mithril-go/internal/manifest/manifest.go

// Package manifest verifies downloaded Mithril snapshot files against the
// per-immutable digest manifest extracted from the snapshot's `digests`
// archive.
//
// The manifest is a flat JSON array of {immutable_file_name, digest}
// entries — one per file in the chain DB's immutable directory. Each
// digest is the SHA-256 of the file contents.
package manifest

import (
	"crypto/sha256"
	"encoding/hex"
	"encoding/json"
	"fmt"
	"io"
	"os"
	"path/filepath"
	"sort"
	"strings"
)

// Entry is one record from the digests JSON.
type Entry struct {
	ImmutableFileName string `json:"immutable_file_name"`
	Digest            string `json:"digest"` // hex-encoded SHA-256
}

// Load reads + parses a digests.json file.
func Load(path string) ([]Entry, error) {
	f, err := os.Open(path)
	if err != nil {
		return nil, err
	}
	defer f.Close()
	var out []Entry
	if err := json.NewDecoder(f).Decode(&out); err != nil {
		return nil, err
	}
	return out, nil
}

// LocateDigests finds the digests.json file inside the extracted digests
// directory (named like `<network>-e<epoch>-i<imm>.digests.json`).
func LocateDigests(digestsDir string) (string, error) {
	entries, err := os.ReadDir(digestsDir)
	if err != nil {
		return "", err
	}
	for _, e := range entries {
		if !e.IsDir() && strings.HasSuffix(e.Name(), ".digests.json") {
			return filepath.Join(digestsDir, e.Name()), nil
		}
	}
	return "", fmt.Errorf("no digests.json found under %s", digestsDir)
}

// Result captures the per-file outcome of a manifest verification pass.
type Result struct {
	TotalEntries  int      `json:"total_entries"`
	Verified      int      `json:"verified"`
	Missing       []string `json:"missing,omitempty"`       // files in manifest, not on disk
	Mismatched    []string `json:"mismatched,omitempty"`    // SHA didn't match
	Unverifiable  []string `json:"unverifiable,omitempty"`  // failed to read
	Extra         []string `json:"extra,omitempty"`         // files on disk, not in manifest
	OK            bool     `json:"ok"`
}

// Verify walks the manifest and checks each immutable file under dbDir.
// Files in the immutable subdirectory are at `<dbDir>/immutable/<name>`.
func Verify(entries []Entry, dbDir string) (*Result, error) {
	r := &Result{TotalEntries: len(entries)}
	immDir := filepath.Join(dbDir, "immutable")

	// Build set of expected names for the extra-file diff.
	expected := make(map[string]struct{}, len(entries))
	for _, e := range entries {
		expected[e.ImmutableFileName] = struct{}{}
	}

	for _, e := range entries {
		path := filepath.Join(immDir, e.ImmutableFileName)
		fi, err := os.Stat(path)
		if err != nil {
			if os.IsNotExist(err) {
				r.Missing = append(r.Missing, e.ImmutableFileName)
				continue
			}
			r.Unverifiable = append(r.Unverifiable, e.ImmutableFileName+": "+err.Error())
			continue
		}
		if fi.IsDir() {
			r.Unverifiable = append(r.Unverifiable, e.ImmutableFileName+": is a directory")
			continue
		}
		got, err := sha256OfFile(path)
		if err != nil {
			r.Unverifiable = append(r.Unverifiable, e.ImmutableFileName+": "+err.Error())
			continue
		}
		if got != e.Digest {
			r.Mismatched = append(r.Mismatched, fmt.Sprintf("%s: got %s, want %s",
				e.ImmutableFileName, got, e.Digest))
			continue
		}
		r.Verified++
	}

	// Identify any extra files in immutable/ that aren't in the manifest.
	if entries, err := os.ReadDir(immDir); err == nil {
		for _, fe := range entries {
			if fe.IsDir() {
				continue
			}
			if _, ok := expected[fe.Name()]; !ok {
				r.Extra = append(r.Extra, fe.Name())
			}
		}
	}

	sort.Strings(r.Missing)
	sort.Strings(r.Mismatched)
	sort.Strings(r.Unverifiable)
	sort.Strings(r.Extra)
	r.OK = len(r.Missing) == 0 && len(r.Mismatched) == 0 && len(r.Unverifiable) == 0
	return r, nil
}

func sha256OfFile(path string) (string, error) {
	f, err := os.Open(path)
	if err != nil {
		return "", err
	}
	defer f.Close()
	h := sha256.New()
	if _, err := io.Copy(h, f); err != nil {
		return "", err
	}
	return hex.EncodeToString(h.Sum(nil)), nil
}