"Unable to get Metadata"

[sdsean]

All of the sudden (no idea why), I cannot use BMT to search for metadata and then use that search for the update (i.e. clicking "Find Metada" so that I can get fanart, etc).

All I ever get anytime I select anything from the search results that it finds is "Unable to get Metadata".

I have attached a log file for persual. . .looks like I'm getting a SAX parser error from IMDB (I should also say there are a whole lot of other exceptions happening which seem like they could be bad news. . .)

[stuckless]

Originally Posted by sdsean

All of the sudden (no idea why), I cannot use BMT to search for metadata and then use that search for the update (i.e. clicking "Find Metada" so that I can get fanart, etc).

All I ever get anytime I select anything from the search results that it finds is "Unable to get Metadata".

I have attached a log file for persual. . .looks like I'm getting a SAX parser error from IMDB (I should also say there are a whole lot of other exceptions happening which seem like they could be bad news. . .)

It appears the the IMDB provider is broken I think imdb changed their layout, which breaks the provider.. That's the downside to "scaping" a site

If you edit your provider list, and move tmdb to the front, then that should make the lookups work again (using themoviedb).

I also noticed that the selection of providers in the web ui doesn't actually work. No matter what you choose, it will always use what is configured

[sdsean]

OK cool that seems to work. . .

I should note also the that the IMDB fetch metadata plugin for the Sage UI does still seem to work. . .not sure how its getting its data if its a pure web scrape or not. . .

Are you using a SAX XML parser from Java? Is it custom? I only ask b/c (i'm sure you're well aware) that HTML is not XML safe.

Any idea how soon you'll have a look at it to fix it?

[stuckless]

Originally Posted by sdsean

I should note also the that the IMDB fetch metadata plugin for the Sage UI does still seem to work. . .not sure how its getting its data if its a pure web scrape or not. . .

Unfortunately, scraping the only way to get data, short of downloading the entire imdb database locally

If someone fixes nielms before I fix mine, then I may just switch over to his... or use the xbmc one, if it's still working. I really hate screen scraping

Are you using a SAX XML parser from Java? Is it custom? I only ask b/c (i'm sure you're well aware) that HTML is not XML safe.

I use an html parser that behaves like a SAX parser but on "real world" html. I can then use traditional sax handling apis to traverse the document. I haven't looked at this yet, so I don't know if it's the parser or something else. If it's the parser, then I'm hosed, since it hasn't been updated since 2006.

Any idea how soon you'll have a look at it to fix it?

I'm not sure... I know I can't really look at it this weekend

[heffneil]

I am having the same problem. When I manually search by the title I only have IMDB results. Any idea how to change this? I choose TMDB and do the search but all the results are still sourced from IMDB?

Thanks

[heffneil]

Weird I tested other movies and they found results on TMDB. I am trying to find alvin and the chipmunks the squeaquel. But doesn't show with the TMDB search - go figure

[stuckless]

Originally Posted by heffneil

I am having the same problem. When I manually search by the title I only have IMDB results. Any idea how to change this? I choose TMDB and do the search but all the results are still sourced from IMDB?

Thanks

There is a bug in the web ui. When doing a search, it will use whatever your configured provider is, and totally ignore what you select from the list. So, if imdb is configured first, then that's what will be used, always.

[megamojo]

Looks like if you add /combined to the end of the imdb url you get the old style page, does that make it easy for you to fix the scraping?

example:
http://www.imdb.com/title/tt0892769/combined

[stuckless]

Originally Posted by megamojo

Looks like if you add /combined to the end of the imdb url you get the old style page, does that make it easy for you to fix the scraping?

example:
http://www.imdb.com/title/tt0892769/combined

Probably... but I already rewrote the imdb detail scraper It won't be available for a week or so, mainly because I'm the middle of a phoenix development cycle, and need to do more testing before I can release all the phoenix changes, etc.

[mkanet]

Sean, you suggestion to switch TMDB to be primary worked. However, I noticed now, the automatic scanner (and possibly the manual one) will not recognize the movie title unless I remove the year at the end for example

Solitary Man (2009).mkv will completely be skipped by the automatic scanner when using TMDB. As soon as I rename it to Solitary Man.mkv, a properties file gets generated immediately.

Is this something that I can fix or is this a limitation of the database provider?

Let us know as soon as we can switch back to IMDB.

[stuckless]

Originally Posted by mkanet

Sean, you suggestion to switch TMDB to be primary worked. However, I noticed now, the automatic scanner (and possibly the manual one) will not recognize the movie title unless I remove the year at the end for example

Solitary Man (2009).mkv will completely be skipped by the automatic scanner when using TMDB. As soon as I rename it to Solitary Man.mkv, a properties file gets generated immediately.

Is this something that I can fix or is this a limitation of the database provider?

Let us know as soon as we can switch back to IMDB.

Michael, it should work either way. The parser should extract the title and year from the filename and query the tmdb provider using only the title. I'll have to run some tests.

[mkanet]

Sean, I'm sure you might have already guessed it, but ignoring the ™ symbol is a bad thing to do in BMT. I took it out her your suggestion; leaving just the Blu-ray word only; and all of a sudden things started working again. I havent tested to see what happens when adding a new bluray title to see if ignoring Blu-ray is good enough to recognize the title.

Originally Posted by stuckless

Michael, it should work either way. The parser should extract the title and year from the filename and query the tmdb provider using only the title. I'll have to run some tests.

[crusing]

It also seems like the IMDB seach returns results fine with ' and & where the TMDB search will not return anything unless the special characters are removed.

[stuckless]

Originally Posted by crusing

It also seems like the IMDB seach returns results fine with ' and & where the TMDB search will not return anything unless the special characters are removed.

Thanks... that was noted elsewhere as well... it's been fixed, and I'm hoping to get a build out this weekend.

[gcm281p]

I'm not really that familiar with this whole thing, I just kind of muddle through, and I want to make sure I understand this.

I'm currently on Sage 6.6 and don't plan on upgrading the 7 right away. I see this is fixed but it looks like it requires Sage 7 to work. Am I reading that right?

So does that mean I'm pretty much out of luck with fixing this unless I upgrade?

[stuckless]

Originally Posted by gcm281p

I'm not really that familiar with this whole thing, I just kind of muddle through, and I want to make sure I understand this.

I'm currently on Sage 6.6 and don't plan on upgrading the 7 right away. I see this is fixed but it looks like it requires Sage 7 to work. Am I reading that right?

So does that mean I'm pretty much out of luck with fixing this unless I upgrade?

I just don't have the time to support v6 and v7 of this tool/plugin. So yeah, the issue still exists in the bmt 4.0-beta21 build, which is the last Sage6 compatible build.

[megamojo]

Any way you can test to see if adding /combined to the url will actually make the scraper work? If you don't have time to make another release just to please us legacy users, can you point me to the right file to edit myself? I'm using BMT 3.1 currently but wouldn't mind updating to the latest standalone-compatible release to make it easier.

[Xsabre]

Originally Posted by gcm281p

I'm not really that familiar with this whole thing, I just kind of muddle through, and I want to make sure I understand this.

I'm currently on Sage 6.6 and don't plan on upgrading the 7 right away. I see this is fixed but it looks like it requires Sage 7 to work. Am I reading that right?

So does that mean I'm pretty much out of luck with fixing this unless I upgrade?

I am in the same boat as you. Looks like we are stuck. Don't know if the fix is easy enough that someone else can step up.

I can't fault Stuckless as we wouldn't have anything at all if it wasn't for him.

[stuckless]

Originally Posted by megamojo

Any way you can test to see if adding /combined to the url will actually make the scraper work? If you don't have time to make another release just to please us legacy users, can you point me to the right file to edit myself? I'm using BMT 3.1 currently but wouldn't mind updating to the latest standalone-compatible release to make it easier.

Originally Posted by Xsabre

I am in the same boat as you. Looks like we are stuck. Don't know if the fix is easy enough that someone else can step up.

I can't fault Stuckless as we wouldn't have anything at all if it wasn't for him.

Here's an update for 3.1... If I had a clean 4.0 zip then I'd be able to provider an update for it as well. The /combined did the trick, but I don't have an older 4.0 build lying around.

[megamojo]

You're the greatest, thanks for supporting us old farts!