agola/internal/objectstorage/posixflat/posixflat.go
Simone Gotti b3867fb7ca objectstorage: add posix standard storage
rename the previous posix storage to posixflat and make it currently not user
selectable (since I'm not sure it's really worth using it).

The new posix storage uses the filesystem without any escaping so it's not a
real flat namespace.

This isn't a real issue since also minio is not a flat namespace and we are so
forced to use it like a hierarchycal filesystem.
2019-05-21 15:17:53 +02:00

436 lines
9.2 KiB
Go

// Copyright 2019 Sorint.lab
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied
// See the License for the specific language governing permissions and
// limitations under the License.
package posixflat
import (
"io"
"os"
"path"
"path/filepath"
"strconv"
"strings"
"unicode/utf8"
"github.com/sorintlab/agola/internal/objectstorage/common"
"github.com/sorintlab/agola/internal/objectstorage/types"
"github.com/pkg/errors"
)
const (
dataDirName = "data"
tmpDirName = "tmp"
splitLength = 8
)
func shouldEscape(c rune) bool {
return c == '/' || c == '%'
}
// escape does percent encoding to '/' and adds a slash every 8 (of the original
// string) chars
func escape(s string) string {
sepCount, hexCount := 0, 0
nc := 0
for _, c := range s {
nc++
if shouldEscape(c) {
hexCount++
}
if nc%splitLength == 0 {
sepCount++
}
}
if sepCount == 0 && hexCount == 0 {
return s
}
hasFileMarker := nc%splitLength == 0
l := len(s) + sepCount + 2*hexCount
// if the string length is a multiple of 8 then we have to add a file marker
// ".f" to not ovverride a possible directory in our fs representation
if hasFileMarker {
l++
}
t := make([]byte, l)
j := 0
nc = 0
for _, c := range s {
nc++
switch {
case shouldEscape(c):
t[j] = '%'
t[j+1] = "0123456789ABCDEF"[c>>4]
t[j+2] = "0123456789ABCDEF"[c&15]
j += 3
default:
s := string(c)
for i := 0; i < len(s); i++ {
t[j] = s[i]
j++
}
}
if nc%splitLength == 0 {
t[j] = '/'
j++
}
}
// add file marker
if hasFileMarker {
t[j-1] = '.'
t[j] = 'f'
}
return string(t)
}
func ishex(c byte) bool {
switch {
case '0' <= c && c <= '9':
return true
case 'a' <= c && c <= 'f':
return true
case 'A' <= c && c <= 'F':
return true
}
return false
}
func unhex(c byte) byte {
switch {
case '0' <= c && c <= '9':
return c - '0'
case 'a' <= c && c <= 'f':
return c - 'a' + 10
case 'A' <= c && c <= 'F':
return c - 'A' + 10
}
return 0
}
type EscapeError string
func (e EscapeError) Error() string {
return "invalid URL escape " + strconv.Quote(string(e))
}
func unescape(s string) (string, bool, error) {
// number of percent encoded
n := 0
// number of slashes
ns := 0
// number of char in the unescaped string
nc := 0
for i := 0; i < len(s); {
r, width := utf8.DecodeRuneInString(s[i:])
if r == utf8.RuneError {
return "", false, errors.Errorf("bad UTF-8 string")
}
switch r {
case '%':
n++
if i+2 >= len(s) || !ishex(s[i+1]) || !ishex(s[i+2]) {
s = s[i:]
if len(s) > 3 {
s = s[:3]
}
return "", false, EscapeError(s)
}
i += 3
nc++
case '/':
ns++
if nc%splitLength != 0 {
return "", false, EscapeError(s)
}
i++
default:
i += width
nc++
}
}
// check and remove trailing file marker
hasFileMarker := false
if nc > splitLength && nc%splitLength == 2 && s[len(s)-2:] == ".f" {
hasFileMarker = true
s = s[:len(s)-2]
nc -= 2
}
if n == 0 && ns == 0 {
return s, hasFileMarker, nil
}
// destination string is
// the length of the escaped one (with the ending file marker already removed) - number of percent * 2 - number os slashes
t := make([]byte, len(s)-n*2-ns)
j := 0
for i := 0; i < len(s); {
r, width := utf8.DecodeRuneInString(s[i:])
if r == utf8.RuneError {
return "", false, errors.Errorf("bad UTF-8 string")
}
switch r {
case '%':
t[j] = unhex(s[i+1])<<4 | unhex(s[i+2])
j++
i += 3
case '/':
// skip "/"
i++
default:
for k := 0; k < width; k++ {
t[j] = s[i]
j++
i++
}
}
}
return string(t), hasFileMarker, nil
}
type PosixFlatStorage struct {
dataDir string
tmpDir string
}
func New(baseDir string) (*PosixFlatStorage, error) {
if err := os.MkdirAll(baseDir, 0770); err != nil {
return nil, err
}
dataDir := filepath.Join(baseDir, dataDirName)
tmpDir := filepath.Join(baseDir, tmpDirName)
if err := os.MkdirAll(dataDir, 0770); err != nil {
return nil, errors.Wrapf(err, "failed to create data dir")
}
if err := os.MkdirAll(tmpDir, 0770); err != nil {
return nil, errors.Wrapf(err, "failed to create tmp dir")
}
return &PosixFlatStorage{
dataDir: dataDir,
tmpDir: tmpDir,
}, nil
}
func (s *PosixFlatStorage) fsPath(p string) (string, error) {
if p == "" {
return "", errors.Errorf("empty key name")
}
return filepath.Join(s.dataDir, escape(p)), nil
}
func (s *PosixFlatStorage) Stat(p string) (*types.ObjectInfo, error) {
fspath, err := s.fsPath(p)
if err != nil {
return nil, err
}
fi, err := os.Stat(fspath)
if err != nil {
if os.IsNotExist(err) {
return nil, types.ErrNotExist
}
return nil, err
}
return &types.ObjectInfo{Path: p, LastModified: fi.ModTime()}, nil
}
func (s *PosixFlatStorage) ReadObject(p string) (types.ReadSeekCloser, error) {
fspath, err := s.fsPath(p)
if err != nil {
return nil, err
}
f, err := os.Open(fspath)
if err != nil && os.IsNotExist(err) {
return nil, types.ErrNotExist
}
return f, err
}
func (s *PosixFlatStorage) WriteObject(p string, data io.Reader, size int64, persist bool) error {
fspath, err := s.fsPath(p)
if err != nil {
return err
}
if err := os.MkdirAll(path.Dir(fspath), 0770); err != nil {
return err
}
return common.WriteFileAtomicFunc(fspath, s.dataDir, s.tmpDir, 0660, persist, func(f io.Writer) error {
_, err := io.Copy(f, data)
return err
})
}
func (s *PosixFlatStorage) DeleteObject(p string) error {
fspath, err := s.fsPath(p)
if err != nil {
return err
}
if err := os.Remove(fspath); err != nil {
if os.IsNotExist(err) {
return types.ErrNotExist
}
return err
}
// try to remove parent empty dirs
// TODO(sgotti) if this fails we ignore errors and the dirs will be left as
// empty, clean them asynchronously
pdir := filepath.Dir(fspath)
for {
if pdir == s.dataDir || !strings.HasPrefix(pdir, s.dataDir) {
break
}
f, err := os.Open(pdir)
if err != nil {
return nil
}
_, err = f.Readdirnames(1)
if err == io.EOF {
f.Close()
if err := os.Remove(pdir); err != nil {
return nil
}
} else {
f.Close()
break
}
pdir = filepath.Dir(pdir)
}
return nil
}
func (s *PosixFlatStorage) List(prefix, startWith, delimiter string, doneCh <-chan struct{}) <-chan types.ObjectInfo {
objectCh := make(chan types.ObjectInfo, 1)
if len(delimiter) > 1 {
objectCh <- types.ObjectInfo{Err: errors.Errorf("wrong delimiter %q", delimiter)}
return objectCh
}
if startWith != "" && !strings.Contains(startWith, prefix) {
objectCh <- types.ObjectInfo{Err: errors.Errorf("wrong startwith value %q for prefix %q", startWith, prefix)}
return objectCh
}
recursive := delimiter == ""
// remove leading slash from prefix
if strings.HasPrefix(prefix, "/") {
prefix = strings.TrimPrefix(prefix, "/")
}
fprefix := filepath.Join(s.dataDir, escape(prefix))
root := filepath.Dir(fprefix)
if len(root) < len(s.dataDir) {
root = s.dataDir
}
// remove leading slash
if strings.HasPrefix(startWith, "/") {
startWith = strings.TrimPrefix(startWith, "/")
}
go func(objectCh chan<- types.ObjectInfo) {
var prevp string
defer close(objectCh)
err := filepath.Walk(root, func(ep string, info os.FileInfo, err error) error {
if err != nil && !os.IsNotExist(err) {
return err
}
if os.IsNotExist(err) {
return nil
}
p := ep
// get the path with / separator
p = filepath.ToSlash(p)
p, err = filepath.Rel(s.dataDir, p)
if err != nil {
return err
}
p, _, err = unescape(p)
if err != nil {
return err
}
if !recursive && len(p) > len(prefix) {
rel := strings.TrimPrefix(p, prefix)
skip := strings.Contains(rel, delimiter)
if info.IsDir() && skip {
return filepath.SkipDir
}
if skip {
return nil
}
}
// don't list dirs if there's not a file with the same name (with filemarker)
// it's not an issue if the file in the meantime has been removed, it won't
// just be listed
hasFile := true
_, err = os.Stat(ep + ".f")
if err != nil && !os.IsNotExist(err) {
return err
}
if os.IsNotExist(err) {
hasFile = false
}
if info.IsDir() && !hasFile {
return nil
}
if strings.HasPrefix(p, prefix) && p > startWith {
// skip keys smaller than the previously returned one. This happens when we
// receive a file with a file marker that we already returned previously
// when we received a dir with the same name
// it'not an issue if the dir has been removed since we already returned the file
if p > prevp {
select {
// Send object content.
case objectCh <- types.ObjectInfo{Path: p, LastModified: info.ModTime()}:
// If receives done from the caller, return here.
case <-doneCh:
return io.EOF
}
}
prevp = p
}
return nil
})
if err != nil && err != io.EOF {
objectCh <- types.ObjectInfo{
Err: err,
}
return
}
}(objectCh)
return objectCh
}