1
0
mirror of https://github.com/go-gitea/gitea.git synced 2024-12-04 14:46:57 -05:00
gitea/vendor/github.com/miekg/dns/scan.go

1369 lines
29 KiB
Go
Raw Normal View History

package dns
import (
"bufio"
"fmt"
"io"
"os"
"path/filepath"
"strconv"
"strings"
)
const maxTok = 2048 // Largest token we can return.
// The maximum depth of $INCLUDE directives supported by the
// ZoneParser API.
const maxIncludeDepth = 7
// Tokinize a RFC 1035 zone file. The tokenizer will normalize it:
// * Add ownernames if they are left blank;
// * Suppress sequences of spaces;
// * Make each RR fit on one line (_NEWLINE is send as last)
// * Handle comments: ;
// * Handle braces - anywhere.
const (
// Zonefile
zEOF = iota
zString
zBlank
zQuote
zNewline
zRrtpe
zOwner
zClass
zDirOrigin // $ORIGIN
zDirTTL // $TTL
zDirInclude // $INCLUDE
zDirGenerate // $GENERATE
// Privatekey file
zValue
zKey
zExpectOwnerDir // Ownername
zExpectOwnerBl // Whitespace after the ownername
zExpectAny // Expect rrtype, ttl or class
zExpectAnyNoClass // Expect rrtype or ttl
zExpectAnyNoClassBl // The whitespace after _EXPECT_ANY_NOCLASS
zExpectAnyNoTTL // Expect rrtype or class
zExpectAnyNoTTLBl // Whitespace after _EXPECT_ANY_NOTTL
zExpectRrtype // Expect rrtype
zExpectRrtypeBl // Whitespace BEFORE rrtype
zExpectRdata // The first element of the rdata
zExpectDirTTLBl // Space after directive $TTL
zExpectDirTTL // Directive $TTL
zExpectDirOriginBl // Space after directive $ORIGIN
zExpectDirOrigin // Directive $ORIGIN
zExpectDirIncludeBl // Space after directive $INCLUDE
zExpectDirInclude // Directive $INCLUDE
zExpectDirGenerate // Directive $GENERATE
zExpectDirGenerateBl // Space after directive $GENERATE
)
// ParseError is a parsing error. It contains the parse error and the location in the io.Reader
// where the error occurred.
type ParseError struct {
file string
err string
lex lex
}
func (e *ParseError) Error() (s string) {
if e.file != "" {
s = e.file + ": "
}
s += "dns: " + e.err + ": " + strconv.QuoteToASCII(e.lex.token) + " at line: " +
strconv.Itoa(e.lex.line) + ":" + strconv.Itoa(e.lex.column)
return
}
type lex struct {
token string // text of the token
err bool // when true, token text has lexer error
value uint8 // value: zString, _BLANK, etc.
torc uint16 // type or class as parsed in the lexer, we only need to look this up in the grammar
line int // line in the file
column int // column in the file
}
// ttlState describes the state necessary to fill in an omitted RR TTL
type ttlState struct {
ttl uint32 // ttl is the current default TTL
isByDirective bool // isByDirective indicates whether ttl was set by a $TTL directive
}
// NewRR reads the RR contained in the string s. Only the first RR is returned.
// If s contains no records, NewRR will return nil with no error.
//
// The class defaults to IN and TTL defaults to 3600. The full zone file syntax
// like $TTL, $ORIGIN, etc. is supported. All fields of the returned RR are
// set, except RR.Header().Rdlength which is set to 0.
func NewRR(s string) (RR, error) {
if len(s) > 0 && s[len(s)-1] != '\n' { // We need a closing newline
return ReadRR(strings.NewReader(s+"\n"), "")
}
return ReadRR(strings.NewReader(s), "")
}
// ReadRR reads the RR contained in r.
//
// The string file is used in error reporting and to resolve relative
// $INCLUDE directives.
//
// See NewRR for more documentation.
func ReadRR(r io.Reader, file string) (RR, error) {
zp := NewZoneParser(r, ".", file)
zp.SetDefaultTTL(defaultTtl)
zp.SetIncludeAllowed(true)
rr, _ := zp.Next()
return rr, zp.Err()
}
// ZoneParser is a parser for an RFC 1035 style zonefile.
//
// Each parsed RR in the zone is returned sequentially from Next. An
// optional comment can be retrieved with Comment.
//
// The directives $INCLUDE, $ORIGIN, $TTL and $GENERATE are all
// supported. Although $INCLUDE is disabled by default.
// Note that $GENERATE's range support up to a maximum of 65535 steps.
//
// Basic usage pattern when reading from a string (z) containing the
// zone data:
//
// zp := NewZoneParser(strings.NewReader(z), "", "")
//
// for rr, ok := zp.Next(); ok; rr, ok = zp.Next() {
// // Do something with rr
// }
//
// if err := zp.Err(); err != nil {
// // log.Println(err)
// }
//
// Comments specified after an RR (and on the same line!) are
// returned too:
//
// foo. IN A 10.0.0.1 ; this is a comment
//
// The text "; this is comment" is returned from Comment. Comments inside
// the RR are returned concatenated along with the RR. Comments on a line
// by themselves are discarded.
//
// Callers should not assume all returned data in an Resource Record is
// syntactically correct, e.g. illegal base64 in RRSIGs will be returned as-is.
type ZoneParser struct {
c *zlexer
parseErr *ParseError
origin string
file string
defttl *ttlState
h RR_Header
// sub is used to parse $INCLUDE files and $GENERATE directives.
// Next, by calling subNext, forwards the resulting RRs from this
// sub parser to the calling code.
sub *ZoneParser
osFile *os.File
includeDepth uint8
includeAllowed bool
generateDisallowed bool
}
// NewZoneParser returns an RFC 1035 style zonefile parser that reads
// from r.
//
// The string file is used in error reporting and to resolve relative
// $INCLUDE directives. The string origin is used as the initial
// origin, as if the file would start with an $ORIGIN directive.
func NewZoneParser(r io.Reader, origin, file string) *ZoneParser {
var pe *ParseError
if origin != "" {
origin = Fqdn(origin)
if _, ok := IsDomainName(origin); !ok {
pe = &ParseError{file, "bad initial origin name", lex{}}
}
}
return &ZoneParser{
c: newZLexer(r),
parseErr: pe,
origin: origin,
file: file,
}
}
// SetDefaultTTL sets the parsers default TTL to ttl.
func (zp *ZoneParser) SetDefaultTTL(ttl uint32) {
zp.defttl = &ttlState{ttl, false}
}
// SetIncludeAllowed controls whether $INCLUDE directives are
// allowed. $INCLUDE directives are not supported by default.
//
// The $INCLUDE directive will open and read from a user controlled
// file on the system. Even if the file is not a valid zonefile, the
// contents of the file may be revealed in error messages, such as:
//
// /etc/passwd: dns: not a TTL: "root:x:0:0:root:/root:/bin/bash" at line: 1:31
// /etc/shadow: dns: not a TTL: "root:$6$<redacted>::0:99999:7:::" at line: 1:125
func (zp *ZoneParser) SetIncludeAllowed(v bool) {
zp.includeAllowed = v
}
// Err returns the first non-EOF error that was encountered by the
// ZoneParser.
func (zp *ZoneParser) Err() error {
if zp.parseErr != nil {
return zp.parseErr
}
if zp.sub != nil {
if err := zp.sub.Err(); err != nil {
return err
}
}
return zp.c.Err()
}
func (zp *ZoneParser) setParseError(err string, l lex) (RR, bool) {
zp.parseErr = &ParseError{zp.file, err, l}
return nil, false
}
// Comment returns an optional text comment that occurred alongside
// the RR.
func (zp *ZoneParser) Comment() string {
if zp.parseErr != nil {
return ""
}
if zp.sub != nil {
return zp.sub.Comment()
}
return zp.c.Comment()
}
func (zp *ZoneParser) subNext() (RR, bool) {
if rr, ok := zp.sub.Next(); ok {
return rr, true
}
if zp.sub.osFile != nil {
zp.sub.osFile.Close()
zp.sub.osFile = nil
}
if zp.sub.Err() != nil {
// We have errors to surface.
return nil, false
}
zp.sub = nil
return zp.Next()
}
// Next advances the parser to the next RR in the zonefile and
// returns the (RR, true). It will return (nil, false) when the
// parsing stops, either by reaching the end of the input or an
// error. After Next returns (nil, false), the Err method will return
// any error that occurred during parsing.
func (zp *ZoneParser) Next() (RR, bool) {
if zp.parseErr != nil {
return nil, false
}
if zp.sub != nil {
return zp.subNext()
}
// 6 possible beginnings of a line (_ is a space):
//
// 0. zRRTYPE -> all omitted until the rrtype
// 1. zOwner _ zRrtype -> class/ttl omitted
// 2. zOwner _ zString _ zRrtype -> class omitted
// 3. zOwner _ zString _ zClass _ zRrtype -> ttl/class
// 4. zOwner _ zClass _ zRrtype -> ttl omitted
// 5. zOwner _ zClass _ zString _ zRrtype -> class/ttl (reversed)
//
// After detecting these, we know the zRrtype so we can jump to functions
// handling the rdata for each of these types.
st := zExpectOwnerDir // initial state
h := &zp.h
for l, ok := zp.c.Next(); ok; l, ok = zp.c.Next() {
// zlexer spotted an error already
if l.err {
return zp.setParseError(l.token, l)
}
switch st {
case zExpectOwnerDir:
// We can also expect a directive, like $TTL or $ORIGIN
if zp.defttl != nil {
h.Ttl = zp.defttl.ttl
}
h.Class = ClassINET
switch l.value {
case zNewline:
st = zExpectOwnerDir
case zOwner:
name, ok := toAbsoluteName(l.token, zp.origin)
if !ok {
return zp.setParseError("bad owner name", l)
}
h.Name = name
st = zExpectOwnerBl
case zDirTTL:
st = zExpectDirTTLBl
case zDirOrigin:
st = zExpectDirOriginBl
case zDirInclude:
st = zExpectDirIncludeBl
case zDirGenerate:
st = zExpectDirGenerateBl
case zRrtpe:
h.Rrtype = l.torc
st = zExpectRdata
case zClass:
h.Class = l.torc
st = zExpectAnyNoClassBl
case zBlank:
// Discard, can happen when there is nothing on the
// line except the RR type
case zString:
ttl, ok := stringToTTL(l.token)
if !ok {
return zp.setParseError("not a TTL", l)
}
h.Ttl = ttl
if zp.defttl == nil || !zp.defttl.isByDirective {
zp.defttl = &ttlState{ttl, false}
}
st = zExpectAnyNoTTLBl
default:
return zp.setParseError("syntax error at beginning", l)
}
case zExpectDirIncludeBl:
if l.value != zBlank {
return zp.setParseError("no blank after $INCLUDE-directive", l)
}
st = zExpectDirInclude
case zExpectDirInclude:
if l.value != zString {
return zp.setParseError("expecting $INCLUDE value, not this...", l)
}
neworigin := zp.origin // There may be optionally a new origin set after the filename, if not use current one
switch l, _ := zp.c.Next(); l.value {
case zBlank:
l, _ := zp.c.Next()
if l.value == zString {
name, ok := toAbsoluteName(l.token, zp.origin)
if !ok {
return zp.setParseError("bad origin name", l)
}
neworigin = name
}
case zNewline, zEOF:
// Ok
default:
return zp.setParseError("garbage after $INCLUDE", l)
}
if !zp.includeAllowed {
return zp.setParseError("$INCLUDE directive not allowed", l)
}
if zp.includeDepth >= maxIncludeDepth {
return zp.setParseError("too deeply nested $INCLUDE", l)
}
// Start with the new file
includePath := l.token
if !filepath.IsAbs(includePath) {
includePath = filepath.Join(filepath.Dir(zp.file), includePath)
}
r1, e1 := os.Open(includePath)
if e1 != nil {
var as string
if !filepath.IsAbs(l.token) {
as = fmt.Sprintf(" as `%s'", includePath)
}
msg := fmt.Sprintf("failed to open `%s'%s: %v", l.token, as, e1)
return zp.setParseError(msg, l)
}
zp.sub = NewZoneParser(r1, neworigin, includePath)
zp.sub.defttl, zp.sub.includeDepth, zp.sub.osFile = zp.defttl, zp.includeDepth+1, r1
zp.sub.SetIncludeAllowed(true)
return zp.subNext()
case zExpectDirTTLBl:
if l.value != zBlank {
return zp.setParseError("no blank after $TTL-directive", l)
}
st = zExpectDirTTL
case zExpectDirTTL:
if l.value != zString {
return zp.setParseError("expecting $TTL value, not this...", l)
}
if err := slurpRemainder(zp.c); err != nil {
return zp.setParseError(err.err, err.lex)
}
ttl, ok := stringToTTL(l.token)
if !ok {
return zp.setParseError("expecting $TTL value, not this...", l)
}
zp.defttl = &ttlState{ttl, true}
st = zExpectOwnerDir
case zExpectDirOriginBl:
if l.value != zBlank {
return zp.setParseError("no blank after $ORIGIN-directive", l)
}
st = zExpectDirOrigin
case zExpectDirOrigin:
if l.value != zString {
return zp.setParseError("expecting $ORIGIN value, not this...", l)
}
if err := slurpRemainder(zp.c); err != nil {
return zp.setParseError(err.err, err.lex)
}
name, ok := toAbsoluteName(l.token, zp.origin)
if !ok {
return zp.setParseError("bad origin name", l)
}
zp.origin = name
st = zExpectOwnerDir
case zExpectDirGenerateBl:
if l.value != zBlank {
return zp.setParseError("no blank after $GENERATE-directive", l)
}
st = zExpectDirGenerate
case zExpectDirGenerate:
if zp.generateDisallowed {
return zp.setParseError("nested $GENERATE directive not allowed", l)
}
if l.value != zString {
return zp.setParseError("expecting $GENERATE value, not this...", l)
}
return zp.generate(l)
case zExpectOwnerBl:
if l.value != zBlank {
return zp.setParseError("no blank after owner", l)
}
st = zExpectAny
case zExpectAny:
switch l.value {
case zRrtpe:
if zp.defttl == nil {
return zp.setParseError("missing TTL with no previous value", l)
}
h.Rrtype = l.torc
st = zExpectRdata
case zClass:
h.Class = l.torc
st = zExpectAnyNoClassBl
case zString:
ttl, ok := stringToTTL(l.token)
if !ok {
return zp.setParseError("not a TTL", l)
}
h.Ttl = ttl
if zp.defttl == nil || !zp.defttl.isByDirective {
zp.defttl = &ttlState{ttl, false}
}
st = zExpectAnyNoTTLBl
default:
return zp.setParseError("expecting RR type, TTL or class, not this...", l)
}
case zExpectAnyNoClassBl:
if l.value != zBlank {
return zp.setParseError("no blank before class", l)
}
st = zExpectAnyNoClass
case zExpectAnyNoTTLBl:
if l.value != zBlank {
return zp.setParseError("no blank before TTL", l)
}
st = zExpectAnyNoTTL
case zExpectAnyNoTTL:
switch l.value {
case zClass:
h.Class = l.torc
st = zExpectRrtypeBl
case zRrtpe:
h.Rrtype = l.torc
st = zExpectRdata
default:
return zp.setParseError("expecting RR type or class, not this...", l)
}
case zExpectAnyNoClass:
switch l.value {
case zString:
ttl, ok := stringToTTL(l.token)
if !ok {
return zp.setParseError("not a TTL", l)
}
h.Ttl = ttl
if zp.defttl == nil || !zp.defttl.isByDirective {
zp.defttl = &ttlState{ttl, false}
}
st = zExpectRrtypeBl
case zRrtpe:
h.Rrtype = l.torc
st = zExpectRdata
default:
return zp.setParseError("expecting RR type or TTL, not this...", l)
}
case zExpectRrtypeBl:
if l.value != zBlank {
return zp.setParseError("no blank before RR type", l)
}
st = zExpectRrtype
case zExpectRrtype:
if l.value != zRrtpe {
return zp.setParseError("unknown RR type", l)
}
h.Rrtype = l.torc
st = zExpectRdata
case zExpectRdata:
2021-02-28 18:08:33 -05:00
var (
rr RR
parseAsRFC3597 bool
)
if newFn, ok := TypeToRR[h.Rrtype]; ok {
rr = newFn()
*rr.Header() = *h
2021-02-28 18:08:33 -05:00
// We may be parsing a known RR type using the RFC3597 format.
// If so, we handle that here in a generic way.
//
// This is also true for PrivateRR types which will have the
// RFC3597 parsing done for them and the Unpack method called
// to populate the RR instead of simply deferring to Parse.
if zp.c.Peek().token == "\\#" {
parseAsRFC3597 = true
}
} else {
rr = &RFC3597{Hdr: *h}
}
_, isPrivate := rr.(*PrivateRR)
if !isPrivate && zp.c.Peek().token == "" {
// This is a dynamic update rr.
// TODO(tmthrgd): Previously slurpRemainder was only called
// for certain RR types, which may have been important.
if err := slurpRemainder(zp.c); err != nil {
return zp.setParseError(err.err, err.lex)
}
return rr, true
} else if l.value == zNewline {
return zp.setParseError("unexpected newline", l)
}
2021-02-28 18:08:33 -05:00
parseAsRR := rr
if parseAsRFC3597 {
parseAsRR = &RFC3597{Hdr: *h}
}
if err := parseAsRR.parse(zp.c, zp.origin); err != nil {
// err is a concrete *ParseError without the file field set.
// The setParseError call below will construct a new
// *ParseError with file set to zp.file.
2021-02-28 18:08:33 -05:00
// err.lex may be nil in which case we substitute our current
// lex token.
if err.lex == (lex{}) {
return zp.setParseError(err.err, l)
}
return zp.setParseError(err.err, err.lex)
}
2021-02-28 18:08:33 -05:00
if parseAsRFC3597 {
err := parseAsRR.(*RFC3597).fromRFC3597(rr)
if err != nil {
return zp.setParseError(err.Error(), l)
}
}
return rr, true
}
}
// If we get here, we and the h.Rrtype is still zero, we haven't parsed anything, this
// is not an error, because an empty zone file is still a zone file.
return nil, false
}
type zlexer struct {
br io.ByteReader
readErr error
line int
column int
comBuf string
comment string
l lex
cachedL *lex
brace int
quote bool
space bool
commt bool
rrtype bool
owner bool
nextL bool
eol bool // end-of-line
}
func newZLexer(r io.Reader) *zlexer {
br, ok := r.(io.ByteReader)
if !ok {
br = bufio.NewReaderSize(r, 1024)
}
return &zlexer{
br: br,
line: 1,
owner: true,
}
}
func (zl *zlexer) Err() error {
if zl.readErr == io.EOF {
return nil
}
return zl.readErr
}
// readByte returns the next byte from the input
func (zl *zlexer) readByte() (byte, bool) {
if zl.readErr != nil {
return 0, false
}
c, err := zl.br.ReadByte()
if err != nil {
zl.readErr = err
return 0, false
}
// delay the newline handling until the next token is delivered,
// fixes off-by-one errors when reporting a parse error.
if zl.eol {
zl.line++
zl.column = 0
zl.eol = false
}
if c == '\n' {
zl.eol = true
} else {
zl.column++
}
return c, true
}
func (zl *zlexer) Peek() lex {
if zl.nextL {
return zl.l
}
l, ok := zl.Next()
if !ok {
return l
}
if zl.nextL {
// Cache l. Next returns zl.cachedL then zl.l.
zl.cachedL = &l
} else {
// In this case l == zl.l, so we just tell Next to return zl.l.
zl.nextL = true
}
return l
}
func (zl *zlexer) Next() (lex, bool) {
l := &zl.l
switch {
case zl.cachedL != nil:
l, zl.cachedL = zl.cachedL, nil
return *l, true
case zl.nextL:
zl.nextL = false
return *l, true
case l.err:
// Parsing errors should be sticky.
return lex{value: zEOF}, false
}
var (
str [maxTok]byte // Hold string text
com [maxTok]byte // Hold comment text
stri int // Offset in str (0 means empty)
comi int // Offset in com (0 means empty)
escape bool
)
if zl.comBuf != "" {
comi = copy(com[:], zl.comBuf)
zl.comBuf = ""
}
zl.comment = ""
for x, ok := zl.readByte(); ok; x, ok = zl.readByte() {
l.line, l.column = zl.line, zl.column
if stri >= len(str) {
l.token = "token length insufficient for parsing"
l.err = true
return *l, true
}
if comi >= len(com) {
l.token = "comment length insufficient for parsing"
l.err = true
return *l, true
}
switch x {
case ' ', '\t':
if escape || zl.quote {
// Inside quotes or escaped this is legal.
str[stri] = x
stri++
escape = false
break
}
if zl.commt {
com[comi] = x
comi++
break
}
var retL lex
if stri == 0 {
// Space directly in the beginning, handled in the grammar
} else if zl.owner {
// If we have a string and its the first, make it an owner
l.value = zOwner
l.token = string(str[:stri])
// escape $... start with a \ not a $, so this will work
switch strings.ToUpper(l.token) {
case "$TTL":
l.value = zDirTTL
case "$ORIGIN":
l.value = zDirOrigin
case "$INCLUDE":
l.value = zDirInclude
case "$GENERATE":
l.value = zDirGenerate
}
retL = *l
} else {
l.value = zString
l.token = string(str[:stri])
if !zl.rrtype {
tokenUpper := strings.ToUpper(l.token)
if t, ok := StringToType[tokenUpper]; ok {
l.value = zRrtpe
l.torc = t
zl.rrtype = true
} else if strings.HasPrefix(tokenUpper, "TYPE") {
t, ok := typeToInt(l.token)
if !ok {
l.token = "unknown RR type"
l.err = true
return *l, true
}
l.value = zRrtpe
l.torc = t
zl.rrtype = true
}
if t, ok := StringToClass[tokenUpper]; ok {
l.value = zClass
l.torc = t
} else if strings.HasPrefix(tokenUpper, "CLASS") {
t, ok := classToInt(l.token)
if !ok {
l.token = "unknown class"
l.err = true
return *l, true
}
l.value = zClass
l.torc = t
}
}
retL = *l
}
zl.owner = false
if !zl.space {
zl.space = true
l.value = zBlank
l.token = " "
if retL == (lex{}) {
return *l, true
}
zl.nextL = true
}
if retL != (lex{}) {
return retL, true
}
case ';':
if escape || zl.quote {
// Inside quotes or escaped this is legal.
str[stri] = x
stri++
escape = false
break
}
zl.commt = true
zl.comBuf = ""
if comi > 1 {
// A newline was previously seen inside a comment that
// was inside braces and we delayed adding it until now.
com[comi] = ' ' // convert newline to space
comi++
if comi >= len(com) {
l.token = "comment length insufficient for parsing"
l.err = true
return *l, true
}
}
com[comi] = ';'
comi++
if stri > 0 {
zl.comBuf = string(com[:comi])
l.value = zString
l.token = string(str[:stri])
return *l, true
}
case '\r':
escape = false
if zl.quote {
str[stri] = x
stri++
}
// discard if outside of quotes
case '\n':
escape = false
// Escaped newline
if zl.quote {
str[stri] = x
stri++
break
}
if zl.commt {
// Reset a comment
zl.commt = false
zl.rrtype = false
// If not in a brace this ends the comment AND the RR
if zl.brace == 0 {
zl.owner = true
l.value = zNewline
l.token = "\n"
zl.comment = string(com[:comi])
return *l, true
}
zl.comBuf = string(com[:comi])
break
}
if zl.brace == 0 {
// If there is previous text, we should output it here
var retL lex
if stri != 0 {
l.value = zString
l.token = string(str[:stri])
if !zl.rrtype {
tokenUpper := strings.ToUpper(l.token)
if t, ok := StringToType[tokenUpper]; ok {
zl.rrtype = true
l.value = zRrtpe
l.torc = t
}
}
retL = *l
}
l.value = zNewline
l.token = "\n"
zl.comment = zl.comBuf
zl.comBuf = ""
zl.rrtype = false
zl.owner = true
if retL != (lex{}) {
zl.nextL = true
return retL, true
}
return *l, true
}
case '\\':
// comments do not get escaped chars, everything is copied
if zl.commt {
com[comi] = x
comi++
break
}
// something already escaped must be in string
if escape {
str[stri] = x
stri++
escape = false
break
}
// something escaped outside of string gets added to string
str[stri] = x
stri++
escape = true
case '"':
if zl.commt {
com[comi] = x
comi++
break
}
if escape {
str[stri] = x
stri++
escape = false
break
}
zl.space = false
// send previous gathered text and the quote
var retL lex
if stri != 0 {
l.value = zString
l.token = string(str[:stri])
retL = *l
}
// send quote itself as separate token
l.value = zQuote
l.token = "\""
zl.quote = !zl.quote
if retL != (lex{}) {
zl.nextL = true
return retL, true
}
return *l, true
case '(', ')':
if zl.commt {
com[comi] = x
comi++
break
}
if escape || zl.quote {
// Inside quotes or escaped this is legal.
str[stri] = x
stri++
escape = false
break
}
switch x {
case ')':
zl.brace--
if zl.brace < 0 {
l.token = "extra closing brace"
l.err = true
return *l, true
}
case '(':
zl.brace++
}
default:
escape = false
if zl.commt {
com[comi] = x
comi++
break
}
str[stri] = x
stri++
zl.space = false
}
}
if zl.readErr != nil && zl.readErr != io.EOF {
// Don't return any tokens after a read error occurs.
return lex{value: zEOF}, false
}
var retL lex
if stri > 0 {
// Send remainder of str
l.value = zString
l.token = string(str[:stri])
retL = *l
if comi <= 0 {
return retL, true
}
}
if comi > 0 {
// Send remainder of com
l.value = zNewline
l.token = "\n"
zl.comment = string(com[:comi])
if retL != (lex{}) {
zl.nextL = true
return retL, true
}
return *l, true
}
if zl.brace != 0 {
l.token = "unbalanced brace"
l.err = true
return *l, true
}
return lex{value: zEOF}, false
}
func (zl *zlexer) Comment() string {
if zl.l.err {
return ""
}
return zl.comment
}
// Extract the class number from CLASSxx
func classToInt(token string) (uint16, bool) {
offset := 5
if len(token) < offset+1 {
return 0, false
}
class, err := strconv.ParseUint(token[offset:], 10, 16)
if err != nil {
return 0, false
}
return uint16(class), true
}
// Extract the rr number from TYPExxx
func typeToInt(token string) (uint16, bool) {
offset := 4
if len(token) < offset+1 {
return 0, false
}
typ, err := strconv.ParseUint(token[offset:], 10, 16)
if err != nil {
return 0, false
}
return uint16(typ), true
}
// stringToTTL parses things like 2w, 2m, etc, and returns the time in seconds.
func stringToTTL(token string) (uint32, bool) {
var s, i uint32
for _, c := range token {
switch c {
case 's', 'S':
s += i
i = 0
case 'm', 'M':
s += i * 60
i = 0
case 'h', 'H':
s += i * 60 * 60
i = 0
case 'd', 'D':
s += i * 60 * 60 * 24
i = 0
case 'w', 'W':
s += i * 60 * 60 * 24 * 7
i = 0
case '0', '1', '2', '3', '4', '5', '6', '7', '8', '9':
i *= 10
i += uint32(c) - '0'
default:
return 0, false
}
}
return s + i, true
}
// Parse LOC records' <digits>[.<digits>][mM] into a
// mantissa exponent format. Token should contain the entire
// string (i.e. no spaces allowed)
func stringToCm(token string) (e, m uint8, ok bool) {
if token[len(token)-1] == 'M' || token[len(token)-1] == 'm' {
token = token[0 : len(token)-1]
}
s := strings.SplitN(token, ".", 2)
var meters, cmeters, val int
var err error
switch len(s) {
case 2:
if cmeters, err = strconv.Atoi(s[1]); err != nil {
return
}
2021-02-28 18:08:33 -05:00
// There's no point in having more than 2 digits in this part, and would rather make the implementation complicated ('123' should be treated as '12').
// So we simply reject it.
// We also make sure the first character is a digit to reject '+-' signs.
if len(s[1]) > 2 || s[1][0] < '0' || s[1][0] > '9' {
return
}
if len(s[1]) == 1 {
// 'nn.1' must be treated as 'nn-meters and 10cm, not 1cm.
cmeters *= 10
}
if s[0] == "" {
2021-02-28 18:08:33 -05:00
// This will allow omitting the 'meter' part, like .01 (meaning 0.01m = 1cm).
break
}
fallthrough
case 1:
if meters, err = strconv.Atoi(s[0]); err != nil {
return
}
2021-02-28 18:08:33 -05:00
// RFC1876 states the max value is 90000000.00. The latter two conditions enforce it.
if s[0][0] < '0' || s[0][0] > '9' || meters > 90000000 || (meters == 90000000 && cmeters != 0) {
return
}
case 0:
// huh?
return 0, 0, false
}
ok = true
if meters > 0 {
e = 2
val = meters
} else {
e = 0
val = cmeters
}
2021-02-28 18:08:33 -05:00
for val >= 10 {
e++
val /= 10
}
m = uint8(val)
return
}
func toAbsoluteName(name, origin string) (absolute string, ok bool) {
// check for an explicit origin reference
if name == "@" {
// require a nonempty origin
if origin == "" {
return "", false
}
return origin, true
}
// require a valid domain name
_, ok = IsDomainName(name)
if !ok || name == "" {
return "", false
}
// check if name is already absolute
if IsFqdn(name) {
return name, true
}
// require a nonempty origin
if origin == "" {
return "", false
}
return appendOrigin(name, origin), true
}
func appendOrigin(name, origin string) string {
if origin == "." {
return name + origin
}
return name + "." + origin
}
// LOC record helper function
func locCheckNorth(token string, latitude uint32) (uint32, bool) {
2021-02-28 18:08:33 -05:00
if latitude > 90*1000*60*60 {
return latitude, false
}
switch token {
case "n", "N":
return LOC_EQUATOR + latitude, true
case "s", "S":
return LOC_EQUATOR - latitude, true
}
return latitude, false
}
// LOC record helper function
func locCheckEast(token string, longitude uint32) (uint32, bool) {
2021-02-28 18:08:33 -05:00
if longitude > 180*1000*60*60 {
return longitude, false
}
switch token {
case "e", "E":
return LOC_EQUATOR + longitude, true
case "w", "W":
return LOC_EQUATOR - longitude, true
}
return longitude, false
}
// "Eat" the rest of the "line"
func slurpRemainder(c *zlexer) *ParseError {
l, _ := c.Next()
switch l.value {
case zBlank:
l, _ = c.Next()
if l.value != zNewline && l.value != zEOF {
return &ParseError{"", "garbage after rdata", l}
}
case zNewline:
case zEOF:
default:
return &ParseError{"", "garbage after rdata", l}
}
return nil
}
// Parse a 64 bit-like ipv6 address: "0014:4fff:ff20:ee64"
// Used for NID and L64 record.
func stringToNodeID(l lex) (uint64, *ParseError) {
if len(l.token) < 19 {
return 0, &ParseError{l.token, "bad NID/L64 NodeID/Locator64", l}
}
// There must be three colons at fixes positions, if not its a parse error
if l.token[4] != ':' && l.token[9] != ':' && l.token[14] != ':' {
return 0, &ParseError{l.token, "bad NID/L64 NodeID/Locator64", l}
}
s := l.token[0:4] + l.token[5:9] + l.token[10:14] + l.token[15:19]
u, err := strconv.ParseUint(s, 16, 64)
if err != nil {
return 0, &ParseError{l.token, "bad NID/L64 NodeID/Locator64", l}
}
return u, nil
}