Provide an Alpine-based Beta Image for Testing

[jojo141185]

Would it be possible to provide an Alpine-based beta image for testing purposes? The current scratch-based image is quite limiting for testing, as it lacks a shell (like sh or bash), making it impossible to connect to the container's console for debugging or exploration.

[euzu]

@jojo141185 yes the beta image could be based on alpine. Will change it.

[jojo141185]

@euzu Thanks for your quick reply.
If you could provide an image of the current beta version, that would be very helpful!

[euzu]

i am currently not home, i will try to push an image tomorrow.

[jojo141185]

There's absolutely no rush. Enjoy your time!

[aniel300]

i thought alpine was meant to be even smaller. what is the difference between the two flavors?

[euzu]

scratch contains nothing except m3u-filter,
alpine has additionally tools.

[euzu]

there is no debian or ubuntu base image used for m3u-filter

[aniel300]

what are u using?

[aniel300]

oh u are using this? https://hub.docker.com/_/scratch

seems this is even smaller than alpine?

[euzu]

Scratch is like a blank page whrre you write your own text and nothing more.
Alpine is a like page with some headers,
ubuntu is a book where you can add your pages.

[aniel300]

wow interesting...i have so much to learn, thank you bud for the explanation.

[euzu]

@jojo141185 the latest beta should be alpine.

[jojo141185]

@euzu Thank you very much for providing the new Alpine-based image! I've had a chance to test it briefly, and it seems to be working well – having shell access makes some testing much easier. 👍

By the way, while updating and reviewing the information in the CHANGELOG.md for the latest release, I encountered a couple of minor points I wanted to share:

1. EPG Option Placement: The new epg option for multi-EPG configuration now needs to be nested under a specific input section, rather than being at a higher level.
2. EPG Field Renamed: The sub-field within epg used to enable the processing/optimization options has been renamed from smart_assign to smart_match.

(It might be helpful to update the changelog or documentation slightly to make these points clearer if they aren't already explicit.)

---

Additionally, I wanted to report an error that seems pre-existing but which I hadn't focused on before. When processing information for a small number of specific series, these errors occasionally appear in the log:

[2025-04-13T18:40:28Z ERROR m3u_filter::processing::processor::playlist] Failed to process series info for MYSERIES1 (YEAR) duplicate field `release_date`
[2025-04-13T18:40:28Z ERROR m3u_filter::processing::processor::playlist] Failed to process series info for MYSERIES2 (YEAR) invalid type: sequence, expected a map

[euzu]

@jojo141185 the reason for the placement is that an EPG is defined per individual input and not intended for use across multiple inputs. Consequently, each EPG is directly associated with its respective input. In cases involving multiple inputs, the corresponding EPGs are consolidated during the merging process. Or am I wrong?

I will update the changelog , thanks for reporting.

I think the errors are comming from malformed json. If you can identify some json i can try to handle this cases.