Pictures are showed in music format

[Wandelstock]

on this site

https://www.androidpolice.com/firefox-focus-93-comes-with-new-interface-icon-and-features-inspired-by-real-firefox/

the pictures are shown as music (see screenshots)

original site:

in Feedme

Version 3.19.2 release 218
System Chromebook Lenovo IdeaPad Flex 3 (Codename Octopus), Android 9 (Vanilla)

[c8c8]

I have lots of sites that happens for.

[seazon]

FeedMe uses http://htmlcleaner.sourceforge.net/index.php to clean html.

In this case (https://www.androidpolice.com/firefox-focus-93-comes-with-new-interface-icon-and-features-inspired-by-real-firefox/ ), the original code is

<picture>
  <source ...>
  <source ...>
  <source ...>
  <source ...>
  <img ...>
</picture>

after parsing, the code changes to

<picture>
  <audio>
    <source ...>
    <source ...>
    <source ...>
    <source ...>
    <img ...>
  </audio>
</picture>

FeedMe simplify the content by cleaning up the html code, HtmlCleaner is the best solution which I found.

[c8c8]

Is there any update on if this is going to be fixed? 8 months later and I haven't seen it mentioned again. It's a pretty big impairment for a program designed to read things on the internet for it to not be able to display images. I often feel like I'm using Lynx in 1996 again. I might have liked Lynx in 1996, but I sure wouldn't go back to it in 2022. FocusReader's subscription model sucks supremely, but I have to say that its mobilizer is flawless and it can pull images to use for the article list (two things FeedMe can't do) and it's making a compelling argument for change. FeedMe is the only RSS reader I've ever seen that struggles parsing code this way. If there's not going to be a move away from HTMLCleaner, why not just write a line of code into the program to strip out the audio tags if they didn't exist pre-parse?

[seazon]

Will fix in next version(FeedMe 4.0.0 Canary 9)

[seazon]

Please try with 4.0.0-Canary-9.
If you upgrade from Canary 8, need update Settings - Cache - Reading List to avoid crash again.

[hdhoang]

sorry, i still see this on substack->inoreader-> canary10. article is fetched by this version, not canary8.

feed: https://kamilkazani.substack.com/feed
item: they all have multiple images, only the thumbnail/hero shows.

cache dir's content, including a web-view fetch https://u.pcloud.link/publink/show?code=XZtzjNVZIcJ3t3HvGTFhWUaR2l7s8fTkVfdy

[Otiel]

Same issue with https://loadingartist.com/index.xml

[seazon]

Ok, it's ok on feedly, but wrong on inoreader since I fixed it with feedly. Will try to fix it for inoreader.

[seazon]

please verify with 4.0.0-Canary-11

[hdhoang]

i confirm it's ok for inoreader+substack. thanks!

[bartlibert]

Seems to still be an issue with tiny tiny rss :(

[c8c8]

Definitely not fixed, in fact, we seem to be going backwards a bit.

Using Feedbin as my aggregator.

The Feedbin parser in FeedMe does not appear to be working at all. The FeedMe parser shows all images as music notes.

[seazon]

Let me check ttrss and feedbin later.

[seazon]

please try 4.0.4

[c8c8]

Continues to be broken in 4.0.4. Easily seen in the Android Police feed. Music notes everywhere.

[seazon]

Deprecate htmlcleaner and use jsoup instead.
will fix in 4.1.

[seazon]

please try 4.1

[c8c8]

4.1 appears to make great strides forward pulling article text out of webpages. For the first time in two years, the images are pulled as images, not music notes (*if they get pulled). I feel hope and anticipation for FeedMe. That said, the images are only pulling 40-50% of the time from what I see browsing my RSS feed today (and in some feeds, it loads images early in the article, but not later). Consider FeedMe's rendering of this feed (using the FeedMe mobilizer): https://www.inspiredtaste.net/feed/

Compared to FocusReader's rendering of the same:

We're moving in the right direction though 👍