TheTVDB.com

We are continously getting invalid xml errors from api searches because there is plain html included in the xml response. Here is the most recent example:

http://www.thetvdb.com/api/GetSeries.ph ... iesname=tv

(opening that link with google chrome shows the error in a big red box)

In the hope that there are only a few cases I have been manually updating the descriptions of the offending shows by removing all html-like tags, but it seems that this is an uphill battle and people continue to paste new descriptions with html tags.

My API client is not even using the description at all, but still the xml parser chokes on the invalid xml.

Would it be possible to at least add a quickfix for this such as using strip_tags() like suggested here: viewtopic.php?f=17&t=8262

or to simply wrap the content of the Overview element in cdata blocks like this:

<Overview><![CDATA[
...
]]></Overview>

?

Thanks, Til

This is still an issue unfortunately

Here are some current examples where the API returns invalid xml:

http://www.thetvdb.com/api/GetSeries.ph ... sname=born

http://www.thetvdb.com/api/GetSeries.ph ... name=trust

http://www.thetvdb.com/api/GetSeries.ph ... sname=rich

http://www.thetvdb.com/api/GetSeries.ph ... ame=on+the

http://www.thetvdb.com/api/GetSeries.ph ... e=twilight

There's unlikely to be any changes to the current API while the new site is being developed so no quick fix unfortunately.
I've have though fixed those and will fix any others you post. You are of course welcome to fix them yourself.

I've also messaged the user adding code to the description.

_________________

Find Fanart Please link to the series/episode page and your sources. Wikipedia is NOT a valid source.

Ok. I'll consider adding a workaround to the tvdb_party ruby lib for this then.

Hi Guys,
This is still an issue unfortunately

Here are some current examples where the API returns invalid xml:

http://www.thetvdb.com/api/GetSeries.ph ... =the+river

Thanks

I've fixed the series overview that was causing it this time by removing the html tags from the overview (it was unlocked so anyone should be able to fix most similar issues if needed themselves). I understand that this is frustrating/annoying but as has been said no work is being done on the current api/site, all efforts are going towards the new site so don't see an automatic fix for this until then

_________________
---

Please watch this video to help LSU Computer Society try to win the RBS ESSA society competition! http://youtu.be/8JrhmoSXLXw - only 1:36min

Thanks

Ok thanks, looking forward to the new site.

For the record, here's a hacky workaround for the ruby gem tvdb_party: https://github.com/Ubertweek/tvdb_party