package pipeline

import (
	"context"

	"gitea.maximumdirect.net/eric/seriatim/internal/config"
	"gitea.maximumdirect.net/eric/seriatim/internal/model"
	"gitea.maximumdirect.net/eric/seriatim/internal/report"
	"gitea.maximumdirect.net/eric/seriatim/schema"
)

// ModelState identifies which representation a preprocessing module consumes.
type ModelState string

const (
	StateRaw       ModelState = "raw"
	StateCanonical ModelState = "canonical"
)

// PreprocessState carries transcript data as it moves from raw to canonical.
type PreprocessState struct {
	State     ModelState
	Raw       []model.RawTranscript
	Canonical []model.CanonicalTranscript
}

// InputReader loads external input specs into raw transcript documents.
type InputReader interface {
	Name() string
	Read(ctx context.Context, cfg config.Config) ([]model.RawTranscript, []report.Event, error)
}

// Preprocessor transforms preprocessing state.
type Preprocessor interface {
	Name() string
	Requires() ModelState
	Produces() ModelState
	Process(ctx context.Context, in PreprocessState, cfg config.Config) (PreprocessState, []report.Event, error)
}

// Merger combines canonical transcripts into one merged transcript.
type Merger interface {
	Name() string
	Merge(ctx context.Context, in []model.CanonicalTranscript, cfg config.Config) (model.MergedTranscript, []report.Event, error)
}

// Postprocessor transforms a merged transcript.
type Postprocessor interface {
	Name() string
	Process(ctx context.Context, in model.MergedTranscript, cfg config.Config) (model.MergedTranscript, []report.Event, error)
}

// OutputWriter emits final artifacts.
type OutputWriter interface {
	Name() string
	Write(ctx context.Context, out schema.Transcript, rpt report.Report, cfg config.Config) ([]report.Event, error)
}