find_uniqueids_in_text fails to extract IMDb ID from URLs with language prefix (e.g. /de/)

[chfury]

The regex in scraper_datahelper.py used to extract IMDb IDs from NFO files fails when the URL contains a language prefix like /de/.

Affected regex:

res = re.search(r'imdb....?/title/tt([0-9]+)', input_text)

Example URL that fails:

https://www.imdb.com/de/title/tt14280366/

The regex expects imdb.com/title/... but the URL has imdb.com/de/title/..., so no ID is extracted. Kodi then falls back to a title search, which fails for movies whose filename contains transliterated umlauts (e.g. Kuechenbrigade instead of Küchenbrigade).

Proposed fix:

res = re.search(r'imdb....?/(?:[a-z]+/)?title/tt([0-9]+)', input_text)

The (?:[a-z]+/)? makes the optional language prefix match correctly.

Steps to reproduce:

1. Have an NFO file containing an IMDb URL with a language prefix, e.g. https://www.imdb.com/de/title/tt14280366/
2. Scan the file into Kodi library using the TMDB Python scraper
3. Kodi logs: Find movie with title '...' from year '...' — meaning the ID was not found and it fell back to title search