Words trimmed from filename

[tormento]

In my toml file, I have:

output_template = "{title} [{lang}] {opt-delay}"

Launching

deezy.exe encode ddp "splendente [ita] 51.dts"

I get

splendente [ita].ec3

I thought that Deezy got confused by the number, perhaps thinking that it is the delay but, actually, Deezy seems to ignore everything after the first word. If I try to encode

deezy.exe encode ddp "splendente [ita] cinquepuntouno.dts"

I get, again

splendente [ita].ec3

or

deezy.exe encode ddp "splendente cinquepuntouno [ita].dts"

I get, again

splendente [ita].ec3

[jessielw]

This is expected actually.

INPUT: splendente [ita] 51.dts

If you use the template {title} [{lang}] {opt-delay}:

• {title}: The title of the input file. This is detected like a movie/tv show or the start of the file. In this case it extracts splendente.
• {[lang]}: Extracts lang from the files mediainfo and falls back to parsing it from the filename if needed/exists.

In this case you get a newly constructed output file name:
splendente [ita].ec3.

Which is the expected behavior.

[tormento]

Mmm… I see your point.

In that case, would you please add a new tag with the complete file name but the other recognizable other toggles parts?

Sometimes a video has more than one track, such as 2.0 and 5.1 and I usually need to distinguish which is which.

[jessielw]

It becomes a lot more complex and error prone picking apart a file in the middle. There is already a built in token that should be able to achieve your goal.

INPUT: splendente [ita] 51.dts

See {stem}:

• This is going to return the stem of the file, which is everything up to the extension splendente [ita] 51.

The only down side is with this approach, it is left as is. Meaning [ita] will be left in place and if you use other tokens behind it those will be appended based on the location of the token. If we modified the stem programmatically you'd have to consider every potential input that could exist, which is obviously not really possible. Of course we could come close, but you'd have room for errors for sure.

For instance if you had splendente 1990 23 xyz [ita+jp] 51 3423413 [DELAY -10ms].dts. Sure we could get the delay no problem, maybe even the year 1990. But channels etc would start to become confusing. Then when do remove parts of it and when do we keep parts of it? I'll have to do some thinking on this to see if there are any other approaches to this.

[tormento]

I've looked at the new {stem-cleaned} tag in the documentation.

That is nice but wouldn't be more intuitive to let {stem} be used according to the remove delay setting in toml file?

I mean: if {stem} is used, just simply copy the whole filename and keep or not the DELAY part according to the toml preferences.

Thank you :)

[jessielw]

Hello,

The issue with {stem} is that you don't have full titles audio files. Which I discussed here. It isn't really possible to use stem if you don't have proper filenames coming in. There isn't a lot of data to work with there.

I'm going to need more information, step by step of what you expect your filenames to be and what you expect them to come out as. It's possible we could add another token to only find and modify/remove delay. But it can't cherry pick everything like channels, etc.

[tormento]

I'd like to replicate the very same deew behaviour, i.e. it leaves the very same filename but the delay part, removed when/if the delay is removed. I think that it's the best compromise for me and a lot of people used to deew.

[jessielw]

What you want can now be achieved with {stem} https://jessielw.github.io/DeeZy/templates.html#stem-token on the latest release

[tormento]

Works perfectly, thank you!

P.S: Would it leave the DELAY part if delay parsing is set to false?

[jessielw]

Np! Also no, you must use another token with it {opt-delay} to retain/add delay into the file name.