ianaindex.go

Documentation: golang.org/x/text/encoding/ianaindex

     1  // Copyright 2015 The Go Authors. All rights reserved.
     2  // Use of this source code is governed by a BSD-style
     3  // license that can be found in the LICENSE file.
     4  
     5  //go:generate go run gen.go
     6  
     7  // Package ianaindex maps names to Encodings as specified by the IANA registry.
     8  // This includes both the MIME and IANA names.
     9  //
    10  // See http://www.iana.org/assignments/character-sets/character-sets.xhtml for
    11  // more details.
    12  package ianaindex
    13  
    14  import (
    15  	"errors"
    16  	"sort"
    17  	"strings"
    18  
    19  	"golang.org/x/text/encoding"
    20  	"golang.org/x/text/encoding/charmap"
    21  	"golang.org/x/text/encoding/internal/identifier"
    22  	"golang.org/x/text/encoding/japanese"
    23  	"golang.org/x/text/encoding/korean"
    24  	"golang.org/x/text/encoding/simplifiedchinese"
    25  	"golang.org/x/text/encoding/traditionalchinese"
    26  	"golang.org/x/text/encoding/unicode"
    27  )
    28  
    29  // TODO: remove the "Status... incomplete" in the package doc comment.
    30  // TODO: allow users to specify their own aliases?
    31  // TODO: allow users to specify their own indexes?
    32  // TODO: allow canonicalizing names
    33  
    34  // NOTE: only use these top-level variables if we can get the linker to drop
    35  // the indexes when they are not used. Make them a function or perhaps only
    36  // support MIME otherwise.
    37  
    38  var (
    39  	// MIME is an index to map MIME names.
    40  	MIME *Index = mime
    41  
    42  	// IANA is an index that supports all names and aliases using IANA names as
    43  	// the canonical identifier.
    44  	IANA *Index = iana
    45  
    46  	// MIB is an index that associates the MIB display name with an Encoding.
    47  	MIB *Index = mib
    48  
    49  	mime = &Index{mimeName, ianaToMIB, ianaAliases, encodings[:]}
    50  	iana = &Index{ianaName, ianaToMIB, ianaAliases, encodings[:]}
    51  	mib  = &Index{mibName, ianaToMIB, ianaAliases, encodings[:]}
    52  )
    53  
    54  // Index maps names registered by IANA to Encodings.
    55  // Currently different Indexes only differ in the names they return for
    56  // encodings. In the future they may also differ in supported aliases.
    57  type Index struct {
    58  	names func(i int) string
    59  	toMIB []identifier.MIB // Sorted slice of supported MIBs
    60  	alias map[string]int
    61  	enc   []encoding.Encoding
    62  }
    63  
    64  var (
    65  	errInvalidName = errors.New("ianaindex: invalid encoding name")
    66  	errUnknown     = errors.New("ianaindex: unknown Encoding")
    67  	errUnsupported = errors.New("ianaindex: unsupported Encoding")
    68  )
    69  
    70  // Encoding returns an Encoding for IANA-registered names. Matching is
    71  // case-insensitive.
    72  //
    73  // If the provided name doesn't match a IANA-registered charset, an error is
    74  // returned. If the name matches a IANA-registered charset but isn't supported,
    75  // a nil encoding and a nil error are returned.
    76  func (x *Index) Encoding(name string) (encoding.Encoding, error) {
    77  	name = strings.TrimSpace(name)
    78  	// First try without lowercasing (possibly creating an allocation).
    79  	i, ok := x.alias[name]
    80  	if !ok {
    81  		i, ok = x.alias[strings.ToLower(name)]
    82  		if !ok {
    83  			return nil, errInvalidName
    84  		}
    85  	}
    86  	return x.enc[i], nil
    87  }
    88  
    89  // Name reports the canonical name of the given Encoding. It will return an
    90  // error if the e is not associated with a known encoding scheme.
    91  func (x *Index) Name(e encoding.Encoding) (string, error) {
    92  	id, ok := e.(identifier.Interface)
    93  	if !ok {
    94  		return "", errUnknown
    95  	}
    96  	mib, _ := id.ID()
    97  	if mib == 0 {
    98  		return "", errUnknown
    99  	}
   100  	v := findMIB(x.toMIB, mib)
   101  	if v == -1 {
   102  		return "", errUnsupported
   103  	}
   104  	return x.names(v), nil
   105  }
   106  
   107  // TODO: the coverage of this index is rather spotty. Allowing users to set
   108  // encodings would allow:
   109  // - users to increase coverage
   110  // - allow a partially loaded set of encodings in case the user doesn't need to
   111  //   them all.
   112  // - write an OS-specific wrapper for supported encodings and set them.
   113  // The exact definition of Set depends a bit on if and how we want to let users
   114  // write their own Encoding implementations. Also, it is not possible yet to
   115  // only partially load the encodings without doing some refactoring. Until this
   116  // is solved, we might as well not support Set.
   117  // // Set sets the e to be used for the encoding scheme identified by name. Only
   118  // // canonical names may be used. An empty name assigns e to its internally
   119  // // associated encoding scheme.
   120  // func (x *Index) Set(name string, e encoding.Encoding) error {
   121  // 	panic("TODO: implement")
   122  // }
   123  
   124  func findMIB(x []identifier.MIB, mib identifier.MIB) int {
   125  	i := sort.Search(len(x), func(i int) bool { return x[i] >= mib })
   126  	if i < len(x) && x[i] == mib {
   127  		return i
   128  	}
   129  	return -1
   130  }
   131  
   132  const maxMIMENameLen = '0' - 1 // officially 40, but we leave some buffer.
   133  
   134  func mimeName(x int) string {
   135  	n := ianaNames[x]
   136  	// See gen.go for a description of the encoding.
   137  	if n[0] <= maxMIMENameLen {
   138  		return n[1:n[0]]
   139  	}
   140  	return n
   141  }
   142  
   143  func ianaName(x int) string {
   144  	n := ianaNames[x]
   145  	// See gen.go for a description of the encoding.
   146  	if n[0] <= maxMIMENameLen {
   147  		return n[n[0]:]
   148  	}
   149  	return n
   150  }
   151  
   152  func mibName(x int) string {
   153  	return mibNames[x]
   154  }
   155  
   156  var encodings = [numIANA]encoding.Encoding{
   157  	enc3:    asciiEnc,
   158  	enc106:  unicode.UTF8,
   159  	enc1015: unicode.UTF16(unicode.BigEndian, unicode.UseBOM),
   160  	enc1013: unicode.UTF16(unicode.BigEndian, unicode.IgnoreBOM),
   161  	enc1014: unicode.UTF16(unicode.LittleEndian, unicode.IgnoreBOM),
   162  	enc2028: charmap.CodePage037,
   163  	enc2011: charmap.CodePage437,
   164  	enc2009: charmap.CodePage850,
   165  	enc2010: charmap.CodePage852,
   166  	enc2046: charmap.CodePage855,
   167  	enc2089: charmap.CodePage858,
   168  	enc2048: charmap.CodePage860,
   169  	enc2013: charmap.CodePage862,
   170  	enc2050: charmap.CodePage863,
   171  	enc2052: charmap.CodePage865,
   172  	enc2086: charmap.CodePage866,
   173  	enc2102: charmap.CodePage1047,
   174  	enc2091: charmap.CodePage1140,
   175  	enc4:    charmap.ISO8859_1,
   176  	enc5:    charmap.ISO8859_2,
   177  	enc6:    charmap.ISO8859_3,
   178  	enc7:    charmap.ISO8859_4,
   179  	enc8:    charmap.ISO8859_5,
   180  	enc9:    charmap.ISO8859_6,
   181  	enc81:   charmap.ISO8859_6E,
   182  	enc82:   charmap.ISO8859_6I,
   183  	enc10:   charmap.ISO8859_7,
   184  	enc11:   charmap.ISO8859_8,
   185  	enc84:   charmap.ISO8859_8E,
   186  	enc85:   charmap.ISO8859_8I,
   187  	enc12:   charmap.ISO8859_9,
   188  	enc13:   charmap.ISO8859_10,
   189  	enc109:  charmap.ISO8859_13,
   190  	enc110:  charmap.ISO8859_14,
   191  	enc111:  charmap.ISO8859_15,
   192  	enc112:  charmap.ISO8859_16,
   193  	enc2084: charmap.KOI8R,
   194  	enc2088: charmap.KOI8U,
   195  	enc2027: charmap.Macintosh,
   196  	enc2109: charmap.Windows874,
   197  	enc2250: charmap.Windows1250,
   198  	enc2251: charmap.Windows1251,
   199  	enc2252: charmap.Windows1252,
   200  	enc2253: charmap.Windows1253,
   201  	enc2254: charmap.Windows1254,
   202  	enc2255: charmap.Windows1255,
   203  	enc2256: charmap.Windows1256,
   204  	enc2257: charmap.Windows1257,
   205  	enc2258: charmap.Windows1258,
   206  	enc18:   japanese.EUCJP,
   207  	enc39:   japanese.ISO2022JP,
   208  	enc17:   japanese.ShiftJIS,
   209  	enc38:   korean.EUCKR,
   210  	enc114:  simplifiedchinese.GB18030,
   211  	enc113:  simplifiedchinese.GBK,
   212  	enc2085: simplifiedchinese.HZGB2312,
   213  	enc2026: traditionalchinese.Big5,
   214  }
   215
View as plain text