npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

streaming-data

v0.0.3

Published

This is a library for reading text from a stream and processing it into data in various ways. Although I'm sure people will find other uses for it, my original intent was to allow command-line tools (primarily my [slack-send][slack-send] tool) to read a

Downloads

2

Readme

streaming-data

This is a library for reading text from a stream and processing it into data in various ways. Although I'm sure people will find other uses for it, my original intent was to allow command-line tools (primarily my slack-send tool) to read a stream of data and process it in useful ways before using it.

Usage

To use it, you simply import it into your program, pipe a stream of text into it, and then pipe it's output into a suitable stream reader.

// To process
const createStream = require( 'streaming-data' );
const streamer = createStream();
streamer.on( 'data', data => console.log( data ) );
stream.on( 'close', () => console.log( 'CLOSED' ) );

Input Format

The original purpose of this library was to allow for a standard means of providing structured data to a logging utility that simply reads from stdin and then sends the data that it reads off to a logging aggregator. When I started also sending these logs to Slack, I wanted a way for the tool producing the logs to produce attachments and other nicely formatted output that the slack-send tool can consume.

These are the types of data you can produce, and the rules that you must follow in order for this library to correctly recognize your data.

YAML

YAML is supported primarily in a block mode. You can start a YAML block with a line that consists of nothing except for three dashes. To end YAML mode, emit a line that contains only three periods:

---
- this is yaml
...

The YAML parser also supports a continuation mode. If you start a new YAML block without ending the previous one, the previous one will be processed and emitted and then a new block begins.

---
this: is a yaml object
---
- this is a yaml array
...

It also supports a single-line mode. If given a line to process while not currently in a YAML block, if the line begins with --- (three dashes followed by a space), then the parser will attempt to parse that line as a single-line YAML document. If the parsing fails then it will ignore it and allow the handler chain to continue (so that it will get emitted by another handler if it looked like YAML but wasn't).

--- [ a, single, line, yaml, array ]
--- { this: "is a single-line yaml object" }

JSON

JSON is handled in much the same way as YAML.

It has a block mode that is started by a line containing only a { or [. That block mode is ended by a line containing only the matching } or ].

It also has a single-line mode. It will attempt to parse a line as a single-line if it either begins with { and ends with } or it begins with [ and ends with ]. If the parsing fails it will assume that it wasn't actually JSON and let the handler chain continue.

The JSON handler uses the json5 parser, which parses the annoying JSON that you know and love, but also lets you write "JSON for Humans", which removes some of the annoying traits of hand-writing JSON. Some of the differences are:

  • You can include extra whitespace anywhere you want
  • You can use comments (both single line, starting with // and multi-line, surrounded by /* and */).
  • Object keys don't need to be quoted if they don't contain whitespace or other odd characters.
  • Objects and arrays can have trailing commas
  • Strings can be single or double quoted
  • Strings can include character escapes
  • Strings can span multiple lines by escaping the newline
  • Numbers can have leading or trailing decimal points
  • You can include Infinity, -Infinity and NaN as numbers
  • Numbers can be hexadecimal
  • Numbers can start with an explicit plus sign

See https://www.npmjs.com/package/json5 for more details.

Verbatim

The verbatim handler is very useful when you are piping in text where you know that it could contain things that would otherwise get processed by the handler rules above. Verbatim mode is entered by a line that starts with a leader. The default leader is StreamingData.Verbatim<<. When the handler sees a line that starts with this string, it enters verbatim mode and records whatever was on the rest of the line as the trailer for this mode. It will then stay in verbatim mode until it sees that trailer text. This ensures that you can pick a trailer that will never appear in your output.

So, for example, if you had a shell script that was running a bunch of commands and you wanted the output of those commands to be processed by slack-send, you might do something like this:

#!/bin/bash
(
  prepare-build
  install-prerequisites
  run-unit-tests
  run-acceptance-tests
  bump-version-numbers
  publish-packages
) | slack-send --channel=builds --text="Beginning build" --stream

However, if the bump-version-numbers script has output that looks like JSON, but you don't want it to get parsed as JSON, you can make sure that it stays as plain text by doing this:

#!/bin/bash
(
  prepare-build
  install-prerequisites
  run-unit-tests
  run-acceptance-tests
  echo 'StreamingData.Verbatim\<\<END_OF_VERSIONING'
  bump-version-numbers
  echo END_OF_VERSIONING
  publish-packages
) | slack-send --channel=builds --text="Beginning build" --stream

Ignore

The ignore handler is useful when you are piping in text that you don't want to be processed at all. It's very similar to the verbatim handler, but instead of producing a block of text, it produces nothing.

The default leader is for the ignore handler is StreamingData.Ignore<<.

Configuration

By default when you create a new stream it is configured with some default handlers. If you want to configure it with a different set of handlers you can pass an object with a handlers property:

const stream = createStream( { handlers : [ 'JSON', 'YAML' ] } );

Since configuring just the handlers is a very common occurrence, there is also a shortcut for that. If you pass an array instead of an object it will be assumed that you are just passing a list of handlers:

// produces exactly the same result as the one above
const stream = createStream( [ 'JSON', 'YAML' ] );

To create a handler, the configuration you provide for it must indicate the type of handler that it is. If you provide an object with a class property, then that property indicates the type. If you provide an object with a name but not a class then the name will also be assumed to be the class. If you provide a class but not a name then the name will be set to the class as well. If you provide just a string then it will be used as both.

// These all produce identical results:
streamer.addHandler( 'JSON' );
streamer.addHandler( { class : 'json' } );
streamer.addHandler( { name : 'json' } );
streamer.addHandler( { name : 'json', class : 'JSON' } );
// Note that the case difference don't matter, because class and name
// are both internally converted to lowercase

The default handlers (and their default configurations) are:

JSON:
  single: true
  multi: true
YAML:
  single: true
  multi: true
Verbatim:
  leader: "StreamingData.Verbatim<<"
  keep_markers: false
Ignore:
  leader: "StreamingData.Ignore<<"