stash/pkg/scene/scan.go
SmallCoccinelle 401660e6a3
Hoist context, enable errchkjson (#2488)
* Make the script scraper context-aware

Connect the context to the command execution. This means command
execution can be aborted if the context is canceled. The context is
usually bound to user-interaction, i.e., a scraper operation issued
by the user. Hence, it seems correct to abort a command if the user
aborts.

* Enable errchkjson

Some json marshal calls are *safe* in that they can never fail. This is
conditional on the types of the the data being encoded. errchkjson finds
those calls which are unsafe, and also not checked for errors.

Add logging warnings to the place where unsafe encodings might happen.
This can help uncover usage bugs early in stash if they are tripped,
making debugging easier.

While here, keep the checker enabled in the linter to capture future
uses of json marshalling.

* Pass the context for zip file scanning.

* Pass the context in scanning

* Pass context, replace context.TODO()

Where applicable, pass the context down toward the lower functions in
the call stack. Replace uses of context.TODO() with the passed context.

This makes the code more context-aware, and you can rely on aborting
contexts to clean up subsystems to a far greater extent now.

I've left the cases where there is a context in a struct. My gut feeling
is that they have solutions that are nice, but they require more deep
thinking to unveil how to handle it.

* Remove context from task-structs

As a rule, contexts are better passed explicitly to functions than they
are passed implicitly via structs. In the case of tasks, we already
have a valid context in scope when creating the struct, so remove ctx
from the struct and use the scoped context instead.

With this change it is clear that the scanning functions are under a
context, and the task-starting caller has jurisdiction over the context
and its lifetime. A reader of the code don't have to figure out where
the context are coming from anymore.

While here, connect context.TODO() to the newly scoped context in most
of the scan code.

* Remove context from autotag struct too

* Make more context-passing explicit

In all of these cases, there is an applicable context which is close
in the call-tree. Hook up to this context.

* Simplify context passing in manager

The managers context handling generally wants to use an outer context
if applicable. However, the code doesn't pass it explicitly, but stores
it in a struct. Pull out the context from the struct and use it to
explicitly pass it.

At a later point in time, we probably want to handle this by handing
over the job to a different (program-lifetime) context for background
jobs, but this will do for a start.
2022-04-15 11:34:53 +10:00

334 lines
9.3 KiB
Go

package scene
import (
"context"
"database/sql"
"fmt"
"os"
"strconv"
"strings"
"time"
"github.com/stashapp/stash/pkg/ffmpeg"
"github.com/stashapp/stash/pkg/file"
"github.com/stashapp/stash/pkg/fsutil"
"github.com/stashapp/stash/pkg/logger"
"github.com/stashapp/stash/pkg/models"
"github.com/stashapp/stash/pkg/models/paths"
"github.com/stashapp/stash/pkg/plugin"
"github.com/stashapp/stash/pkg/utils"
)
const mutexType = "scene"
type videoFileCreator interface {
NewVideoFile(path string, stripFileExtension bool) (*ffmpeg.VideoFile, error)
}
type Scanner struct {
file.Scanner
StripFileExtension bool
UseFileMetadata bool
FileNamingAlgorithm models.HashAlgorithm
CaseSensitiveFs bool
TxnManager models.TransactionManager
Paths *paths.Paths
Screenshotter screenshotter
VideoFileCreator videoFileCreator
PluginCache *plugin.Cache
MutexManager *utils.MutexManager
}
func FileScanner(hasher file.Hasher, fileNamingAlgorithm models.HashAlgorithm, calculateMD5 bool) file.Scanner {
return file.Scanner{
Hasher: hasher,
CalculateOSHash: true,
CalculateMD5: fileNamingAlgorithm == models.HashAlgorithmMd5 || calculateMD5,
}
}
func (scanner *Scanner) ScanExisting(ctx context.Context, existing file.FileBased, file file.SourceFile) (err error) {
scanned, err := scanner.Scanner.ScanExisting(existing, file)
if err != nil {
return err
}
s := existing.(*models.Scene)
path := scanned.New.Path
interactive := getInteractive(path)
oldHash := s.GetHash(scanner.FileNamingAlgorithm)
changed := false
var videoFile *ffmpeg.VideoFile
if scanned.ContentsChanged() {
logger.Infof("%s has been updated: rescanning", path)
s.SetFile(*scanned.New)
videoFile, err = scanner.VideoFileCreator.NewVideoFile(path, scanner.StripFileExtension)
if err != nil {
return err
}
videoFileToScene(s, videoFile)
changed = true
} else if scanned.FileUpdated() || s.Interactive != interactive {
logger.Infof("Updated scene file %s", path)
// update fields as needed
s.SetFile(*scanned.New)
changed = true
}
// check for container
if !s.Format.Valid {
if videoFile == nil {
videoFile, err = scanner.VideoFileCreator.NewVideoFile(path, scanner.StripFileExtension)
if err != nil {
return err
}
}
container := ffmpeg.MatchContainer(videoFile.Container, path)
logger.Infof("Adding container %s to file %s", container, path)
s.Format = models.NullString(string(container))
changed = true
}
if changed {
// we are operating on a checksum now, so grab a mutex on the checksum
done := make(chan struct{})
if scanned.New.OSHash != "" {
scanner.MutexManager.Claim(mutexType, scanned.New.OSHash, done)
}
if scanned.New.Checksum != "" {
scanner.MutexManager.Claim(mutexType, scanned.New.Checksum, done)
}
if err := scanner.TxnManager.WithTxn(ctx, func(r models.Repository) error {
defer close(done)
qb := r.Scene()
// ensure no clashes of hashes
if scanned.New.Checksum != "" && scanned.Old.Checksum != scanned.New.Checksum {
dupe, _ := qb.FindByChecksum(s.Checksum.String)
if dupe != nil {
return fmt.Errorf("MD5 for file %s is the same as that of %s", path, dupe.Path)
}
}
if scanned.New.OSHash != "" && scanned.Old.OSHash != scanned.New.OSHash {
dupe, _ := qb.FindByOSHash(scanned.New.OSHash)
if dupe != nil {
return fmt.Errorf("OSHash for file %s is the same as that of %s", path, dupe.Path)
}
}
s.Interactive = interactive
s.UpdatedAt = models.SQLiteTimestamp{Timestamp: time.Now()}
_, err := qb.UpdateFull(*s)
return err
}); err != nil {
return err
}
// Migrate any generated files if the hash has changed
newHash := s.GetHash(scanner.FileNamingAlgorithm)
if newHash != oldHash {
MigrateHash(scanner.Paths, oldHash, newHash)
}
scanner.PluginCache.ExecutePostHooks(ctx, s.ID, plugin.SceneUpdatePost, nil, nil)
}
// We already have this item in the database
// check for thumbnails, screenshots
scanner.makeScreenshots(path, videoFile, s.GetHash(scanner.FileNamingAlgorithm))
return nil
}
func (scanner *Scanner) ScanNew(ctx context.Context, file file.SourceFile) (retScene *models.Scene, err error) {
scanned, err := scanner.Scanner.ScanNew(file)
if err != nil {
return nil, err
}
path := file.Path()
checksum := scanned.Checksum
oshash := scanned.OSHash
// grab a mutex on the checksum and oshash
done := make(chan struct{})
if oshash != "" {
scanner.MutexManager.Claim(mutexType, oshash, done)
}
if checksum != "" {
scanner.MutexManager.Claim(mutexType, checksum, done)
}
defer close(done)
// check for scene by checksum and oshash - MD5 should be
// redundant, but check both
var s *models.Scene
if err := scanner.TxnManager.WithReadTxn(ctx, func(r models.ReaderRepository) error {
qb := r.Scene()
if checksum != "" {
s, _ = qb.FindByChecksum(checksum)
}
if s == nil {
s, _ = qb.FindByOSHash(oshash)
}
return nil
}); err != nil {
return nil, err
}
sceneHash := oshash
if scanner.FileNamingAlgorithm == models.HashAlgorithmMd5 {
sceneHash = checksum
}
interactive := getInteractive(file.Path())
if s != nil {
exists, _ := fsutil.FileExists(s.Path)
if !scanner.CaseSensitiveFs {
// #1426 - if file exists but is a case-insensitive match for the
// original filename, then treat it as a move
if exists && strings.EqualFold(path, s.Path) {
exists = false
}
}
if exists {
logger.Infof("%s already exists. Duplicate of %s", path, s.Path)
} else {
logger.Infof("%s already exists. Updating path...", path)
scenePartial := models.ScenePartial{
ID: s.ID,
Path: &path,
Interactive: &interactive,
}
if err := scanner.TxnManager.WithTxn(ctx, func(r models.Repository) error {
_, err := r.Scene().Update(scenePartial)
return err
}); err != nil {
return nil, err
}
scanner.makeScreenshots(path, nil, sceneHash)
scanner.PluginCache.ExecutePostHooks(ctx, s.ID, plugin.SceneUpdatePost, nil, nil)
}
} else {
logger.Infof("%s doesn't exist. Creating new item...", path)
currentTime := time.Now()
videoFile, err := scanner.VideoFileCreator.NewVideoFile(path, scanner.StripFileExtension)
if err != nil {
return nil, err
}
// Override title to be filename if UseFileMetadata is false
if !scanner.UseFileMetadata {
videoFile.SetTitleFromPath(scanner.StripFileExtension)
}
newScene := models.Scene{
Checksum: sql.NullString{String: checksum, Valid: checksum != ""},
OSHash: sql.NullString{String: oshash, Valid: oshash != ""},
Path: path,
FileModTime: models.NullSQLiteTimestamp{
Timestamp: scanned.FileModTime,
Valid: true,
},
Title: sql.NullString{String: videoFile.Title, Valid: true},
CreatedAt: models.SQLiteTimestamp{Timestamp: currentTime},
UpdatedAt: models.SQLiteTimestamp{Timestamp: currentTime},
Interactive: interactive,
}
videoFileToScene(&newScene, videoFile)
if scanner.UseFileMetadata {
newScene.Details = sql.NullString{String: videoFile.Comment, Valid: true}
_ = newScene.Date.Scan(videoFile.CreationTime)
}
if err := scanner.TxnManager.WithTxn(ctx, func(r models.Repository) error {
var err error
retScene, err = r.Scene().Create(newScene)
return err
}); err != nil {
return nil, err
}
scanner.makeScreenshots(path, videoFile, sceneHash)
scanner.PluginCache.ExecutePostHooks(ctx, retScene.ID, plugin.SceneCreatePost, nil, nil)
}
return retScene, nil
}
func videoFileToScene(s *models.Scene, videoFile *ffmpeg.VideoFile) {
container := ffmpeg.MatchContainer(videoFile.Container, s.Path)
s.Duration = sql.NullFloat64{Float64: videoFile.Duration, Valid: true}
s.VideoCodec = sql.NullString{String: videoFile.VideoCodec, Valid: true}
s.AudioCodec = sql.NullString{String: videoFile.AudioCodec, Valid: true}
s.Format = sql.NullString{String: string(container), Valid: true}
s.Width = sql.NullInt64{Int64: int64(videoFile.Width), Valid: true}
s.Height = sql.NullInt64{Int64: int64(videoFile.Height), Valid: true}
s.Framerate = sql.NullFloat64{Float64: videoFile.FrameRate, Valid: true}
s.Bitrate = sql.NullInt64{Int64: videoFile.Bitrate, Valid: true}
s.Size = sql.NullString{String: strconv.FormatInt(videoFile.Size, 10), Valid: true}
}
func (scanner *Scanner) makeScreenshots(path string, probeResult *ffmpeg.VideoFile, checksum string) {
thumbPath := scanner.Paths.Scene.GetThumbnailScreenshotPath(checksum)
normalPath := scanner.Paths.Scene.GetScreenshotPath(checksum)
thumbExists, _ := fsutil.FileExists(thumbPath)
normalExists, _ := fsutil.FileExists(normalPath)
if thumbExists && normalExists {
return
}
if probeResult == nil {
var err error
probeResult, err = scanner.VideoFileCreator.NewVideoFile(path, scanner.StripFileExtension)
if err != nil {
logger.Error(err.Error())
return
}
logger.Infof("Regenerating images for %s", path)
}
at := float64(probeResult.Duration) * 0.2
if !thumbExists {
logger.Debugf("Creating thumbnail for %s", path)
makeScreenshot(scanner.Screenshotter, *probeResult, thumbPath, 5, 320, at)
}
if !normalExists {
logger.Debugf("Creating screenshot for %s", path)
makeScreenshot(scanner.Screenshotter, *probeResult, normalPath, 2, probeResult.Width, at)
}
}
func getInteractive(path string) bool {
_, err := os.Stat(GetFunscriptPath(path))
return err == nil
}