[yee379]

so i've been looking into a way of getting my mythtv recordings better integrated with xbmc. the myth:// method works fine, but because of an issue with schedules direct and the lack of database support in mythtv itself; important information like the series and episode numbers are missing.

my goal is to be able to use the TV Shows pane with all the nice graphics and extra info provided by the scrapers, rather than the drab Video's list.

i currently use mythrename.pl to generate readable filenames for all my mythtv recordings. this filename can then have info such as the original air date, syndicated episode numbers and the episode name.

eg:

Code:

Fringe - Unleashed - (2009-04-14 @ 2101).mpg

from there i can run a separate job to search say http://epguides.com to lookup the syndicated episode id and the original air date and generate a xml nfo file that xbmc can use.

so i have:

Code:

Fringe - Unleashed - (2009-04-14 @ 2101).nfo

which contains a partial nfo:

Code:

<tvshow>

  <title>Fringe</title>
  <episodeguide>&lt;episodeguide&gt;&lt;url cache=&quot;82066.xml&quot;&gt;http://www.thetvdb.com/api/1D62F2F90030C444/series/82066/all/en.zip&lt;/url&gt;&lt;/episodeguide&gt;</episodeguide>
  
  <episodedetails>
    <title>Unleashed</title>
    <season>1</season>
    <episode>16</episode>
  </episodedetails>
  
</tvshow>

in the same directory (tvshows/Fringe); i add that to my Video's sources, set the content to tv show and scraper to use thetvdb.com. do a manual scan, go to TV Shows... and nothing!

can anyone help? cheers,

[spiff]

that's not the correct format of the nfo's.

either you want

Code:

tvshow>
  <title>Fringe</title>
  <episodeguide>&lt;episodeguide&gt;&lt;url cache=&quot;82066.xml&quot;&gt;http://www.thetvdb.com/api/1D62F2F90030C444/series/82066/all/en.zip&lt;/url&gt;&lt;/episodeguide&gt;</episodeguide>  
</tvshow>

for the show itself, or you want

Code:

<episodedetails>
    <title>Unleashed</title>
    <season>1</season>
    <episode>16</episode>
  </episodedetails>

for the episode

[yee379]

thanks for the quick reply!

i still can't get it to work i think i should clarify what i'm trying to do; i do not want to do a full scrape of data and dump it into the nfo's, i only want to populate the episode and series numbers so that the xbmc scapers can have sufficient data so that it gets the remaining information about the tv show.

can it work like this? or do i have to populate the entire <episodedetail/> myself?

thanks!

[spiff]

if you do it like this you need to populate the entire info yourself.
plus currently it won't work as the filenames needs to enumerate, i.e. match one of the regular expressions, for us to find the nfo files

[yee379]

hmm.. so it sounds like i'm probably better off renaming the files to have the series and episode data instead.

i was hoping not to have to write my own scraper, but i guess i'll have to get my hands dirty

i noticed from the wiki that there is a scape.exe where i can test my xml files (which i'm lead to believe is out of date anyway). is there a version available for linux? i tried a simple gcc -o scraper on a couple of the files, but they all come back with errors. is there a makefile available?

[spiff]

we lost the source to that program :/

in any case, you don't need to write a new scraper. you need to change the xbmc code to accept any nfo files, then read the episode and season number from those nfo files. based on those you can do a normal scrape