Editing Integer (computer science) (section)

== Common integral data types ==
{| class="wikitable"
|-
! rowspan="2" | Bits
! rowspan="2" | Name
! rowspan="2" | Range (assuming [[two's complement]] for [[Signed number representations|signed]])
! rowspan="2" | Decimal digits
! rowspan="2" | Uses
! colspan="8" scope="col" | Implementations
|-
! [[C (programming language)|C]]/[[C++]]
! [[C Sharp (programming language)|C#]]
! [[Pascal (programming language)|Pascal]] and [[Delphi (programming language)|Delphi]]
! [[Java (programming language)|Java]]
! [[SQL]]{{efn |name=notesqla |Not all SQL dialects have unsigned datatypes.<ref name=SybDT>{{cite web|url=http://infocenter.sybase.com/help/index.jsp?topic=/com.sybase.infocenter.dc36271.1550/html/blocks/blocks20.htm |title=Sybase Adaptive Server Enterprise 15.5: Exact Numeric Datatypes}}</ref><ref name=MySQLDT>{{cite web|url=http://dev.mysql.com/doc/refman/5.6/en/numeric-types.html |title=MySQL 5.6 Numeric Datatypes}}</ref>}}
! [[FORTRAN]]
! [[D (programming language)|D]]
! [[Rust (programming language)|Rust]]
|-
| rowspan="2" style="text-align: right;" | 4
| rowspan="2" | [[nibble]], semioctet
| ''[[Signed number representations|Signed:]]'' From −8 to 7, from −(2<sup>3</sup>) to 2<sup>3</sup>&nbsp;−&nbsp;1
| style="text-align: right;" | {{#expr:3*ln2/ln10 round 2}}
| rowspan="2" | [[Binary-coded decimal]], single decimal digit {{shy|repre|sen|ta|tion}}
| colspan="8" rowspan="2" style="text-align: center; background: #eaecf0;" | &mdash;
|-
| ''Unsigned:'' From 0 to 15, which equals 2<sup>4</sup>&nbsp;−&nbsp;1
| style="text-align: right;" | {{#expr: 4 * ln(2) / ln(10) round 2}}
|-
| rowspan="2" style="text-align:right;"| 8
| rowspan="2" | [[byte]], [[octet (computing)|octet]], i8, u8
| ''Signed:'' From −128 to 127, from −(2<sup>7</sup>) to 2<sup>7</sup>&nbsp;−&nbsp;1
| style="text-align: right;" | {{#expr: 7 * ln(2) / ln(10) round 2}}
| rowspan="2" | [[ASCII]] characters, [[code unit]]s in the [[UTF-8]] [[character encoding]]
| {{mono|int8_t}}, {{mono|signed char}}{{efn|name=notescb|garbage=8 char}}
| {{mono|sbyte}}
| {{mono|Shortint}}
| {{mono|byte}}
| {{mono|tinyint}}
| {{mono|INTEGER}}{{efn|name=fortran|Fortan uses 'kinds' to control the size of integers.  Parameterized constants defining the available kinds are available in the iso_fortran_env intrinsic module.  Constants defining C compatible kinds are available in the iso_c_binding intrinsic module.}}
| {{mono|byte}}
| {{mono|i8}}
|-
| ''Unsigned:'' From 0 to 255, which equals 2<sup>8</sup>&nbsp;−&nbsp;1
| style="text-align: right;" | {{#expr: 8 * ln(2) / ln(10) round 2}}
| {{mono|uint8_t}}, {{mono|unsigned char}}{{efn|name=notescb|garbage=8 char}}
| {{mono|byte}}
| {{mono|Byte}}
| {{n/a}}
| {{mono|unsigned tinyint}}
| {{n/a}}
| {{mono|ubyte}}
| {{mono|u8}}
|-
| rowspan="2" style="text-align: right;" | 16
| rowspan="2" | halfword, [[Word (data type)|word]], short, i16, u16
| ''Signed:'' From −32,768 to 32,767, from −(2<sup>15</sup>) to 2<sup>15</sup>&nbsp;−&nbsp;1
| style="text-align: right;" | {{#expr: 15 * ln(2) / ln(10) round 2}}
| rowspan="2" | [[Universal Character Set|UCS-2]] characters, [[code unit]]s in the [[UTF-16]] [[character encoding]]
| {{mono|int16_t}}, {{mono|short}},{{efn|name=notescb|garbage=16 short}} {{mono|int}}{{efn|name=notescb|garbage=16 int}}
| {{mono|short}}
| {{mono|Smallint}}
| {{mono|short}}
| {{mono|smallint}}
| {{mono|INTEGER}}{{efn|name=fortran}}
| {{mono|short}}
| {{mono|i16}}
|-
| ''Unsigned:'' From 0 to 65,535, which equals 2<sup>16</sup>&nbsp;−&nbsp;1
| style="text-align: right;" | {{#expr:16 * ln(2) / ln(10) round 2}}
| {{mono|uint16_t, unsigned}},{{efn|name=notescb|garbage=16 int}} {{mono|unsigned int}}{{efn|name=notescb|garbage=16 int}}
| {{mono|ushort}}
| {{mono|Word}}
| {{mono|char}}{{efn|name=notejavad}}
| {{mono|unsigned smallint}}
| {{n/a}}
| {{mono|ushort}}
| {{mono|u16}}
|-
| rowspan="2" style="text-align: right;" | 32
| rowspan="2" | word, [[Unsigned long integer|long]], doubleword, longword, int, i32, u32
| ''Signed:'' From [[2147483647 (number)|−2,147,483,648 to 2,147,483,647]], from −(2<sup>31</sup>) to 2<sup>31</sup>&nbsp;−&nbsp;1
| style="text-align: right;" | {{#expr: 31 * ln(2) / ln(10) round 2}}
| rowspan="2" | [[UTF-32]] characters, [[24-bit color|true color]] with alpha, [[FourCC]], pointers in [[32-bit computing]]
| {{mono|int32_t}}, {{mono|int}},{{efn|name=notescb|garbage=32 int}} {{mono|long}}{{efn|name=notescb|garbage=32 long}}
| {{mono|int}}
| {{mono|LongInt}}; {{mono|Integer}}{{efn|name=notedelphic|garbage=signed}}
| {{mono|int}}
| {{mono|int}}
| {{mono|INTEGER}}{{efn|name=fortran}}
| {{mono|int}}
| {{mono|i32}}
|-
| ''Unsigned:'' From 0 to 4,294,967,295, which equals 2<sup>32</sup>&nbsp;−&nbsp;1
| style="text-align: right;" | {{#expr: 32 * ln(2) / ln(10) round 2}}
| {{mono|uint32_t, unsigned}},{{efn|name=notescb|garbage=16 int}} {{mono|unsigned int}},{{efn|name=notescb|garbage=16 int}} {{mono|unsigned long}}{{efn|name=notescb|garbage=16 int}}
| {{mono|uint}}
| {{mono|LongWord}}; {{mono|DWord}}; {{mono|Cardinal}}{{efn|name=notedelphic|garbage=unsigned}}
| {{n/a}}
| {{mono|unsigned int}}
| {{n/a}}
| {{mono|uint}}
| {{mono|u32}}
|-
| rowspan="2" style="text-align: right;" | 64
| rowspan="2" | word, doubleword, longword, long, long long, quad, quadword, qword, int64, i64, u64
| ''Signed:'' From {{tooltip|−(2<sup>63</sup>)|−9,223,372,036,854,775,808}} to {{tooltip|2<sup>63</sup>&nbsp;−&nbsp;1|9,223,372,036,854,775,807}}
| style="text-align: right;" | {{#expr: 63 * ln(2) / ln(10) round 2}}
| rowspan="2" | Time (e.g. {{shy|milli|seconds}} since the [[Unix epoch]]), pointers in [[64-bit computing]]
| {{mono|int64_t}}, {{mono|long}},{{efn|name=notescb|garbage=64 long}} {{mono|long long}}{{efn|name=notescb|garbage=64 long long}}
| {{mono|long}}
| {{mono|Int64}}
| {{mono|long}}
| {{mono|bigint}}
| {{mono|INTEGER}}{{efn|name=fortran}}
| {{mono|long}}
| {{mono|i64}}
|-
| ''Unsigned:'' From 0 to {{tooltip|2<sup>64</sup>&nbsp;−&nbsp;1|18,446,744,073,709,551,615}}
| style="text-align: right;" | {{#expr: 64 * ln(2) / ln(10) round 2}}
| {{mono|uint64_t}}, {{mono|unsigned long long}}{{efn|name=notescb|garbage=16 int}}
| {{mono|ulong}}
| {{mono|UInt64}}; {{mono|QWord}}
| {{n/a}}
| {{mono|unsigned bigint}}
| {{n/a}}
| {{mono|ulong}}
| {{mono|u64}}
|-
| rowspan="2" style="text-align: right;" | 128
| rowspan="2"| octaword, double quadword, i128, u128
| ''Signed:'' From {{tooltip|−(2<sup>127</sup>)|−170,141,183,460,469,231,731,687,303,715,884,105,728}} to {{tooltip|2<sup>127</sup>&nbsp;−&nbsp;1|170,141,183,460,469,231,731,687,303,715,884,105,727}}
| style="text-align:right;" | {{#expr:127*ln2/ln10 round 2}}
| rowspan="2" | Complex scientific {{shy|cal|cula|tions}},
[[IPv6]] addresses,
[[GUID]]s
| colspan="6" rowspan="2" style="text-align: center; background: #eaecf0;" | Only available as non-standard or compiler-specific extensions
| {{mono|cent}}{{efn|name=notede|garbage=reserved}}
| {{mono|i128}}
|-
| ''Unsigned:'' From 0 to {{tooltip|2<sup>128</sup>&nbsp;−&nbsp;1|340,282,366,920,938,463,463,374,607,431,768,211,455}}
| style="text-align: right;" | {{#expr:128*ln2/ln10 round 2}}
| {{mono|ucent}}{{efn|name=notede|garbage=reserved}}
| {{mono|u128}}
|-
| rowspan="2" style="text-align: right;" | ''n''
| rowspan="2" | ''n''-bit integer<br /> (general case)
| ''Signed:'' −(2<sup>''n''−1</sup>) to (2<sup>''n''−1</sup>&nbsp;−&nbsp;1)
| colspan="2" | (''n''&nbsp;−&nbsp;1) log<sub>10</sub>&nbsp;2
| C23: {{mono|_BitInt(n)}}, {{mono|signed _BitInt(n)}}
| colspan="7" | [[Ada (programming language)|Ada]]: {{code|range -2**(n-1)..2**(n-1)-1|ada}}
|-
| ''Unsigned:'' 0 to (2<sup>''n''</sup> − 1)
| colspan="2" | ''n'' log<sub>10</sub>&nbsp;2
| C23: {{mono|unsigned _BitInt(n)}}
| colspan="7" | Ada: {{code|range 0..2**n-1|ada}}, {{code|mod 2**n|ada}}; standard libraries' or third-party arbitrary arithmetic libraries' BigDecimal or Decimal classes in many languages such as Python, C++, etc.
|}

Different [[CPU]]s support different integral data types. Typically, hardware will support both signed and unsigned types, but only a small, fixed set of widths.

The table above lists integral type widths that are supported in hardware by common processors. High-level programming languages provide more possibilities. It is common to have a 'double width' integral type that has twice as many bits as the biggest hardware-supported type. Many languages also have ''bit-field'' types (a specified number of bits, usually constrained to be less than the maximum hardware-supported width) and ''range'' types (that can represent only the integers in a specified range).

Some languages, such as [[Lisp (programming language)|Lisp]], [[Smalltalk]], [[REXX]], [[Haskell]], [[Python (programming language)|Python]], and [[Raku (programming language)|Raku]], support ''arbitrary precision'' integers (also known as ''infinite precision integers'' or ''[[bignum]]s''). Other languages that do not support this concept as a top-level construct may have libraries available to represent very large numbers using arrays of smaller variables, such as Java's {{mono|BigInteger}} class or [[Perl]]'s "{{mono|bigint}}" package.<ref>{{cite web |url=http://download.oracle.com/javase/6/docs/api/java/math/BigInteger.html |title=BigInteger (Java Platform SE 6) |publisher=Oracle |access-date=2011-09-11 }}</ref> These use as much of the computer's memory as is necessary to store the numbers; however, a computer has only a finite amount of storage, so they, too, can only represent a finite subset of the mathematical integers. These schemes support very large numbers; for example one kilobyte of memory could be used to store numbers up to 2466 decimal digits long.

A [[Boolean data type|Boolean]] type is a type that can represent only two values: 0 and 1, usually identified with ''false'' and ''true'' respectively. <!-- Pascal has them the other way around --> This type can be stored in memory using a single bit, but is often given a full byte for convenience of addressing and speed of access.

A four-bit quantity is known as a ''[[nibble]]'' (when eating, being smaller than a ''bite'') or ''nybble'' (being a pun on the form of the word ''byte''). One nibble corresponds to one digit in [[hexadecimal]] and holds one digit or a sign code in binary-coded decimal.

=== Bytes and octets ===
{{Main article|Byte|Octet (computing)}}
The term ''byte'' initially meant 'the smallest addressable unit of memory'. In the past, 5-, 6-, 7-, 8-, and 9-bit bytes have all been used. There have also been computers that could address individual bits ('bit-addressed machine'), or that could only address 16- or 32-bit quantities ('word-addressed machine'). The term ''byte'' was usually not used at all in connection with bit- and word-addressed machines.

The term ''octet'' always refers to an 8-bit quantity. It is mostly used in the field of [[computer network]]ing, where computers with different byte widths might have to communicate.

In modern usage ''byte'' almost invariably means eight bits, since all other sizes have fallen into disuse; thus ''byte'' has come to be synonymous with ''octet''.

=== Words ===
{{Main article|Word (computer architecture)}}
The term 'word' is used for a small group of bits that are handled simultaneously by processors of a particular [[computer architecture|architecture]]. The size of a word is thus CPU-specific. Many different word sizes have been used, including 6-, 8-, 12-, 16-, 18-, 24-, 32-, 36-, 39-, 40-, 48-, 60-, and 64-bit. Since it is architectural, the size of a ''word'' is usually set by the first CPU in a family, rather than the characteristics of a later compatible CPU. The meanings of terms derived from ''word'', such as ''longword'', ''doubleword'', ''quadword'', and ''halfword'', also vary with the CPU and OS.<ref name="agnerfog" />

Practically all new desktop processors are capable of using 64-bit words, though [[Embedded system|embedded processors]] with 8- and 16-bit word size are still common. The [[36-bit word length]] was common in the early days of computers.

One important cause of non-portability of software is the incorrect assumption that all computers have the same word size as the computer used by the programmer. For example, if a programmer using the C language incorrectly declares as {{mono|int}} a variable that will be used to store values greater than 2<sup>15</sup>−1, the program will fail on computers with 16-bit integers. That variable should have been declared as {{mono|long}}, which has at least 32 bits on any computer. Programmers may also incorrectly assume that a pointer can be converted to an integer without loss of information, which may work on (some) 32-bit computers, but fail on 64-bit computers with 64-bit pointers and 32-bit integers. This issue is resolved by C99 in [[stdint.h]] in the form of {{code|intptr_t}}.

The ''bitness'' of a program may refer to the word size (or bitness) of the processor on which it runs, or it may refer to the width of a memory address or pointer, which can differ between execution modes or contexts. For example, 64-bit versions of [[Microsoft Windows]] support existing 32-bit binaries, and programs compiled for Linux's [[x32 ABI]] run in 64-bit mode yet use 32-bit memory addresses.<ref>{{cite news |author=Thorsten Leemhuis |date=2011-09-13 |title=Kernel Log: x32 ABI gets around 64-bit drawbacks |url=http://www.h-online.com/open/features/Kernel-Log-x32-ABI-gets-around-64-bit-drawbacks-1342061.html |archive-url=https://web.archive.org/web/20111028081253/http://www.h-online.com/open/features/Kernel-Log-x32-ABI-gets-around-64-bit-drawbacks-1342061.html |archive-date=28 October 2011 |access-date=2011-11-01 |publisher=www.h-online.com}}</ref>

=== Standard integer ===
The standard integer size is platform-dependent.

In [[C (programming language)|C]], it is denoted by {{mono|int}} and required to be at least 16 bits. Windows and Unix systems have 32-bit {{mono|int}}s on both 32-bit and 64-bit architectures.

=== Short integer ===
A ''short integer'' can represent a whole number that may take less storage, while having a smaller range, compared with a standard integer on the same machine.

In [[C (programming language)|C]], it is denoted by {{mono|short}}. It is required to be at least 16 bits, and is often smaller than a standard integer, but this is not required.<ref name="c99" /><ref name="drdobbsinteger" /> A conforming program can assume that it can safely store values between −(2<sup>15</sup>−1)<ref name="c-std-6.2.6.2p2">{{cite web |url=http://www.open-std.org/JTC1/SC22/WG14/www/docs/n1570.pdf |at=section 6.2.6.2, paragraph 2|title=ISO/IEC 9899:201x |access-date=2016-06-20 |publisher=open-std.org}}</ref> and 2<sup>15</sup>−1,<ref name="c-std-5.2.4.2.1">{{cite web |url=http://www.open-std.org/JTC1/SC22/WG14/www/docs/n1570.pdf |at=section 5.2.4.2.1|title=ISO/IEC 9899:201x |access-date=2016-06-20 |publisher=open-std.org}}</ref> but it may not assume that the range is not larger. In [[Java (programming language)|Java]], a {{mono|short}} is ''always'' a 16-bit integer. In the [[Windows API]], the datatype {{mono|SHORT}} is defined as a 16-bit signed integer on all machines.<ref name="agnerfog" />

{| class="wikitable"
|+ Common short integer sizes
|-
! [[Programming language]]
! Data type name
! [[Signedness]]
! Size in [[bytes]]
! Minimum value
! Maximum value
|-
| rowspan="2" | [[C (programming language)|C]] and [[C++]]
| {{mono|short}}
| signed
| style="text-align: right;" | 2
| style="text-align: right;" | −32,767{{efn|The ISO C standard allows implementations to reserve the value with sign bit 1 and all other bits 0 (for sign–magnitude and two's complement representation) or with all bits 1 (for ones' complement) for use as a "trap" value, used to indicate (for example) an overflow.<ref name="c-std-6.2.6.2p2" />}}
| style="text-align: right;" | +32,767
|-
| {{mono|unsigned short}}
| unsigned
| style="text-align: right;" | 2
| style="text-align: right;" | 0
| style="text-align: right;" | 65,535
|-
| rowspan="2" | [[C Sharp (programming language)|C#]]
| {{mono|short}}
| signed
| style="text-align: right;" | 2
| style="text-align: right;" | −32,768
| style="text-align: right;" | +32,767
|-
| {{mono|ushort}}
| unsigned
| style="text-align: right;" | 2
| style="text-align: right;" | 0
| style="text-align: right;" | 65,535
|-
| [[Java (programming language)|Java]]
| {{mono|short}}
| signed
| style="text-align: right;" | 2
| style="text-align: right;" | −32,768
| style="text-align: right;" | +32,767
|-
| [[SQL]]
| {{mono|smallint}}
| signed
| style="text-align: right;" | 2 <!-- (implementation-defined) -->
| style="text-align: right;" | −32,768
| style="text-align: right;" | +32,767
|}

=== Long integer ===
A ''long integer'' can represent a whole [[integer]] whose [[range (computer science)|range]] is greater than or equal to that of a standard integer on the same machine.

In [[C (programming language)|C]], it is denoted by {{mono|long}}. It is required to be at least 32 bits, and may or may not be larger than a standard integer. A conforming program can assume that it can safely store values between −(2<sup>31</sup>−1)<ref name="c-std-6.2.6.2p2" /> and 2<sup>31</sup>−1,<ref name="c-std-5.2.4.2.1" /> but it may not assume that the range is not larger.
{{Table alignment}}
{| class="wikitable col6center"
|+ Common long integer sizes
|-
! [[Programming language]]
! Approval Type
! [[Computing platform|Platform]]s
! Data type name
! Storage in [[bytes]]
! [[Signedness|Signed]] range
! [[Signedness|Unsigned]] range
|-
| [[C (programming)|C]] ISO/ANSI C99
| International Standard
| [[Unix]], 16/32-bit systems<ref name="agnerfog" /><br />[[Windows]], 16/32/64-bit systems<ref name="agnerfog" />
| {{mono|long}}
| 4<br />(minimum require{{shy}}ment 4)
| {{val|−2147483647| to |+2147483647|fmt=commas}}
| {{val|0|to|4294967295|fmt=commas}}<br />(minimum require{{shy}}ment)
|-
| [[C (programming)|C]] ISO/ANSI C99
| International Standard
| [[Unix]],<br />64-bit systems<ref name="agnerfog" /><ref name="drdobbsinteger" />
| {{mono|long}}
| 8<br />(minimum require{{shy}}ment 4)
| {{val|−9223372036854775807|to|+9223372036854775807|fmt=commas}}
| {{val|0|to|18446744073709551615|fmt=commas}}
|-
| [[C++]] ISO/ANSI
| International Standard
| [[Unix]], [[Windows]],<br />16/32-bit system
| {{mono|long}}
| 4 <ref>{{cite web |title=Fundamental types in C++|url=http://cppreference.com/wiki/language/types|publisher=cppreference.com|access-date=5 December 2010}}</ref><br />(minimum require{{shy}}ment 4)
| {{val|−2147483648|to|+2147483647|fmt=commas}}<br />
| {{val|0|to|4294967295|fmt=commas}}<br />(minimum require{{shy}}ment)
|-
| [[C++/CLI]]
| International Standard<br />[[ECMA-372]]
| [[Unix]], [[Windows]],<br />16/32-bit systems
| {{mono|long}}
| 4 <ref>{{cite web| url=http://www.ecma-international.org/publications/files/ECMA-ST/ECMA-372.pdf |title=Chapter 8.6.2 on page 12|publisher=ecma-international.org}}</ref><br />(minimum require{{shy}}ment 4)
| {{val|−2147483648|to|+2147483647|fmt=commas}}<br />
| {{val|0| to |4294967295|fmt=commas}}<br />(minimum require{{shy}}ment)
|-
| [[Visual Basic (classic)|VB]]
| Company Standard
| [[Windows]]
| {{mono|Long}}
| 4 <ref>VB 6.0 help file</ref>
| {{val|−2147483648|to|+2147483647|fmt=commas}}
| {{n/a}}
|-
| [[Visual Basic for Applications|VBA]]
| Company Standard
| [[Windows]], [[Mac OS X]]
| {{mono|Long}}
| 4<ref>{{cite web |url=http://msdn2.microsoft.com/en-us/library/aa164754(office.10).aspx |title=The Integer, Long, and Byte Data Types (VBA) |access-date=2006-12-19 |publisher=microsoft.com}}</ref>
| {{val|−2147483648| to |+2147483647|fmt=commas}}
| {{n/a}}
|-
| [[Microsoft SQL Server|SQL Server]]
| Company Standard
| [[Windows]]
| {{mono|BigInt}}
| 8
| {{val|−9223372036854775808|to|+9223372036854775807|fmt=commas}}
| {{val|0|to|18446744073709551615|fmt=commas}}
|-
| [[C Sharp (programming language)|C#]]/ [[VB.NET]]
| ECMA International Standard
| [[Microsoft .NET]]
| {{mono|long}} or {{mono|Int64}}
| 8
| {{val|−9223372036854775808|to|+9223372036854775807|fmt=commas}}
| {{val|0|to|18446744073709551615|fmt=commas}}
|-
| [[Java (programming language)|Java]]
| International/Company Standard
| [[Java platform]]
| {{mono|long}}
| 8
| {{val|−9223372036854775808|to|+9223372036854775807|fmt=commas}}
| {{n/a}}
|-
| [[Pascal (programming language)|Pascal]]
| ?
| [[Windows]], [[UNIX]]
| {{mono|int64}}
| 8
| {{val|−9223372036854775808|to|+9223372036854775807|fmt=commas}}
| {{val|0|to|18446744073709551615|fmt=commas}} (Qword type)
|}

=== Long long ===
{{Redirect|long long||Long (disambiguation)}}
In the [[C99]] version of the [[C programming language]] and the [[C++11]] version of [[C++]], a <CODE>long long</CODE> type is supported that has double the minimum capacity of the standard <CODE>long</CODE>. This type is not supported by compilers that require C code to be compliant with the previous C++ standard, C++03, because the {{mono|long long}} type did not exist in C++03. For an ANSI/ISO compliant compiler, the minimum requirements for the specified ranges, that is, −(2<sup>63</sup>&minus;1)<ref name="c-std-6.2.6.2p2" /> to 2<sup>63</sup>−1 for signed and 0 to 2<sup>64</sup>−1 for unsigned,<ref name="c-std-5.2.4.2.1" /> must be fulfilled; however, extending this range is permitted.<ref>{{cite web| url=http://www.ericgiguere.com/articles/ansi-c-summary.html|title=The ANSI Standard: A Summary for the C Programmer|first=Eric|last=Giguere|date=December 18, 1987|access-date=2010-09-04}}</ref><ref>{{cite web|url=http://flash-gordon.me.uk/ansi.c.txt|title=American National Standard Programming Language C specifies the syntax and semantics of programs written in the C programming language.|access-date=2010-09-04|url-status=dead|archive-url=https://web.archive.org/web/20100822072551/http://flash-gordon.me.uk/ansi.c.txt|archive-date=2010-08-22}}</ref> This can be an issue when exchanging code and data between platforms, or doing direct hardware access. Thus, there are several sets of headers providing platform independent exact width types. The C [[standard library]] provides ''[[stdint.h]]''; this was introduced in C99 and C++11.