Code page is the traditional IBM term used for a specific character encoding table: a mapping in which a sequence of bits, usually a single octet representing integer values 0 through 255, is associated with a specific character. A few code pages use more than 8 bits per character and thus encode more than 256 characters.
Although IBM created and maintained many code pages, the term came to be associated primarily with character maps used by the IBM PC and compatible platforms, especially prior to the advent of Unicode-capable programming languages and operating systems.
To this day, it is typical for PC hardware to support a single 8-bit code page that is, by default, for a particular regional market, and to make available mechanisms for operating systems to switch to other code pages. However, it is now commonplace for operating system vendors to provide their own character encoding and rendering systems that bypass the hardware code pages entirely. These alternative character encodings are sometimes called code pages as well.
Relationship to ASCII
The basis of many code pages is ASCII, originally a 7-bit code representing, at most, 128 characters. In the past, 8-bit representations of ASCII typically either set the top bit to zero, or used it as a parity bit in network data transmissions. When this bit was instead made available for representing character data, another 128 characters could be represented. IBM used this extended range to encode characters used by various languages. No formal standard existed for these ‘extended character sets’; IBM merely referred to the variants as code pages, as it had always done for variants of EBCDIC encodings.
Partial List of IBM Code Pages
Since the original IBM PC code page (number 437) was not really designed for international use, several incompatible variants emerged. Examples include:
- 737 -- Greek characters
- 850 -- Multilingual, most European languages
- 858 -- Multilingual with euro symbol
- 860 -- Portugal
- 863 -- French Canadian
- 865 -- Nordic
- 868 -- Urdu
- 899 -- Symbol
- 904 -- Taiwan
- 1088 -- Revised Korean
- 1114 -- Taiwan
- 1252 -- Superset of ISO 8859-1, used by Microsoft Windows
Other code pages of note are:
- 10000 -- Macintosh Roman (followed by several other Mac character sets)
- 12000 -- Unicode little-endian, 12001 big-endian
- 20000 -- CNS Taiwan, followed by other national character sets
In modern applications, operating systems and programming languages, the IBM code pages have been rendered obsolete by international standards, such as ISO 8859-1 and Unicode.
Microsoft code pages
Microsoft defined a number of proprietary code page extensions which were subtly incompatible with those by other vendors.
The most notable of these is the windows-1252 code page, which contains a range of typographical punctuation characters, the euro sign, and a few other special characters, in character positions which were reserved for control characters in the ISO 8859-1 "latin-1" code page.
Many Microsoft products produce characters in these ranges automatically, notably with ‘smartquotes’. This means that other software has to choose between
- not interoperating with documents produced with Microsoft applications
- mis-rendering the text in question
- adding support for the Microsoft code pages, in effect making Microsoft’s implementation a de facto standard.
These code pages were sometimes viewed as part of Microsoft’s embrace, extend and extinguish strategy towards open standards. On the other hand, when standards bodies decided to not assign characters to the control code positions 80–9F, a precious 12.5% of the available space appeared to be wasted. This, perhaps, was not in users’ best interests, either. Fortunately, the ongoing transition to Unicode support now offers standards-based applications the possibility of full interoperability with the character repertoire of these documents without giving up standards compliance on output.
Private code pages
When, early in the history of personal computers, users didn’t find their character encoding requirements met, private or local codepages were created using Terminate and Stay Resident utilities or by re-programming BIOS EPROMs. In some cases, unofficial code page numbers were invented (e.g., cp895).
When more diverse character set support became available most of those code pages fell into disuse, with some exceptions such as the Kamenicky or KEYBCS2 encoding for the Czech and Slovak alphabets.