Problems with getting RDF from XHTML5 served as application/xhtml+xml?

[christianhujer]

The tool seems to have problems extracting the schema.org RDFa data from the following page: http://nelkinda.com/blog/user-stories-are-not-always-user-stories/
The page is written in XHTML5, delivered as application/xhtml+xml, encoded with gzip, and it seems that pymicrodata is unable to extract any information from it.
I have successfully used the following tools with said page:

• W3C Nu Validator to ensure the page is valid XHTML5 https://validator.w3.org/nu/?doc=http%3A%2F%2Fnelkinda.com%2Fblog%2Fuser-stories-are-not-always-user-stories%2F
• Yandex Structured data validator https://webmaster.yandex.com/tools/microtest/
• Google Structured data testing tool https://search.google.com/structured-data/testing-tool#url=http%3A%2F%2Fnelkinda.com%2Fblog%2Fuser-stories-are-not-always-user-stories%2F
• Sturctured Data Linter http://linter.structured-data.org/?url=http:%2F%2Fnelkinda.com%2Fblog%2Fuser-stories-are-not-always-user-stories%2F

By the way, the tools from Microsoft also have problems with this page.

[christianhujer]

The following attachment is a zip archive with the XHTML page that isn't processed successfully.

sample.zip

[iherman]

@christianhujer: there were two problems. One was yours and the other was mine...

• Your code is not based on microdata; it is in RDFa. It seems that the tools that you refer to interpret both microdata and RDFa, and hence produce proper output. However, pyMicrodata is strictly for microdata and not for RDFa.
• One the other hand: there is an RDFa distiller, too. There is a a service at W3C, and there is also an RDFLib library to handle that, namely pyrdfa3. There was a bug in that code, and it was indeed related to the fact that you served XHTML. I have found that bug, and have updated that repository. The aforementioned service has also been updated, and it does interpret your file, see at http://bit.ly/2dWuXBr

Thanks for the bug report!

[christianhujer]

@iherman Ahaha, thanks for clearing it up! I actually used the service at W3C, but when reporting the bug I must have confused the two libraries (RDF vs microdata). And I can confirm that the bug is now fixed. I have, however, found another small glitch, which I will report at pyrdfa3.