rss.from_as1: remove possible addition of invalid author element

[isnbh0]

I appreciate this project, thanks for your work on it!

Here's a change that I think will fix a few broken cases such as this one that directly publish RSS converted from JSON feeds via granary.io.

In brief, we adhere closer to the RSS spec by removing the use of the default author email value '-' when constructing an RSS item.

(relevant upstream project code: https://github.com/lkiesow/python-feedgen/blob/v1.0.0/feedgen/entry.py#L366-L372)

illustrative example

https://granary.io/url?input=jsonfeed&output=rss&url=https://jamesg.blog/hf-papers.json

The above link produces an RSS file that at the time of writing is technically invalid when checked. This invalidity appears to break some RSS readers.

url.rss excerpt

<?xml version='1.0' encoding='UTF-8'?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" version="2.0">
  <channel>
    <title>Feed for https://jamesg.blog/hf-papers.json</title>
    <link>https://jamesg.blog/</link>
    <description>-</description>
    <atom:link href="https://granary.io/url?input=jsonfeed&amp;output=rss&amp;url=https://jamesg.blog/hf-papers.json" rel="self"/>
    <docs>http://www.rssboard.org/rss-specification</docs>
    <generator>granary</generator>
    <language>en</language>
    <lastBuildDate>Wed, 18 Dec 2024 05:46:10 +0000</lastBuildDate>
    <item>
      <title>Are Your LLMs Capable of Stable Reasoning?</title>
      <link>https://huggingface.co/papers/2412.13147</link>
      <description><![CDATA[The rapid advancement of Large Language Models (LLMs) has demonstrated remarkable progress in complex reasoning tasks. However, a significant discrepancy persists between benchmark performances and real-world applications. We identify this gap as primarily stemming from current evaluation protocols and metrics, which inadequately capture the full spectrum of LLM capabilities, particularly in complex reasoning tasks where both accuracy and consistency are crucial. This work makes two key contributions. First, we introduce G-Pass@k, a novel evaluation metric that provides a continuous assessment of model performance across multiple sampling attempts, quantifying both the model's peak performance potential and its stability. Second, we present LiveMathBench, a dynamic benchmark comprising challenging, contemporary mathematical problems designed to minimize data leakage risks during evaluation. Through extensive experiments using G-Pass@k on state-of-the-art LLMs with LiveMathBench, we provide comprehensive insights into both their maximum capabilities and operational consistency. Our findings reveal substantial room for improvement in LLMs' "realistic" reasoning capabilities, highlighting the need for more robust evaluation methods. The benchmark and detailed results are available at: https://github.com/open-compass/GPassK.]]></description>


      <author>-</author> <!-- INVALID -->


      <guid isPermaLink="true">https://huggingface.co/papers/2412.13147</guid>
    </item>

<!-- etc... -->

[isnbh0]

It seems I've gotten a bit lost in fixing up the tests -- I'll get back to this later, sorry :)

[snarfed]

This looks great! Thank you for the contribution! I think you're almost there, don't let the tests get you down.

[snarfed]

Tests are passing! Awesome! One small comment left, and please also add an entry to the changelog in the readme:

granary/README.md

Lines 339 to 340 in 55b5610

	* `rss`:
	* Support image enclosures, both directions.

...then I think we're ready to merge!

[snarfed]

Love it. Merging, thank you again!

cc @capjamesg, let me know if you install granary from GitHub or if you need a new release on PyPi for this.

[capjamesg]

I have just installed Granary from source for the hosted version of the Papers with Code RSS feed. Thank you everyone for the fix!

[capjamesg]

Oh wait, this pertains to the HF Papers project. I see. @snarfed The project depends on the hosted version of granary at granary.io:

curl -I https://jamesg.blog/hf-papers.xml
HTTP/2 301
server: nginx
date: Mon, 23 Dec 2024 23:24:50 GMT
content-type: text/html
content-length: 162
location: https://granary.io/url?input=jsonfeed&output=rss&url=https://jamesg.blog/hf-papers.json

The hosted version would thus need to be updated to use the latest GitHub version.

[snarfed]

Deployed to granary.io!