lformella / rss-extender Goto Github PK

Readability.php analyses any html site and tries to extract relevant content. It is ported from Mozilla's Readability.js, which is used by firefox "reader view".

I noticed this could also be used for RSS extension:

• As a fallback if Regex rules fail / return empty text
• For feeds with no configured regex (yet) to get maybe not 100% perfect, but basic extension
• For news aggregators, which link to lots of different sites where no one can create regex for

Of course everything should optional.

In fact, I already implemented everything above for personal use. Quick and dirty - but if someone wants to take a look - but don't loot too close ;).

Any interest in those patches?

I really just wrote them for personal use - so no problem at all if you say no. If you are interested, I would clean up my mess, refactor the patches and create a PR. Otherwise just close this issue.

I use the rss-extender feed for pro-linux. Just the default pro-linux.php.

This time rss-extender delivers a empty post → http://www.pro-linux.de/artikel/2/1596/pfiffige-datensicherung-mit-storebackup.html

Hi,

first of all. Thank you for your great work, I rely on rss extender every day and it works great so far. Unfortunately, the img tags from heise posts (especiall from mac&i) are not closed anymore and thus dont work.
Can you reproduce the issue?

Thanks in advance.

Andy

Spiegel still serves their html in encoding="ISO-8859-1". This seems to be an issue with the umlauts when extending the spiegel feed.
Can anyone assist here to fix this issue?

Hi Lars,

I really like RSS Extender a lot and use it everyday, however, since two weeks the heise feeds are no longer extracting the images correctly. For example in the post:

http://www.heise.de/newsticker/meldung/Windows-10-Upgrade-Installation-auch-von-Windows-7-aus-moeglich-2438050.html/from/atom10?wt_mc=rss.ho.beitrag.atom

RSS Extender extracts the image path as:
http://www.heise.de/newsticker/meldung/Windows-10-Upgrade-Installation-auch-von-Windows-7-aus-moeglich-2438050.html?view=zoom;zoom=1

while it should be:
http://2.f.ix.de/imgs/18/1/3/6/9/8/1/0/upgrader-a8c02cd6e11f7516.png

I would deeply appreciate your help

Andy

can you tel me exactly this script can do what? and how it works?

Hi.

Under which license is rss-extender published?

BTW: Thanks a lot for this great piece of code.

low prio warning

Hello,
i got: "Your Rss-Extender rules returned an empty string for link"

maybe a little tutorial to fix that pages for my own would be nice :)

Hi Lars,

I like RSS-Extender a lot and still use it every day. However, I have some problems producing a useable feed for iphoneblog.de. My config looks like this:

<?php
$config['url']                  = "http://www.iphoneblog.de/feed/"; 
$config['base_url']             = "http://www.iphoneblog.de/"; 
$config['content']              = array("<div class=\"beitragstext\">(.*)<div class=\"beitragsmedia\">", 1);
$config['search ']              = array("#<ads>.*<\/ads>#Uis",
                                    "&nbsp;",
                                    "<div class=\"beitragsmedia\">");
$config['replace']              = array("",
                                    " ",
                                    "");
?>

However, the config does not work, because there are still some '&nbsp' tokens in the text. How can this be? Moreover, is there any way how to block the ads from feedsportal in this feed using RSS-Extender?

Cheers

Andy

Hi everybody,

I like your software very much and use it every day. However, I experienced that the heise feed is not always parsed correctly. In particular, videos and pictured embedded in the original articles are not always included in the parsed content.

E.g. the pictures in the following articles are not parsed:

• http://www.heise.de/newsticker/meldung/Apples-Siri-erhaelt-ein-Gesicht-1973092.html?wt_mc=rss.ho.beitrag.atom
• http://www.heise.de/newsticker/meldung/Steam-Machines-Intel-und-Nvidia-sind-drin-AMD-draussen-1973070.html?wt_mc=rss.ho.beitrag.atom

I would deeply appreciate your help

Thanks

Andy

It would be great to see support for http://zeit.de in rss-extender :)

low low prio notice

the article titles in golem feed have problems with the encoding. e.g.:

"Apple: "Steve-Jobs-Patent" für ungültig erklärt"
"Ubisoft: Nächstes Assassins's Creed angeblich in Brasilien"
"Schrei vor Glück: Zalando verdoppelt den Umsatz, aber auch den Verlust"

Content of the article has no problem.

Half of telepolis articles are empty…

if the url of an image does not start with http:// or https:// but only with // , then the image gets the base_url prefixed. instead of that it should not get prefixed at all.

dilbert config currently has a workaround, with search/replace in 57aa244, which needs to be reverted, when this bug is fixed

Hi everybody,

if I create a parsing rule for http://individualicious.com/feed/ I get the error message:
Feed individualicious not found

You can reproduce the issue by creating an emty parsing rule, just with the feed address. I double checked the feed address and if you open the feedaddress in your browser it works too.

Any help is appreciated

Hi,

it seems to me that every Heise Feed is broken, since every article contains html-tagging.

Hi,

I like rss-extender a lot. Is there any possibility to change the favicons of an extended feed? I use rss-extender together with Yahoo Pipes which overwrites each feed with its own favicon.

Thanks in advance

Andy

Hi,

while using RSS-Extender I discovered that it does not work with the subfeed mac & i of heise.de. I used the basic heise.de template and replaced only url and base url. My config file looks like this:

<?php
 include(__DIR__ . "/heise.php"); 
 $config['url'] = "http://www.heise.de/mac-and-i/news-atom.xml";
 ?>

What am I doing wrong? As far as I checked the html code the formatting of these websites look quite similar.