" (TILDE)
indicates a negative value. For monospaced and character cell fonts, this is the width of
all glyphs in the font. | Page: | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
|---|
Version 1.5
X Consortium Standard
X Version 11, Release 6
Jim Flowers
Digital Equipment Corporation
Version 1.5 edited by Stephen Gildea
X Consortium, Inc.
X Window System is a trademark of X Consortium, Inc.
Helvetica and Times are registered trademarks of Linotype Company.
ITC Avant Garde Gothic is a registered trademark of International Typeface Corporation.
Times Roman is a registered trademark of Monotype Corporation.
Bitstream Amerigo is a registered trademark of Bitstream Inc.
Stone is a registered trademark of Adobe Systems Inc.
Copyright © 1988, 1994 x Consortium
Permission is hereby granted, free of change, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
It is a requirement that X client applications must be portable across server implementations, with very different file systems, naming conventions, and font libraries. However, font access requests, as defined by the X Window System Protocol, neither specify server-independent conventions for font names nor provide adequate font properties for logically describing typographic fonts.
X clients must be able to dynamically determine the fonts available on any given server so that understandable information can be presented to the user or that intelligent font fallbacks can be chosen. It is desirable for the most common queries to be accomplished without the overhead of opening each font and inspecting font properties, by means of simple ListFonts requests. For example, if a user selected a Helvetica typeface family, a client application should be able to query the server for all Helvetica fonts and present only those setwidths, weights, slants, point sizes, and character sets available for that family.
This document gives a standard logical font description (hereafter referred to as XLFD) and the conventions to be used in the core protocol so that clients can query and access screen type libraries in a consistent manner across all X servers. In addition to completely specifying a given font by means of its FontName, the XLFD also provides for a standard set of key FontProperties that describe the font in more detail.
The XLFD provides an adequate set of typographic font properties, such as CAP_HEIGHT, X_HEIGHT, RELATIVE_SETWIDTH, for publishing and other applications to do intelligent font matching or substitution when handling documents created on some foreign server that use potentially unknown fonts. In addition, this information is required by certain clients to position subscripts automatically and determine small capital heights, recommended leading, word-space values, and so on.
The XLFD meets the short and long-term goals to have a standard logical font description that:
It should be possible to have font names that are long enough and descriptive enough to have a reasonable probability of being unique without inventing a new registration organization. Resolution and size-dependent font masters, multivendor font libraries, and so on must be anticipated and handled by the font name alone.
The name itself should be structured to be amenable to simple pattern matching and parsing, thus, allowing X clients to restrict font queries to some subset of all possible fonts in the server.
The font name and properties should distinguish between fonts that were supplied by different font vendors but that possibly share the same name. We anticipate a highly competitive font market where users will be able to buy fonts from many sources according to their particular requirements.
A number of font vendors deliver each font with all glyphs designed for that font, where charset mappings are defined by encoding vectors. Some server implementations may force these mappings to proprietary or standard charsets statically in the font data. Others may desire to perform
the mapping dynamically in the server. Provisions must be made in the font name that allows a font request to specify or identify specific charset mappings in server environments where multiple charsets are supported.
If a font source can be scaled to arbitrary size or varied in other ways, it should be possible for an application to determine that fact from the font name, and the application should be able to construct a font name for any specific instance.
Arbitrary two-dimensional linear transformations of fonts should be able to be requested by applications. Since such transformed fonts may be used for special effects requiring a few characters from each of many differently-transformed fonts, it should be possible to request only a few characters from a font for efficiency.
X client applications that require a particular font should be able to use the descriptive name without knowledge of the file system or other repository in use by the server. However, it should be possible for servers to translate a given font name into a file name syntax that it knows how to deal with, without compromising the uniqueness of the font name. This algorithm should be reversible (exactly how this translation is done is implementation dependent).
In addition to the font name, the XLFD should define a standard list of descriptive font properties, with agreed upon fallbacks for all fonts. This allows client applications to derive font-specific formatting or display data and to perform font matching or substitution when asked to handle potentially unknown fonts, as required.
The XLFD must be extensible so that new and/or private descriptive font properties can
be added to conforming fonts without making existing X client or server implementations
obsolete.
XLFD is divided into two basic components: the FontName, which gives all font information needed to uniquely identify a font in X protocol requests (for example, OpenFont, ListFonts, and so on) and a variable list of optional FontProperties, which describe a font in more detail.
The FontName is used in font queries and is returned as data in certain X protocol requests. It is also specified as the data value for the FONT item in the X Consortium Character Bitmap Distribution Format Standard (BDF V2.1).
The FontProperties are supplied on a font-by-font basis and are returned as data in certain X protocol requests as part of the XFontStruct data structure. The names and associated data values for each of the FontProperties may also appear as items of the STARTPROPERTIES...ENDPROPERTIES list in the BDF V2. 1 specification.
Each FontName is logically composed of two string: a FontNameRegistry prefix that is followed by a FontName Suffix. The FontName uses the ISO 8859-l encoding. The FontNameRegistry is an xregistered-name (a name that has been registered with the X Consortium) that identifies the registration authority that owns the specified FontNameSuffix syntax and semantics.
All font names that conform to this specification are to use a FontNameRegistry prefix, which is defined to be the string "-" (HYPHEN). All FontNameRegistry prefixes of the form: +version-, where the specified version indicates some future XLFD specification, are reserved by the X Consortium for future extensions to XLFD font names. If required, extensions to the current XLFD font name shall be constructed by appending new fields to the current structure, each delimited by the existing field delimiter. The availability of other FontNameRegistry prefixes or fonts that support other registries is server implementation dependent.
In the X protocol specification, the FontName is required to be a string; hence, numeric field values are represented in the name as string equivalents. All FontNameSuffix fields are also defined as FontProperties; numeric property values are represented as signed or unsigned integers, as appropriate.
The FontName is a structured, parsable string (of type STRING8) whose Backus-Naur Form syntax description is as follows:
| FontName :: | = | XFontNameRegistry XFontNameSuffix IPrivFontNameRegistryPrivFontNameSuffix |
| XFontNameRegistry :: | = | XFNDelim I XFNExtPrefix Version XFNDelim |
| XFontNameSuffix :: | = | FOUNDRY XFNDelim FAMILY_NAME XFNDelim WEIGHT_NAME |
| XFNDelim SLANT XFNDelim SETWIDTH_NAME XFNDelim ADD_ | ||
| STYLE_NAME XFNDelim PIXEL_SIZE XFNDelim POINT_SIZE | ||
| XFNDelim RESOLUTION_X XFNDelim RESOLUTION_Y XFNDelim | ||
| SPACING XFNDelim AVERAGE_WIDTH XFNDelim | ||
| CHARSET_REGISTRY XFNDelim CHARSET_ENCODING | ||
| Version :: | = | STRING8 - the XLFD version that defines an extension to the font name syntax (for example, " 1.4") |
| XFNExtPrefix :: | = | OCTET - "+" (PLUS) |
| XFNDelim :: | = | OCTET - "_" (HYPHEN) |
| PrivFontNameRegistry :: | = | STRING8 - other than those strings reserved by XLFD |
| PrivFontNameSuffix :: | = | STRING8 |
Field values are constructed as strings of IS08859-1 graphic characters, excluding the following:
Alphabetic case distinctions are allowed but are for human readability concerns only. Conforming X servers will perform matching on font name query or open requests independent of case. The entire font name string must have no more than 255 characters. It is recommended that clients construct font name query patterns by explicitly including all field delimiters to avoid unexpected results. Note that SPACE is a valid character of a FontName field; for example, the string "ITC Avant Garde Gothic" might be a FAMILY_NAME.
This section discusses the FontName:
FOUNDRY is an x-registered-name, the name or identifier of the digital type foundry that digitized and supplied the font data, or if different, the identifier of the organization that last modified the font shape or metric information.
The reason this distinction is necessary is that a given font design may be licensed from one source (for example, ITC) but digitized and sold by any number of different type suppliers. Each digital version of the original design, in general, will be somewhat different in metrics and shape from the idealized original font data, because each font foundry, for better or for worse, has its own standards and practices for tweaking a typeface for a particular generation of output technologies or has its own perception of market needs.
It is up to the type supplier to register with the X Consortium a suitable name for this FontName field according to the registration procedures defined by the Consortium.
The X Consortium shall define procedures for registering foundry and other names and shall maintain and publish, as part of its public distribution, a registry of such registered names for use in XLFD font names and properties.
FAMILY_NAME is a string that identifies the range or "family" of typeface designs that are all variations of one basic typographic style. This must be spelled out in full, with words separated by spaces, as required. This name must be human-understandable and suitable for presentation to a font user to identify the typeface family.
It is up to the type supplier to supply and maintain a suitable string for this field and font property, to secure the proper legal title to a given name, and to guard against the infringement of other's copyrights or trademarks. By convention, FAMILY_NAME is not translated. FAMILY_NAME may include an indication of design ownership if considered a valid part of the typeface family name.
The following are examples of FAMILY_NAME:
WEIGHT_NAME is a string that identifies the font's typographic weight, that is, the nominal blackness of the font, according to the FOUNDRY's judgment. This name must be human-understandable and suitable for presentation to a font user.
The value "0" is used to indicate a polymorphic font; see Section 5.5. The interpretation of this field is somewhat problematic because the typographic judgment of weight has traditionally depended on the overall design of the typeface family in question; that is, it is possible that the DemiBold weight of one font could be almost equivalent in typographic feel to a Bold font from another family.
WEIGHT_NAME is captured as an arbitrary string because it is an important part of a font's complete human-understandable name. However, it should not be used for font matching or substitution. For this purpose, X client applications should use the weight-related font properties (RELATIVE_WEIGHT and WEIGHT) that give the coded relative weight and the calculated weight, respectively.
SLANT is a code-string that indicates the overall posture of the typeface design used in the font. The encoding is as follows:
| Code | EnglishTranslation | Description |
|---|---|---|
| "R" | Roman | Upright design |
| "I" | Italic | Italic design, slanted clockwise from the vertical |
| "O" | Oblique | Obliqued upright design, slanted clockwise from the vertical |
| "RI" | Reverse Italic | Italic design, slanted counterclockwise from the vertical |
| "RO" | Reverse Oblique | Obliqued upright design, slanted counterclockwise from the vertical |
| "OT" | Other | Other |
| numeric | Polymorphic | See Section 5.5 on polymorphic font support. |
The SLANT codes are for programming convenience only and usually are converted into their equivalent human-understandable form before being presented to a user.
SETWIDTH_NAME is a string that gives the font's typographic proportionate width, that is, the nominal width per horizontal unit of the font, according to the FOUNDRY's judgment.
The value "0" is used to indicate a polymorphic font; see Section 5.5.
As with WEIGHT_NAME, the interpretation of this field or font property is somewhat problematic, because the designer's judgment of setwidth has traditionally depended on the overall design of the typeface family in question. For purposes of font matching or substitution, X client applications should either use the RELATIVE_SETWIDTH font property that gives the relative coded proportionate width or calculate the proportionate width.
The following are examples of SETWIDTH_NAME:
ADD_STYLE_NAME is a string that identifies additional typographic style information that is not captured by other fields but is needed to identify the particular font.
The character [ anywhere in the field is used to indicate a polymorphic font; see Section 5.5.
ADD_STYLE_NAME is not a typeface classification field and is only used for uniqueness. Its use, as such, is not limited to typographic style distinctions.
The following are examples of ADD_STYLE_NAME:
PIXEL_SIZE gives the body size of the font at a particular POINT_SIZE and RESOLUTION_Y. PIXEL_SIZE is either an integer-string or a string beginning with "[". A string beginning with "[" represents a matrix; see Section 5.4. PIXEL_SIZE usually incorporates additional vertical spacing that is considered part of the font design. (Note, however, that this value is not necessarily equivalent to the height of the font bounding box.) Zero is used to indicate a scalable font; see Section 5.4.
PIXEL_SIZE usually is used by X client applications that need to query fonts according to device-dependent size, regardless of the point size or vertical resolution for which the font was designed.
POINT_SIZE gives the body size for which the font was designed. POINT_SIZE is either an integer-string or a string beginning with "[". A string beginning with represents a matrix; see Section 5.4. This field usually incorporates additional vertical spacing that is considered part of the font design. (Note, however, that POINT_SIZE is not necessarily equivalent to the height of the font bounding box.) POINT_SIZE is expressed in descriptions (where points are as defined in the X protocol or 72.27 points equal l inch) Zero is used to indicate a scalable font; see Section 5.4.
POINT_SIZE and RESOLUTION_Y are used by X clients to query fonts according to device-independent size to maintain constant text size on the display regardless of the PIXEL_SIZE used for the font.
RESOLUTION_X and RESOLUTION_Y are unsigned integer-strings that give the horizontal and vertical resolution, measured in pixels or dots per inch (dpi), for which the font was designed. Zero is used to indicate a scalable font; see Section 5.4. Horizontal and vertical values are required because a separate bitmap font must be designed for displays with very different aspect ratios (for example, 1:1, 4:3, 2:1, and so on).
The separation of pixel or point size and resolution is necessary because X allows for servers with very different video characteristics (for example, horizontal and vertical resolution, screen and pixel size, pixel shape, and so on) to potentially access the same font library. The font name, for example, must differentiate between a 14 point font designed for 75 dpi (body size of about 14 pixels) or a 14 point font designed for 150 dpi (body size of about 28 pixels). Further, in servers that implement some or all fonts as continuously scaled and scan-converted outlines, POINT_SIZE and RESOLUTION_Y will help the server to differentiate between potentially separate font masters for text, title, and display sizes or for other typographic considerations.
| Home |
|---|
SPACING is a code-string that indicates the escapement class of the font, that is, monospace (fixed pith), proportional (variable pitch), or charcell (a special monospaced font that conforms to the traditional data processing character cell font model). The encoding is as follows:
| Code | English Translation | Description |
| "P" | Proportional | A font whose logical character widths vary for each glyph. Note that no other restrictions are placed on the metrics of a proportional font. |
| "M" | Monospaced | A font whose logical character widths are constant (that is, every glyph in the font has the same logical width). No other restrictions are placed on the metrics of a monospaced font. |
| "C" | CharCell | A monospaced font that follows the standard typewriter character cell model (that is, the glyphs of the font can be modeled by X clients as "boxes" of the same width and height that are imaged side-by-side to form text strings or top-to-bottom to form text lines. By definition, all glyphs have the same logical character width, and no glyphs have "ink" outside of the character cell. There is no kerning (that is, on a per character basis with positive metrics: 0 <= left-bearing <= right bearing <= width; with negative metrics: width <= left-bearing <= right-bearing <= zero). Also, the vertical extents of the font do not exceed the vertical spacing (that is, on a per character basis: ascent <= font-ascent and descent <= font-descent). The cell height = font descent + font-ascent, and the width = AVERAGE_WIDTH. |
AVERAGE_WIDTH is an integer-string typographic metric value that gives the unweighted
arithmetic mean of the absolute value of the width of each glyphs in the font (measured in
tenths of pixels), multiplied by -1 if the dominant writing direction for the font is
right-to-left. A leading "" (TILDE)
indicates a negative value. For monospaced and character cell fonts, this is the width of
all glyphs in the font.
Zero is used to indicate a scalable font; see Section 5.4.
The character set used to encode the glyphs of the font (and implicitly the font's glyph repertoire), as maintained by the X Consortium character set registry. CHARSET_REGISTRY is an xregistered-name that identifies the registration authority that owns the specified encoding. CHARSET_ENCODING is a registered-name that identifies the coded character set as defined by that registration authority and, optionally, a subsetting hint.
Although the X protocol does not explicitly have any knowledge about character set encodings, it is expected that server implementers will prefer to embed knowledge of certain proprietary or standard charsets into their font library for reasons of performance and convenience. The CHARSET_REGISTRY and CHARSET_ENCODING fields or properties allow an X client font request to specify a specific charset mapping in server environments where multiple charsets are supported. The availability of any particular character set is font and server implementation dependent.
To prevent collisions when defining character set names, it is recommended that CHARSET_REGISTRY and CHARSET_ENCODING name pairs be constructed according to the following conventions:
| CharsetRegistry :: | = | StdCharsetRegistryName | PrivCharsetRegistryName |
| CharsetEncoding :: | = | StdCharsetEncodingName | PrivCharsetEncodingName |
| StdCharsetRegistryName :: | = | StdOrganizationId StdNumber | StdOrganizationId StdNumber Dot Year |
| PrivCharsetRegistryName :: | = | OrganizationId STRING8 |
| StdCharsetEncodingName :: | = | STRING8-numeric part number of referenced standard |
| PrivCharsetEncodingName :: | = | STRING8 |
| StdOrganizationId :: | = | STRING8-the registered name or acronym of the referenced standard organization |
| StdNumber :: | = | STRING8-referenced standard number |
| OrganizationId :: | = | STRING8-the registered name or acronym of the organization |
| Dot :: | = | OCTET-"." (F1JLL STOP) |
| Year :: | = | STRING8-numeric year (for example, 1989) |
The X Consortium shall maintain and publish a registry of such character set names for use in X protocol font names and properties as specified in XLFD.
The ISO Latin- 1 character set shall be registered by the X Consortium as the CHARSET_REGISTRY-CHARSET_ENCODING value pair: "IS08859-1".
If the CHARSET_ENCODING contains a "[" (LEFT SQUARE BRACKET), the "[" and the characters after it up to a "]" (RIGHT SQUARE BRACKET) are subsetting hint telling the font source that the client is interested only i a subset of the characters of the font. The font source can, optionally, return a font that contains only those characters or any superset of those characters. The client can expect to obtain valid glyphs and metrics only for those characters, and not for any other characters in the font. The font properties may optionally be calculated by considering only the characters in the subset.
The BNF for the subsetting hint is
| Subset :: | = | LeftBracket RangeList RightBracket |
| RangeList :: | = | Range | Range Space RangeList |
| Range :: | = | Number | Number Underscore Number |
| Number :: | = | "0x" HexNumber | DecNumber |
| HexNumber :: | = | HexDigit | HexDigit HexNumber |
| DecNumber :: | = | DecDigit | DecDigit DecNumber |
| DecDigit :: | = | "0"|"1"|"2"|"3"|"4"|"5"|"6"|"7"|"8"|"9" |
| HexDigit :: | = | DecDigt | "a" | "b" | "c" | "d" | "e" | "f' |
| LeftBracket : | "]" (RIGHT SQUARE BRACKET) | |
| Space :: | = | " " (SPACE) |
| Underscore :: | = | "_" (LOW LINE) |
Each Range specifies characters that are to be part of the subset included in the font. A range containing two Numbers specifies the first and last character, inclusively, of a rage of characters. A Range that is a single Number specifies a single character to be included in the font. A HexNumber is interpreted as a hexadecimal number. A DecNumber is interpreted as a decimal number. The font consists of the union of all the Ranges in the RangeList.
tells the font source that the client is interested only in characters 65, 70, and 80-90.
The following examples of font names are derived from the screen fonts shipped with the MIT X distribution.
| Font | XFontName |
| 75 dpi Fonts | |
| Charter 12 pt | -Bitstream-Charter-Medium-R-Normal-- 12-1 20-75-75-P-68-lSO8859- 1 |
| Charter Bold 12 pt | -Bitstream-Charter-Bold-R-Normal--12-120-75-75-P-76-lSO8859-1 |
| Charter Bold Italic 12 pt | -Bitstream-Chaner-Bold-I-Normal--12-120-75-75-P-75-lSO8859-1 |
| Charter Italic 12 pt | -Bitstream-Chaner-Medium-I-Normal--12-120-75-75-P-66-lSO8859-1 |
| Courier 8 pt | -Adobe-Courier-Medium-R-Normal--8-80-75-75-M-50-lSO8859-1 |
| Courier 10 pt | -Adobe-Courier-Medium -R-Normal-- 10-100-75-75-M-60-lSO8859- 1 |
| Courier 12 pt | -Adobe-Courier-Medium-R-Normal-- 12-120-75-75-M-70-lSO8859- 1 |
| Courier 24 pt | -Adobe-Courier-Medium-R-Normal--24-240-75-75-M-150-lSO8859-1 |
| Courier Bold 10 pt | -Adobe-Courier-Bold-R-Normal-- 10-1 00-75-75-M-60-lSO8859- 1 |
| Courier Bold Oblique 10 pt | -Adobe-Courier-Bold-O-Normal--10-100-75-75-M-60-lSO8859-1 |
| Courier Oblique 10 pt | -Adobe-Courier-Medium-O-Normal--10-100-75-75-M-60-lSO8859-1 |
| 100 dpi Fonts | |
| Symbol 8 pt | -Adobe-Symbol-Medium-R-Normal-- I 1-80-100-1 00-P-6 1 -Adobe-FONTSPECIFIC |
| Symbol 10 pt | -Adobe-Symbol-Medium-R-Normal-- 14-100-100-1 00-P-85-Adobe-FONTSPECIFIC |
| Symbol 14 pt | -Adobe-Symbol-Medium-R-Normal--20- 140-100-1 00-P- 107-Adobe-FONTSPECIFIC |
| Symbol 18 pt | -Adobe-Symbol-Medium-R-Normal--25 - I 80-1 00-1 00-P- 142-Adobe-FONTSPECIFIC |
| Symbol 24 pt | -Adobe-Symbol-Medium-R-Normal--34-240- 100-1 00-P- 191 -Adobe-FONTSPEC IFIC |
| Times Bold 10 pt | -Adobe-Times-Bold-R-Normal -- 1 4-1 00-100-100-P-76-lSO8859- 1 |
| Times Bold Italic 10 pt | -Adobe-Times-Bold-I-Normal--14-100-100-100-P-77-lSO8859-1 |
| Times Italic 10 pt | -Adobe-Times-Medium-I-Normal-- 14-100-100-1 00-P-73-lSO8859- 1 |
| Times Roman 10 pt | -Adobe-Times-Medium-R-Normal-- 14-100-100-1 00-P-74-lSO8859- 1 |
All font properties are optional but will generally include the font name fields and, on a font-by font basis, any other useful font descriptive and use information that may be required to use the font intelligently. The XLFD specifies an extensive set of standard X font properties, their interpretation, and fallback rules when the property is not defined for a given font. The goal is to provide client applications with enough font information to be able to make automatic formatting and display decisions with good typographic results. Additional standard X font property definitions may be defined in the future and private proper ties may exist in X fonts at any time. Private font properties should be defined to conform to the general mechanism defined in the X protocol to prevent overlap of name space and ambiguous property names, that is, private font property names are of the form: ISO8859-1 UNDERSCORE property names, that is, private font property names are of the from: "_" (LOW LINE), followed by the organizational identifier, followed by "_" (LOW LINE), and terminated with the property name.
The Backus-Naur Form syntax description of X font properties is as follows:
| Properties | :: | = | OptFontPropList |
| OptFontPropList | :: | = | NULL | OptFontProp OptFontPropList |
| OptFontProp | :: | = | PrivateFontProp | XFontProp |
| PrivateFontProp | :: | = | STRING8 | Underscore OrganizationId Underscore STRING8 |
| XFontProp | :: | = | FOUNDRY | FAMLY_NAME | WEIGHT_NAME | SLANT | SETWIDTH_NAME | ADD_STYLE_NAME | PIXEL_SIZE | POINT_SIZE | RESOLUTION_X | RESOLUTION_Y | SPACING | AVERAGE_WIDTH | CHARSET_REGISTRY | CHARSET_ENCODING | QUAD_WIDTH | RESOLUTION | MIN_SPACE | NORM_SPACE | MAX_SPACE | END_SPACE | SUPERSCRIPT_X | SUPERSCRIPT_Y | SUBSCRIPT_X | SUBSCRIPT_Y | UNDERLINE_POSITION | UNDERLINE_THICKNESS | STRIKEOUT_ASCENT | STRIKEOUT_DESCENT | ITALIC_ANGLE | X_HEIGHT | WEIGHT | FACE_NAME | FIJLL-NAME | FONT | COPYRIGHT | AVG_CAPITAL WIDTH | AVG_LOWERCASE_WIDTH | RELATIVE_SETWIDTH | RELATIVE_WEIGHT | CAP_HEIGHT | SUPERSCRIPT_ SIZE | FIGURE_WIDTH | SUBSCRIPT_SIZE | SMALL_CAP_SIZE | NOTICE | DESTINATION | FONT_TYPE | FONT | VERSION | RASTERIZER_ NAME | RASTERIZER_VERSION | RAW_ASCENT | RAW_DESCENT | RAW_* | AXIS_NAME | AXIS_LIMIT | AXIS_TYPES |
| Underscore | :: | = | OCTET-"_" (LOW LINE) |
| Organizationld | :: | = | STRING8-the registered name of the organization |
FOUNDRY is as defined in the FontName except that the property type is ATOM. FOUNDRY cannot be calculated or defaulted if not supplied as a font property.
FAMILY_NAME is as defined in the FontName except that the property type is ATOM. FAMILY_NAME cannot be calculated or defaulted if not supplied as a font property.
if (WEIGHT_NAME undefined) then
WEIGHT_NAME = ATOM("Medium")
if (SLANT undefined) then
SLANT = ATOM("R")
if (SETWIDTH_NAME undefined) then
SETWIDTH_NAME = ATOM("Normal")
if (ADD_STYLE_NAME undefined) then
ADD_STYLE_NAME = ATOM(" ")
PIXEL_SIZE is as defined in the FontName except that the property type is INT32. X clients requiring pixel values for the various typographic fixed spaces (em space, en space and thin space), can use the following algorithm for computing these values from other properties specified for a font:
where a slash "/" denotes real division, the asterisk "*" denotes real multiplication, and "ROUND" denotes a function that rounds its real argument 'a' up or down to the next integer. This rounding is done according to X = FLOOR (a + O.5), where FLOOR is a function that rounds its real argument down to the nearest integer.
DeciPointsPerInch = 722.7 if (PIXEL_SIZE undefined) then PIXEL_SIZE = ROUND ((RESOLUTION_Y * PONT_SIZE) / DeciPointsPerlnch)
POINT_SIZE is as defined in the FontName except that the property type is INT32.
Design POINT_SIZE cannot be calculated or approximated.
RESOLUTION_X is as defined in the FontName except that the property type is CARD32.
RESOLUTION_X cannot be calculated or approximated.
RESOLUTION_Y is as defined in the FontName except that the property type is CARD32.
RESOLUTION_X cannot be calculated or approximated.
SPACING is as defined in the FontName except that the property type is ATOM.
SPACING can be calculated if not supplied as a font property, according to the definitions given above for the FontName.
AVERAGE_WIDTH is as defined in the FontName except that the property type is INT32.
AVERAGE_WIDTH can be calculated if not provided as a font property, according to the following algorithm:
if (AVERAGE_WIDTH undefined)then AVERGE_WIDTH=ROUND(MEAN(ABS(width of each glyph widths in font))*10) *(if(dominant writing direction L-to-R)then1else-1)
where MEAN is a function that returns the arithmetic mean of its arguments.
if (SPACING not proportional)then CharPitch=(RESOLUTION_X * 10) / AVERAGE_WIDTH
CHARSET_REGISTRY is as defined in the FontName except that the property type is ATOM.
CHARSET_REGISTRY cannot be defaulted if not supplied as a font property.
CHARSET_ENCODING is as defined in the FontName except that the property type is ATOM.
CHARSET_ENCODING cannot be defaulted if not supplied as a font property.
MIN_SPACE is an integer value (of type INT32) that gives the recommended minimum word-space value to be used with this font.
if (MIN_SPACE undefined)then MIN_SPACE = ROUND(0.75 * NORM_SPACE)
NORM_SPACE is an integer value (of type INT32) that gives the recommended normal word-space value to be used with this font.
NORM_SPACE can be approximated if not provided as a font property, according to the following algorithm:
DeciPointsPerInch = 722.7
if (NORM_SPACE undefined)then
if (SPACE glyph exists)then
NORM_SPACE = width of SPACE
else NORM_SPACE = ROUND((0.33 * RESOLUTION_X * POINT_SIZE)/ DeciPointsPerInch)
MAX_SPACE is an integer value (of type INT32) that gives the recommended maximum word-space value to be used with this font.
if (MAX_SPACE undefined) then MAX_SPACE= ROUND(1.5 * NORM_SPACE)
END_SPACE is an integer value (of type INT32) that gives the recommended spacing at the end of sentences.
if (END_SPACE undefined)then END_SPACE = NORM_SPACE
AVG_CAPITAL_WIDTH is an integer value (of type INT32) that gives the unweighted arithmetic mean of the absolute value of the width of each capital glyph in the font, in tenths of pixels, multiplied by -1 if the dominant writing direction for the font is right-to-left. This property applies to both Latin and non-Latin fonts. For Latin fonts, capitals are the glyphs A through Z. This property is usually used for font matching or substitution.
AVG_CAPITAL_WIDTH can be calculated if not provided as a font property, according to the following algorithm:
if (AVG_CAPITAL_WIDTH undefined)then
if (capitals exist)then
AVG_CAPITAL_WIDTH=ROUND(MEAN
(ABS (Width of each capital glyph))*10)
*(if(dominant writing direction L-to-R)then1else-1)
else AVG_CAPITAL_WIDTH undefined
if (AVG_LOWERCASE_WIDTH undefined)then
if (lowercase exists)then
AVG_LOWERCASE_WIDTH=ROUND(MEAN
(ABS(width of each lowercase glyph))*10)
*(if(dominant writing direction L-to-R)then1else-1)
else AVG_LOWERCASE_WIDTH undefined
QUAD_WIDTH is an integer typographic metric (of type INT32) that gives the width of a quad (em) space.
| Because all typographic fixed spaces (em, en, and thin) are constant for a given font size (that is, they do not vary according to setwidth), the use of this font property has been deprecated. X clients that require typographic fixed space values are encouraged to discontinue use of QUAD_WIDTH and compute these values from other font properties (for example, PIXEL_SIZE). X clients that require a font-dependent width value should use either the FIGURE_WIDTH or one of the average character width font properties (AVERAGE_WIDTH, AVG_CAPITAL_WIDTH or AVG_LOWERCASE_WIDTH). |
FIGURE_WIDTH is an integer typographic metric (of type INT32) that gives the width of the tabular figures and the dollar sign, if suitable for tabular setting (all widths equal). For Latin fonts, these tabular figures are the arabic numerals 0 through 9.
if (numerals and DOLLAR sign are defined & widths are equal)then FIGURE_WIDTH = width of DOLLAR else FIGURE_WIDTH property undefined
SUPERSCRIPT_X is an integer value (of type ST32) that gives the recommended horizontal offset in pixels from the position point to the X origin of synthetic superscript text. If the current position point is at [X,Y], then superscripts should begin at [X + SUPERSCRIPT X, Y SUPERSCRIPT_Y].
SUPERSCRIPT_X can be approximated if not provided as a font property, according to the following algorithm:
if (SUPERSCRIPT_X undefined)then
if (TANGENT(ITALIC_ANGLE) defined)then
SUPERSCRIPT_X = ROUND((0.40 * CAP_HEIGHT) / TANGENT(ITALIC_ANGLE))
else SUPERSCRIPT_X = ROUND(0.40 * CAP_HEIGHT)
where TANGENT is a trigonometric function that returns the tangent of its argument,
which is in l/64 degrees.
SUPERSCRIPT_Y is an integer value (of type INT32) that gives the recommended vertical offset in pixels from the position point to the Y origin of synthetic superscript text. If the current position point is at [X,Y], then superscripts should begin at [X + SUPERSCRIPT_X, Y SUPERSCRIPT_Y].
SUPERSCRIPT_Y can be approximated if not provided as a font property, according to the following algorithm:
if (SUPERSCRIPT_Y undefined)then SUPERSCRIPT_Y = ROUND(0.40 * CAP_HEIGHT)
| Home |
|---|
SUBSCRIPT_X is an integer value (of type INT32) that gives the recommended horizontal offset in pixels from the position point to the X origin of synthetic subscript text. If the current position point is at [X,Y], then subscripts should begin at [X + SUBSCRIPT_X, Y + SUBSCRIPT_Y].
SUBSCRIPT_X can be approximated if not provided as a font property, according to the following algorithm:
if (SUBSCRIPT_X undefined)then
if (TANGENT(ITALIC_ANGLE) defined)then
SUBSCRIPT_X = ROUND((0.40 * CAP_HEIGHT) / TANGENT(ITALIC_ANGLE))
else SUBSCRIPT_X = ROUND(0.40 * CAP_HEIGHT)
SUBSCRIPT_Y is an integer value (of type ST32) that gives the recommended vertical offset in pixels from the position point to the Y origin of synthetic subscript text. If the current position point is at [X,Y], then subscripts should begin at [X + SUBSCRIPT_X, Y + SUBSCRIPT_Y].
SUBSCRIPT_Y can be approximated if not provided as a font property, according to the following algorithm:
if (SUBSCRIPT_Y undefined)then SUBSCRIPT_Y = ROUND(0.40 * CAP_HEIGHT)
SUPERSCRIPT_SIZE is an integer value (of type INT32) that gives the recommended body size of synthetic superscripts to be used with this font, in pixels. This will generally be smaller than the size of the current font, that is, superscripts are imaged from a smaller font offset according to SUPERSCRIPT_X and SUPERSCRIPT_Y .
SUPERSCRIPT_SIZE can be approximated if not provided as a font property, according to the following algorithm:
if (SUPERSCRIPT_SIZE undefined)then SUPERSCRIPT_SIZE = ROUND(0.60 * PIXEL_SIZE)
SUBSCRIPT_SIZE is an integer value (of type INT32) that gives the recommended body size of synthetic subscripts to be used with this font, in pixels. As with SUPERSCRIPT_SIZE, this will generally be smaller than the size of the current font; that is, subscripts are imaged from a smaller font offset according to SUB SCRIPT_X and SUB SCRIPT_Y .
SUBSCRIPT_SIZE can be approximated if not provided as a font property, according to the algorithm:
if (SUBSCRIPT_SIZE undefined)then SUBSCRIPT_SIZE = ROUND(0.60 * PIXEL_SIZE)
SMALL_CAP_SIZE is an integer value (of type INT32) that gives the recommended body size of synthetic small capitals to be used with this font, in pixels. Mall capitals are generally imaged from a smaller font of slightly more weight. No offset [X, Y] is necessary.
SMALL CAP_SIZE can be approximated if not provided as a font property, according to the following algorithm:
if (SMALL_CAP_SIZE undefined)then
SMALL_CAP_SIZE = ROUND(PIXEL_SIZE * ((X_HEIGHT
+ ((CAP_HEIGHT- X_HEIGHT) / 3)) / CAP_HEIGHT))
UNDERLINE_POSITION is an integer value (of type 1NT32) that gives the recommended vertical offset in pixels from the baseline to the top of the underline. If the current position point is at [X, Y], the top of the baseline is given by [X, Y +UNDERLINE_POSITION].
UNDERLINE_POSITION can be approximated if not provided as a font property, according to the following algorithm:
if (UNDERLINE_POSITION undefined)then UNDERLINE_POSITION = ROUND((maximum descent) / 2)
where maximum descent is the maximum descent (below the baseline) in pixels of any
glyph in the font.
UNDERLINE_POSITION is an integer value (of type IN132) that gives the recommended underline thickness, in pixels.
CapStemWidth = average width of the stems of capitals if (UNDERLINE_THICKNESS undefined)then UNDERLINE_THICKNESS = CapStemWidth
STRIKEOUT_ASCENT is an integer value (of type INT32) that gives the vertical ascent for boxing or voiding glyphs in this font. If the current position is at [X,Y] and the string extent is EXTENT, the upper-left corner of the strikeout box is at [X, Y - STRIKEOUT_ASCENT] and the lower-right corner of the box is at [X + EXTENT, Y + STRIKEOUT_DESCENT].
STRIKEOUT_ASCENT can be approximated if not provided as a font property, according to the following algorithm:
if (STRIKEOUT_ASCENT undefined) STRIKEOUT_ASCENT = maximum ascent
where maximum ascent is the maximum ascent (above the baseline) in pixels of any glyph
in the font.
STRIKEOUT_DESCENT is an integer value (of type INT32) that gives the vertical descent for boxing or voiding glyphs in this font. If the current position is at [X,Y] and the string extent is EXTENT, the upper-left corner of the strikeout box is at [X, Y - STRIKEOUT_ASCENT] and the lower-right corner of the box is at [X + EXTENT, Y + STRIKEOUT_DESCENT].
STRIKEOUT_DESCENT can be approximated if not provided as a font property, according to
the following algorithm:
if (STRIKEOUT_DESCENT undefined) STRIKEOUT_DESCENT = maximum descent
where maximum descent is the maximum descent (below the baseline) in pixels of any glyph in the font.
ITALIC_ANGLE is an integer value (of type INT32) that gives the nominal posture angle of the typeface design, in 1/64 degrees, measured from the glyph origin counterclockwise from the three o'clock position.
ITALIC_ANGLE can be defaulted if not provided as a font property, according to the following algorithm:
if (ITALIC_ANGLE undefined)then ITALIC_ANGLE = (90 * 64)
CAP_HEIGHT is an integer value (of type INT32) that gives the nominal height of the capital letters contained in the font, as specified by the FOUNDRY or typeface designer. Where applicable, it is defined to be the height of the Latin uppercase letter X.
Certain clients require CAP_HEIGHT to compute scale factors and positioning offsets for synthesized glyphs where this information or designed glyphs are not explicitly provided by the font (for example, small capitals, superiors, inferiors, and so on). CAP_HEIGHT is also a critical factor in font matching and substitution.
CAP_HEIGHT can be approximated if not provided as a font property, according to the following algorithm:
if (CAP_HEIGHT undefined)then
if (latin font) then
CAP_HEIGHT = XCharStruct.ascent[glyph X]
else if (capitals exist) then
CAP_HEIGHT = XCharStruct.ascent[some capital glyph]
else CAP_HEIGHT undefined
X_HEIGHT is an integer value (of type INT32) that gives the nominal height above the
baseline of the lowercase glyphs contained in the font, as specified by the FOUNDRY or
typeface designer.
As with CAP_HEIGHT, X_HEIGHT is required by certain clients to compute scale factors for synthesized small capitals where this information is not explicitly provided by the font resource. X_HEIGHT is a critical factor in font matching and substitution.
X_HEIGHT can be approximated if not provided as a font property, according to the following algorithm:
if (X_HEIGHT undefined)then
if (latin font) then
X_HEIGHT = XCharStruct.ascent[glyph x]
else if (lowercase exists)then
X_HEIGHT=XCharStruct.ascent[some unaccented 1c glyph without an ascender]
else X_HEIGHT undefined
RELATIVE_SETWIDTH is an unsigned integer value (of type CARD32) that gives the coded proportionate width of the font, relative to all known fonts of the same typeface family, according to the type designer's or FOUNDRY's judgment.
RELATIVE_SETWIDTH ranges from 10 to 90, or is 0 if undefined or unknown. The following
reference values are defined:
| Code | English Translation | Description |
| 0 | Undefined | Undefined or unknown |
| 10 | UltraCondensed | The lowest ratio of average width to height |
| 20 | ExtraCondensed | |
| 30 | Condensed | Condensed, Narrow, Compressed,... |
| 40 | SemiCondensed | |
| 50 | Medium | Medium, Normal, Regular,... |
| 60 | SemiExpanded | SemiExpanded, DemiExpanded,... |
| 70 | Expanded | |
| 80 | ExtraExpanded | ExtraExpanded, Wide,... |
| 90 | UltraExpanded | The highest ratio of average width to height |
RELATIVE_SETWIDTH can be defaulted if not provided as a font property, according to the following algorithm:
if (RELATIVE_SETWIDTH undefined)then RELATIVE_SETWIDTH = 50
For polymorphic fonts, RELATIVE_SETWIDTH is not necessarily a linear function of the
font's setwidth axis.
X clients that want to obtain a calculated proportionate width of the font (that is, a font-independent way of identifying the proportionate width across all fonts and all font vendors) can use the following algorithm:
where SETWIDTH is a real number with zero being the narrowest calculated setwidth.
RELATIVE_WEIGHT is an unsigned integer value (of type CARD32) that gives the coded weight of the font, relative to all known fonts of the same typeface family, according to the type designer's or FOUNDRY's judgment.
RELATIVE WEIGHT ranges from 1 0 to 90, or is 0 if undefined or unknown. The following
reference values are defined:
| Code | EnglishTranslation | Description |
| 0 | Undefined | Undefined or unknown |
| 10 | UltraLight | The lowest ratio of stem width to height |
| 20 | ExtraLight | |
| 30 | Light | |
| 40 | SemiLight | SemiLight, Book,... |
| 50 | Medium | Medium,Normal, Regular,... |
| 60 | SemiBold | SemiBold, DemiBold,... |
| 70 | Bold | |
| 80 | ExtraBold | ExtraBold, Heavy,... |
| 90 | UltraBold | UltraBold, Black,..., the highest ratio of stem width to height |
RELATIVE_WEIGHT can be defaulted if not provided as a font property, according to the following algorithm:
if (RELATIVE_WEIGHT undefined)then RELATIVE_WEIGHT = 50
For polymorphic fonts, RELATIVE_WEIGHT is not necessarily a linear function of the
font's weight axis.
Calculated WEIGHT is an unsigned integer value (of type CARD32) that gives the calculated weight of the font, computed as the ratio of capital stem width to CAP_HEIGHT, in the range 0 to 1000, where 0 is the lightest weight.
WEIGHT can be calculated if not supplied as a font property, according to the following algorithm:
CapStemWidth = average width of the stems of capitals
if (WEIGHT undefined) then
WEIGHT = ROUND ((CapStemWidth * 1000) / CAP_HEIGHT)
A calculated value for weight is necessary when matching fonts from different families because both the RELATIVE_WEIGHT and the WEIGHT_NAME are assigned by the typeface supplier, according to its tradition and practice, and therefore, are somewhat subjective. Calculated WEIGHT provides a font-independent way of identifying the weight across all fonts and all font vendors.
RESOLUTION is an integer value (of type INT32) that gives the resolution for which this font was created, measured in 1/100 pixels per point.
| As independent horizontal and vertical design resolution components are required to accommodate displays with nonsquare aspect ratios, the use of this font property has been deprecated, and independent RESOLUTION_X and RESOLUTION_Y font name fields/properties have been defined (see Section 5.2.1.2.9 and Section 5.2.1.2.10). X clients are encouraged to discontinue use of the RESOLUTION property and are encouraged to use the appropriate X,Y resolution properties, as required. |
FONT is a string (of type ATOM) that gives the full XLFD name of the font-that is, the value can be used to open another instance of the same font.
If not provided, the FONT property cannot be calculated.
FACE_NAME is a human-understandable string (of type ATOM) that gives the full device-independent typeface name, including the owner, weight, slant, set, and so on but not the resolution, size, and so on. This property may be used as feedback during font selection.
FACE_NAME cannot be calculated or approximated if not provided as a font property.
FULL_NAME is the same as FACE_NAME. Its use is deprecated, but it is found on some old fonts.
COPYRIGHT is a human-understandable string (of type ATOM) that gives the copyright information of the legal owner of the digital font data.
This information is a required component of a font but is independent of the particular format used to represent it (that is, it cannot be captured as a comment that could later be "thrown away" for efficiency reasons).
COPYRIGHT cannot be calculated or approximated if not provided as a font property.
NOTICE is a human-understandable string (of type ATOM) that gives the copyright information of the legal owner of the font design or, if not applicable, the trademark information for the typeface FAMILY_NAME.
Typeface design and trademark protection laws vary from country to country, the USA having no design copyright protection currently while various countries in Europe offer both design and typeface family name trademark protection. As with COPYRIGHT, this information is a required component of a font but is independent of the particular format used to represent it.
NOTICE cannot be calculated or approximated if not provided as a font property.
DESTINATION is an unsigned integer code (of type CARD32) that gives the font design destination, that is, whether it was designed as a screen proofing font to match printer font glyph widths (WYSIWYG), as an optimal video font (possibly with corresponding printer font) for extended screen viewing (video text), and so on.
The font design considerations are very different, and at current display resolutions, the readability and legibility of these two kinds of screen fonts are very different. DESTINATION allows publishing clients that use X to model the printed page and video text clients, such as on-line documentation browsers, to query for X screen fonts that suit their particular requirements.
The encoding is as follows:
| Code | EnglishTranslation | Description |
|---|---|---|
| 0 | WYSIWYG | The font is optimized to match the typographic design and metrics of an equivalent printer font |
| 1 | Video text | The font is optimized for screen legibility and readability |
FONT_TYPE is a human-understandable string (of type ATOM) which describes the format of the font data as it is read from permanent storage by the current font source. It is a static attribute of the source data. It can be used by clients to select a type of bitmap or outline font without regard to the rasterizer used to render the font.
Predefined values are
| Value | When applicable |
| "Bitmap" | Hand-tuned bitmap fonts. Some attempt has been made to optimized the visual appearance of the font for the requested size and resolution. |
| "Prebuilt" | All bitmap format fonts which cannot be described as "Bitmap", that is hand-tuned. For example, a bitmap format font which was generated mechanically using a scalable font rasterizer would be considered "Prebuilt", not "Bitmap". |
| "Type 1" | Any Type 1 font |
| "TrueType" | Any TrueType font. |
| "Speedo" | Any Speedo font. |
| "F3" | Any F3 font. |
Other values may be registered with the X Consortium.
FONT_VERSION is a human-understandable string (of type ATOM) which describes the formal or informal version of the font. None is a valid value.
RASTERIZER_NAME is a human-understandable string (of type ATOM) which is the specific name of the rasterizer that has performed some rasterization operation (such as scaling from outlines) on the font.
To define a RASTERIZER_NAME, the following format is recommended:
| RasterizerName :: | = | Organizationld Space Rasterizer |
| Organizationld :: | = | STRING8-the X Registry ORGANIZATION name of the rasterizer implementor or maintainer. |
| Rasterizer :: | = | the case-sensitive, human-understandable product name of the rasterizer. Words within this name should be separated by a single SPACE. |
| Space :: | = | OCTET-" " (SPACE) |
If RASTERIZER_NAME is not defined, or is None, no rasterization operation has been applied to the FONT_TYPE.
RASTERIZER_VERSION is a human-understandable string (of type ATOM) which represents the formal or informal version of a font rasterizer. The RASTERIZER_VERSION should match the corresponding product version number known to users, when applicable.
For a font with a transformation matrix, RAW_ASCENT is the font ascent in 1000 pixel metrics. See Section 5.3.1.
For a font with a transformation matrix, RAW_DESCENT is the font descent in 1000 pixel metrics. See Section 5.3.1.
For a font with a transformation matrix, all font properties that represent horizontal or vertical sizes or displacements will be accompanied by a new property, named as the original except prefixed with "RAW_", that is computed as described in Section 5.3.1, Metrics and Font Properties.
AXIS_NAMES is a list of all the names of the axes for a polymorphic font, separated by a null (0) byte. These names are suitable for presentation in a user interface. See Section 5.5, Polymorphic Fonts.
AXIS_LIMITS is a list of integers, two for each axis, giving the minimum and maximum allowable values for that axis of polymorphic font. See Section 5.5, Polymorphic Fonts.
AXIS_TYPES is like AXIS_NAMES, but can be registered as having specific semantics. See Section 5.5, Polymorphic Fonts.
The following font property atom definitions were predefined in the initial version of the core protocol:
| Font Property/Atom Name | Property type |
| MIN_SPACE | INT32 |
| NORM_SPACE | INT32 |
| MAX_SPACE | INT32 |
| END_SPACE | INT32 |
| SUPERSCRIPT_X | INT32 |
| SUPERSCRITP_Y | INT32 |
| SUBSCRIPT_X | INT32 |
| SUBSCRIPT_Y | INT32 |
| UNDERLINE_POSITION | INT32 |
| UNDERLINE_THICKNESS | INT32 |
| STRIKEOUT_ASCENT | INT32 |
| STRIKEOUT_DESCENT | INT32 |
| FONT_ASCENT | INT32 |
| FONT_DESCENT | INT32 |
| ITALIC_ANGLE | INT32 |
| X_HEIGHT | INT32 |
| QUAD_WIDTH | INT32 - deprecated |
| WEIGHT | CARD32 |
| POINT_SIZE | INT32 |
| RESOLUTION | CARD32 - deprecated |
| COPYRIGHT | ATOM |
| FULL_NAME | ATOM - deprecated |
| FAMILY_NAME | ATOM |
| DEFAULT_CHAR | CARD32 |
An XLFD name presented to the server can have the POINT_SIZE or PIXEL_SIZE field begin with the character'[". If the first character of the field is "[", the character must be followed with ASCII representations of four floating point numbers and a trailing"]", with white space separating the numbers and optional whit space separating the number from the "[" and "]" characters. Numbers use standard floating point syntax but use the character "~" to represent a minus sign in the mantissa or exponent.
The BNF for a matrix transformation string is
| MatrixString :: | = | LeftBracket OptionalSpace Float Space Flat Space |
| Float Space Float OptionalSpace RightBracket | ||
| OptionalSpace :: | = | "" | Space |
| Space :: | = | SpaceChar | SpaceChar Space |
| Float :: | = | Mantissa | Mantissa Exponent |
| Mantissa :: | = | Sign Number | Number |
| Sign :: | = | Plus | Tilde |
| Number :: | = | Integer | Integer Dot Integer | Dot Integer |
| Integer :: | = | Digit | Digit Integer |
| Digit : : | = | "0"|"1"|"2"|"3"|"4"|"5"|&quo;6"|"7"|"8"|"9" |
| Exponent :: | = | "e" SignedInteger | "E" SignedInteger |
| SignedInteger :: | = | Sign Integer | Integer |
| LeftBracket :: | = | OCTET-"[" (LEFT SQUARE BRACKET) |
| RightBracket :: | = | OCTET-"]" (RIGHT SQUARE BRACKET) |
| SpaceChar :: | = | OCTET-" " (SPACE) |
| Tide :: | = | OCTET-"" (TILDE) |
| Plus:: | = | OCTET-"+" (PLUS) |
| Dot :: | = | OCTET-"." (FULL STOP) |
The string "[a b c d]" represents a graphical transformation of the glyphs in the font by the matrix
[ a b 0 ] [ c d 0 ] [ 0 0 1 ]
All transformations occur around the origin of the glyph. The relationship between the current scalar values and the matrix transformation values is that the scalar value "N" in the POINT_SIZE filed produces the same glyphs as the matrix "[N/10 0 0 N/10]" in that filed, and the scalar value "N" in the PIXEL_SIZE field produces the same glyphs as the matrix "[N*RESOLUTION_X/RESOLUTION_Y 0 0 N]" in that field.
If matrices are specified for both the POINT_SIZE and PIXEL_SIZE, they must bear the following relationship to each other within an implementation-specific tolerance:
If either the POINT_SIZE or PIXEL_SIZE field is unspecified (either "0" or
wildcarded) the preceding formulas can be used to compute one from the other.
In this section, the phrase " 1000 pixel metrics" means the metrics that would be obtained if the rasterizer took the base untransformed design used to generate the transformed font and scaled it linearly to a height of 1000 pixels, with no rotation component. Note that there may be no way for the application to actually request this font since the rasterizer may use different outlines or rasterization techniques at that size than the ones used to generate the transformed font.
Notes on properties and metrics:
The per-char ink metrics (lbearing, rbearing, ascent, and descent) represent the ink extent of the transformed glyph around its origin.
The per-char width is the x component of the transformed character width.
The font ascent and descent are the y component of the transformed font ascent or descent.
The FONT property returns a name reflecting the matrix being used-that is, the name returned can be used to open another instance of the same font. The returned name is not necessarily an exact coy of the requested name. If, for example, the user requests
the resulting FONT property might be
The FONT property will always include matrices in both the PIXEL_SIZE and the POINT_SIZE fields.
To allow accurate client positioning of transformed characters, the attributes field of the XChar-Info contains the width of the character in 100 pixel metrics. This attributes field should be interpreted as a signed integer.
There will always be 2 new font properties defined, RAW_ASCENT and RAW_DESCENT, that hold the ascent and descent in 1000 pixel metrics.
All font properties that represent horizontal widths or displacements have as their value the x component of the transformed width or displacement All font properties that represent vertical heights or displacement have as their value the y component of the transformed height or displacement. Each such property will be accompanied by a new property, named as the original except prefixed with "RAW_", that gives the value of the width, height, or displacement in 1000 pixel metrics.
The XLFD is designed to support scalable fonts. A scalable font is a font source from which instances of arbitrary size can be derived. A scalable font source might be one or more outlines together with zero or more hand-tuned bitmap fonts at specific sizes and resolutions, or it might be a programmatic description together with zero or more bitmap fonts, or some other format (perhaps even just a single bitmap font).
The following definitions are useful for discussing scalable fonts:
The global list consists of font names derived from font sources. If a single font source can support multiple character sets (specified in the CHARSET_REGISTRY and CHARSET_ENCODING fields), each such character set should be used to form a separate font name in the list. For a nonscalable font source, the simple font name for each character set is included in the global list. For a scalable font source, a scalable font name for each character set is included in the list. In addition to the scalable font name, specific derived instance names may also be included in the list. The relative order of derived instances with respect to the scalable font name is not constrained. Finally, font name aliases may also be included in the list. The relative order of aliases with respect to the "real" font name is not constrained.
The values of the RESOLUTION_X and RESOLUTION_Y fields of a scalable font name are implementation dependent, but to maximize backward compatibility they should be reasonable non-zero values, for example, a resolution close to that provided by the screen (in a single-screen server). Since some existing applications rely on seeing a collection of point and pixel sizes, server vendors are strongly encouraged in the near term to provide a mechanism for including, for each scalable font name, a set of specific derived instance names. For font sources that contain a collection of hand-tuned bitmap fonts, including names of these instances in the global list is recommended and sufficient.
The X protocol request OpenFont on a scalable font name returns a font corresponding to an implementation-dependent derived instance of that font name.
The X protocol request ListFonts on a well-formed XLFD pattern returns the following. Start with the global list. If the actual pattern argument has values containing no wildcards in scalable fields, then substitute each such field into the corresponding field in each scalable font name in the list. For each resulting font name, if the remaining scalable fields cannot be replaced with values to produce a derived instance, remove the font name from the list. Now take the modified list, and perform a simple pattern match against the pattern argument. ListFonts returns the resulting list.
ListFonts on a pattern containing wildcards that is not a well-formed XLFD pattern is only required to return the list obtained by performing a simple pattern match against the global list. X servers are permitted, but not required, to use a more sophisticated matching algorithm.
Fonts that can be varied in ways other than size or resolution are called polymorphic fonts. Multiple Master Type l font programs are one type of a polymorphic font. Current examples of axes along which the fonts can be varied are width, weight, an optical size; others might include formality or x-height.
To support polymorphic fonts, special values indication governability are defined for the following XLFD fields:
The string "0" is the special polymorphic value. In the WEIGHT._NAME, SLANT, or SETWIDTH_NAME field, "0" must be the entire field. There may be multiple polymorphic values in the ADD_STYLE_NAME field. They are surrounded by "[" and "]" and separated by a Space, as "[0 0]". The polymorphic values may coexist with other data in the field. It is recommended that the polymorphic values be at the end of the ADD_STYLE_NAME field.
The font matching algorithms for a font with polymorphic fields are identical to the matching algorithms for a font with scalable fields.
There are three new font properties to describe the axes of variation, AXIS_NAMES, AXIS_LIMITS, and AXIS_TYPES. AXIS_NAMES is a list of all the names of the axes for the font, separated by null (0) byte. These names are suitable for presentation in a user interface. AXIS_LIMITS is a list of integers, two for each axis, giving the minimum and maximum allowable values for that axis.; AXIS_TYPES is like AXIS_NAMES, but can be registered as having specific semantics.
The axes are listed in the properties in the same order as they appear in the font name. They are matched with font name fields by looking for the special polymorphic values in the font name.
Examples:
The Adobe Myriad MM font program has width and weight axes. Weight can vary form 215 to
830, and width form 300 to 700.
The Adobe Minion MM Italic font program has width, weight, and optical size axes.
The Adobe Minion MM Swash Italic font program has the same axes and values. This shows how "[0]" in the ADD_STYLE_NAME field can coexist with other words.
The XYZ Abc font, a hypothetical font, has optical size and x-height axes. This shows how there can be more than one polymorphic value in the ADD_STYLE_NAME field.
If an axis allows negative values, a client requests a negative value by using "~" (TILDE) as a minus sign. Axis types can be registered with the X Consortium, along with their semantics. If a font name that contains the polymorphic value or a wildcard in a polymorphic field is presented to a font source, the font source is free to substitute any value that is convenient. However, font sources should t.~ to use a value that would be considered normal or medium for the particular font. For example, if an optical size variable is unresolved, the font source should provide a value appropriate to the size of the font. The result of specifying an out-of-range value for a polymorphic field is undefined. The font source may treat this as a BadName error, treat the value as if it were the closest legal value, or extrapolate to try to accommodate the value.
The following X protocol requests must support the XLFD conventions:
In addition, the following Xlib functions must support the XLFD conventions:
The bitmap font distribution and interchange format adopted by the X Consortium (BDF V2.1) provides a general mechanism for identifying the font name of an X font and a variable list of font properties, but it does not mandate the syntax or semantics of the font name or the semantics of the font properties that might be provided in a BDF font. This section identifies the requirements for BDF fonts that conform to XLFD.
A BDF font conforms to the XLFD specification if and only if the following conditions are satisfied:
The value for the BDF item FONT conforms to the syntax and semantic definition of a XLFD FontName string.
The FontName begins with the X FontNameRegistry prefix: "-".
All XLFD FontName fields are defined.
Any FontProperties provided conform in name and semantics to the XLFD FontProperty definitions.
A simple method of testing for conformance would entail verifying that the FontNameRegistry prefix is the string "-", that the number of field delimiters in the string and coded field values are valid, and that each font property name either matches a standard XLFD property name or follows the definition of a private property.
FONT_ASCENT, FONT_DESCENT, and DEFAULT_CHAR are provided in the BDF specification as properties that are moved to the XFontStruct by the BDF font compiler in generating the X server-specific binary font encoding. If present, these properties shall comply with the following semantic definitions.
FONT_ASCENT is an integer value (of type INT32) that gives the recommended typographic ascent above the baseline for determining interline spacing. Specific glyphs of the font may extend beyond this. If the current position point for line n is at [X,Y], then the origin of the next line n+l (allowing for a possible font change) is [X, Y + !FONT_DESCENT Tn +!FONT_ ASCENT Tm].
if (FONT_ASCENT undefined) then FONT_ASCENT= maximum ascent
where maximum ascent is the maximum ascent (above the baseline) in pixels of any glyph in the font.
FONT_DESCENT is an integer value (of type INT32) that gives the recommended typographic descent below the baseline for determining interline spacing. Specific glyphs of the font may extend beyond this. If the current position point for line n is at [X,Y], then the origin of the next line n+l (allowing for a possible font change) is [X, Y + !FONT_DESCENT Tn + !FONT_ ASCENT Tm].
The logical extent of the font is inclusive between the Y-coordinate values: Y - FONT_ASCENT and Y + FONT_DESCENT + 1.
if (FONT_DESCENT undefined) then FONT_DESCENT= maximum descent
where maximum descent is the maximum descent (below the baseline) in pixels of any glyph in the font.
The DEFAULT_CHAR is a unsigned integer value (of type CARD32) that specifies the index of the default character to be used by the X server when an attempt is made to display an undefined or nonexistent character in the font. (For a font using two byte matrix format, the index bytes are encoded in the integer as byte1 * 65536 + byte2.) If the DEFAULT_CHAR itself specifies an
undefined or nonexistent character in the font, then no display is performed. DEFAULT_CHAR cannot be approximated if not provided as a font property.
| Home |
|---|
| Contents | Previous Chapter | Next Chapter |