Selkie – Opinionated TUI Framework for Raku

7 points by apogee 2 months ago · 12 comments

Reader

Nice. To get your dependency Notcurses::Native built via zef on my OpenSUSE Tumbleweed, I had to run

zypper in ffmpeg-7-libavdevice-devel ffmpeg-7-libavutil-devel ffmpeg-7-libavcodec-devel libunistring-devel libdeflate-devel

which pulled in several dependencies, resulting in the overall package list:

ffmpeg-7-libavcodec-devel ffmpeg-7-libavutil-devel ffmpeg-7-libswresample-devel ffmpeg-7-libavformat-devel ffmpeg-7-libavdevice-devel ffmpeg-7-libavfilter-devel ffmpeg-7-libpostproc-devel ffmpeg-7-libswscale-devel libdeflate-devel libunistring-devel

Maybe just a subset of these is necessary. Conversely, I've already had ffmpeg-7 and libavcodec57 and a number of other devel packages installed from before, so some other packages might also be neccessary. Still, this might help you in adding instructions for installing your Notcurses::Native on OpenSUSE.

apogeeOP 2 months ago

Thanks, I will get this into the next version of the Readme, I think the minimum would be roughly:

  sudo zypper in \                                                                                                                                                         
    ncurses-devel libunistring-devel libdeflate-devel \                                                                                                                    
    ffmpeg-7-libavcodec-devel ffmpeg-7-libavformat-devel \                                                                                                                 
    ffmpeg-7-libavutil-devel ffmpeg-7-libavdevice-devel \                                                                                                                  
    ffmpeg-7-libswscale-devel \                                                                                                                                            
    cmake pkg-config gcc

apogeeOP 2 months ago

I've set up GitHub actions to prebuild all the deps so on any modern Mac/Windows/Linux it should just be 'zef install selkie' now.

librasteve 2 months ago

Oh I found this https://raku.land/zef:antononcube/DSL::Examples

So Raku has a module for “few shot” LLM training for DSLs … would be cool to see an example DSL interpreter in Selkie (eg a window for DSL code and a window for output…

antononcube 2 months ago

There is a set of Raku modules that leverage LLMs for different tasks (mostly code generation) using different techniques:
- https://raku.land/zef:antononcube/LLM::Resources : Uses agentic LLM-graphs with asynchronous execution
- https://raku.land/zef:antononcube/ML::FindTextualAnswer : Finds answers to questions over provided texts (e.g. natural language code generation commands)
- https://raku.land/zef:antononcube/ML::NLPTemplateEngine : Fills-in predefined code templates based on natural language code descriptions/commands
- https://raku.land/zef:antononcube/DSL::Examples : Example translations of natural language commands to executable code
- apogeeOP 2 months ago
  
  I've got a few LLM modules too, mostly for handling context management:
  - https://raku.land/zef:apogee/LLM::Character implements CCv3 which is a standard for managing characters (system prompts) and lorebooks (injected snippets)
  - https://raku.land/zef:apogee/LLM::Chat handles context shifting for long contexts, sampler settings, templating for text completion & inferencing with or without streaming using supply/tap
  - https://raku.land/zef:apogee/LLM::Data::Inference adds retries, JSON parsing & multi-model route handling to LLM::Chat
  - https://raku.land/zef:apogee/LLM::Data::Pipeline allows you to declaratively build multi-step pipelines (simple agentic LLM use)
  - https://raku.land/zef:apogee/HuggingFace::API is a partial wrapper around HF API for grabbing tokenizers.json & tokenizer_config.json
  - https://raku.land/zef:apogee/Template::Jinja2 is a near-complete impl of Jinja2 for parsing LLM text completion templates (can be used for anything you'd use Jinja2 for)
  - https://raku.land/zef:apogee/Tokenizers is a thin wrapper around HF tokenizers, for token counting mostly
apogeeOP 2 months ago

Yeah I have thought about doing this with Roaring::Tags (https://raku.land/zef:apogee/Roaring::Tags) so you type a Roaring::Tags query on one side and it shows the built Raku on the other.
Wouldn't be too difficult:
- FileBrowser widget to get the serialised bitmap.
- MultiLineInput for the query text input.
- ListView to show tags and fields.
- RichText for the AST pretty printer.

librasteve 2 months ago

I tried this on macOS … very smooth install and looks great.

But why would I switch to Raku just to get a TUI framework?

apogeeOP 2 months ago

The tl;dr is: because you want native performance and don't want to focus on blitting pixels or terminal internals.
Selkie has an event/effect model similar to Elm & re-frame, you just declare your widgets and renders/updates are handled by the lib.
As for why Raku:
- grammars allow for easy parsing of complex DSLs
- supply/tap are a natural fit for the event/effect model, asynchronous thread-safe programming
- roles make composing widgets simple
- it's fun to write :D
- librasteve 2 months ago
  
  Ok cool, thanks!
  - apogeeOP 2 months ago
    
    np, what kind of TUIs do you write?

Settings

Selkie – Opinionated TUI Framework for Raku

Keyboard Shortcuts