streamy-data

Utility functions to work on data collection in Stream format.

streamy-data

Inspired by gulp.js, this library helps you work on data collection in the form of Node Stream (in object mode). If you are not familiar with gulp.js, here is an introduction:

No Need To Grunt, Take A Gulp Of Fresh Air

The pros of this approach are similar to gulp.js:

  • Leverage the power of Stream.
    • No intermidiate file I/O.
    • Highly granular plugin functions.
    • Straightforward spec. (They're just object mode streams.)
  • More code, less config.

The cons are:

  • Require basic knowledge of Stream.

Install

npm install streamy-data

Usage

var streamy = require('streamy-data');
 
streamy.ptt.board('food', { limit: 2 }) // start with a readable stream which emits post URLs 
    .pipe(streamy.ptt.post()) // a transform stream which map the URL to post content 
    .on('data', function (data) {
        // do something in an explicit handler 
    })
    .on('end', function () {
        // or do something when the stream ends 
        console.log('= END =');
    });

API

###streamy.array(array)

Return a readable string which iterates through the array.

streamy.array(['A', 'B', 'C'])
    .on('data', function (data) {
        console.log(data);
    });

####array Type: Array

The source array.

###streamy.map(func)

Return an asynchronous transform stream.

someReadableStream
    .pipe(streamy.map(function (datacallback) {
        // do something 
        if (error) {
            callback(error);
            return;
        }
        // do some transformation 
        // data = data + 1 
        callback(null, data);
    }))
    .on('data', function (data) {
        console.log(data);
    });

####func Type: Function

The transform function.

###streamy.map.sync(func)

The synchronous version of streamy.map.

someReadableStream
    .pipe(streamy.map.sync(function (data) {
        // do some transformation 
        // data = data + 1 
        return data;
    }))
    .on('data', function (data) {
        console.log(data);
    });

####func Type: Function: * -> *

The transform function.

###streamy.http()

Return a transform stream which maps input to the result of its http request.

Input format:

See the first parameter of npm request.

Output format:

{
    url: (input url),
    response: (the response object),
    body: (html body string)
}

###streamy.ptt.board(name, [options])

Return a readable stream which emits post links in the PTT board, in descending order.

streamy.ptt.board('food', { limit: 2 })
    .on('data', function (data) {
        console.log(data);
    });

Output format:

{
    board: (board name string),
    author: (author id string),
    post: (post id string),
    title: (post title string),
    href: (post link url string)
}

####name

The board name.

####options.start

The start page index (high, inclusive).

####options.end

The end page index (low, exclusive).

####options.limit

The number of pages to scrape. When limit and end are both specified, both conditions are respected.

###streamy.ptt.post(options)

Return a transform stream which maps board-post pair to post content.

Input format: (compatible with streamy.ptt.board() output)

{
    board: (board name string),
    post: (post id string)
}

Output format:

{
    board: (board name string),
    post: (post id string),
    title: (post title string),
    raw: (the html body string), // if the raw flag is true 
    meta: [
        { tag: '作者', value: ... },
        { tag: '站內', value: ... },
        { tag: '標題', value: ... },
        { tag: '時間', value: ... }
    ],
    content: [
        (content),
        { type: comment, subtype: (1/2/3), author: ..., content: ... },
        { type: comment, subtype: (1/2/3), author: ..., content: ... }
        ...
    ]
}
streamy.ptt.board('food', { limit: 2 })
    .pipe(streamy.ptt.post())
    .on('data', function (data) {
        console.log(data);
    });

####options.raw Type: boolean, default: false

The raw field is included in the output if the value true.

See also