go-mtree/keywords.go

package mtree

import (
	"archive/tar"
	"crypto/md5"
	"crypto/sha1"
	"crypto/sha256"
	"crypto/sha512"
	"fmt"
	"hash"
	"io"
	"os"
	"strings"

	"golang.org/x/crypto/ripemd160"
)

// KeywordFunc is the type of a function called on each file to be included in
// a DirectoryHierarchy, that will produce the string output of the keyword to
// be included for the file entry. Otherwise, empty string.
// io.Reader `r` is to the file stream for the file payload. While this
// function takes an io.Reader, the caller needs to reset it to the beginning
// for each new KeywordFunc
type KeywordFunc func(path string, info os.FileInfo, r io.Reader) (string, error)

// Keyword is the string name of a keyword, with some convenience functions for
// determining whether it is a default or bsd standard keyword.
type Keyword string

// Default returns whether this keyword is in the default set of keywords
func (k Keyword) Default() bool {
	return inSlice(string(k), DefaultKeywords)
}

// Bsd returns whether this keyword is in the upstream FreeBSD mtree(8)
func (k Keyword) Bsd() bool {
	return inSlice(string(k), BsdKeywords)
}

// KeyVal is a "keyword=value"
type KeyVal string

// Keyword is the mapping to the available keywords
func (kv KeyVal) Keyword() string {
	if !strings.Contains(string(kv), "=") {
		return ""
	}
	chunks := strings.SplitN(strings.TrimSpace(string(kv)), "=", 2)[0]
	if !strings.Contains(chunks, ".") {
		return chunks
	}
	return strings.SplitN(chunks, ".", 2)[0]
}

// KeywordSuffix is really only used for xattr, as the keyword is a prefix to
// the xattr "namespace.key"
func (kv KeyVal) KeywordSuffix() string {
	if !strings.Contains(string(kv), "=") {
		return ""
	}
	chunks := strings.SplitN(strings.TrimSpace(string(kv)), "=", 2)[0]
	if !strings.Contains(chunks, ".") {
		return ""
	}
	return strings.SplitN(chunks, ".", 2)[1]
}

// Value is the data/value portion of "keyword=value"
func (kv KeyVal) Value() string {
	if !strings.Contains(string(kv), "=") {
		return ""
	}
	return strings.SplitN(strings.TrimSpace(string(kv)), "=", 2)[1]
}

// ChangeValue changes the value of a KeyVal
func (kv KeyVal) ChangeValue(newval string) string {
	return fmt.Sprintf("%s=%s", kv.Keyword(), newval)
}

// KeyValEqual returns whether two KeyVals are equivalent. This takes
// care of certain odd cases such as tar_mtime, and should be used over
// using == comparisons directly unless you really know what you're
// doing.
func KeyValEqual(a, b KeyVal) bool {
	// TODO: Implement handling of tar_mtime.
	return a.Keyword() == b.Keyword() && a.Value() == b.Value()
}

// keywordSelector takes an array of "keyword=value" and filters out that only the set of words
func keywordSelector(keyval, words []string) []string {
	retList := []string{}
	for _, kv := range keyval {
		if inSlice(KeyVal(kv).Keyword(), words) {
			retList = append(retList, kv)
		}
	}
	return retList
}

// NewKeyVals constructs a list of KeyVal from the list of strings, like "keyword=value"
func NewKeyVals(keyvals []string) KeyVals {
	kvs := make(KeyVals, len(keyvals))
	for i := range keyvals {
		kvs[i] = KeyVal(keyvals[i])
	}
	return kvs
}

// KeyVals is a list of KeyVal
type KeyVals []KeyVal

// Has the "keyword" present in the list of KeyVal, and returns the
// corresponding KeyVal, else an empty string.
func (kvs KeyVals) Has(keyword string) KeyVal {
	for i := range kvs {
		if kvs[i].Keyword() == keyword {
			return kvs[i]
		}
	}
	return emptyKV
}

var emptyKV = KeyVal("")

// MergeSet takes the current setKeyVals, and then applies the entryKeyVals
// such that the entry's values win. The union is returned.
func MergeSet(setKeyVals, entryKeyVals []string) KeyVals {
	retList := NewKeyVals(append([]string{}, setKeyVals...))
	eKVs := NewKeyVals(entryKeyVals)
	seenKeywords := []string{}
	for i := range retList {
		word := retList[i].Keyword()
		if ekv := eKVs.Has(word); ekv != emptyKV {
			retList[i] = ekv
		}
		seenKeywords = append(seenKeywords, word)
	}
	for i := range eKVs {
		if !inSlice(eKVs[i].Keyword(), seenKeywords) {
			retList = append(retList, eKVs[i])
		}
	}
	return retList
}

var (
	// DefaultKeywords has the several default keyword producers (uid, gid,
	// mode, nlink, type, size, mtime)
	DefaultKeywords = []string{
		"size",
		"type",
		"uid",
		"gid",
		"mode",
		"link",
		"nlink",
		"time",
	}

	// DefaultTarKeywords has keywords that should be used when creating a manifest from
	// an archive. Currently, evaluating the # of hardlinks has not been implemented yet
	DefaultTarKeywords = []string{
		"size",
		"type",
		"uid",
		"gid",
		"mode",
		"link",
		"tar_time",
	}

	// BsdKeywords is the set of keywords that is only in the upstream FreeBSD mtree
	BsdKeywords = []string{
		"cksum",
		"device",
		"flags", // this one is really mostly BSD specific ...
		"ignore",
		"gid",
		"gname",
		"link",
		"md5",
		"md5digest",
		"mode",
		"nlink",
		"nochange",
		"optional",
		"ripemd160digest",
		"rmd160",
		"rmd160digest",
		"sha1",
		"sha1digest",
		"sha256",
		"sha256digest",
		"sha384",
		"sha384digest",
		"sha512",
		"sha512digest",
		"size",
		"tags",
		"time",
		"type",
		"uid",
		"uname",
	}

	// SetKeywords is the default set of keywords calculated for a `/set` SpecialType
	SetKeywords = []string{
		"uid",
		"gid",
	}
	// KeywordFuncs is the map of all keywords (and the functions to produce them)
	KeywordFuncs = map[string]KeywordFunc{
		"size":            sizeKeywordFunc,                                     // The size, in bytes, of the file
		"type":            typeKeywordFunc,                                     // The type of the file
		"time":            timeKeywordFunc,                                     // The last modification time of the file
		"link":            linkKeywordFunc,                                     // The target of the symbolic link when type=link
		"uid":             uidKeywordFunc,                                      // The file owner as a numeric value
		"gid":             gidKeywordFunc,                                      // The file group as a numeric value
		"nlink":           nlinkKeywordFunc,                                    // The number of hard links the file is expected to have
		"uname":           unameKeywordFunc,                                    // The file owner as a symbolic name
		"mode":            modeKeywordFunc,                                     // The current file's permissions as a numeric (octal) or symbolic value
		"cksum":           cksumKeywordFunc,                                    // The checksum of the file using the default algorithm specified by the cksum(1) utility
		"md5":             hasherKeywordFunc("md5digest", md5.New),             // The MD5 message digest of the file
		"md5digest":       hasherKeywordFunc("md5digest", md5.New),             // A synonym for `md5`
		"rmd160":          hasherKeywordFunc("ripemd160digest", ripemd160.New), // The RIPEMD160 message digest of the file
		"rmd160digest":    hasherKeywordFunc("ripemd160digest", ripemd160.New), // A synonym for `rmd160`
		"ripemd160digest": hasherKeywordFunc("ripemd160digest", ripemd160.New), // A synonym for `rmd160`
		"sha1":            hasherKeywordFunc("sha1digest", sha1.New),           // The SHA1 message digest of the file
		"sha1digest":      hasherKeywordFunc("sha1digest", sha1.New),           // A synonym for `sha1`
		"sha256":          hasherKeywordFunc("sha256digest", sha256.New),       // The SHA256 message digest of the file
		"sha256digest":    hasherKeywordFunc("sha256digest", sha256.New),       // A synonym for `sha256`
		"sha384":          hasherKeywordFunc("sha384digest", sha512.New384),    // The SHA384 message digest of the file
		"sha384digest":    hasherKeywordFunc("sha384digest", sha512.New384),    // A synonym for `sha384`
		"sha512":          hasherKeywordFunc("sha512digest", sha512.New),       // The SHA512 message digest of the file
		"sha512digest":    hasherKeywordFunc("sha512digest", sha512.New),       // A synonym for `sha512`

		"flags": flagsKeywordFunc, // NOTE: this is a noop, but here to support the presence of the "flags" keyword.

		// This is not an upstreamed keyword, but used to vary from "time", as tar
		// archives do not store nanosecond precision. So comparing on "time" will
		// be only seconds level accurate.
		"tar_time": tartimeKeywordFunc, // The last modification time of the file, from a tar archive mtime

		// This is not an upstreamed keyword, but a needed attribute for file validation.
		// The pattern for this keyword key is prefixed by "xattr." followed by the extended attribute "namespace.key".
		// The keyword value is the SHA1 digest of the extended attribute's value.
		// In this way, the order of the keys does not matter, and the contents of the value is not revealed.
		"xattr":  xattrKeywordFunc,
		"xattrs": xattrKeywordFunc,
	}
)

var (
	modeKeywordFunc = func(path string, info os.FileInfo, r io.Reader) (string, error) {
		permissions := info.Mode().Perm()
		if os.ModeSetuid&info.Mode() > 0 {
			permissions |= (1 << 11)
		}
		if os.ModeSetgid&info.Mode() > 0 {
			permissions |= (1 << 10)
		}
		if os.ModeSticky&info.Mode() > 0 {
			permissions |= (1 << 9)
		}
		return fmt.Sprintf("mode=%#o", permissions), nil
	}
	sizeKeywordFunc = func(path string, info os.FileInfo, r io.Reader) (string, error) {
		if sys, ok := info.Sys().(*tar.Header); ok {
			if sys.Typeflag == tar.TypeSymlink {
				return fmt.Sprintf("size=%d", len(sys.Linkname)), nil
			}
		}
		return fmt.Sprintf("size=%d", info.Size()), nil
	}
	cksumKeywordFunc = func(path string, info os.FileInfo, r io.Reader) (string, error) {
		if !info.Mode().IsRegular() {
			return "", nil
		}
		sum, _, err := cksum(r)
		if err != nil {
			return "", err
		}
		return fmt.Sprintf("cksum=%d", sum), nil
	}
	hasherKeywordFunc = func(name string, newHash func() hash.Hash) KeywordFunc {
		return func(path string, info os.FileInfo, r io.Reader) (string, error) {
			if !info.Mode().IsRegular() {
				return "", nil
			}
			h := newHash()
			if _, err := io.Copy(h, r); err != nil {
				return "", err
			}
			return fmt.Sprintf("%s=%x", name, h.Sum(nil)), nil
		}
	}
	tartimeKeywordFunc = func(path string, info os.FileInfo, r io.Reader) (string, error) {
		return fmt.Sprintf("tar_time=%d.%9.9d", info.ModTime().Unix(), 0), nil
	}
	timeKeywordFunc = func(path string, info os.FileInfo, r io.Reader) (string, error) {
		tSec := info.ModTime().Unix()
		tNano := info.ModTime().Nanosecond()
		return fmt.Sprintf("time=%d.%9.9d", tSec, tNano), nil
	}
	linkKeywordFunc = func(path string, info os.FileInfo, r io.Reader) (string, error) {
		if sys, ok := info.Sys().(*tar.Header); ok {
			if sys.Linkname != "" {
				return fmt.Sprintf("link=%s", sys.Linkname), nil
			}
			return "", nil
		}

		if info.Mode()&os.ModeSymlink != 0 {
			str, err := os.Readlink(path)
			if err != nil {
				return "", err
			}
			return fmt.Sprintf("link=%s", str), nil
		}
		return "", nil
	}
	typeKeywordFunc = func(path string, info os.FileInfo, r io.Reader) (string, error) {
		if info.Mode().IsDir() {
			return "type=dir", nil
		}
		if info.Mode().IsRegular() {
			return "type=file", nil
		}
		if info.Mode()&os.ModeSocket != 0 {
			return "type=socket", nil
		}
		if info.Mode()&os.ModeSymlink != 0 {
			return "type=link", nil
		}
		if info.Mode()&os.ModeNamedPipe != 0 {
			return "type=fifo", nil
		}
		if info.Mode()&os.ModeDevice != 0 {
			if info.Mode()&os.ModeCharDevice != 0 {
				return "type=char", nil
			}
			return "type=device", nil
		}
		return "", nil
	}
)