8

I have been using a feed for a specific tag ([set-theory]) of the Mathematics site on my WP blog. Right now it returns the following error message:

RSS Error: This XML document is invalid, likely due to invalid characters. XML error: not well-formed (invalid token) at line 805, column 34

I looked at the text dumped at the RSS link and the offending line has a "^L" character (that's the way it shows in Emacs) right before the last "filters" word. It looks like this:

<ol>
<li>The union of a chain of ^Lfilters is a ^Lfilter.
^O2. A maximal ^Lfilter is an ultra^Lfilter. 
How I can use Zorn's lemma to find the maximal filter?</li>
</ol>

The complete entry looks like this as plain text:

<entry>
    <id>https://math.stackexchange.com/q/1472767</id>
    <re:rank scheme="http://math.stackexchange.com">1</re:rank>
    <title type="text">Show that every proper filter on a set X can be extended to a proper prime filter?</title>
        <category scheme="https://math.stackexchange.com/tags" term="set-theory" />
        <category scheme="https://math.stackexchange.com/tags" term="filters" />
    <author>
        <name>Almoh</name>
        <uri>https://math.stackexchange.com/users/278782</uri>
    </author>
    <link rel="alternate" href="https://math.stackexchange.com/questions/1472767/show-that-every-proper-filter-on-a-set-x-can-be-extended-to-a-proper-prime-filte" />
    <published>2015-10-10T03:31:20Z</published>
    <updated>2015-10-10T05:15:44Z</updated>
    <summary type="html">


        &lt;p&gt;Are the following enough to complete the proof &lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;The union of a chain of filters is a filter.
2. A maximal filter is an ultrafilter. 
How I can use Zorn&#39;s lemma to find the maximal filter?&lt;/li&gt;
 &lt;/ol&gt;

    </summary>
</entry>

I assume that when this entry is deep enough it won't bother any more, but anyway there was a problem

  • I cannot reproduce the issue. Using Liferea v1.10.16; Gnome 3.14.4; Fedora 22 Linux. – Diminutive Colossus Oct 25 '15 at 21:11
  • I repro-ed it on this question feeds (before it was edited to fix) – rene Oct 25 '15 at 22:07
3
+50

I did some digging after my comments and can confirm that there is something funky going on with that post. Firefox, which has generally been pretty good to me as far as Unicode support goes shows me this when editing the question:

edit post screenshot

As you can see there are several characters that aren't parsing correctly. I'm assuming that they are - characters, or other punctuation that got corrupted somewhere between typing the question in, and the question hitting the database. When I grab the question feed for that question, my XML parser gives up.

parsing errors

I suggested an edit to the original post that removes the invalid characters, that has been approved. Now that the characters are no longer in the generated RSS XML data, the feed parses correctly.

enter image description here

I'll leave it to a developer to make the call about whether the behavior is a fluke or whether there is a sanitization bug somewhere between question submission and feed generation. So at least until another user with the same keyboard layout asks a question the feeds should be safe. :)

  • Glad someone looked at this. Specially the fact that it could be traced back to where the problem was. I'm a little geeky but not enough to have this insight. Thx! – Pedro Sánchez Terraf Oct 26 '15 at 0:29
  • In the future, when you run into a problem like it, and you want to see if it's your setup (Wordpress or whatever) you can do what I did and use an online parser. The one in the screenshots above is Code Beautify's XML viewer. – theB Oct 26 '15 at 0:38
  • We'll have to wait on a developer to see if there's anything between the user entering the text in the post editor and the final output to RSS that could have mangled the characters in the post. – theB Oct 26 '15 at 0:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .