'Episode' Number Handling

[shelfactor]

Interested in more information on how SportScanner Agent handles numbering episodes, and/or how it handles the Ordering

Issue: SportScanner only sorts based on the episode number (for oldest/newest), not the 'available date'

When adding MLB game files to plex, an episode number is generated loosely based on the game date located in the file name.

Current Sort Order: Oldest First

Game 1
File: MLB 2025-03-27 San Diego Padres vs Atlanta Braves.mkv
Episode 250327953
Game 2
File: MLB 2025-03-29 San Diego Padres vs Atlanta Braves (3-28).mkv
Episode 250329847
NOTE: This game aired on 3-28 local time, though TheSportsDB recognizes UST, thus the naming has to show 3-29
Game 3
File: MLB 2025-03-29 San Diego Padres vs Atlanta Braves.mkv
Episode 250329617

How does SportScanner Agent / Plex determine those last three digits?

The Sorting will change from oldest first, to newest first (reversing the current example). However, Game 2 is always viewed as "newer" than Game 3, as its Episode Number is the higher of the two.

[jbalcorn]

It's a random number -- Justin B. Alcorn The views expressed are not necessarily my own, much less anyone else's PGP Fingerprint CCEB F776 C3FD 1050 C8DB 532E B8B9 BED7 7764 406C

…

On Tue, Apr 8, 2025 at 12:55 PM shelfactor ***@***.***> wrote: Interested in more information on how SportScanner Agent handles numbering episodes, and/or how it handles the Ordering *Issue: SportScanner only sorts based on the episode number (for oldest/newest), not the 'available date'* When adding MLB game files to plex, an episode number is generated loosely based on the game date located in the file name. Current Sort Order: Oldest First Game 1 File: MLB 2025-03-27 San Diego Padres vs Atlanta Braves.mkv Episode 250327*953* Game 2 File: MLB 2025-03-29 San Diego Padres vs Atlanta Braves (3-28).mkv Episode 250329*847* NOTE: This game aired on 3-28 local time, though TheSportsDB recognizes UST, thus the naming has to show 3-29 Game 3 File: MLB 2025-03-29 San Diego Padres vs Atlanta Braves.mkv Episode 250329*617* How does SportScanner Agent / Plex determine those last three digits? image.png (view on web) <https://github.com/user-attachments/assets/ea078603-12b2-4551-ab54-f114ed1a1194> The Sorting will change from oldest first, to newest first (reversing the current example). However, Game 2 is always viewed as "newer" than Game 3, as its Episode Number is the higher of the two. — Reply to this email directly, view it on GitHub <#50>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABXEEZGP7SPV57YIASHXLA32YP5QLAVCNFSM6AAAAAB2W2MJLGVHI2DSMVQWIX3LMV43ASLTON2WKOZSHE4DAMZZHEYDQOI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> *shelfactor* created an issue (mmmmmtasty/SportScanner#50) <#50> Interested in more information on how SportScanner Agent handles numbering episodes, and/or how it handles the Ordering *Issue: SportScanner only sorts based on the episode number (for oldest/newest), not the 'available date'* When adding MLB game files to plex, an episode number is generated loosely based on the game date located in the file name. Current Sort Order: Oldest First Game 1 File: MLB 2025-03-27 San Diego Padres vs Atlanta Braves.mkv Episode 250327*953* Game 2 File: MLB 2025-03-29 San Diego Padres vs Atlanta Braves (3-28).mkv Episode 250329*847* NOTE: This game aired on 3-28 local time, though TheSportsDB recognizes UST, thus the naming has to show 3-29 Game 3 File: MLB 2025-03-29 San Diego Padres vs Atlanta Braves.mkv Episode 250329*617* How does SportScanner Agent / Plex determine those last three digits? image.png (view on web) <https://github.com/user-attachments/assets/ea078603-12b2-4551-ab54-f114ed1a1194> The Sorting will change from oldest first, to newest first (reversing the current example). However, Game 2 is always viewed as "newer" than Game 3, as its Episode Number is the higher of the two. — Reply to this email directly, view it on GitHub <#50>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABXEEZGP7SPV57YIASHXLA32YP5QLAVCNFSM6AAAAAB2W2MJLGVHI2DSMVQWIX3LMV43ASLTON2WKOZSHE4DAMZZHEYDQOI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[shelfactor]

As far as I can tell, the number while potentially meaningless, is not completely random.

I can remove the file, do full plex dance, re-add the file... and it will retain the exact same episode number.

Further testing with altered file name creates same result:

1. Removed the file
2. Library rescan
3. Stop plex
4. Start plex
5. Add same file back, with altered file name.
6. Episode number is different than prior example: E250328942

7. Change file name back to correct naming convention for TheSportsDB.
8. Rescan library
9. Episode number is the same as original example: E250329847

There must be some logic that the Scanner Agent follows that produces the same result each cycle.

[jbalcorn]

I'm sorry, yes you're right, I forgot about this. Some sports - I think Formula 1 is the best example - have sub-episode information that is used. Most, however, use the file hash. It used to be ep = int('%s%02d%02d%04d' % (year[-2:],month, day, abs(hash(file)) % (10 ** 4))) But that broke in 2022 when the calculation that created the integer overflowed a 32-bit integer. It was changed to this: ep = int('%s%02d%02d%03d' % (year[-2:],month, day, filehash % (10 ** 3))) So a 3-digit number dependent on the file hash. https://github.com/mmmmmtasty/SportScanner/pull/42/files

…

-- Justin B. Alcorn The views expressed are not necessarily my own, much less anyone else's PGP Fingerprint CCEB F776 C3FD 1050 C8DB 532E B8B9 BED7 7764 406C

On Tue, Apr 8, 2025 at 3:11 PM shelfactor ***@***.***> wrote: As far as I can tell, the number while potentially meaningless, is not completely random. I can remove the file, do full plex dance, re-add the file... and it will retain the exact same episode number. Further testing with altered file name creates same result: 1. Removed the file 2. Library rescan 3. Stop plex 4. Start plex 5. Add same file back, with altered file name. 6. *Episode number is different than prior example: E250328942* image.png (view on web) <https://github.com/user-attachments/assets/f77662b6-228b-423f-ae90-45adb835eaa4> 7. Change file name back to correct naming convention for TheSportsDB. 8. Rescan library 9. *Episode number is the same as original example:* E250329*847* image.png (view on web) <https://github.com/user-attachments/assets/8f42c0e2-a36a-48da-b537-5660bff4bc46> - There must be some logic that the Scanner Agent follows that produces the same result each cycle. — Reply to this email directly, view it on GitHub <#50 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABXEEZATDMJQJBP4TCN42XD2YQNMJAVCNFSM6AAAAAB2W2MJLGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOOBXGQZDQMRRGM> . You are receiving this because you commented.Message ID: ***@***.***> *shelfactor* left a comment (mmmmmtasty/SportScanner#50) <#50 (comment)> As far as I can tell, the number while potentially meaningless, is not completely random. I can remove the file, do full plex dance, re-add the file... and it will retain the exact same episode number. Further testing with altered file name creates same result: 1. Removed the file 2. Library rescan 3. Stop plex 4. Start plex 5. Add same file back, with altered file name. 6. *Episode number is different than prior example: E250328942* image.png (view on web) <https://github.com/user-attachments/assets/f77662b6-228b-423f-ae90-45adb835eaa4> 7. Change file name back to correct naming convention for TheSportsDB. 8. Rescan library 9. *Episode number is the same as original example:* E250329*847* image.png (view on web) <https://github.com/user-attachments/assets/8f42c0e2-a36a-48da-b537-5660bff4bc46> - There must be some logic that the Scanner Agent follows that produces the same result each cycle. — Reply to this email directly, view it on GitHub <#50 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABXEEZATDMJQJBP4TCN42XD2YQNMJAVCNFSM6AAAAAB2W2MJLGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOOBXGQZDQMRRGM> . You are receiving this because you commented.Message ID: ***@***.***>

[Francis2b]

I think there's still a bug in the ep number generation... at least for the Formula 1 because for eg this weekend GP the FP2 is sorted before the FP1. And for last weekend Qualifying comes before FP3...

[pkjensen]

Is there a way to turn this off globally? My library is mostly US college sports, and those three digits don't add any value, and in fact make it harder to decipher the date from the episode number.

[jbalcorn]

The value it adds is that the original code that generated the number overflowed a 32 bit integer in 2022, causing the numbering to be completely out of order after that date. -- Justin B. Alcorn The views expressed are not necessarily my own, much less anyone else's PGP Fingerprint CCEB F776 C3FD 1050 C8DB 532E B8B9 BED7 7764 406C

…

On Wed, Aug 13, 2025 at 1:10 PM pkjensen ***@***.***> wrote: *pkjensen* left a comment (mmmmmtasty/SportScanner#50) <#50 (comment)> Is there a way to turn this off globally? My library is mostly US college sports, and those three digits don't add any value, and in fact make it harder to decipher the date from the episode number. — Reply to this email directly, view it on GitHub <#50 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABXEEZGRSB4VCXEHDDRV2QD3NNWRFAVCNFSM6AAAAAB2W2MJLGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTCOBUG43DKMJUGI> . You are receiving this because you commented.Message ID: ***@***.***>