Want to take part in these discussions? Sign in if you have an account, or apply for one below
Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.
The thread Introductions to category theory in physics is currently showing “XML Parsing Error” when I try to view it.
What browser?
Firefox 3.6
The problem is with ampersands. Meanwhile, you may have just locked yourself out of this topic.
Ampersands are problems (I believe) when using Markdown. I used Text to post that.
Firefox does not work for me either, but google’s Chrome does, it displays the error message:
“error on line 788 at column 35: Entity ’Foote’ not defined”
but what part of a comment did generate “Foote” I do not know :-) Or did someone enter this explicitly?
That was really unfortunate chose of punctuation! Mike wrote &Foote; which matches the pattern for an entity. The Markdown filter allows entities but doesn’t check that they are valid so let it through. The XML Validator doesn’t check validity of entities either; maybe it should. Anyway, so everything looked at ’&Foote;’ and assumed it was a valid entity, only it wasn’t, so Proper Browsers (like Firefox) complained about invalid XML. I’ve fixed it by altering the ampersand to its correct entity.
In summary, it was the conjunction of an ampersand and a semi-colon.
Sorry about that. (-: I think the XML validator should really check validity of entities, especially if some browsers will refuse to display the whole page because of a single invalid entity. I think there was just recently a similar issue with Todd writing an incorrect entity in Bourbaki set theory that then locked him out of the page. Chrome gives me an error at the top of the page but still displays it.
so Proper Browsers (like Firefox) complained about invalid XML.
Proper Browsers should also have Proper Error Handling (this is part of the Acid2 and Acid3 tests). If the only pages that rendered were those that validated, your browser wouldn't be able to display the vast majority of websites.
Firefox can display the vast majority of websites, since these are not served as XHTML to begin with. In my experience, the vast majority of websites that are serves as XHTML do validate.
However, I agree with Harry about how Firefox handles the few websites that are served as XHTML but do not validate. It does follow the written standard (or so I understand), but it is still wrong, even inexcusable. (And that makes the written standard inexcusable.) One needs at least a button ‘Display this page as tag soup.’ to push as an option.
@ Andrew
If you read this, can you manually edit the original Bourbaki thread (lined in Mike’s comment #10 above) to change &ecute; to é? Then we could read it again.
Toby said:
If you read this, can you manually edit the original Bourbaki thread (lined in Mike’s comment #10 above) to change … Then we could read it again.
The IE displays the thread, but does not let me post (version 8), but the Chrome browser by google both displays the page and lets me add comments, so you could try that one for the time being.
Opera also has no problems displaying or posting.
Original Bourbaki thread is now viewable again. I’ll have a think about the best way to parse entities - problem is that there’s just so many of them that it might slow things down a bit to go through the lot. I’ll try out some tests.
In the meantime, I fixed it using Firefox. Even if a page doesn’t display correctly, you can still “view source” and from that you can find the link to the edit-comment link (assuming that you are allowed to edit that comment).
I’ve added a step to the parser to escape unrecognised entities. So if I write something like &ecute; or &Foote; then it’ll escape them, but if I write a valid one such as é then it’ll be okay.
Thank you very much, Andrew – that’s wonderful.
Has anyone told you today that you rock?
1 to 18 of 18