git-lfs/lfs/gitscanner.go
brian m. carlson 087db1de70
Set package version to v3
Since we're about to do a v3.0.0 release, let's bump the version to v3.

Make this change automatically with the following command to avoid any
missed items:

  git grep -l github.com/git-lfs/git-lfs/v2 | \
  xargs sed -i -e 's!github.com/git-lfs/git-lfs/v2!github.com/git-lfs/git-lfs/v3!g'
2021-09-02 20:41:08 +00:00

312 lines
9.1 KiB
Go

package lfs
import (
"errors"
"fmt"
"sync"
"time"
"github.com/git-lfs/git-lfs/v3/config"
"github.com/git-lfs/git-lfs/v3/filepathfilter"
"github.com/rubyist/tracerx"
)
var missingCallbackErr = errors.New("no callback given")
// IsCallbackMissing returns a boolean indicating whether the error is reporting
// that a GitScanner is missing a required GitScannerCallback.
func IsCallbackMissing(err error) bool {
return err == missingCallbackErr
}
// GitScanner scans objects in a Git repository for LFS pointers.
type GitScanner struct {
Filter *filepathfilter.Filter
FoundPointer GitScannerFoundPointer
FoundLockable GitScannerFoundLockable
PotentialLockables GitScannerSet
remote string
skippedRefs []string
closed bool
started time.Time
mu sync.Mutex
cfg *config.Configuration
}
type GitScannerFoundPointer func(*WrappedPointer, error)
type GitScannerFoundLockable func(filename string)
type GitScannerSet interface {
Contains(string) bool
}
// NewGitScanner initializes a *GitScanner for a Git repository in the current
// working directory.
func NewGitScanner(cfg *config.Configuration, cb GitScannerFoundPointer) *GitScanner {
return &GitScanner{started: time.Now(), FoundPointer: cb, cfg: cfg}
}
// Close stops exits once all processing has stopped, and all resources are
// tracked and cleaned up.
func (s *GitScanner) Close() {
s.mu.Lock()
defer s.mu.Unlock()
if s.closed {
return
}
s.closed = true
tracerx.PerformanceSince("scan", s.started)
}
// RemoteForPush sets up this *GitScanner to scan for objects to push to the
// given remote. Needed for ScanLeftToRemote().
func (s *GitScanner) RemoteForPush(r string) error {
s.mu.Lock()
defer s.mu.Unlock()
if len(s.remote) > 0 && s.remote != r {
return fmt.Errorf("trying to set remote to %q, already set to %q", r, s.remote)
}
s.remote = r
s.skippedRefs = calcSkippedRefs(r)
return nil
}
// ScanRangeToRemote scans through all commits starting at the left ref but not
// including the right ref (if given)that the given remote does not have. See
// RemoteForPush().
func (s *GitScanner) ScanRangeToRemote(left, right string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
s.mu.Lock()
if len(s.remote) == 0 {
s.mu.Unlock()
return fmt.Errorf("unable to scan starting at %q: no remote set", left)
}
s.mu.Unlock()
return scanLeftRightToChan(s, callback, left, right, s.cfg.GitEnv(), s.cfg.OSEnv(), s.opts(ScanRangeToRemoteMode))
}
// ScanMultiRangeToRemote scans through all commits starting at the left ref but
// not including the right ref (if given) that the given remote does not have.
// See RemoteForPush().
func (s *GitScanner) ScanMultiRangeToRemote(left string, rights []string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
s.mu.Lock()
if len(s.remote) == 0 {
s.mu.Unlock()
return fmt.Errorf("unable to scan starting at %q: no remote set", left)
}
s.mu.Unlock()
return scanMultiLeftRightToChan(s, callback, left, rights, s.cfg.GitEnv(), s.cfg.OSEnv(), s.opts(ScanRangeToRemoteMode))
}
// ScanRefs through all commits reachable by refs contained in "include" and
// not reachable by any refs included in "excluded"
func (s *GitScanner) ScanRefs(include, exclude []string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
opts := s.opts(ScanRefsMode)
opts.SkipDeletedBlobs = false
return scanRefsToChan(s, callback, include, exclude, s.cfg.GitEnv(), s.cfg.OSEnv(), opts)
}
// ScanRefRange scans through all commits from the given left and right refs,
// including git objects that have been modified or deleted.
func (s *GitScanner) ScanRefRange(left, right string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
opts := s.opts(ScanRefsMode)
opts.SkipDeletedBlobs = false
return scanLeftRightToChan(s, callback, left, right, s.cfg.GitEnv(), s.cfg.OSEnv(), opts)
}
// ScanRefRangeByTree scans through all trees from the given left and right
// refs.
func (s *GitScanner) ScanRefRangeByTree(left, right string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
opts := s.opts(ScanRefsMode)
opts.SkipDeletedBlobs = false
opts.CommitsOnly = true
return scanRefsByTree(s, callback, []string{right}, []string{left}, s.cfg.GitEnv(), s.cfg.OSEnv(), opts)
}
// ScanRefWithDeleted scans through all objects in the given ref, including
// git objects that have been modified or deleted.
func (s *GitScanner) ScanRefWithDeleted(ref string, cb GitScannerFoundPointer) error {
return s.ScanRefRange(ref, "", cb)
}
// ScanRef scans through all objects in the current ref, excluding git objects
// that have been modified or deleted before the ref.
func (s *GitScanner) ScanRef(ref string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
opts := s.opts(ScanRefsMode)
opts.SkipDeletedBlobs = true
return scanLeftRightToChan(s, callback, ref, "", s.cfg.GitEnv(), s.cfg.OSEnv(), opts)
}
// ScanRefByTree scans through all trees in the current ref.
func (s *GitScanner) ScanRefByTree(ref string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
opts := s.opts(ScanRefsMode)
opts.SkipDeletedBlobs = true
opts.CommitsOnly = true
return scanRefsByTree(s, callback, []string{ref}, []string{}, s.cfg.GitEnv(), s.cfg.OSEnv(), opts)
}
// ScanAll scans through all objects in the git repository.
func (s *GitScanner) ScanAll(cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
opts := s.opts(ScanAllMode)
opts.SkipDeletedBlobs = false
return scanLeftRightToChan(s, callback, "", "", s.cfg.GitEnv(), s.cfg.OSEnv(), opts)
}
// ScanTree takes a ref and returns WrappedPointer objects in the tree at that
// ref. Differs from ScanRefs in that multiple files in the tree with the same
// content are all reported.
func (s *GitScanner) ScanTree(ref string) error {
callback, err := firstGitScannerCallback(s.FoundPointer)
if err != nil {
return err
}
return runScanTree(callback, ref, s.Filter, s.cfg.GitEnv(), s.cfg.OSEnv())
}
// ScanUnpushed scans history for all LFS pointers which have been added but not
// pushed to the named remote. remote can be left blank to mean 'any remote'.
func (s *GitScanner) ScanUnpushed(remote string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
return scanUnpushed(callback, remote)
}
// ScanStashed scans for all LFS pointers referenced solely by a stash
func (s *GitScanner) ScanStashed(cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
return scanStashed(callback, s)
}
// ScanPreviousVersions scans changes reachable from ref (commit) back to since.
// Returns channel of pointers for *previous* versions that overlap that time.
// Does not include pointers which were still in use at ref (use ScanRefsToChan
// for that)
func (s *GitScanner) ScanPreviousVersions(ref string, since time.Time, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
return logPreviousSHAs(callback, ref, since)
}
// ScanIndex scans the git index for modified LFS objects.
func (s *GitScanner) ScanIndex(ref string, cb GitScannerFoundPointer) error {
callback, err := firstGitScannerCallback(cb, s.FoundPointer)
if err != nil {
return err
}
return scanIndex(callback, ref, s.Filter, s.cfg.GitEnv(), s.cfg.OSEnv())
}
func (s *GitScanner) opts(mode ScanningMode) *ScanRefsOptions {
s.mu.Lock()
defer s.mu.Unlock()
opts := newScanRefsOptions()
opts.ScanMode = mode
opts.RemoteName = s.remote
opts.skippedRefs = s.skippedRefs
return opts
}
func firstGitScannerCallback(callbacks ...GitScannerFoundPointer) (GitScannerFoundPointer, error) {
for _, cb := range callbacks {
if cb == nil {
continue
}
return cb, nil
}
return nil, missingCallbackErr
}
type ScanningMode int
const (
ScanRefsMode = ScanningMode(iota) // 0 - or default scan mode
ScanAllMode = ScanningMode(iota)
ScanRangeToRemoteMode = ScanningMode(iota)
)
type ScanRefsOptions struct {
ScanMode ScanningMode
RemoteName string
SkipDeletedBlobs bool
CommitsOnly bool
skippedRefs []string
nameMap map[string]string
mutex *sync.Mutex
}
func (o *ScanRefsOptions) GetName(sha string) (string, bool) {
o.mutex.Lock()
name, ok := o.nameMap[sha]
o.mutex.Unlock()
return name, ok
}
func (o *ScanRefsOptions) SetName(sha, name string) {
o.mutex.Lock()
o.nameMap[sha] = name
o.mutex.Unlock()
}
func newScanRefsOptions() *ScanRefsOptions {
return &ScanRefsOptions{
nameMap: make(map[string]string, 0),
mutex: &sync.Mutex{},
}
}