Apply correction proposals to transcripts

2026-05-11 13:23:59 +00:00
parent c58d307ba7
commit f461922b9b
2 changed files with 448 additions and 0 deletions
--- a/internal/framework/proposals/apply.go
+++ b/internal/framework/proposals/apply.go
@@ -0,0 +1,155 @@
+package proposals
+
+import (
+	"sort"
+
+	"gitea.maximumdirect.net/eric/audita/internal/core/schema"
+)
+
+// AppliedChange records a proposal that was successfully applied.
+type AppliedChange struct {
+	ProposalIndex    int    `json:"proposal_index"`
+	ModuleKey        string `json:"module_key"`
+	ModuleInstance   string `json:"module_instance"`
+	TargetSegmentID  int    `json:"target_segment_id"`
+	OriginalText     string `json:"original_text"`
+	CorrectedText    string `json:"corrected_text"`
+	ReplacementCount int    `json:"replacement_count"`
+}
+
+// SkippedChange records a proposal that was not applied safely.
+type SkippedChange struct {
+	ProposalIndex   int                `json:"proposal_index"`
+	ModuleKey       string             `json:"module_key"`
+	ModuleInstance  string             `json:"module_instance"`
+	TargetSegmentID int                `json:"target_segment_id"`
+	OriginalText    string             `json:"original_text"`
+	CorrectedText   string             `json:"corrected_text"`
+	SkipReason      ProposalSkipReason `json:"skip_reason"`
+	Message         string             `json:"message"`
+}
+
+// ApplyResult is the deterministic output of applying proposals.
+type ApplyResult struct {
+	Transcript *schema.Transcript `json:"transcript"`
+	Applied    []AppliedChange    `json:"applied_changes"`
+	Skipped    []SkippedChange    `json:"skipped_changes"`
+}
+
+// ApplyProposals applies approved proposals against a working transcript copy.
+// Proposals are applied in ascending proposal_index order. Later proposals run
+// against segment text produced by earlier proposals.
+func ApplyProposals(
+	transcript *schema.Transcript,
+	proposals []EnrichedCorrectionProposal,
+	policy ReplacementPolicy,
+) ApplyResult {
+	working := cloneTranscript(transcript)
+	ordered := sortProposalsByIndex(proposals)
+
+	applied := make([]AppliedChange, 0, len(ordered))
+	skipped := make([]SkippedChange, 0)
+
+	segmentIndexByID := make(map[int]int, len(working.Segments))
+	for i, seg := range working.Segments {
+		segmentIndexByID[seg.ID] = i
+	}
+
+	for _, proposal := range ordered {
+		segmentIdx, found := segmentIndexByID[proposal.TargetSegmentID]
+		if !found {
+			skipped = append(skipped, newSkippedChange(proposal, SkipReasonMissingSegment))
+			continue
+		}
+
+		preview := PreviewProposalForSegment(&working.Segments[segmentIdx], proposal.CorrectionProposal, policy)
+		if !preview.Applicable {
+			skipped = append(skipped, newSkippedChange(proposal, preview.SkipReason))
+			continue
+		}
+
+		working.Segments[segmentIdx].Text = preview.CorrectedSegmentText
+		applied = append(applied, AppliedChange{
+			ProposalIndex:    proposal.ProposalIndex,
+			ModuleKey:        proposal.ModuleKey,
+			ModuleInstance:   proposal.ModuleInstance,
+			TargetSegmentID:  proposal.TargetSegmentID,
+			OriginalText:     proposal.OriginalText,
+			CorrectedText:    proposal.CorrectedText,
+			ReplacementCount: preview.ReplacementCount,
+		})
+	}
+
+	return ApplyResult{
+		Transcript: working,
+		Applied:    applied,
+		Skipped:    skipped,
+	}
+}
+
+func sortProposalsByIndex(proposals []EnrichedCorrectionProposal) []EnrichedCorrectionProposal {
+	ordered := make([]EnrichedCorrectionProposal, len(proposals))
+	copy(ordered, proposals)
+
+	sort.SliceStable(ordered, func(i, j int) bool {
+		return ordered[i].ProposalIndex < ordered[j].ProposalIndex
+	})
+
+	return ordered
+}
+
+func cloneTranscript(t *schema.Transcript) *schema.Transcript {
+	if t == nil {
+		return &schema.Transcript{}
+	}
+
+	segments := make([]schema.Segment, len(t.Segments))
+	for i, s := range t.Segments {
+		var categories []string
+		if s.Categories != nil {
+			categories = make([]string, len(s.Categories))
+			copy(categories, s.Categories)
+		}
+
+		segments[i] = schema.Segment{
+			ID:         s.ID,
+			Speaker:    s.Speaker,
+			Start:      s.Start,
+			End:        s.End,
+			Text:       s.Text,
+			Categories: categories,
+		}
+	}
+
+	return &schema.Transcript{Segments: segments}
+}
+
+func newSkippedChange(proposal EnrichedCorrectionProposal, reason ProposalSkipReason) SkippedChange {
+	return SkippedChange{
+		ProposalIndex:   proposal.ProposalIndex,
+		ModuleKey:       proposal.ModuleKey,
+		ModuleInstance:  proposal.ModuleInstance,
+		TargetSegmentID: proposal.TargetSegmentID,
+		OriginalText:    proposal.OriginalText,
+		CorrectedText:   proposal.CorrectedText,
+		SkipReason:      reason,
+		Message:         skipReasonMessage(reason),
+	}
+}
+
+func skipReasonMessage(reason ProposalSkipReason) string {
+	switch reason {
+	case SkipReasonMissingSegment:
+		return "target segment was not found"
+	case SkipReasonMissingOriginalText:
+		return "original_text was not found in current segment text"
+	case SkipReasonAmbiguousOriginal:
+		return "original_text matched multiple spans under require_unique policy"
+	case SkipReasonNoEffect:
+		return "original_text and corrected_text are identical"
+	case SkipReasonInvalidProposal:
+		return "proposal failed structural or policy validation"
+	default:
+		return "proposal could not be applied safely"
+	}
+}