Kodi Community Forum

Full Version: Workaround for scraping from STB
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Hi,

I have a Humax DVB-S STB which I have SMB access to. I've been trying a few methods to integrate the recordings this box makes into my Kodi library but am having some odd results. I was hoping someone on here would be able to help.


The STB makes .ts recordings (which are playable in Kodi) using the following conventions.

title/title_YYYYMMDD_XXXX.ts

where "XXXX" seems to be some kind of index number used by the STB.

I have written a shell script which parses the file name and date into a corresponding .nfo file for each episode.

An example:

Files:
Code:
Danger Mouse/Danger Mouse_20151006_0725.nfo
Danger Mouse/Danger Mouse_20151006_0725.ts

.nfo file contents:
Code:
<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<episodedetails>
    <title>Danger Mouse_20151006</title>
    <showtitle>Danger Mouse</showtitle>
    <season>2015</season>
    <episode>1006</episode>
    <aired>2015-10-06</aired>
</episodedetails>

However, TheTVDB scraper seems to ignore the season / episode tags contained in the .nfo file, and scans the file to the library according to the filename. Confusingly, using the above example, Kodi scans a single file in to the library twice:

Code:
Season 7 > Episode 25 (with correct episode title tag)
Season 201510 > Episode 06 (with correct episode title tag)

I am using TheTVDB scraper because it seems to be able to fetch the TV show metadata ok from the folder name. Using purely the .nfo scraper (with a generated tvshow.nfo file in the folder) yields no scanned episodes into my library.

If someone could spot a fault in something that I'm doing, or even suggest a better method then I'd be very grateful. I'm simply looking for a very rudimentary way of scanning these recordings into Kodi, so the usual fully-featured metadata is not really necessary - I had investigated renaming files so that TheTVDB scraper would index using air dates, but it was a whole can of worms!
If anyone is interested in this, I have come up with a solution and posted here:

2410151 (post)