View source for GBK (character encoding)

{{Short description|Simplified Chinese character encoding}}
{{More citations needed|date=October 2016}}<!-- We need some good GBK 1.0 & 1993 GBK data. All I can find is CP936. -->
{{Infobox character encoding
| name = Guójiā Biāozhǔn Kuòzhǎn (GBK)
| mime = GBK
| alias = [[Code page 1386|CP936]], MS936, windows-936, csGBK
| image = GBK encoding.svg
| caption = Layout of GBK ([[#Layout diagram|see below]] for a larger copy of this diagram)
| standard = GBK 1.0
| lang = Web browsers, decode as [[GB 18030]], supporting all languages, while the encoding (and other software decoders) is primarily used for [[Simplified Chinese]], but also supports [[Traditional Chinese]], [[Japanese language|Japanese]], [[English language|English]], [[Russian language|Russian]] and (partially) [[Greek language|Greek]].
| status = 
| extends = [[EUC-CN]]
| prev = [[GB 2312]]
| next = [[GB 18030]]
| classification = [[Extended ASCII]],{{efn|Not in the strictest sense of the term, as ASCII bytes can appear as trail bytes.}} [[variable-width encoding]], [[CJK characters|CJK encoding]]
|extra = <div style="text-align: left;">{{notelist}}</div>
}}
'''GBK''' is an extension of the [[GB 2312]] [[character set]] for [[Simplified Chinese characters]], used in the [[People's Republic of China]]. It includes all unified [[CJK characters]] found in {{nowrap|GB 13000.1-93}}, i.e. ISO/IEC 10646:1993, or Unicode 1.1. Since its initial release in 1993, GBK has been extended by Microsoft in '''[[Code page 1386|Code page 936/1386]]''', which was then extended into '''[[#GBK 1.0|GBK 1.0]]'''. '''GBK''' is also the IANA-registered internet name for the Microsoft mapping,<ref name=iana>{{cite web|title=Character Sets|url=https://www.iana.org/assignments/character-sets/character-sets.xhtml|access-date=3 October 2016}}</ref> which differs from other implementations primarily by the single-byte [[euro sign]] at 0x80.

''GB'' abbreviates [[guobiao|Guójiā Biāozhǔn]], which means ''national standard'' in Chinese, while ''K'' stands for ''Extension'' (扩展 ''kuòzhǎn''). GBK not only extended the old standard {{nowrap|GB 2312}} with Traditional Chinese characters, but also with Chinese characters that were simplified after the establishment of {{nowrap|GB 2312}} in 1981. With the arrival of GBK, certain names with characters formerly unrepresentable, like the 镕 (''róng'') character in former Chinese Premier [[Zhu Rongji|Zhu Rongji's]] name, are now representable.<ref>{{cite web |url=http://www.microsoft.com/typography/unicode/936.txt |archive-url=https://web.archive.org/web/20021001194325/http://www.microsoft.com/typography/unicode/936.txt |title=Code Page 936 - PRC GBK (XGB) |website=[[Microsoft]] |archive-date=2002-10-01 |url-status=dead}} Conversion map between Codepage 936 and Unicode. Need manually selecting GB 18030 or GBK in browser to view it correctly.</ref>

{{As of|2022|10}}, GBK is the third-most popular encoding served from China and territories (after [[UTF-8]] and the subset {{nowrap|GB 2312}}), with 1.9% of web servers serving a page that declares GBK.<ref name=ChinaDist>{{Cite web|title=Distribution of Character Encodings among websites that use China and territories|url=https://w3techs.com/technologies/segmentation/sl-cnter-/character_encoding|website=w3techs.com|access-date=2022-10-25}}</ref> However, all major web browsers decode GB2312-marked documents as if they were marked GBK, except for Safari and Edge on the label <code>GB_2312</code>.<ref>{{Cite web|url=https://www.w3.org/International/tests/repo/results/encoding-dbl-byte-labels.en.html|title=Encoding: Summarized test results|website=www.w3.org|access-date=2019-11-15}}</ref> Together, GBK and {{nowrap|GB 2312}} encodings have a combined <!-- 5.1 + 1.9 = 7.0 declaring it, note numbers do not add to 100%, because some pages declare more than one encoding, so this overestimates; using 100-94.5 = --> 5.5% presence in China and territories.<ref name=ChinaDist/> Globally, GBK accounts for less than 0.07% of all web pages and GBK+GB2312 for <!-- 0.07% + 0.15% --> 0.2%.<ref>{{Cite web|title=Historical trends in the usage statistics of character encodings for websites, October 2022|url=https://w3techs.com/technologies/history_overview/character_encoding|access-date=2022-10-25|website=w3techs.com}}</ref>

==History==
In 1993, the Unicode 1.1 standard was released, including 20,902 characters used in [[mainland China]], [[Taiwan]], [[Japan]] and [[Korea]].  Following this, China released {{nowrap|GB 13000.1-93}}, the [[Guobiao standard]] equivalent of Unicode 1.1.

{{anchor|1993}}The '''GBK''' character set was defined in 1993 as an extension of {{nowrap|GB 2312-80}}, while also including the characters of GB 13000.1-93 through the unused codepoints available in GB 2312.  Hence GBK is backward compatible with GB 2312. GBK was defined in a normative annex to GB 13000.1-93.<ref>{{cite web |url=https://www.unicode.org/versions/Unicode15.0.0/ch18.pdf#page=28 |p=763 |title=18.2: Ideographic Description Characters |work=The Unicode Standard |version=Version 15.0.0 |year=2022 |quotation=The Ideographic Description characters are found in GBK—an extension to GB 2312-80 that added all 20,902 Unicode Version 1.1 ideographs not already in GB 2312-80. GBK is defined as a normative annex of GB 13000.1-93.}}</ref>

{{anchor|cp936}}Microsoft implemented GBK in [[Windows 95]] and [[Windows NT 3.51]] as '''Code Page 936'''.  While GBK was never an official standard, widespread usage of Windows 95 led to GBK becoming the ''de facto'' standard. While GBK included all the Chinese characters defined in Unicode 1.1 and GB 13000.1-93, these standards used different code tables. The primary reason for its existence was simply to bridge the gap between GB 2312-80 and GB 13000.1-93.

{{anchor|GBK 1.0}}In 1995, China National Information Technology Standardization Technical Committee set down the Chinese Internal Code Extension Specification ({{lang-zh|s=汉字内码扩展规范 (GBK)|p=Hànzì Nèimǎ Kuòzhǎn Guīfàn (GBK)}}), Version 1.0, known as '''GBK&nbsp;1.0''', which is a slight extension of Codepage 936. The newly added 95 characters were not found in GB&nbsp;13000.1-1993, and were provisionally assigned Unicode [[Private Use Areas|PUA]] code points.<ref name="gb18030-2005">{{Cite book|url=https://archive.org/details/GB18030-2005|title=GB 18030-2005: Information Technology—Chinese coded character set|last=Standardization Administration of China (SAC)|date=2005-11-18}}</ref>{{Rp|534}}

Microsoft later added the [[euro sign]] to Code page 936 and assigned the code 0x80 to it. This is not a valid code point in GBK&nbsp;1.0.

In 2000, the {{nowrap|GB 18030-2000}} standard was released, superseding yet maintaining compatibility with GBK&nbsp;1.0. It increased the number of definitions of Chinese characters and extended the number of possible characters through the implementation of four-byte character spaces. The subset of GB&nbsp;18030 consisting of one-byte and two-byte characters is sometimes also referred to as '''GBK'''. Mapping to Unicode has been slightly changed, though, as some characters are now defined in Unicode. In the most up-to-date form of the standard, GB&nbsp;18030-2005, only 24<ref>GB 18030-2005 Standard p.9, 79</ref> characters are still mapped to Unicode PUA (see [[GB 18030#PUA]].)

In 2002, '''GBK''' was registered as an IANA charset; the registration uses code page 936 mapping as well as CP936/MS936 aliases, but refers to GBK&nbsp;1.0 specification.<ref name=iana/> {{anchor|W3C}}[[W3C]]'s technical recommendation published in 2015<ref>{{cite web|title=Encoding Standard # gbk-encoder|url=https://www.w3.org/TR/encoding/#gbk-encoder|website=W3C|access-date=2016-10-02}}</ref> defines a '''GBK''' ''encoder'' as a GB&nbsp;18030 encoder with a single-byte euro sign and without four-byte sequences (while W3C's '''GBK''' ''decoder'' specification has no such limitation, decodes as {{nowrap|GB 18030}}, i.e. with same range of letters as all of [[Unicode]]).

==Encoding==
A character is encoded as 1 or 2 bytes.  A byte in the range <code>00</code>–<code>7F</code> is a single byte that means the same thing as it does in [[ASCII]].  Strictly speaking, there are 95 characters and 33 control codes in this range.

A byte with the high bit set indicates that it is the first of 2 bytes.  Loosely speaking, the first byte is in the range <code>81</code>–<code>FE</code> (that is, never <code>80</code> or <code>FF</code>), and the second byte is <code>40</code>–<code>A0</code> except <code>7F</code> for some areas and <code>A1</code>–<code>FE</code> for others.

More specifically, the following ranges of bytes are defined:

{|class="wikitable"
|+GBK Encoding Ranges
|-
!rowspan="2"|range || rowspan="2"|byte 1 || rowspan="2"|byte 2 || rowspan="2"|code points || colspan="4"|characters
|-
!GB 18030 || GBK 1.0 || Codepage 936 || GB 2312
|-
|Level [[GB2312 Blocks#Block 1-9: Signs|GBK/1]] || <code>A1</code>–<code>A9</code> || <code>A1</code>–<code>FE</code>
|align="right"|846 || style="text-align:right;"|718<ref name=gb18030-2005/>{{rp|8&ndash;10}} || style="text-align:right;"|717 || style="text-align:right;"|715 || style="text-align:right;"|682
|-
|Level [[GBK/2]] || <code>B0</code>–<code>F7</code> || <code>A1</code>–<code>FE</code> || style="text-align:right;"|6,768 || colspan="2" style="text-align:right;"|6,763 || style="text-align:right;"|6,763 || style="text-align:right;"|6,763
|-
|Level GBK/3 || <code>81</code>–<code>A0</code> || <code>40</code>–<code>FE</code> except <code>7F</code>|| style="text-align:right;"|6,080 || colspan="2" style="text-align:right;"|6,080 || style="text-align:right;"|6,080 || rowspan="6"|
|-
|Level GBK/4 || <code>AA</code>–<code>FE</code> || <code>40</code>–<code>A0</code> except <code>7F</code> || style="text-align:right;"|8,160 || colspan="2" style="text-align:right;"|8,160 || style="text-align:right;"|8,080
|-
|Level [[GBK/5]] || <code>A8</code>–<code>A9</code> || <code>40</code>–<code>A0</code> except <code>7F</code> || style="text-align:right;"|192 || colspan="2" style="text-align:right;"|166 || style="text-align:right;"|153
|-
|user-defined 1<ref name=gb18030-2005/> || <code>AA</code>–<code>AF</code> || <code>A1</code>–<code>FE</code> || style="text-align:right;"|564 || colspan="4" rowspan="3"|
|-
|user-defined 2 || <code>F8</code>–<code>FE</code> || <code>A1</code>–<code>FE</code> || style="text-align:right;"|658
|-
|user-defined 3 || <code>A1</code>–<code>A7</code> || <code>40</code>–<code>A0</code> except <code>7F</code> || style="text-align:right;"|672
|-
!total: || || || style="text-align:right;"|23,940 || style="text-align:right;"|21,887 || style="text-align:right;"|21,886 || style="text-align:right;"|21,791 || style="text-align:right;"|7,445
|}

===Layout diagram===
In graphical form, the following figure shows the space of all 64K possible 2-byte codes.  Green and yellow areas are assigned GBK codepoints, red are for user-defined characters.  The uncolored areas are invalid byte combinations.

[[Image:GBK encoding.svg]]

==Relationship to other encodings==
The areas indicated in the previous section as GBK/1 and GBK/2, taken by themselves, is simply {{nowrap|GB 2312-80}} in its usual encoding, GBK/1 being the non-hanzi region and GBK/2 the hanzi region.  GB 2312, or more properly the EUC-CN encoding thereof, takes a pair of bytes from the range <code>A1</code>–<code>FE</code>, like any 94² ISO-2022 character set loaded into GR.  This corresponds to the lower-right quarter of the illustration above.  However, GB 2312 does not assign any code points to the rows located at <code>AA</code>–<code>B0</code> and <code>F8</code>–<code>FE</code>, even though it had staked out the territory.  GBK added extensions to these rows.  You can see that the two gaps were filled in with user-defined areas.

More significantly, GBK extended the range of the bytes.  Having two-byte characters in the ISO-2022 GR range gives a limit of 94²=8,836 possibilities.  Abandoning the ISO-2022 model of strict regions for graphics and control characters, but retaining the feature of low bytes being 1-byte characters and pairs of high bytes denoting a character, you could potentially have 128²=16,384 positions.  GBK takes part of that, extending the range from <code>A1</code>–<code>FE</code> (94 choices for each byte) to <code>81</code>–<code>FE</code> (126 choices) for the first byte and <code>40</code>–<code>FE</code> (191 choices) for the second byte, for a total of 24,066 positions.

Microsoft's Code Page 936 is generally thought of as being GBK.<ref name=iana/>  However, the [[GB 18030#History|95 PUA characters]] added in GBK 1.0 are not included in Code Page 936.  Code Page 936 also has a single-byte [[euro sign]] at 0x80 which GBK 1.0 doesn't have.<ref>{{cite web |last1=Scherer |first1=Markus |title=Re: Fun with GBK & GB2312 |url=http://unicode.org/mail-arch/unicode-ml/y2002-m01/0042.html |website=Unicode Mail List Archive |access-date=4 March 2020 |date=4 January 2002}}</ref>

GBK's successor, {{nowrap|GB 18030-2000}}, uses the remaining range available to the second byte ({{code|30}}&ndash;{{code|39}}) to further expand the number of possibilities while retaining GBK as a subset.

==References==
{{reflist}}

==Notes==
{{notelist}}

==External links==
*[https://www.unicode.org/irg/docs/n0278-GBKv1.pdf A scan of the GBK 1.0 specification] provided by the [[Ideographic Research Group]]
*[http://icu-project.org/repos/icu/data/trunk/charset/source/gb18030/gbkuni30.txt ICU's Authoritative GBK mapping] - part of [https://ssl.icu-project.org/repos/icu/data/trunk/charset/source/gb18030/gb18030.html GB18030 data] {{Webarchive|url=https://web.archive.org/web/20161031024951/https://ssl.icu-project.org/repos/icu/data/trunk/charset/source/gb18030/gb18030.html |date=2016-10-31 }}
*[http://www.microsoft.com/globaldev/reference/dbcs/936.mspx Microsoft Reference page for GBK]
*[https://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP936.TXT  Mapping of GBK to Unicode ] N.B.: this is Microsoft code page 936, which contains entries for 21791 double-byte code points, 96 single-byte graphic characters, and 33 control characters.  This is not exactly the same as GBK which has 21886 characters.
*[http://www.khngai.com/chinese/charmap/tblgbk.php?page=0 GBK Code Table] N.B. This gbk-encoded page shows the available coding space totally populated except for 2 places, for a total of 32256 glyphs (32352 with the implied single-byte ASCII codes not illustrated), which is more than 23940 or 21886. Actual rendering of this table depends on your browser's GBK decoder.

{{CJK computing}}
{{Character encoding}}

[[Category:Windows code pages]]
[[Category:Chinese character encodings]]