review-bot/review/parser.go

package review

import (
	"encoding/json"
	"fmt"
	"strings"
)

// Finding represents a single code review finding.
type Finding struct {
	Severity string `json:"severity"`
	File     string `json:"file"`
	Line     int    `json:"line"`
	Finding  string `json:"finding"`
}

// ReviewResult is the structured output from the LLM.
type ReviewResult struct {
	Verdict        string    `json:"verdict"`
	Summary        string    `json:"summary"`
	Findings       []Finding `json:"findings"`
	Recommendation string    `json:"recommendation"`
}

// ParseResponse parses the LLM response into a ReviewResult.
func ParseResponse(response string) (*ReviewResult, error) {
	// Try to extract JSON from the response — the LLM might wrap it in markdown fences
	cleaned := extractJSON(response)

	var result ReviewResult
	if err := json.Unmarshal([]byte(cleaned), &result); err != nil {
		// LLMs sometimes produce JSON with unescaped quotes inside string values.
		// Try to repair before giving up.
		repaired := repairJSON(cleaned)
		if err2 := json.Unmarshal([]byte(repaired), &result); err2 != nil {
			return nil, fmt.Errorf("parse LLM response as JSON: %w\nRaw response: %s", err, response)
		}
	}

	// Validate verdict
	switch result.Verdict {
	case "APPROVE", "REQUEST_CHANGES":
		// valid
	default:
		return nil, fmt.Errorf("invalid verdict %q (must be APPROVE or REQUEST_CHANGES)", result.Verdict)
	}

	// Validate finding severities
	for i, f := range result.Findings {
		switch f.Severity {
		case "MAJOR", "MINOR", "NIT":
			// valid
		default:
			return nil, fmt.Errorf("finding %d has invalid severity %q", i, f.Severity)
		}
	}

	return &result, nil
}

// extractJSON attempts to pull JSON from a potentially markdown-wrapped response.
func extractJSON(s string) string {
	s = strings.TrimSpace(s)

	// Remove markdown code fences if present
	if strings.HasPrefix(s, "```") {
		lines := strings.Split(s, "\n")
		// Remove first line (```json or ```)
		if len(lines) > 2 {
			lines = lines[1:]
		}
		// Remove last line (```)
		if len(lines) > 0 && strings.TrimSpace(lines[len(lines)-1]) == "```" {
			lines = lines[:len(lines)-1]
		}
		s = strings.Join(lines, "\n")
	}

	s = strings.TrimSpace(s)
	return s
}

// repairJSON attempts to fix common LLM JSON issues:
// - Unescaped double quotes inside string values
//
// Strategy: walk the JSON character by character, tracking whether we're inside
// a string value. When we find a quote that doesn't look like a structural
// delimiter (not preceded by \, not followed by : , } ] or whitespace+structural),
// escape it.
func repairJSON(s string) string {
	var out strings.Builder
	out.Grow(len(s) + 64)

	inString := false
	runes := []rune(s)

	for i := 0; i < len(runes); i++ {
		c := runes[i]

		if !inString {
			out.WriteRune(c)
			if c == '"' {
				inString = true
			}
			continue
		}

		// We're inside a string
		if c == '\\' {
			// Escape sequence — pass through both characters
			out.WriteRune(c)
			if i+1 < len(runes) {
				i++
				out.WriteRune(runes[i])
			}
			continue
		}

		if c == '"' {
			// Is this the end of the string, or an unescaped interior quote?
			// Look ahead: skip whitespace, then check for structural character
			j := i + 1
			for j < len(runes) && (runes[j] == ' ' || runes[j] == '\t' || runes[j] == '\n' || runes[j] == '\r') {
				j++
			}
			if j < len(runes) {
				next := runes[j]
				if next == ',' || next == '}' || next == ']' || next == ':' {
					// Structural — this is really the end of the string
					out.WriteRune(c)
					inString = false
					continue
				}
			} else {
				// End of input — must be closing quote
				out.WriteRune(c)
				inString = false
				continue
			}
			// Not a structural close — this is an unescaped interior quote
			out.WriteRune('\\')
			out.WriteRune('"')
			continue
		}

		out.WriteRune(c)
	}

	return out.String()
}