Update golangci-lint to v2.12.2 with canonical config (closes #60)
All checks were successful
check / check (push) Successful in 35s

- Add canonical .golangci.yml (v2 schema, default: all, project
  thresholds for lll/funlen/cyclop/dupl)
- Bump golangci-lint pins from v2.0.2 to v2.12.2 in Makefile
  (go install, new /v2 module path) and Dockerfile (tagged+digest
  Debian image pin)
- Fix all lint findings surfaced by the new linter set across
  cmd/mfer, internal/bork, internal/cli, internal/log, and mfer:
  static sentinel errors (err113), context-aware HTTP and exec
  (noctx), guarded integer conversions and stricter permissions
  (gosec), named constants (mnd, goconst), function decomposition
  (funlen, cyclop, gocognit, nestif), declaration ordering
  (funcorder), t.Parallel/t.TempDir/t.Setenv adoption in tests
  (paralleltest, usetesting), protobuf getters (protogetter), plus
  formatting and style cleanups (wsl_v5, nlreturn, lll, revive,
  testifylint, and others)
- Serialize CLI runs in tests behind a mutex so parallel tests do
  not cross-wire the process-global logger's captured output

The decompositions are behavior-preserving. In particular:

- REPO_POLICIES.md is untouched and stays byte-identical to the
  authoritative copy in the prompts repo
- the mfer.manifest type stays unexported; whether to export it is an
  open owner design question (README question 13)
- directories created by fetch keep mode 0755, because fetched trees
  are content meant to be readable by other uids
- an absent MFFilePath.Mtime is handled explicitly and identically in
  freshen, list, and export rather than being read as the Unix epoch,
  which would classify every entry as changed and rewrite the manifest
  on every freshen
- every user-visible error message renders byte-identically to what it
  did before, with the err113 sentinels wrapped mid-sentence where
  needed; the rendered strings are now pinned by tests

Also fixes an argument-injection defect the lint pass surfaced: key IDs
reach gpg as bare positional arguments, so a key ID beginning with "-"
was parsed by gpg as an option. All positional arguments now follow an
explicit "--" end-of-options marker.

The symlink-escape gap in fetch's path handling, which sanitizePath
does not and cannot address, is filed separately as #86.
This commit is contained in:
2026-08-09 02:16:37 +00:00
parent 6d19de74e7
commit 3bfbb3fbe2
40 changed files with 3999 additions and 1799 deletions

View File

@@ -1,6 +1,7 @@
package cli
import (
"errors"
"fmt"
"os"
"os/signal"
@@ -16,9 +17,78 @@ import (
"sneak.berlin/go/mfer/mfer"
)
func (mfa *CLIApp) generateManifestOperation(ctx *cli.Context) error {
log.Debug("generateManifestOperation()")
var (
// errPathNotExist indicates an input path that does not exist.
errPathNotExist = errors.New("path does not exist")
// errOutputExists indicates the output file already exists and
// --force was not given. It is wrapped mid-sentence so that the
// rendered message stays exactly as mfer has always printed it.
errOutputExists = errors.New(
"already exists (use --force to overwrite)")
)
// reportEnumProgress renders enumeration progress until the channel
// closes.
func reportEnumProgress(progress <-chan mfer.EnumerateStatus, wg *sync.WaitGroup) {
defer wg.Done()
for status := range progress {
log.Progressf("Enumerating: %d files, %s",
status.FilesFound,
humanize.IBytes(safeUint64(int64(status.BytesFound))))
}
log.ProgressDone()
}
// reportScanProgress renders scan progress until the channel closes.
func reportScanProgress(progress <-chan mfer.ScanStatus, wg *sync.WaitGroup) {
defer wg.Done()
for status := range progress {
if status.ETA > 0 {
log.Progressf("Scanning: %d/%d files, %s/s, ETA %s",
status.ScannedFiles,
status.TotalFiles,
humanize.IBytes(safeRateUint64(status.BytesPerSec)),
status.ETA.Round(time.Second))
} else {
log.Progressf("Scanning: %d/%d files, %s/s",
status.ScannedFiles,
status.TotalFiles,
humanize.IBytes(safeRateUint64(status.BytesPerSec)))
}
}
log.ProgressDone()
}
// collectInputPaths validates the input path arguments and returns them
// as absolute paths.
func (mfa *CLIApp) collectInputPaths(args cli.Args) ([]string, error) {
paths := make([]string, 0, args.Len())
for i := range args.Len() {
inputPath := args.Get(i)
ap, err := filepath.Abs(inputPath)
if err != nil {
return nil, fmt.Errorf("generate: invalid path %q: %w", inputPath, err)
}
// Validate path exists before adding to list
if exists, _ := afero.Exists(mfa.Fs, ap); !exists {
return nil, fmt.Errorf("%w: %s", errPathNotExist, inputPath)
}
log.Debugf("enumerating path: %s", ap)
paths = append(paths, ap)
}
return paths, nil
}
// buildScannerOptions constructs scanner options from the CLI flags.
func (mfa *CLIApp) buildScannerOptions(ctx *cli.Context) *mfer.ScannerOptions {
opts := &mfer.ScannerOptions{
IncludeDotfiles: ctx.Bool("include-dotfiles"),
FollowSymLinks: ctx.Bool("follow-symlinks"),
@@ -29,6 +99,7 @@ func (mfa *CLIApp) generateManifestOperation(ctx *cli.Context) error {
// Set seed for deterministic UUID if provided
if seed := ctx.String("seed"); seed != "" {
opts.Seed = seed
log.Infof("using deterministic seed for manifest UUID")
}
@@ -40,136 +111,167 @@ func (mfa *CLIApp) generateManifestOperation(ctx *cli.Context) error {
log.Infof("signing manifest with GPG key: %s", signKey)
}
s := mfer.NewScannerWithOptions(opts)
// Phase 1: Enumeration - collect paths and stat files
args := ctx.Args()
showProgress := ctx.Bool("progress")
// Set up enumeration progress reporting
var enumProgress chan mfer.EnumerateStatus
var enumWg sync.WaitGroup
if showProgress {
enumProgress = make(chan mfer.EnumerateStatus, 1)
enumWg.Add(1)
go func() {
defer enumWg.Done()
for status := range enumProgress {
log.Progressf("Enumerating: %d files, %s",
status.FilesFound,
humanize.IBytes(uint64(status.BytesFound)))
}
log.ProgressDone()
}()
}
return opts
}
// enumerateInputs runs the enumeration phase over the argument paths,
// or the current directory when no arguments are given.
func (mfa *CLIApp) enumerateInputs(
s *mfer.Scanner, args cli.Args, enumProgress chan mfer.EnumerateStatus,
) error {
if args.Len() == 0 {
// Default to current directory
if err := s.EnumeratePath(".", enumProgress); err != nil {
return fmt.Errorf("generate: failed to enumerate current directory: %w", err)
}
} else {
// Collect and validate all paths first
paths := make([]string, 0, args.Len())
for i := 0; i < args.Len(); i++ {
inputPath := args.Get(i)
ap, err := filepath.Abs(inputPath)
if err != nil {
return fmt.Errorf("generate: invalid path %q: %w", inputPath, err)
}
// Validate path exists before adding to list
if exists, _ := afero.Exists(mfa.Fs, ap); !exists {
return fmt.Errorf("path does not exist: %s", inputPath)
}
log.Debugf("enumerating path: %s", ap)
paths = append(paths, ap)
}
if err := s.EnumeratePaths(enumProgress, paths...); err != nil {
return fmt.Errorf("generate: failed to enumerate paths: %w", err)
err := s.EnumeratePath(".", enumProgress)
if err != nil {
return fmt.Errorf(
"generate: failed to enumerate current directory: %w", err)
}
return nil
}
// Collect and validate all paths first
paths, err := mfa.collectInputPaths(args)
if err != nil {
return err
}
err = s.EnumeratePaths(enumProgress, paths...)
if err != nil {
return fmt.Errorf("generate: failed to enumerate paths: %w", err)
}
return nil
}
// cleanupOnSignal installs a handler that removes the temp output file
// and exits when the process is interrupted. It returns the signal
// channel so the caller can stop and close it when done.
func (mfa *CLIApp) cleanupOnSignal(outFile afero.File, tmpPath string) chan os.Signal {
sigChan := make(chan os.Signal, 1)
signal.Notify(sigChan, os.Interrupt, syscall.SIGTERM)
go func() {
sig, ok := <-sigChan
if !ok || sig == nil {
return // Channel closed normally, not a signal
}
_ = outFile.Close()
_ = mfa.Fs.Remove(tmpPath)
os.Exit(1)
}()
return sigChan
}
// runEnumeratePhase enumerates all input paths with optional progress
// reporting and logs the totals.
func (mfa *CLIApp) runEnumeratePhase(ctx *cli.Context, s *mfer.Scanner) error {
// Set up enumeration progress reporting
var (
enumProgress chan mfer.EnumerateStatus
enumWg sync.WaitGroup
)
if ctx.Bool("progress") {
enumProgress = make(chan mfer.EnumerateStatus, 1)
enumWg.Add(1)
go reportEnumProgress(enumProgress, &enumWg)
}
err := mfa.enumerateInputs(s, ctx.Args(), enumProgress)
if err != nil {
return err
}
enumWg.Wait()
log.Infof("enumerated %d files, %s total", s.FileCount(), humanize.IBytes(uint64(s.TotalBytes())))
log.Infof("enumerated %d files, %s total", s.FileCount(),
humanize.IBytes(safeUint64(int64(s.TotalBytes()))))
return nil
}
func (mfa *CLIApp) generateManifestOperation(ctx *cli.Context) error {
log.Debug("generateManifestOperation()")
s := mfer.NewScannerWithOptions(mfa.buildScannerOptions(ctx))
// Phase 1: Enumeration - collect paths and stat files
err := mfa.runEnumeratePhase(ctx, s)
if err != nil {
return err
}
showProgress := ctx.Bool("progress")
// Check if output file exists
outputPath := ctx.String("output")
if exists, _ := afero.Exists(mfa.Fs, outputPath); exists {
if !ctx.Bool("force") {
return fmt.Errorf("output file %s already exists (use --force to overwrite)", outputPath)
}
if exists, _ := afero.Exists(mfa.Fs, outputPath); exists && !ctx.Bool("force") {
return fmt.Errorf("output file %s %w", outputPath, errOutputExists)
}
// Create temp file for atomic write
tmpPath := outputPath + ".tmp"
outFile, err := mfa.Fs.Create(tmpPath)
if err != nil {
return fmt.Errorf("failed to create temp file: %w", err)
}
// Set up signal handler to clean up temp file on Ctrl-C
sigChan := make(chan os.Signal, 1)
signal.Notify(sigChan, os.Interrupt, syscall.SIGTERM)
go func() {
sig, ok := <-sigChan
if !ok || sig == nil {
return // Channel closed normally, not a signal
}
_ = outFile.Close()
_ = mfa.Fs.Remove(tmpPath)
os.Exit(1)
}()
sigChan := mfa.cleanupOnSignal(outFile, tmpPath)
// Clean up temp file on any error or interruption
success := false
defer func() {
signal.Stop(sigChan)
close(sigChan)
_ = outFile.Close()
if !success {
_ = mfa.Fs.Remove(tmpPath)
}
}()
// Phase 2: Scan - read file contents and generate manifest
var scanProgress chan mfer.ScanStatus
var scanWg sync.WaitGroup
var (
scanProgress chan mfer.ScanStatus
scanWg sync.WaitGroup
)
if showProgress {
scanProgress = make(chan mfer.ScanStatus, 1)
scanWg.Add(1)
go func() {
defer scanWg.Done()
for status := range scanProgress {
if status.ETA > 0 {
log.Progressf("Scanning: %d/%d files, %s/s, ETA %s",
status.ScannedFiles,
status.TotalFiles,
humanize.IBytes(uint64(status.BytesPerSec)),
status.ETA.Round(time.Second))
} else {
log.Progressf("Scanning: %d/%d files, %s/s",
status.ScannedFiles,
status.TotalFiles,
humanize.IBytes(uint64(status.BytesPerSec)))
}
}
log.ProgressDone()
}()
go reportScanProgress(scanProgress, &scanWg)
}
err = s.ToManifest(ctx.Context, outFile, scanProgress)
scanWg.Wait()
if err != nil {
return fmt.Errorf("failed to generate manifest: %w", err)
}
// Close file before rename to ensure all data is flushed
if err := outFile.Close(); err != nil {
err = outFile.Close()
if err != nil {
return fmt.Errorf("failed to close temp file: %w", err)
}
// Atomic rename
if err := mfa.Fs.Rename(tmpPath, outputPath); err != nil {
err = mfa.Fs.Rename(tmpPath, outputPath)
if err != nil {
return fmt.Errorf("failed to rename temp file: %w", err)
}
@@ -177,7 +279,9 @@ func (mfa *CLIApp) generateManifestOperation(ctx *cli.Context) error {
elapsed := time.Since(mfa.startupTime).Seconds()
rate := float64(s.TotalBytes()) / elapsed
log.Infof("wrote %d files (%s) to %s in %.1fs (%s/s)", s.FileCount(), humanize.IBytes(uint64(s.TotalBytes())), outputPath, elapsed, humanize.IBytes(uint64(rate)))
log.Infof("wrote %d files (%s) to %s in %.1fs (%s/s)", s.FileCount(),
humanize.IBytes(safeUint64(int64(s.TotalBytes()))), outputPath, elapsed,
humanize.IBytes(safeRateUint64(rate)))
return nil
}