Files
knowledge-governance/internal/workflows/sourcescan/file_scanner.go
T
zheromiku dcdc8d1ddd feat: 知识治理平台(含已验证发布/回滚契约)
全栈知识治理平台首次入库(P1-P10 + 发布链路模拟验证闭环):

- 治理核心:源目录/变更集/人工审核/发布批次/发布执行(fail-closed 闸口)
- 测试通道:WeKnora 测试 KB 上传/轮询/验收;poll 完成时写入
  parser_generation/index_generation,修复 manual/seed 路径发布闸口永久拦截
- 发布适配器 v071(本地 source=测试角色 / 云端 target=生产角色 模拟验证通过):
  * ES 导出/导入 mapping 自适应(.keyword 子字段、可选 tenant_id)
  * OSS locator 支持嵌套 scheme;chunks.seq_id 由目标序列分配
  * bundle 归档同身份重试原子替换;verifier 按 bundle chunk_type 计数
  * 回滚 job payload 支持 prior_execution_id(strict 解码)
- 三通道发布+回滚演练实测:publish 4/4 verified,检索命中,
  rollback 4/4 rolled_back,目标端 PG is_enabled=false、ES is_enabled=false
- 前端 A·Data-console 风格全站统一,120 用例通过
- 门禁:go vet / go test 全包通过,gofmt 干净
2026-09-30 10:54:06 +08:00

195 lines
5.6 KiB
Go

package sourcescan
import (
"bytes"
"context"
"crypto/sha256"
"encoding/hex"
"encoding/json"
"errors"
"fmt"
"io"
"os"
"path/filepath"
ppconnector "knowledge-governance/internal/connectors/pp"
)
const maxPPMetadataBytes int64 = 64 * 1024 * 1024
type FilePPScannerConfig struct {
AllowedRoot string
LedgerPath string
ManifestPath string
}
type FilePPScanner struct {
root string
ledgerPath string
manifestPath string
}
func NewFilePPScanner(config FilePPScannerConfig) (*FilePPScanner, error) {
root, err := filepath.Abs(config.AllowedRoot)
if err != nil || config.AllowedRoot == "" {
return nil, errors.New("PP scan allowed root is required")
}
resolvedRoot, err := filepath.EvalSymlinks(root)
if err != nil || resolvedRoot != root {
return nil, errors.New("PP scan allowed root must be a canonical directory")
}
info, err := os.Lstat(root)
if err != nil || !info.IsDir() || info.Mode()&os.ModeSymlink != 0 {
return nil, errors.New("PP scan allowed root must be a canonical directory")
}
if config.LedgerPath == "" || config.ManifestPath == "" {
return nil, errors.New("PP ledger and manifest paths are required")
}
return &FilePPScanner{root: root, ledgerPath: config.LedgerPath, manifestPath: config.ManifestPath}, nil
}
func (s *FilePPScanner) Scan(ctx context.Context) (ppconnector.Snapshot, string, error) {
if err := ctx.Err(); err != nil {
return ppconnector.Snapshot{}, "", err
}
ledger, err := s.readFile(s.ledgerPath, maxPPMetadataBytes)
if err != nil {
return ppconnector.Snapshot{}, "", fmt.Errorf("read PP ledger: %w", err)
}
manifest, err := s.readFile(s.manifestPath, maxPPMetadataBytes)
if err != nil {
return ppconnector.Snapshot{}, "", fmt.Errorf("read PP manifest: %w", err)
}
if err := s.validateManifestPaths(manifest); err != nil {
return ppconnector.Snapshot{}, "", err
}
snapshot, err := ppconnector.FoldLedger(bytes.NewReader(ledger))
if err != nil {
return ppconnector.Snapshot{}, "", err
}
snapshot, _, err = ppconnector.RecoverMissingSHA256(snapshot, bytes.NewReader(manifest))
if err != nil {
return ppconnector.Snapshot{}, "", err
}
snapshot, _, err = ppconnector.MergeOverLimitManifest(snapshot, bytes.NewReader(manifest))
if err != nil {
return ppconnector.Snapshot{}, "", err
}
if err := ppconnector.ValidateSnapshot(snapshot); err != nil {
return ppconnector.Snapshot{}, "", err
}
digest := sha256.New()
_, _ = digest.Write(ledger)
_, _ = digest.Write([]byte{0})
_, _ = digest.Write(manifest)
return snapshot, hex.EncodeToString(digest.Sum(nil)), nil
}
func (s *FilePPScanner) validateManifestPaths(body []byte) error {
var manifest struct {
Results []struct {
LocalPath string `json:"local_path"`
Status string `json:"status"`
} `json:"results"`
}
if err := json.Unmarshal(body, &manifest); err != nil {
return fmt.Errorf("decode PP manifest paths: %w", err)
}
for _, item := range manifest.Results {
if item.LocalPath == "" {
return errors.New("PP manifest artifact local_path is required")
}
if item.Status == string(ppconnector.StatusOverLimit) {
if err := s.validateDeclaredPath(item.LocalPath); err != nil {
return fmt.Errorf("validate PP over-limit artifact path: %w", err)
}
file, err := s.openRegular(item.LocalPath)
if errors.Is(err, os.ErrNotExist) {
continue
}
if err != nil {
return fmt.Errorf("validate PP over-limit artifact: %w", err)
}
if err := file.Close(); err != nil {
return errors.New("close PP over-limit artifact")
}
continue
}
file, err := s.openRegular(item.LocalPath)
if err != nil {
return fmt.Errorf("validate PP manifest artifact: %w", err)
}
if err := file.Close(); err != nil {
return errors.New("close PP manifest artifact")
}
}
return nil
}
func (s *FilePPScanner) readFile(path string, maxBytes int64) ([]byte, error) {
file, err := s.openRegular(path)
if err != nil {
return nil, err
}
body, readErr := io.ReadAll(io.LimitReader(file, maxBytes+1))
closeErr := file.Close()
if readErr != nil {
return nil, readErr
}
if closeErr != nil {
return nil, closeErr
}
if int64(len(body)) > maxBytes {
return nil, errors.New("PP metadata file exceeds size limit")
}
return body, nil
}
func (s *FilePPScanner) validateDeclaredPath(path string) error {
absolute, err := filepath.Abs(path)
if err != nil {
return errors.New("invalid PP source path")
}
relative, err := filepath.Rel(s.root, absolute)
if err != nil || relative == ".." || filepath.IsAbs(relative) || len(relative) >= 3 && relative[:3] == ".."+string(filepath.Separator) {
return errors.New("PP source path is outside allowed root")
}
parent := filepath.Dir(absolute)
resolvedParent, err := filepath.EvalSymlinks(parent)
if err != nil || resolvedParent != parent {
return errors.New("PP source path parent must not contain symlinks")
}
return nil
}
func (s *FilePPScanner) openRegular(path string) (*os.File, error) {
if err := s.validateDeclaredPath(path); err != nil {
return nil, err
}
absolute, _ := filepath.Abs(path)
resolved, err := filepath.EvalSymlinks(absolute)
if err != nil {
return nil, err
}
if resolved != absolute {
return nil, errors.New("PP source path must not contain symlinks")
}
before, err := os.Lstat(absolute)
if err != nil {
return nil, err
}
if !before.Mode().IsRegular() || before.Mode()&os.ModeSymlink != 0 {
return nil, errors.New("PP source path must be a regular file")
}
file, err := os.Open(absolute)
if err != nil {
return nil, err
}
after, err := file.Stat()
if err != nil || !after.Mode().IsRegular() || !os.SameFile(before, after) {
_ = file.Close()
return nil, errors.New("PP source file changed while opening")
}
return file, nil
}