kilabit.info
| Ask Me Anything | Build | GitHub | Mastodon | SourceHut

I just add a new section in this site called 'Home - Recent Activities’, as part of WUI [1] project, which showed my recent activities on the Internet; a site I have been submitted (read and like) and what changes has been made to all of my projects.

Before delving more into this journal, I should state the goal of this task:

  • parsing feed content (using JavaScript),

  • sort all feed entry by date,

  • and display it in html page.

Now, you may wonder why it’s called "poor man’s feed parser" because I tackled all the code in reverse order, instead of read the feed specification first, I download each of the feed and looks for the pattern.

Disclaimer: To all experienced Web Developer out there, please excuse my poor solution. As far I as hate web development, this is the best way I can do :)

Parsing Feed Content

There is two feed that I need to parsing, github public feed [2] and reddit submission [3].

Parsing Atom Feed Content

First is github public feed. The feed content is look like this,

<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xmlns:media="http://search.yahoo.com/mrss/" xml:lang="en-US">
  <id>tag:github.com,2008:/shuLhan</id>
  <link type="text/html" rel="alternate" href="https://github.com/shuLhan"/>
  <link type="application/atom+xml" rel="self" href="https://github.com/shuLhan.atom"/>
  <title>shuLhan's Activity</title>
  <updated>2011-01-07T21:58:02-08:00</updated>
  <entry>
    <id>tag:github.com,2008:PushEvent/1070663767</id>
    <published>2011-01-07T21:58:02-08:00</published>
    <updated>2011-01-07T21:58:02-08:00</updated>
    <link type="text/html" rel="alternate" href="https://github.com/shuLhan/kilabit.info/compare/09a6024d91...284d3800b0"/>
    <title>shuLhan pushed to master at shuLhan/kilabit.info</title>
    <author>
      <name>shuLhan</name>
          <email>ms@kilabit.info</email>
          <uri>https://github.com/shuLhan</uri>
    </author>
    <media:thumbnail height="30" width="30" url="https://secure.gravatar.com/avatar/80e039afe2e0ecb9bbe7e78fef270ede?s=30&d=https%3A%2F%2Fgithub.com%2Fimages%2Fgravatars%2Fgravatar-140.png"/>
    <content type="html">
	...
  </entry>
  <entry>
    <id>tag:github.com,2008:PushEvent/1069771445</id>
    <published>2011-01-07T10:20:19-08:00</published>
    <updated>2011-01-07T10:20:19-08:00</updated>
    <link type="text/html" rel="alternate" href="https://github.com/shuLhan/kilabit.info/compare/9fb7c7418a...09a6024d91"/>
    <title>shuLhan pushed to master at shuLhan/kilabit.info</title>
    <author>
      <name>shuLhan</name>
          <email>ms@kilabit.info</email>
          <uri>https://github.com/shuLhan</uri>
    </author>
    <media:thumbnail height="30" width="30" url="https://secure.gravatar.com/avatar/80e039afe2e0ecb9bbe7e78fef270ede?s=30&d=https%3A%2F%2Fgithub.com%2Fimages%2Fgravatars%2Fgravatar-140.png"/>
    <content type="html">
	...
    </content>
  </entry>
	...

(Some of feed content has been cut and replaced by "…​")

Just by quick look you can see that each feed content is enclosed by entry tag. I don’t care the content of entry right now, because that will be handled in the third task.

JavaScript function to parsing Atom feed,

function feed_atom_parsing()
{
	var entries = _xml.getElementsByTagName('entry');

	for (var i = 0; i < entries.length; i++) {
		var entry = {};

		entry.type = 'atom';

		for (var c = 0; c < entries[i].childNodes.length; c++) {
			var child = entries[i].childNodes[c];

			if (child.nodeType != 1) {
				continue;
			}

			switch (child.nodeName) {
			case 'link':
				entry[child.nodeName] = child.getAttribute('href');
				break;
			case 'published':
				entry['my_date'] = new Date(child.textContent);
				/* no break */
			default:
				entry[child.nodeName] = child.textContent;
			}
		}

		_feeds[_feeds.length] = entry;
	}

Quick explanation,

_xml is XMLHttpRequest responseXML value.

var entry = {}, this is where we will save the feed entry, as an object.

child.nodeType != 1, skip child content which content is "#text".

entry['my_date’], we need feed date in Date object for sorting the feed later.

_feeds[_feeds.length] = entry;, save the feed entry in an array.

Parsing RSS Feed Content

Second feed is reddit submission. To bad, reddit is not using Atom but RSS version 2.0, so we need another function to parsing it. A snippet example of RSS feed content,

<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:media="http://search.yahoo.com/mrss/" version="2.0">
  <channel>
    <title>reddit: the voice of the internet -- news before it
happens</title>
    <link>http://www.reddit.com/</link>
    <description/>
    <image>
      <url>http://static.reddit.com/reddit.com.header.png</url>
      <title>reddit: the voice of the internet -- news before it
happens</title>
      <link>http://www.reddit.com/</link>
    </image>
    <item>
      <title>Turning Info Tech into Clean Tech</title>
      <link>http://www.reddit.com/r/technology/comments/evv4d/turning_info_tech_into_clean_tech/</link>
      <guid
isPermaLink="true">http://www.reddit.com/r/technology/comments/evv4d/turning_info_tech_into_clean_tech/</guid>
      <pubDate>Mon, 03 Jan 2011 23:38:41 -0700</pubDate>
      <description>
	...
	</description>
    </item>
    <item>
      <title>Archive Testing: Comparison On Close-to-realistic Tasks.</title>
      <link>http://www.reddit.com/r/linux/comments/esk1g/archive_testing_comparison_on_closetorealistic/</link>
      <guid
isPermaLink="true">http://www.reddit.com/r/linux/comments/esk1g/archive_testing_comparison_on_closetorealistic/</guid>
      <pubDate>Tue, 28 Dec 2010 07:36:02 -0700</pubDate>
      <description>
	...
</description>
    </item>
	...

(Some of feed content has been cut and replaced by "…​")

And by quick look the RSS feed content is enclosed by item tag.

A function to parsing RSS (version 2.0),

function feed_rss20_parsing()
{
	var entries = _xml.getElementsByTagName('item');

	for (var i = 0; i < entries.length; i++) {
		var entry = {};

		entry.type = 'rss20';

		for (var c = 0; c < entries[i].childNodes.length; c++) {
			var child = entries[i].childNodes[c];

			if (child.nodeType != 1) {
				continue;
			}
			if (child.nodeName == 'pubDate') {
				entry['my_date'] = feed_rss_pubdate_to_date(child.textContent);
			}

			entry[child.nodeName] = child.textContent;
		}

		_feeds[_feeds.length] = entry;
	}
}

The code is almost like Atom feed parser, except for additional function feed_rss_pubdate_to_date(), because of format of RSS date is not easily converted to JavaScript Date object, we need a function to convert it.

RSS publication date (pubDate) is like this:

Mon, 03 Jan 2011 23:38:41 -0700

and I need to parsing it and rewrite it back just like Atom published date value format,

YYYY-MM-DDTHH:MM:SS<GMT-format>

So, here is the function to convert RSS pubDate to Atom date format,

function feed_rss_pubdate_to_date(pubDate)
{
	var mm_to_m	= {
			  Jan:'01', Feb:'02', Mar:'03', Apr:'04', May:'05'
			, Jun:'06', Jul:'07', Aug:'08', Sep:'09', Oct:'10'
			, Nov:'11', Dec:'12'
			};
	var arr_date	= pubDate.split(' ');
	var d		= arr_date[1];
	var mm		= arr_date[2];
	var y		= arr_date[3];
	var time	= arr_date[4];
	var gmt_l	= arr_date[5].substring(0,3);
	var gmt_r	= arr_date[5].substring(3,5);
	var sdate	= '';

	sdate = y +"-"+ mm_to_m[mm] +"-"+ d +"T"+ time
		+ gmt_l +":"+ gmt_r;

	return new Date(sdate);
}

Sort All Feed Entry by Date

After I have got all the feed contents in array (_feeds), I need to sort all the feeds by date (descending). This is how I do it,

function feed_sort(a, b)
{
	return b.my_date - a.my_date;
}

...
	_feeds.sort(feed_sort);
...

Display It in HTML Page

The feed is ready and sorted and now I need create a HTML output for display all feed.

function feed_create_output()
{
	for (var i = 0; i < _feeds.length; i++) {
		switch(_feeds[i].type) {
		case 'atom':
			_act	+= "<div class='activity'>"
				+ "<div class='activity_header'>"
				+ "<a href='"+ _feeds[i].link +"'>"
				+ _feeds[i].title
				+ "</a>"
				+ "<span class='activity_date'>"
				+ feed_convert_date(_feeds[i].my_date)
				+ "</span>"
				+ "</div>"
				+ _feeds[i].content
				+ "</div>";
			break;
		case 'rss20':
			_act	+= "<div class='activity'>"
				+ "<div class='activity_header'>"
				+ "<a href='"+ _feeds[i].link +"'>"
				+ _feeds[i].title
				+ "</a>"
				+ "<span class='activity_date'>"
				+ feed_convert_date(_feeds[i].my_date)
				+ "</span>"
				+ "</div>"
				+ _feeds[i].description
				+ "</div>";
			break;
		}
	}
}

Later, I can insert it to any element on web page,

...
	var e;
	e = document.getElementById('my_activity');
	e.innerHTML = _act;
...

Deployment

Since browser does not allow JavaScript request across domain, I need to pull all the feed from the web server and save it in www public folder so it can be accessed by JavaScript.

By using cron jobs I can make it automatically refresh once a hour.

#!/bin/bash

curl -k -o ~/www/my_github_feed.atom https://github.com/shuLhan.atom
curl -o ~/www/my_reddit_feed.rss http://www.reddit.com/user/rv77ax/submitted/.rss

In user browser, JavaScript just need to request only from original server using the pull output file.

function wui_get_feed(url)
{
	var type;

	wui_get(url);

	type = _xml.firstChild;

	switch (type.nodeName) {
	case 'feed':
		feed_atom_parsing();
		break;
	case 'rss':
		var version = type.getAttribute('version');
		switch (version) {
		case '2.0':
			feed_rss20_parsing();
			break;
		}
		break;
	}
}

function feed_init()
{
	wui_get_feed("/my_github_feed.atom");
	wui_get_feed("/my_reddit_feed.rss");

	_feeds.sort(feed_sort);

	feed_create_output();
}