1
0
mirror of https://github.com/golang/go synced 2024-11-19 17:44:43 -07:00
go/src/mime/mediatype.go
Brad Fitzpatrick 48db2c01b4 all: use strings.Builder instead of bytes.Buffer where appropriate
I grepped for "bytes.Buffer" and "buf.String" and mostly ignored test
files. I skipped a few on purpose and probably missed a few others,
but otherwise I think this should be most of them.

Updates #18990

Change-Id: I5a6ae4296b87b416d8da02d7bfaf981d8cc14774
Reviewed-on: https://go-review.googlesource.com/102479
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
2018-03-26 23:05:53 +00:00

387 lines
9.1 KiB
Go

// Copyright 2010 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
package mime
import (
"errors"
"fmt"
"sort"
"strings"
"unicode"
)
// FormatMediaType serializes mediatype t and the parameters
// param as a media type conforming to RFC 2045 and RFC 2616.
// The type and parameter names are written in lower-case.
// When any of the arguments result in a standard violation then
// FormatMediaType returns the empty string.
func FormatMediaType(t string, param map[string]string) string {
var b strings.Builder
if slash := strings.Index(t, "/"); slash == -1 {
if !isToken(t) {
return ""
}
b.WriteString(strings.ToLower(t))
} else {
major, sub := t[:slash], t[slash+1:]
if !isToken(major) || !isToken(sub) {
return ""
}
b.WriteString(strings.ToLower(major))
b.WriteByte('/')
b.WriteString(strings.ToLower(sub))
}
attrs := make([]string, 0, len(param))
for a := range param {
attrs = append(attrs, a)
}
sort.Strings(attrs)
for _, attribute := range attrs {
value := param[attribute]
b.WriteByte(';')
b.WriteByte(' ')
if !isToken(attribute) {
return ""
}
b.WriteString(strings.ToLower(attribute))
b.WriteByte('=')
if isToken(value) {
b.WriteString(value)
continue
}
b.WriteByte('"')
offset := 0
for index, character := range value {
if character == '"' || character == '\\' {
b.WriteString(value[offset:index])
offset = index
b.WriteByte('\\')
}
if character&0x80 != 0 {
return ""
}
}
b.WriteString(value[offset:])
b.WriteByte('"')
}
return b.String()
}
func checkMediaTypeDisposition(s string) error {
typ, rest := consumeToken(s)
if typ == "" {
return errors.New("mime: no media type")
}
if rest == "" {
return nil
}
if !strings.HasPrefix(rest, "/") {
return errors.New("mime: expected slash after first token")
}
subtype, rest := consumeToken(rest[1:])
if subtype == "" {
return errors.New("mime: expected token after slash")
}
if rest != "" {
return errors.New("mime: unexpected content after media subtype")
}
return nil
}
// ErrInvalidMediaParameter is returned by ParseMediaType if
// the media type value was found but there was an error parsing
// the optional parameters
var ErrInvalidMediaParameter = errors.New("mime: invalid media parameter")
// ParseMediaType parses a media type value and any optional
// parameters, per RFC 1521. Media types are the values in
// Content-Type and Content-Disposition headers (RFC 2183).
// On success, ParseMediaType returns the media type converted
// to lowercase and trimmed of white space and a non-nil map.
// If there is an error parsing the optional parameter,
// the media type will be returned along with the error
// ErrInvalidMediaParameter.
// The returned map, params, maps from the lowercase
// attribute to the attribute value with its case preserved.
func ParseMediaType(v string) (mediatype string, params map[string]string, err error) {
i := strings.Index(v, ";")
if i == -1 {
i = len(v)
}
mediatype = strings.TrimSpace(strings.ToLower(v[0:i]))
err = checkMediaTypeDisposition(mediatype)
if err != nil {
return "", nil, err
}
params = make(map[string]string)
// Map of base parameter name -> parameter name -> value
// for parameters containing a '*' character.
// Lazily initialized.
var continuation map[string]map[string]string
v = v[i:]
for len(v) > 0 {
v = strings.TrimLeftFunc(v, unicode.IsSpace)
if len(v) == 0 {
break
}
key, value, rest := consumeMediaParam(v)
if key == "" {
if strings.TrimSpace(rest) == ";" {
// Ignore trailing semicolons.
// Not an error.
return
}
// Parse error.
return mediatype, nil, ErrInvalidMediaParameter
}
pmap := params
if idx := strings.Index(key, "*"); idx != -1 {
baseName := key[:idx]
if continuation == nil {
continuation = make(map[string]map[string]string)
}
var ok bool
if pmap, ok = continuation[baseName]; !ok {
continuation[baseName] = make(map[string]string)
pmap = continuation[baseName]
}
}
if _, exists := pmap[key]; exists {
// Duplicate parameter name is bogus.
return "", nil, errors.New("mime: duplicate parameter name")
}
pmap[key] = value
v = rest
}
// Stitch together any continuations or things with stars
// (i.e. RFC 2231 things with stars: "foo*0" or "foo*")
var buf strings.Builder
for key, pieceMap := range continuation {
singlePartKey := key + "*"
if v, ok := pieceMap[singlePartKey]; ok {
if decv, ok := decode2231Enc(v); ok {
params[key] = decv
}
continue
}
buf.Reset()
valid := false
for n := 0; ; n++ {
simplePart := fmt.Sprintf("%s*%d", key, n)
if v, ok := pieceMap[simplePart]; ok {
valid = true
buf.WriteString(v)
continue
}
encodedPart := simplePart + "*"
v, ok := pieceMap[encodedPart]
if !ok {
break
}
valid = true
if n == 0 {
if decv, ok := decode2231Enc(v); ok {
buf.WriteString(decv)
}
} else {
decv, _ := percentHexUnescape(v)
buf.WriteString(decv)
}
}
if valid {
params[key] = buf.String()
}
}
return
}
func decode2231Enc(v string) (string, bool) {
sv := strings.SplitN(v, "'", 3)
if len(sv) != 3 {
return "", false
}
// TODO: ignoring lang in sv[1] for now. If anybody needs it we'll
// need to decide how to expose it in the API. But I'm not sure
// anybody uses it in practice.
charset := strings.ToLower(sv[0])
if len(charset) == 0 {
return "", false
}
if charset != "us-ascii" && charset != "utf-8" {
// TODO: unsupported encoding
return "", false
}
encv, err := percentHexUnescape(sv[2])
if err != nil {
return "", false
}
return encv, true
}
func isNotTokenChar(r rune) bool {
return !isTokenChar(r)
}
// consumeToken consumes a token from the beginning of provided
// string, per RFC 2045 section 5.1 (referenced from 2183), and return
// the token consumed and the rest of the string. Returns ("", v) on
// failure to consume at least one character.
func consumeToken(v string) (token, rest string) {
notPos := strings.IndexFunc(v, isNotTokenChar)
if notPos == -1 {
return v, ""
}
if notPos == 0 {
return "", v
}
return v[0:notPos], v[notPos:]
}
// consumeValue consumes a "value" per RFC 2045, where a value is
// either a 'token' or a 'quoted-string'. On success, consumeValue
// returns the value consumed (and de-quoted/escaped, if a
// quoted-string) and the rest of the string. On failure, returns
// ("", v).
func consumeValue(v string) (value, rest string) {
if v == "" {
return
}
if v[0] != '"' {
return consumeToken(v)
}
// parse a quoted-string
buffer := new(strings.Builder)
for i := 1; i < len(v); i++ {
r := v[i]
if r == '"' {
return buffer.String(), v[i+1:]
}
// When MSIE sends a full file path (in "intranet mode"), it does not
// escape backslashes: "C:\dev\go\foo.txt", not "C:\\dev\\go\\foo.txt".
//
// No known MIME generators emit unnecessary backslash escapes
// for simple token characters like numbers and letters.
//
// If we see an unnecessary backslash escape, assume it is from MSIE
// and intended as a literal backslash. This makes Go servers deal better
// with MSIE without affecting the way they handle conforming MIME
// generators.
if r == '\\' && i+1 < len(v) && !isTokenChar(rune(v[i+1])) {
buffer.WriteByte(v[i+1])
i++
continue
}
if r == '\r' || r == '\n' {
return "", v
}
buffer.WriteByte(v[i])
}
// Did not find end quote.
return "", v
}
func consumeMediaParam(v string) (param, value, rest string) {
rest = strings.TrimLeftFunc(v, unicode.IsSpace)
if !strings.HasPrefix(rest, ";") {
return "", "", v
}
rest = rest[1:] // consume semicolon
rest = strings.TrimLeftFunc(rest, unicode.IsSpace)
param, rest = consumeToken(rest)
param = strings.ToLower(param)
if param == "" {
return "", "", v
}
rest = strings.TrimLeftFunc(rest, unicode.IsSpace)
if !strings.HasPrefix(rest, "=") {
return "", "", v
}
rest = rest[1:] // consume equals sign
rest = strings.TrimLeftFunc(rest, unicode.IsSpace)
value, rest2 := consumeValue(rest)
if value == "" && rest2 == rest {
return "", "", v
}
rest = rest2
return param, value, rest
}
func percentHexUnescape(s string) (string, error) {
// Count %, check that they're well-formed.
percents := 0
for i := 0; i < len(s); {
if s[i] != '%' {
i++
continue
}
percents++
if i+2 >= len(s) || !ishex(s[i+1]) || !ishex(s[i+2]) {
s = s[i:]
if len(s) > 3 {
s = s[0:3]
}
return "", fmt.Errorf("mime: bogus characters after %%: %q", s)
}
i += 3
}
if percents == 0 {
return s, nil
}
t := make([]byte, len(s)-2*percents)
j := 0
for i := 0; i < len(s); {
switch s[i] {
case '%':
t[j] = unhex(s[i+1])<<4 | unhex(s[i+2])
j++
i += 3
default:
t[j] = s[i]
j++
i++
}
}
return string(t), nil
}
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
}