: String || null,
maintainers: [
{
username: String,
email: String
},
...etc
] || null,
keywords: [String] || null,
date: Date || null
}
For getting results in one chunk, see search
.
Example
search.stream('libnpm').on('data', console.log)
=>
// entry 1
{
name: 'libnpm',
description: 'programmatic npm API',
...etc
}
// entry 2
{
name: 'libnpmsearch',
description: 'Programmatic API for searching in npm and compatible registries',
...etc
}
// etc