rclone/backend/union/entry.go

146 lines
3.6 KiB
Go
Raw Normal View History

package union
import (
"context"
"io"
"sync"
"time"
"github.com/pkg/errors"
"github.com/rclone/rclone/backend/union/upstream"
"github.com/rclone/rclone/fs"
)
// Object describes a union Object
//
// This is a wrapped object which returns the Union Fs as its parent
type Object struct {
*upstream.Object
fs *Fs // what this object is part of
co []upstream.Entry
}
// Directory describes a union Directory
//
// This is a wrapped object contains all candidates
type Directory struct {
*upstream.Directory
cd []upstream.Entry
}
type entry interface {
upstream.Entry
candidates() []upstream.Entry
}
// UnWrap returns the Object that this Object is wrapping or
// nil if it isn't wrapping anything
func (o *Object) UnWrap() *upstream.Object {
return o.Object
}
// Fs returns the union Fs as the parent
func (o *Object) Fs() fs.Info {
return o.fs
}
func (o *Object) candidates() []upstream.Entry {
return o.co
}
func (d *Directory) candidates() []upstream.Entry {
return d.cd
}
// Update in to the object with the modTime given of the given size
//
// When called from outside an Fs by rclone, src.Size() will always be >= 0.
// But for unknown-sized objects (indicated by src.Size() == -1), Upload should either
// return an error or update the object properly (rather than e.g. calling panic).
func (o *Object) Update(ctx context.Context, in io.Reader, src fs.ObjectInfo, options ...fs.OpenOption) error {
entries, err := o.fs.actionEntries(o.candidates()...)
if err != nil {
return err
}
if len(entries) == 1 {
obj := entries[0].(*upstream.Object)
return obj.Update(ctx, in, src, options...)
}
// Multi-threading
readers, errChan := multiReader(len(entries), in)
errs := Errors(make([]error, len(entries)+1))
multithread(len(entries), func(i int) {
if o, ok := entries[i].(*upstream.Object); ok {
err := o.Update(ctx, readers[i], src, options...)
errs[i] = errors.Wrap(err, o.UpstreamFs().Name())
} else {
errs[i] = fs.ErrorNotAFile
}
})
errs[len(entries)] = <-errChan
return errs.Err()
}
// Remove candidate objects selected by ACTION policy
func (o *Object) Remove(ctx context.Context) error {
entries, err := o.fs.actionEntries(o.candidates()...)
if err != nil {
return err
}
errs := Errors(make([]error, len(entries)))
multithread(len(entries), func(i int) {
if o, ok := entries[i].(*upstream.Object); ok {
err := o.Remove(ctx)
errs[i] = errors.Wrap(err, o.UpstreamFs().Name())
} else {
errs[i] = fs.ErrorNotAFile
}
})
return errs.Err()
}
// SetModTime sets the metadata on the object to set the modification date
func (o *Object) SetModTime(ctx context.Context, t time.Time) error {
entries, err := o.fs.actionEntries(o.candidates()...)
if err != nil {
return err
}
var wg sync.WaitGroup
errs := Errors(make([]error, len(entries)))
multithread(len(entries), func(i int) {
if o, ok := entries[i].(*upstream.Object); ok {
err := o.SetModTime(ctx, t)
errs[i] = errors.Wrap(err, o.UpstreamFs().Name())
} else {
errs[i] = fs.ErrorNotAFile
}
})
wg.Wait()
return errs.Err()
}
// ModTime returns the modification date of the directory
// It returns the latest ModTime of all candidates
func (d *Directory) ModTime(ctx context.Context) (t time.Time) {
entries := d.candidates()
times := make([]time.Time, len(entries))
multithread(len(entries), func(i int) {
times[i] = entries[i].ModTime(ctx)
})
for _, ti := range times {
if t.Before(ti) {
t = ti
}
}
return t
}
// Size returns the size of the directory
// It returns the sum of all candidates
func (d *Directory) Size() (s int64) {
for _, e := range d.candidates() {
s += e.Size()
}
return s
}