EXT: RSS Newsfeed Export¶

Table of Contents¶

EXT: My Extension 1

Introduction 1

What does it do? 1

Screenshots 1

Users manual 2

FAQ 3

Configuration 3

Reference 3

Known problems 4

To-Do list 4

Changelog 4

Important guidelines 5

Issues with Open Office documentation for TYPO3 5

Inserting images 5

Paragraph styles 5

Linking 6

Meta data and updates 6

Introduction¶

Fostered by the popularity of the BLOG-community, syndication of content has become a major issue. No major player, be it wired.com, slashdot.com, zdnet.com, the leading German computer magazine publisher heise.de or leading newspapers like nytimes.com, usatoday.com or the German weekly magazine spiegel.de dares to NOT offer content syndication. Thousands of smaller sites profit from this syndication, since they can inlcude content from “the big guys” in a perfectly legal way, and thus boost their own content while the original syndicator gets the credit for it. A win-win situation.

Recently, a new trend started to evolve. Web-citizens, tired of spam- ridden email and of pseudo-newsletters they have never subscribed to or can't unsubscribe from, are switching to actively request information while avoiding newsletters and email-contacts. The mature and experienced web-visitors expects that webmasters provide this information in a no-nonsense, straightforward way. Especially commercial webmasters would better provide this information.

The tool used in both scenarioes is the same: a standardized content- feed in an XML-format known as RSS.

The abbreviation stands for “Rich Site Summary” or “Real Simple Syndication” or “RDF Site Summary” - regarding who is talking to you. This acronym confusion shows already that the story behind RSS is not as simple as it seems. The main problem is that there are conflicting standards, developed and promoted by two different organizations. I don't intend to go into this right now. A very good comparison and historic documentation can be found here http://www.xml.com/pub/a/2002/12/18/dive-into-xml.html

Users manual¶

Install the extension via the EM.

You can then insert a new content element type “RSS Newsfeed Export” on any page:

Enter a header title and tick the checkbox “Protect external access to this newsfeed?” if you wish it to be protected.

Protection in this case means that

a user needs to be logged in as a valid FE user in order to be presented the URL to the feed

the URL for external access to the raw XML data will contain the user's username and his md5'ed password hash

The URL a user is shown in order to view our XML feed basically looks like this:

1. non-protected feed:

index.php?id =<id_of_this_page>&type=333&feed_id=<content_element_id>

2. protected feed:

index.php?id =<id_of_this_page>&type=333&fe ed_id=<content_element_id>&user=<fe_username>&pass=<md5_hash>

An extension template needs to be created on the page the newsfeed resides on to make some adjustments that are necessary in order to get properly formatted RSS link tags etc.

An exemplary template setup should look somewhat like this:

pmk_rssnewsexport >
pmk_rssnewsexport = PAGE
pmk_rssnewsexport {
  typeNum = 334
  config.disableAllHeaderCode = 1
  config.additionalHeaders = Content-type:text/xml
  includeLibs.pmk_rssnewsexport = EXT:pmk_rssnewsexport/pi1/class.tx_pmkrssnewsexport_pi1.php
  10 = USER
  10 {
    userFunc = tx_pmkrssnewsexport_pi1->make_xml

    # title of your newsfeed, maximum 100 chars
    feedTitle = My RSS newsfeed

    # Copyright notice
    feedCopyright = 2004 My Company

    # Who is the person in charge of the newsfeed-content (email address)
    feedManagingEditor = editor@mycompany.com

    #  Who is the technical contact for problems with the newsfeed (email address)
    feedWebMaster = webmaster@mycompany.com

    #  If you don't want the individal author's names appear in the feed, you can override them here
    feedAuthorOverride = info@mycompany.com

    # main URL to include in your newsfeed, maximum 500 chars (just for user information)
    feedLink = http://www.mycompany.com

    # language in which the information is written, optional for RSS 2.0, required for 0.91
    # by default from typoscript variable "config.language" - might be overridden
    # default = en
    feedLanguage < config.language

    # description of your newsfeed, maximum 500 chars
    feedDescription = Great news about ym company

    # image included in the feed, optional for RSS 2.0, required for RSS 0.91
    # feedImage = http://www.mycompany.com/fileadmin/template/images/Logo.gif

    # prefix for direct tt_news item links
    # set up a news details page (code=SINGLE) with an extension template
    # that sets the correct backPID value to a news overview page!
    # (simulateStaticDocuments should be on, since &-signs violate RDF charspace!)
    feedItemLinkPrefix = http://your.domain.tld/newsfeeddetail.html=

    # shorten tt_news short + bodytext to this length in XML item description
    # default=100, maximum 500 in RSS 0.91
    feedItemDescLength = 200

    # select news from these tt_news sysfolders, default=all
    # newsPidList =

    # which news categories to show, default=all
    # newsCatList = 1,2,3,5

    # how many items to show in XML feed, default=all, maximum of 15 for RSS 0.91
    newsItemCount = 5

    # SQL sorting by tt_news field name, default=no ordering
    newsItemOrderBy = crdate DESC

  }

}

Modify this to your situation, save and view the page. This should be it. A reference of the possible TS params is available below.

For testing, use a modern browser that is capable of showing inline XML structures. Mozilla's DOM inspector is a great tool for that purpose.

Otherwise, you could use a desktop RSS reader like f.ex. FeedReader ( www.feedreader.com , for Windows), KNewsTicker (for Unix, included in KDE2+) or MulleNews (www.mulle- kybernetik.com/software/MulleNews/, for MacOS X) to access and examine your own newsfeed.

Frank Becker < fb@alien8.de > provides the following input, which I haven't tested for myself (yet):

Make sure "config.xhtml_cleaning=all" is not set in your TS. Otherwise Typo3 will replace each <link> tag with <link />

Configuration¶

This section refers to the possible TS parameters that can be used to configure this extension.

<link>feedItemLinkPrefix<tt_news_item_id></link>

[crit1] [crit2] [crit3] ... [DESC] | [ASC]

newsItemOrderBy = crdate uid DESC

Known problems¶

Although this extension delivers properly formatted RSS feeds, there is currently one limitation:

Links to any news items (each tt_news item has a direct link like index.php?id=123&backPID=321&tt_news=44) contain “&” signs that are not allowed in the RSS “link” field. As a matter of fact, you need a way to get RSS viewers to be able to open a direct link in a browser that points to the actual tt_news item in the feed. But you can't use the &tt_news=... GET variable in the TS param “feedItemLinkPrefix”...

A possible workaround is to use simulateStaticDocuments and set up a .htaccess for mod_rewrite in your T3 root that rewrites URLs to a news page, which is then accessible as something “newsfeeddetail.html”.

Put a line like this in your .htaccess to enable address rewriting:

RewriteEngine On
RewriteCond %{REQUEST_FILENAME} !-f
RewriteRule   ^[^/]*\.html$  index.php

So, set up a page that is being used to display single news (code=SINGLE), with an alias of “newsfeeddetail”.

Then, set up an extension template on the newsfeed details page that contains a way to transfer the backPID variable to the links in the news plugin on this page.

Put this into the TS constants of the newly created news detail page:

plugin.tt_news {
  backPid = 34
}

And, don't forget to adapt the TS value “feedItemLinkPrefix” in the RSS template to something like:

feedItemLinkPrefix = http://your.domain.tld/newsfeeddetail.html?tt_news=

This way, when someone clicks on the item in his RSS reader to open the item's link, mod_rewrite will transfer to the correct tt_news PHP page.

To-Do list¶

• Better way of linking to tt_news items
• Wrapping images in newsitems into <enclosure> elements of RSS 2.0
• Getting the feed to validate (mostly character encoding issues)
• Finding a way to generate plaintext-URL's (instead of the parameter passing) with XML-ending
• Providing functionality to embed auto-aggregate tag on pages without the plugin

Changelog¶

• 0.0.0.: initial version based on cm_rdfexport
• 0.0.1.: first non-public version for testing only
• 0.0.2.: second version with minor bugfixes, non-public
• 0.0.3.: first public beta. Should work as intended.
• 0.0.4.: <docs> section according to selected RSS version. If only teaser(short) text on tt_news exist, an empty <description> was given – fixed.
• 0.0.5.: Finnish translation :-) Version was more or less generated by error :-)
• 0.1.0.: Taken care of most of the character encoding. Auto-aggregation code is inserted into the <head> element if the feed is public.
• 0.1.1.: Page type changed from 333 to 334 in order to avoid conflicts with the extension cm_rdfexport.
• 0.1.2.: Version generated by error (I'm still a bit confused with the publishing settings in the TER...)
• 0.1.3.: Norwegian translation
• 0.1.4.: Override option for the individual news item author names in favour of a global name

EXT: RSS Newsfeed Export - 7