Error While Loading XML

[vletroye]

Hi,

I have the latest version 4.5 from 08/10/2014.
I run on Windows Server 2012 Essentials (x64)

I wanted to execute filebot -script fn:sysinfo in a filebot console.
But when I run that console, it runs automatically filebot (the GUI) as a modal window. I.e.: I cannot type anything in the console ?!


Filebot fails to match a Serie that I stored in a folder named "Dragons - Le Chevalier de Beurk" (in french)...

The reason for the problem seems to be that filebot picks this serie: http://thetvdb.com/?tab=series&id=286672&lid=17 (named "Dragons : Cavaliers de Beurk") and this one is empty and most probably incorrect on thetvdb...

The right one to be picked is IMO http://thetvdb.com/?tab=series&id=261202&lid=17 (named "Dragons: Cavaliers de Beurk")

Notice that in this last one, there is no blank before the :

If filebot fails because of an actual issue in the xml sent by thetvdb, ok...
But the problem is that I have no way to force it to pick the right serie ?!
See here after... The only title available is "Dragons : Cavaliers de Beurk" (serie 286672)



What can I do ? OR what can be made on your side for such a case ?

Thx a lot in adv.


[EDIT]I also tried to rename the folder with the title in english (Dragons Riders of Berk) which seems to have no duplicate on thetvdb.. No success
FYI, searching on "Dragons Riders of Berk" in french, within the "Episodes" tab of Filebot, works fine...

[rednoah]

Have you tried clicking on the very first result?

@see http://thetvdb.com/?tab=series&id=261202

You can run manual searches and see what's what in Episodes panel.

[vletroye]

Thanks and sorry, I didn't try indeed this title as I was convinced "Dragons: cavalier de Beurk" (french) was the one I was looking for.

I didn't realize that "Dragons" was the English name (lid=7) of my serie.
I.e.: you suggested me to pick the title Dragon (http://thetvdb.com/?tab=series&id=261202&lid=7 (<== lid=7)
while I was looking for that serie in french (lid=17) http://thetvdb.com/?tab=series&id=261202&lid=17


It's not the first time that I feel confused because the "Select" list displays a mix of titles in English and French.
I know this question is not related to the original topic, but why is there a mix of languages in that list ?
Couldn't you display the title in the preferred language (the one set via the preferences) - possible based on an option ?


And, just for my own info, the xml error was due to an issue on TheTVDB side... right ?!


Thx again

V.

[rednoah]

1.
id286672 only exists because some French person added the show again. It's clearly a bad entry and it's probably gonna be removed by some admin eventually. If only one title is entered, it'll show you that title, whatever language that is.

2.
Fetching full episode data will respect your language preferences. Search will not. Search will try to find the closest match. Like if I google "Dragons - Le Chevalier de Beurk" and it'll show me French results, even though my google language preferences are English.

[vletroye]

Thx for the explanations!
(I sent a few biers to Reinhard.. hope he will share with you

Maybe to help users (within a future release ?) the Search results list could loop on the results and fetch each title in user's preferred language ?!

E.g.: I had a list with 27 matches for "Danny.the.Dog.1080p.Multi.DTS" (in french)... Not easy to guess which one is the right match...
Searching on the web, I found that in Quebeck (french too), it's "enchainer"... And from then, that in USA it's "Unleashed" (6th in the search result list).
But it's actually weird for me that "Danny The Dog" is not in the result list... Indeed, if I pick "Unleashed", the new name generated by FileBot is "Danny The Dog"...

[rednoah]

The above has the obvious issue that even for you as a smart human it took research and googling to figure this one out. So if the name doesn't match, and if the year doesn't match. It's a really really hard to make this match.

Translating search results its possible, but for a list like this it would take at least 20 HTTP requests, which will take at least a few seconds, and with 10 requests per 20 seconds request limiting it could take 40 seconds. Then except for the one entry you want all those requests will have been wasted. All this for language preferences which don't even work reliable cause the data is missing for most languages and will default to English anyway in most cases.

Also the correct match should always be at #1 but in this case it's just bad luck.

PS: I like sharing beers with myself.


EDIT:
The "bad luck" in this case was that there's a release group called "Danny" so that part was ignored given that the rest "the Dog" matched another movie perfectly and there was no year to suggest otherwise.

[vletroye]

as you said, even for a human, it's indeed not always obvious to match... name and nicknames (the mismatch was on Tapei vs Austria)

Thx a lot for the extra details! I understand the limitation...
And fortunately, I don't have to many cases like those so solve...

Keep up your great tool

V.