X Window System Protocol

Every request contains an 8-bit major opcode and a 16-bit length field expressed in units of four bytes. Every request consists of four bytes of a header (containing the major opcode, the length field, and a data byte) followed by zero or more additional bytes of data. The length field defines the total length of the request, including the header. The length field in a request must equal the minimum length required to contain the request. If the specified length is smaller or larger than the required length, an error is generated. Unused bytes in a request are not required to be zero. Major opcodes 128 through 255 are reserved for extensions. Extensions are intended to contain multiple requests, so extension requests typically have an additional minor opcode encoded in the second data byte in the request header. However, the placement and interpretation of this minor opcode and of all other fields in extension requests are not defined by the core protocol. Every request on a given connection is implicitly assigned a sequence number, starting with one, that is used in replies, errors, and events.

Every reply contains a 32-bit length field expressed in units of four bytes. Every reply consists of 32 bytes followed by zero or more additional bytes of data, as specified in the length field. Unused bytes within a reply are not guaranteed to be zero. Every reply also contains the least significant 16 bits of the sequence number of the corresponding request.

Error reports are 32 bytes long. Every error includes an 8-bit error code. Error codes 128 through 255 are reserved for extensions. Every error also includes the major and minor opcodes of the failed request and the least significant 16 bits of the sequence number of the request. For the following errors (see section 4), the failing resource ID is also returned: Colormap, Cursor, Drawable, Font, GContext, IDChoice, Pixmap and Window. For Atom errors, the failing atom is returned. For Value errors, the failing value is returned. Other core errors return no additional data. Unused bytes within an error are not guaranteed to be zero.

Events are 32 bytes long. Unused bytes within an event are not guaranteed to be zero. Every event contains an 8-bit type code. The most significant bit in this code is set if the event was generated from a SendEvent request. Event codes 64 through 127 are reserved for extensions, although the core protocol does not define a mechanism for selecting interest in such events. Every core event (with the exception of KeymapNotify) also contains the least significant 16 bits of the sequence number of the last request issued by the client that was (or is currently being) processed by the server.

The client must send an initial byte of data to identify the byte order to be employed. The value of the byte must be octal 102 or 154. The value 102 (ASCII uppercase B) means values are transmitted most significant byte first, and value 154 (ASCII lowercase l) means values are transmitted least significant byte first. Except where explicitly noted in the protocol, all 16-bit and 32-bit quantities sent by the client must be transmitted with this byte order, and all 16-bit and 32-bit quantities returned by the server will be transmitted with this byte order.

Following the byte-order byte, the client sends the following information at connection setup:

The version numbers indicate what version of the protocol the client expects the server to implement.

The authorization name indicates what authorization (and authentication) protocol the client expects the server to use, and the data is specific to that protocol. Specification of valid authorization mechanisms is not part of the core X protocol. A server that does not implement the protocol the client expects or that only implements the host-based mechanism may simply ignore this information. If both name and data strings are empty, this is to be interpreted as "no explicit authorization."

The client receives the following information at connection setup:

The client receives the following additional data if the returned success value is Failed, and the connection is not successfully established:

The client receives the following additional data if the returned success value is Authenticate, and further authentication negotiation is required:

The contents of the reason string are specific to the authorization protocol in use. The semantics of this authentication negotiation are not constrained, except that the negotiation must eventually terminate with a reply from the server containing a success value of Failed or Success.

The client receives the following additional data if the returned success value is Success, and the connection is successfully established:

The information that is global to the server is:

The protocol version numbers are an escape hatch in case future revisions of the protocol are necessary. In general, the major version would increment for incompatible changes, and the minor version would increment for small upward compatible changes. Barring changes, the major version will be 11, and the minor version will be 0. The protocol version numbers returned indicate the protocol the server actually supports. This might not equal the version sent by the client. The server can (but need not) refuse connections from clients that offer a different version than the server supports. A server can (but need not) support more than one version simultaneously.

The vendor string gives some identification of the owner of the server implementation. The vendor controls the semantics of the release number.

The resource-id-mask contains a single contiguous set of bits (at least 18). The client allocates resource IDs for types WINDOW, PIXMAP, CURSOR, FONT, GCONTEXT, and COLORMAP by choosing a value with only some subset of these bits set and ORing it with resource-id-base. Only values constructed in this way can be used to name newly created resources over this connection. Resource IDs never have the top three bits set. The client is not restricted to linear or contiguous allocation of resource IDs. Once an ID has been freed, it can be reused. An ID must be unique with respect to the IDs of all other resources, not just other resources of the same type. However, note that the value spaces of resource identifiers, atoms, visualids, and keysyms are distinguished by context, and as such, are not required to be disjoint; for example, a given numeric value might be both a valid window ID, a valid atom, and a valid keysym.

Although the server is in general responsible for byte-swapping data to match the client, images are always transmitted and received in formats (including byte order) specified by the server. The byte order for images is given by image-byte-order and applies to each scanline unit in XY format (bitmap format) and to each pixel value in Z format.

A bitmap is represented in scanline order. Each scanline is padded to a multiple of bits as given by bitmap-scanline-pad. The pad bits are of arbitrary value. The scanline is quantized in multiples of bits as given by bitmap-scanline-unit. The bitmap-scanline-unit is always less than or equal to the bitmap-scanline-pad. Within each unit, the leftmost bit in the bitmap is either the least significant or most significant bit in the unit, as given by bitmap-bit-order. If a pixmap is represented in XY format, each plane is represented as a bitmap, and the planes appear from most significant to least significant in bit order with no padding between planes.

Pixmap-formats contains one entry for each depth value. The entry describes the Z format used to represent images of that depth. An entry for a depth is included if any screen supports that depth, and all screens supporting that depth must support only that Z format for that depth. In Z format, the pixels are in scanline order, left to right within a scanline. The number of bits used to hold each pixel is given by bits-per-pixel. Bits-per-pixel may be larger than strictly required by the depth, in which case the least significant bits are used to hold the pixmap data, and the values of the unused high-order bits are undefined. When the bits-per-pixel is 4, the order of nibbles in the byte is the same as the image byte-order. When the bits-per-pixel is 1, the format is identical for bitmap format. Each scanline is padded to a multiple of bits as given by scanline-pad. When bits-per-pixel is 1, this will be identical to bitmap-scanline-pad.

How a pointing device roams the screens is up to the server implementation and is transparent to the protocol. No geometry is defined among screens.

The server may retain the recent history of pointer motion and do so to a finer granularity than is reported by MotionNotify events. The GetMotionEvents request makes such history available. The motion-buffer-size gives the approximate maximum number of elements in the history buffer.

Maximum-request-length specifies the maximum length of a request accepted by the server, in 4-byte units. That is, length is the maximum value that can appear in the length field of a request. Requests larger than this maximum generate a Length error, and the server will read and simply discard the entire request. Maximum-request-length will always be at least 4096 (that is, requests of length up to and including 16384 bytes will be accepted by all servers).

Min-keycode and max-keycode specify the smallest and largest keycode values transmitted by the server. Min-keycode is never less than 8, and max-keycode is never greater than 255. Not all keycodes in this range are required to have corresponding keys.

The information that applies per screen is:

The allowed-depths specifies what pixmap and window depths are supported. Pixmaps are supported for each depth listed, and windows of that depth are supported if at least one visual type is listed for the depth. A pixmap depth of one is always supported and listed, but windows of depth one might not be supported. A depth of zero is never listed, but zero-depth InputOnly windows are always supported.

Root-depth and root-visual specify the depth and visual type of the root window. Width-in-pixels and height-in-pixels specify the size of the root window (which cannot be changed). The class of the root window is always InputOutput. Width-in-millimeters and height-in-millimeters can be used to determine the physical size and the aspect ratio.

The default-colormap is the one initially associated with the root window. Clients with minimal color requirements creating windows of the same depth as the root may want to allocate from this map by default.

Black-pixel and white-pixel can be used in implementing a monochrome application. These pixel values are for permanently allocated entries in the default-colormap. The actual RGB values may be settable on some screens and, in any case, may not actually be black and white. The names are intended to convey the expected relative intensity of the colors.

The border of the root window is initially a pixmap filled with the black-pixel. The initial background of the root window is a pixmap filled with some unspecified two-color pattern using black-pixel and white-pixel.

Min-installed-maps specifies the number of maps that can be guaranteed to be installed simultaneously (with InstallColormap), regardless of the number of entries allocated in each map. Max-installed-maps specifies the maximum number of maps that might possibly be installed simultaneously, depending on their allocations. Multiple static-visual colormaps with identical contents but differing in resource ID should be considered as a single map for the purposes of this number. For the typical case of a single hardware colormap, both values will be 1.

Backing-stores indicates when the server supports backing stores for this screen, although it may be storage limited in the number of windows it can support at once. If save-unders is True, the server can support the save-under mode in CreateWindow and ChangeWindowAttributes, although again it may be storage limited.

The current-input-events is what GetWindowAttributes would return for the all-event-masks for the root window.

The information that applies per visual-type is:

A given visual type might be listed for more than one depth or for more than one screen.

For PseudoColor, a pixel value indexes a colormap to produce independent RGB values; the RGB values can be changed dynamically. GrayScale is treated in the same way as PseudoColor except which primary drives the screen is undefined; thus, the client should always store the same value for red, green, and blue in colormaps. For DirectColor, a pixel value is decomposed into separate RGB subfields, and each subfield separately indexes the colormap for the corresponding value. The RGB values can be changed dynamically. TrueColor is treated in the same way as DirectColor except the colormap has predefined read-only RGB values. These values are server-dependent but provide linear or near-linear increasing ramps in each primary. StaticColor is treated in the same way as PseudoColor except the colormap has predefined read-only RGB values, which are server-dependent. StaticGray is treated in the same way as StaticColor except the red, green, and blue values are equal for any single pixel value, resulting in shades of gray. StaticGray with a two-entry colormap can be thought of as monochrome.

The red-mask, green-mask, and blue-mask are only defined for DirectColor and TrueColor. Each has one contiguous set of bits set to 1 with no intersections. Usually each mask has the same number of bits set to 1.

The bits-per-rgb-value specifies the log base 2 of the number of distinct color intensity values (individually) of red, green, and blue. This number need not bear any relation to the number of colormap entries. Actual RGB values are always passed in the protocol within a 16-bit spectrum, with 0 being minimum intensity and 65535 being the maximum intensity. On hardware that provides a linear zero-based intensity ramp, the following relationship exists:

       hw-intensity = protocol-intensity / (65536 / total-hw-intensities)

Colormap entries are indexed from 0. The colormap-entries defines the number of available colormap entries in a newly created colormap. For DirectColor and TrueColor, this will usually be 2 to the power of the maximum number of bits set to 1 in red-mask, green-mask, and blue-mask.

This request creates an unmapped window and assigns the identifier wid to it.

A class of CopyFromParent means the class is taken from the parent. A depth of zero for class InputOutput or CopyFromParent means the depth is taken from the parent. A visual of CopyFromParent means the visual type is taken from the parent. For class InputOutput, the visual type and depth must be a combination supported for the screen (or a Match error results). The depth need not be the same as the parent, but the parent must not be of class InputOnly (or a Match error results). For class InputOnly, the depth must be zero (or a Match error results), and the visual must be one supported for the screen (or a Match error results). However, the parent can have any depth and class.

The server essentially acts as if InputOnly windows do not exist for the purposes of graphics requests, exposure processing, and VisibilityNotify events. An InputOnly window cannot be used as a drawable (as a source or destination for graphics requests). InputOnly and InputOutput windows act identically in other respects-properties, grabs, input control, and so on.

The coordinate system has the X axis horizontal and the Y axis vertical with the origin [0, 0] at the upper-left corner. Coordinates are integral, in terms of pixels, and coincide with pixel centers. Each window and pixmap has its own coordinate system. For a window, the origin is inside the border at the inside, upper-left corner.

The x and y coordinates for the window are relative to the parent's origin and specify the position of the upper-left outer corner of the window (not the origin). The width and height specify the inside size (not including the border) and must be nonzero (or a Value error results). The border-width for an InputOnly window must be zero (or a Match error results).

The window is placed on top in the stacking order with respect to siblings.

The value-mask and value-list specify attributes of the window that are to be explicitly initialized. The possible values are:

The default values when attributes are not explicitly initialized are:

Only the following attributes are defined for InputOnly windows:

It is a Match error to specify any other attributes for InputOnly windows.

If background-pixmap is given, it overrides the default background-pixmap. The background pixmap and the window must have the same root and the same depth (or a Match error results). Any size pixmap can be used, although some sizes may be faster than others. If background None is specified, the window has no defined background. If background ParentRelative is specified, the parent's background is used, but the window must have the same depth as the parent (or a Match error results). If the parent has background None, then the window will also have background None. A copy of the parent's background is not made. The parent's background is reexamined each time the window background is required. If background-pixel is given, it overrides the default background-pixmap and any background-pixmap given explicitly, and a pixmap of undefined size filled with background-pixel is used for the background. Range checking is not performed on the background-pixel value; it is simply truncated to the appropriate number of bits. For a ParentRelative background, the background tile origin always aligns with the parent's background tile origin. Otherwise, the background tile origin is always the window origin.

When no valid contents are available for regions of a window and the regions are either visible or the server is maintaining backing store, the server automatically tiles the regions with the window's background unless the window has a background of None. If the background is None, the previous screen contents from other windows of the same depth as the window are simply left in place if the contents come from the parent of the window or an inferior of the parent; otherwise, the initial contents of the exposed regions are undefined. Exposure events are then generated for the regions, even if the background is None.

The border tile origin is always the same as the background tile origin. If border-pixmap is given, it overrides the default border-pixmap. The border pixmap and the window must have the same root and the same depth (or a Match error results). Any size pixmap can be used, although some sizes may be faster than others. If CopyFromParent is given, the parent's border pixmap is copied (subsequent changes to the parent's border attribute do not affect the child), but the window must have the same depth as the parent (or a Match error results). The pixmap might be copied by sharing the same pixmap object between the child and parent or by making a complete copy of the pixmap contents. If border-pixel is given, it overrides the default border-pixmap and any border-pixmap given explicitly, and a pixmap of undefined size filled with border-pixel is used for the border. Range checking is not performed on the border-pixel value; it is simply truncated to the appropriate number of bits.

Output to a window is always clipped to the inside of the window, so that the border is never affected.

The bit-gravity defines which region of the window should be retained if the window is resized, and win-gravity defines how the window should be repositioned if the parent is resized (see ConfigureWindow request).

A backing-store of WhenMapped advises the server that maintaining contents of obscured regions when the window is mapped would be beneficial. A backing-store of Always advises the server that maintaining contents even when the window is unmapped would be beneficial. In this case, the server may generate an exposure event when the window is created. A value of NotUseful advises the server that maintaining contents is unnecessary, although a server may still choose to maintain contents while the window is mapped. Note that if the server maintains contents, then the server should maintain complete contents not just the region within the parent boundaries, even if the window is larger than its parent. While the server maintains contents, exposure events will not normally be generated, but the server may stop maintaining contents at any time.

If save-under is True, the server is advised that when this window is mapped, saving the contents of windows it obscures would be beneficial.

When the contents of obscured regions of a window are being maintained, regions obscured by noninferior windows are included in the destination (and source, when the window is the source) of graphics requests, but regions obscured by inferior windows are not included.

The backing-planes indicates (with bits set to 1) which bit planes of the window hold dynamic data that must be preserved in backing-stores and during save-unders. The backing-pixel specifies what value to use in planes not covered by backing-planes. The server is free to save only the specified bit planes in the backing-store or save-under and regenerate the remaining planes with the specified pixel value. Any bits beyond the specified depth of the window in these values are simply ignored.

The event-mask defines which events the client is interested in for this window (or for some event types, inferiors of the window). The do-not-propagate-mask defines which events should not be propagated to ancestor windows when no client has the event type selected in this window.

The override-redirect specifies whether map and configure requests on this window should override a SubstructureRedirect on the parent, typically to inform a window manager not to tamper with the window.

The colormap specifies the colormap that best reflects the true colors of the window. Servers capable of supporting multiple hardware colormaps may use this information, and window managers may use it for InstallColormap requests. The colormap must have the same visual type and root as the window (or a Match error results). If CopyFromParent is specified, the parent's colormap is copied (subsequent changes to the parent's colormap attribute do not affect the child). However, the window must have the same visual type as the parent (or a Match error results), and the parent must not have a colormap of None (or a Match error results). For an explanation of None, see FreeColormap request. The colormap is copied by sharing the colormap object between the child and the parent, not by making a complete copy of the colormap contents.

If a cursor is specified, it will be used whenever the pointer is in the window. If None is specified, the parent's cursor will be used when the pointer is in the window, and any change in the parent's cursor will cause an immediate change in the displayed cursor.

This request generates a CreateNotify event.

The background and border pixmaps and the cursor may be freed immediately if no further explicit references to them are to be made.

Subsequent drawing into the background or border pixmap has an undefined effect on the window state. The server might or might not make a copy of the pixmap.

The value-mask and value-list specify which attributes are to be changed. The values and restrictions are the same as for CreateWindow.

Setting a new background, whether by background-pixmap or background-pixel, overrides any previous background. Setting a new border, whether by border-pixel or border-pixmap, overrides any previous border.

Changing the background does not cause the window contents to be changed. Setting the border or changing the background such that the border tile origin changes causes the border to be repainted. Changing the background of a root window to None or ParentRelative restores the default background pixmap. Changing the border of a root window to CopyFromParent restores the default border pixmap.

Changing the win-gravity does not affect the current position of the window.

Changing the backing-store of an obscured window to WhenMapped or Always or changing the backing-planes, backing-pixel, or save-under of a mapped window may have no immediate effect.

Multiple clients can select input on the same window; their event-masks are disjoint. When an event is generated, it will be reported to all interested clients. However, only one client at a time can select for SubstructureRedirect, only one client at a time can select for ResizeRedirect, and only one client at a time can select for ButtonPress. An attempt to violate these restrictions results in an Access error.

There is only one do-not-propagate-mask for a window, not one per client.

Changing the colormap of a window (by defining a new map, not by changing the contents of the existing map) generates a ColormapNotify event. Changing the colormap of a visible window might have no immediate effect on the screen (see InstallColormap request).

Changing the cursor of a root window to None restores the default cursor.

The order in which attributes are verified and altered is server-dependent. If an error is generated, a subset of the attributes may have been altered.

This request returns the current attributes of the window. A window is Unviewable if it is mapped but some ancestor is unmapped. All-event-masks is the inclusive-OR of all event masks selected on the window by clients. Your-event-mask is the event mask selected by the querying client.

If the argument window is mapped, an UnmapWindow request is performed automatically. The window and all inferiors are then destroyed, and a DestroyNotify event is generated for each window. The ordering of the DestroyNotify events is such that for any given window, DestroyNotify is generated on all inferiors of the window before being generated on the window itself. The ordering among siblings and across subhierarchies is not otherwise constrained.

Normal exposure processing on formerly obscured windows is performed.

If the window is a root window, this request has no effect.

This request performs a DestroyWindow request on all children of the window, in bottom-to-top stacking order.

This request adds or removes the specified window from the client's save-set. The window must have been created by some other client (or a Match error results). For further information about the use of the save-set, see section 10.

When windows are destroyed, the server automatically removes them from the save-set.

If the window is mapped, an UnmapWindow request is performed automatically first. The window is then removed from its current position in the hierarchy and is inserted as a child of the specified parent. The x and y coordinates are relative to the parent's origin and specify the new position of the upper-left outer corner of the window. The window is placed on top in the stacking order with respect to siblings. A ReparentNotify event is then generated. The override-redirect attribute of the window is passed on in this event; a value of True indicates that a window manager should not tamper with this window. Finally, if the window was originally mapped, a MapWindow request is performed automatically.

Normal exposure processing on formerly obscured windows is performed. The server might not generate exposure events for regions from the initial unmap that are immediately obscured by the final map.

A Match error is generated if: The new parent is not on the same screen as the old parent. The new parent is the window itself or an inferior of the window. The new parent is InputOnly, and the window is not. The window has a ParentRelative background, and the new parent is not the same depth as the window.

If the window is already mapped, this request has no effect.

If the override-redirect attribute of the window is False and some other client has selected SubstructureRedirect on the parent, then a MapRequest event is generated, but the window remains unmapped. Otherwise, the window is mapped, and a MapNotify event is generated.

If the window is now viewable and its contents have been discarded, the window is tiled with its background (if no background is defined, the existing screen contents are not altered), and zero or more exposure events are generated. If a backing-store has been maintained while the window was unmapped, no exposure events are generated. If a backing-store will now be maintained, a full-window exposure is always generated. Otherwise, only visible regions may be reported. Similar tiling and exposure take place for any newly viewable inferiors.

This request performs a MapWindow request on all unmapped children of the window, in top-to-bottom stacking order.

If the window is already unmapped, this request has no effect. Otherwise, the window is unmapped, and an UnmapNotify event is generated. Normal exposure processing on formerly obscured windows is performed.

This request performs an UnmapWindow request on all mapped children of the window, in bottom-to-top stacking order.

This request changes the configuration of the window. The value-mask and value-list specify which values are to be given. The possible values are:

The x and y coordinates are relative to the parent's origin and specify the position of the upper-left outer corner of the window. The width and height specify the inside size, not including the border, and must be nonzero (or a Value error results). Those values not specified are taken from the existing geometry of the window. Note that changing just the border-width leaves the outer-left corner of the window in a fixed position but moves the absolute position of the window's origin. It is a Match error to attempt to make the border-width of an InputOnly window nonzero.

If the override-redirect attribute of the window is False and some other client has selected SubstructureRedirect on the parent, a ConfigureRequest event is generated, and no further processing is performed. Otherwise, the following is performed:

If some other client has selected ResizeRedirect on the window and the inside width or height of the window is being changed, a ResizeRequest event is generated, and the current inside width and height are used instead. Note that the override-redirect attribute of the window has no effect on ResizeRedirect and that SubstructureRedirect on the parent has precedence over ResizeRedirect on the window.

The geometry of the window is changed as specified, the window is restacked among siblings, and a ConfigureNotify event is generated if the state of the window actually changes. If the inside width or height of the window has actually changed, then children of the window are affected, according to their win-gravity. Exposure processing is performed on formerly obscured windows (including the window itself and its inferiors if regions of them were obscured but now are not). Exposure processing is also performed on any new regions of the window (as a result of increasing the width or height) and on any regions where window contents are lost.

If the inside width or height of a window is not changed but the window is moved or its border is changed, then the contents of the window are not lost but move with the window. Changing the inside width or height of the window causes its contents to be moved or lost, depending on the bit-gravity of the window. It also causes children to be reconfigured, depending on their win-gravity. For a change of width and height of W and H, we define the [x, y] pairs as:

When a window with one of these bit-gravities is resized, the corresponding pair defines the change in position of each pixel in the window. When a window with one of these win-gravities has its parent window resized, the corresponding pair defines the change in position of the window within the parent. This repositioning generates a GravityNotify event. GravityNotify events are generated after the ConfigureNotify event is generated.

A gravity of Static indicates that the contents or origin should not move relative to the origin of the root window. If the change in size of the window is coupled with a change in position of [X, Y], then for bit-gravity the change in position of each pixel is [-X, -Y] and for win-gravity the change in position of a child when its parent is so resized is [-X, -Y]. Note that Static gravity still only takes effect when the width or height of the window is changed, not when the window is simply moved.

A bit-gravity of Forget indicates that the window contents are always discarded after a size change, even if backing-store or save-under has been requested. The window is tiled with its background (except, if no background is defined, the existing screen contents are not altered) and zero or more exposure events are generated.

The contents and borders of inferiors are not affected by their parent's bit-gravity. A server is permitted to ignore the specified bit-gravity and use Forget instead.

A win-gravity of Unmap is like NorthWest, but the child is also unmapped when the parent is resized, and an UnmapNotify event is generated. UnmapNotify events are generated after the ConfigureNotify event is generated.

If a sibling and a stack-mode are specified, the window is restacked as follows:

If a stack-mode is specified but no sibling is specified, the window is restacked as follows:

It is a Match error if a sibling is specified without a stack-mode or if the window is not actually a sibling.

Note that the computations for BottomIf, TopIf, and Opposite are performed with respect to the window's final geometry (as controlled by the other arguments to the request), not to its initial geometry.

Attempts to configure a root window have no effect.

If some other client has selected SubstructureRedirect on the window, then a CirculateRequest event is generated, and no further processing is performed. Otherwise, the following is performed, and then a CirculateNotify event is generated if the window is actually restacked.

For RaiseLowest, CirculateWindow raises the lowest mapped child (if any) that is occluded by another child to the top of the stack. For LowerHighest, CirculateWindow lowers the highest mapped child (if any) that occludes another child to the bottom of the stack. Exposure processing is performed on formerly obscured windows.

This request returns the root and current geometry of the drawable. The depth is the number of bits per pixel for the object. The x, y, and border-width will always be zero for pixmaps. For a window, the x and y coordinates specify the upper-left outer corner of the window relative to its parent's origin, and the width and height specify the inside size, not including the border.

It is legal to pass an InputOnly window as a drawable to this request.

This request returns the root, the parent, and the children of the window. The children are listed in bottom-to-top stacking order.

This request returns the atom for the given name. If only-if-exists is False, then the atom is created if it does not exist. The string should use the ISO Latin-1 encoding. Uppercase and lowercase matter.

The lifetime of an atom is not tied to the interning client. Atoms remain defined until server reset (see section 10).

This request returns the name for the given atom.

This request alters the property for the specified window. The type is uninterpreted by the server. The format specifies whether the data should be viewed as a list of 8-bit, 16-bit, or 32-bit quantities so that the server can correctly byte-swap as necessary.

If the mode is Replace, the previous property value is discarded. If the mode is Prepend or Append, then the type and format must match the existing property value (or a Match error results). If the property is undefined, it is treated as defined with the correct type and format with zero-length data. For Prepend, the data is tacked on to the beginning of the existing data, and for Append, it is tacked on to the end of the existing data.

This request generates a PropertyNotify event on the window.

The lifetime of a property is not tied to the storing client. Properties remain until explicitly deleted, until the window is destroyed, or until server reset (see section 10).

The maximum size of a property is server-dependent and may vary dynamically.

This request deletes the property from the specified window if the property exists and generates a PropertyNotify event on the window unless the property does not exist.

If the specified property does not exist for the specified window, then the return type is None, the format and bytes-after are zero, and the value is empty. The delete argument is ignored in this case. If the specified property exists but its type does not match the specified type, then the return type is the actual type of the property, the format is the actual format of the property (never zero), the bytes-after is the length of the property in bytes (even if the format is 16 or 32), and the value is empty. The delete argument is ignored in this case. If the specified property exists and either AnyPropertyType is specified or the specified type matches the actual type of the property, then the return type is the actual type of the property, the format is the actual format of the property (never zero), and the bytes-after and value are as follows, given:

	N = actual length of the stored property in bytes
	    (even if the format is 16 or 32)
	I = 4 * long-offset
	T = N - I
	L = MINIMUM(T, 4 * long-length)
	A = N - (I + L)

The returned value starts at byte index I in the property (indexing from 0), and its length in bytes is L. However, it is a Value error if long-offset is given such that L is negative. The value of bytes-after is A, giving the number of trailing unread bytes in the stored property. If delete is True and the bytes-after is zero, the property is also deleted from the window, and a PropertyNotify event is generated on the window.

If the property names in the list are viewed as being numbered starting from zero, and there are N property names in the list, then the value associated with property name I becomes the value associated with property name (I + delta) mod N, for all I from zero to N - 1. The effect is to rotate the states by delta places around the virtual ring of property names (right for positive delta, left for negative delta).

If delta mod N is nonzero, a PropertyNotify event is generated for each property in the order listed.

If an atom occurs more than once in the list or no property with that name is defined for the window, a Match error is generated. If an Atom or Match error is generated, no properties are changed.

This request returns the atoms of properties currently defined on the window.

This request changes the owner, owner window, and last-change time of the specified selection. This request has no effect if the specified time is earlier than the current last-change time of the specified selection or is later than the current server time. Otherwise, the last-change time is set to the specified time with CurrentTime replaced by the current server time. If the owner window is specified as None, then the owner of the selection becomes None (that is, no owner). Otherwise, the owner of the selection becomes the client executing the request. If the new owner (whether a client or None) is not the same as the current owner and the current owner is not None, then the current owner is sent a SelectionClear event.

If the client that is the owner of a selection is later terminated (that is, its connection is closed) or if the owner window it has specified in the request is later destroyed, then the owner of the selection automatically reverts to None, but the last-change time is not affected.

The selection atom is uninterpreted by the server. The owner window is returned by the GetSelectionOwner request and is reported in SelectionRequest and SelectionClear events.

Selections are global to the server.

This request returns the current owner window of the specified selection, if any. If None is returned, then there is no owner for the selection.

If the specified selection has an owner, the server sends a SelectionRequest event to that owner. If no owner for the specified selection exists, the server generates a SelectionNotify event to the requestor with property None. The arguments are passed on unchanged in either of the events.

If PointerWindow is specified, destination is replaced with the window that the pointer is in. If InputFocus is specified and the focus window contains the pointer, destination is replaced with the window that the pointer is in. Otherwise, destination is replaced with the focus window.

If the event-mask is the empty set, then the event is sent to the client that created the destination window. If that client no longer exists, no event is sent.

If propagate is False, then the event is sent to every client selecting on destination any of the event types in event-mask.

If propagate is True and no clients have selected on destination any of the event types in event-mask, then destination is replaced with the closest ancestor of destination for which some client has selected a type in event-mask and no intervening window has that type in its do-not-propagate-mask. If no such window exists or if the window is an ancestor of the focus window and InputFocus was originally specified as the destination, then the event is not sent to any clients. Otherwise, the event is reported to every client selecting on the final destination any of the types specified in event-mask.

The event code must be one of the core events or one of the events defined by an extension (or a Value error results) so that the server can correctly byte-swap the contents as necessary. The contents of the event are otherwise unaltered and unchecked by the server except to force on the most significant bit of the event code and to set the sequence number in the event correctly.

Active grabs are ignored for this request.

This request actively grabs control of the pointer. Further pointer events are only reported to the grabbing client. The request overrides any active pointer grab by this client.

If owner-events is False, all generated pointer events are reported with respect to grab-window and are only reported if selected by event-mask. If owner-events is True and a generated pointer event would normally be reported to this client, it is reported normally. Otherwise, the event is reported with respect to the grab-window and is only reported if selected by event-mask. For either value of owner-events, unreported events are simply discarded.

If pointer-mode is Asynchronous, pointer event processing continues normally. If the pointer is currently frozen by this client, then processing of pointer events is resumed. If pointer-mode is Synchronous, the state of the pointer (as seen by means of the protocol) appears to freeze, and no further pointer events are generated by the server until the grabbing client issues a releasing AllowEvents request or until the pointer grab is released. Actual pointer changes are not lost while the pointer is frozen. They are simply queued for later processing.

If keyboard-mode is Asynchronous, keyboard event processing is unaffected by activation of the grab. If keyboard-mode is Synchronous, the state of the keyboard (as seen by means of the protocol) appears to freeze, and no further keyboard events are generated by the server until the grabbing client issues a releasing AllowEvents request or until the pointer grab is released. Actual keyboard changes are not lost while the keyboard is frozen. They are simply queued for later processing.

If a cursor is specified, then it is displayed regardless of what window the pointer is in. If no cursor is specified, then when the pointer is in grab-window or one of its subwindows, the normal cursor for that window is displayed. Otherwise, the cursor for grab-window is displayed.

If a confine-to window is specified, then the pointer will be restricted to stay contained in that window. The confine-to window need have no relationship to the grab-window. If the pointer is not initially in the confine-to window, then it is warped automatically to the closest edge (and enter/leave events are generated normally) just before the grab activates. If the confine-to window is subsequently reconfigured, the pointer will be warped automatically as necessary to keep it contained in the window.

This request generates EnterNotify and LeaveNotify events.

The request fails with status AlreadyGrabbed if the pointer is actively grabbed by some other client. The request fails with status Frozen if the pointer is frozen by an active grab of another client. The request fails with status NotViewable if grab-window or confine-to window is not viewable or if the confine-to window lies completely outside the boundaries of the root window. The request fails with status InvalidTime if the specified time is earlier than the last-pointer-grab time or later than the current server time. Otherwise, the last-pointer-grab time is set to the specified time, with CurrentTime replaced by the current server time.

This request releases the pointer if this client has it actively grabbed (from either GrabPointer or GrabButton or from a normal button press) and releases any queued events. The request has no effect if the specified time is earlier than the last-pointer-grab time or is later than the current server time.

This request generates EnterNotify and LeaveNotify events.

An UngrabPointer request is performed automatically if the event window or confine-to window for an active pointer grab becomes not viewable or if window reconfiguration causes the confine-to window to lie completely outside the boundaries of the root window.

This request establishes a passive grab. In the future, the pointer is actively grabbed as described in GrabPointer, the last-pointer-grab time is set to the time at which the button was pressed (as transmitted in the ButtonPress event), and the ButtonPress event is reported if all of the following conditions are true: The pointer is not grabbed and the specified button is logically pressed when the specified modifier keys are logically down, and no other buttons or modifier keys are logically down. The grab-window contains the pointer. The confine-to window (if any) is viewable. A passive grab on the same button/key combination does not exist on any ancestor of grab-window.

The interpretation of the remaining arguments is the same as for GrabPointer. The active grab is terminated automatically when the logical state of the pointer has all buttons released, independent of the logical state of modifier keys. Note that the logical state of a device (as seen by means of the protocol) may lag the physical state if device event processing is frozen.

This request overrides all previous passive grabs by the same client on the same button/key combinations on the same window. A modifier of AnyModifier is equivalent to issuing the request for all possible modifier combinations (including the combination of no modifiers). It is not required that all specified modifiers have currently assigned keycodes. A button of AnyButton is equivalent to issuing the request for all possible buttons. Otherwise, it is not required that the button specified currently be assigned to a physical button.

An Access error is generated if some other client has already issued a GrabButton request with the same button/key combination on the same window. When using AnyModifier or AnyButton, the request fails completely (no grabs are established), and an Access error is generated if there is a conflicting grab for any combination. The request has no effect on an active grab.

This request releases the passive button/key combination on the specified window if it was grabbed by this client. A modifiers argument of AnyModifier is equivalent to issuing the request for all possible modifier combinations (including the combination of no modifiers). A button of AnyButton is equivalent to issuing the request for all possible buttons. The request has no effect on an active grab.

This request changes the specified dynamic parameters if the pointer is actively grabbed by the client and the specified time is no earlier than the last-pointer-grab time and no later than the current server time. The interpretation of event-mask and cursor are the same as in GrabPointer. This request has no effect on the parameters of any passive grabs established with GrabButton.

This request actively grabs control of the keyboard. Further key events are reported only to the grabbing client. This request overrides any active keyboard grab by this client.

If owner-events is False, all generated key events are reported with respect to grab-window. If owner-events is True and if a generated key event would normally be reported to this client, it is reported normally. Otherwise, the event is reported with respect to the grab-window. Both KeyPress and KeyRelease events are always reported, independent of any event selection made by the client.

If keyboard-mode is Asynchronous, keyboard event processing continues normally. If the keyboard is currently frozen by this client, then processing of keyboard events is resumed. If keyboard-mode is Synchronous, the state of the keyboard (as seen by means of the protocol) appears to freeze. No further keyboard events are generated by the server until the grabbing client issues a releasing AllowEvents request or until the keyboard grab is released. Actual keyboard changes are not lost while the keyboard is frozen. They are simply queued for later processing.

If pointer-mode is Asynchronous, pointer event processing is unaffected by activation of the grab. If pointer-mode is Synchronous, the state of the pointer (as seen by means of the protocol) appears to freeze. No further pointer events are generated by the server until the grabbing client issues a releasing AllowEvents request or until the keyboard grab is released. Actual pointer changes are not lost while the pointer is frozen. They are simply queued for later processing.

This request generates FocusIn and FocusOut events.

The request fails with status AlreadyGrabbed if the keyboard is actively grabbed by some other client. The request fails with status Frozen if the keyboard is frozen by an active grab of another client. The request fails with status NotViewable if grab-window is not viewable. The request fails with status InvalidTime if the specified time is earlier than the last-keyboard-grab time or later than the current server time. Otherwise, the last-keyboard-grab time is set to the specified time with CurrentTime replaced by the current server time.

This request releases the keyboard if this client has it actively grabbed (as a result of either GrabKeyboard or GrabKey) and releases any queued events. The request has no effect if the specified time is earlier than the last-keyboard-grab time or is later than the current server time.

This request generates FocusIn and FocusOut events.

An UngrabKeyboard is performed automatically if the event window for an active keyboard grab becomes not viewable.

This request establishes a passive grab on the keyboard. In the future, the keyboard is actively grabbed as described in GrabKeyboard, the last-keyboard-grab time is set to the time at which the key was pressed (as transmitted in the KeyPress event), and the KeyPress event is reported if all of the following conditions are true: The keyboard is not grabbed and the specified key (which can itself be a modifier key) is logically pressed when the specified modifier keys are logically down, and no other modifier keys are logically down. Either the grab-window is an ancestor of (or is) the focus window, or the grab-window is a descendent of the focus window and contains the pointer. A passive grab on the same key combination does not exist on any ancestor of grab-window.

The interpretation of the remaining arguments is the same as for GrabKeyboard. The active grab is terminated automatically when the logical state of the keyboard has the specified key released, independent of the logical state of modifier keys. Note that the logical state of a device (as seen by means of the protocol) may lag the physical state if device event processing is frozen.

This request overrides all previous passive grabs by the same client on the same key combinations on the same window. A modifier of AnyModifier is equivalent to issuing the request for all possible modifier combinations (including the combination of no modifiers). It is not required that all modifiers specified have currently assigned keycodes. A key of AnyKey is equivalent to issuing the request for all possible keycodes. Otherwise, the key must be in the range specified by min-keycode and max-keycode in the connection setup (or a Value error results).

An Access error is generated if some other client has issued a GrabKey with the same key combination on the same window. When using AnyModifier or AnyKey, the request fails completely (no grabs are established), and an Access error is generated if there is a conflicting grab for any combination.

This request releases the key combination on the specified window if it was grabbed by this client. A modifiers argument of AnyModifier is equivalent to issuing the request for all possible modifier combinations (including the combination of no modifiers). A key of AnyKey is equivalent to issuing the request for all possible keycodes. This request has no effect on an active grab.

This request releases some queued events if the client has caused a device to freeze. The request has no effect if the specified time is earlier than the last-grab time of the most recent active grab for the client or if the specified time is later than the current server time.

For AsyncPointer, if the pointer is frozen by the client, pointer event processing continues normally. If the pointer is frozen twice by the client on behalf of two separate grabs, AsyncPointer thaws for both. AsyncPointer has no effect if the pointer is not frozen by the client, but the pointer need not be grabbed by the client.

For SyncPointer, if the pointer is frozen and actively grabbed by the client, pointer event processing continues normally until the next ButtonPress or ButtonRelease event is reported to the client, at which time the pointer again appears to freeze. However, if the reported event causes the pointer grab to be released, then the pointer does not freeze. SyncPointer has no effect if the pointer is not frozen by the client or if the pointer is not grabbed by the client.

For ReplayPointer, if the pointer is actively grabbed by the client and is frozen as the result of an event having been sent to the client (either from the activation of a GrabButton or from a previous AllowEvents with mode SyncPointer but not from a GrabPointer), then the pointer grab is released and that event is completely reprocessed, this time ignoring any passive grabs at or above (towards the root) the grab-window of the grab just released. The request has no effect if the pointer is not grabbed by the client or if the pointer is not frozen as the result of an event.

For AsyncKeyboard, if the keyboard is frozen by the client, keyboard event processing continues normally. If the keyboard is frozen twice by the client on behalf of two separate grabs, AsyncKeyboard thaws for both. AsyncKeyboard has no effect if the keyboard is not frozen by the client, but the keyboard need not be grabbed by the client.

For SyncKeyboard, if the keyboard is frozen and actively grabbed by the client, keyboard event processing continues normally until the next KeyPress or KeyRelease event is reported to the client, at which time the keyboard again appears to freeze. However, if the reported event causes the keyboard grab to be released, then the keyboard does not freeze. SyncKeyboard has no effect if the keyboard is not frozen by the client or if the keyboard is not grabbed by the client.

For ReplayKeyboard, if the keyboard is actively grabbed by the client and is frozen as the result of an event having been sent to the client (either from the activation of a GrabKey or from a previous AllowEvents with mode SyncKeyboard but not from a GrabKeyboard), then the keyboard grab is released and that event is completely reprocessed, this time ignoring any passive grabs at or above (towards the root) the grab-window of the grab just released. The request has no effect if the keyboard is not grabbed by the client or if the keyboard is not frozen as the result of an event.

For SyncBoth, if both pointer and keyboard are frozen by the client, event processing (for both devices) continues normally until the next ButtonPress, ButtonRelease, KeyPress, or KeyRelease event is reported to the client for a grabbed device (button event for the pointer, key event for the keyboard), at which time the devices again appear to freeze. However, if the reported event causes the grab to be released, then the devices do not freeze (but if the other device is still grabbed, then a subsequent event for it will still cause both devices to freeze). SyncBoth has no effect unless both pointer and keyboard are frozen by the client. If the pointer or keyboard is frozen twice by the client on behalf of two separate grabs, SyncBoth thaws for both (but a subsequent freeze for SyncBoth will only freeze each device once).

For AsyncBoth, if the pointer and the keyboard are frozen by the client, event processing for both devices continues normally. If a device is frozen twice by the client on behalf of two separate grabs, AsyncBoth thaws for both. AsyncBoth has no effect unless both pointer and keyboard are frozen by the client.

AsyncPointer, SyncPointer, and ReplayPointer have no effect on processing of keyboard events. AsyncKeyboard, SyncKeyboard, and ReplayKeyboard have no effect on processing of pointer events.

It is possible for both a pointer grab and a keyboard grab to be active simultaneously (by the same or different clients). When a device is frozen on behalf of either grab, no event processing is performed for the device. It is possible for a single device to be frozen because of both grabs. In this case, the freeze must be released on behalf of both grabs before events can again be processed. If a device is frozen twice by a single client, then a single AllowEvents releases both.

This request disables processing of requests and close-downs on all connections other than the one this request arrived on.

This request restarts processing of requests and close-downs on other connections.

The root window the pointer is logically on and the pointer coordinates relative to the root's origin are returned. If same-screen is False, then the pointer is not on the same screen as the argument window, child is None, and win-x and win-y are zero. If same-screen is True, then win-x and win-y are the pointer coordinates relative to the argument window's origin, and child is the child containing the pointer, if any. The current logical state of the modifier keys and the buttons are also returned. Note that the logical state of a device (as seen by means of the protocol) may lag the physical state if device event processing is frozen.

This request returns all events in the motion history buffer that fall between the specified start and stop times (inclusive) and that have coordinates that lie within (including borders) the specified window at its present placement. The x and y coordinates are reported relative to the origin of the window.

If the start time is later than the stop time or if the start time is in the future, no events are returned. If the stop time is in the future, it is equivalent to specifying CurrentTime.

The src-x and src-y coordinates are taken relative to src-window's origin and are returned as dst-x and dst-y coordinates relative to dst-window's origin. If same-screen is False, then src-window and dst-window are on different screens, and dst-x and dst-y are zero. If the coordinates are contained in a mapped child of dst-window, then that child is returned.

If dst-window is None, this request moves the pointer by offsets [dst-x, dst-y] relative to the current position of the pointer. If dst-window is a window, this request moves the pointer to [dst-x, dst-y] relative to dst-window's origin. However, if src-window is not None, the move only takes place if src-window contains the pointer and the pointer is contained in the specified rectangle of src-window.

The src-x and src-y coordinates are relative to src-window's origin. If src-height is zero, it is replaced with the current height of src-window minus src-y. If src-width is zero, it is replaced with the current width of src-window minus src-x.

This request cannot be used to move the pointer outside the confine-to window of an active pointer grab. An attempt will only move the pointer as far as the closest edge of the confine-to window.

This request will generate events just as if the user had instantaneously moved the pointer.

This request changes the input focus and the last-focus-change time. The request has no effect if the specified time is earlier than the current last-focus-change time or is later than the current server time. Otherwise, the last-focus-change time is set to the specified time with CurrentTime replaced by the current server time.

If None is specified as the focus, all keyboard events are discarded until a new focus window is set. In this case, the revert-to argument is ignored.

If a window is specified as the focus, it becomes the keyboard's focus window. If a generated keyboard event would normally be reported to this window or one of its inferiors, the event is reported normally. Otherwise, the event is reported with respect to the focus window.

If PointerRoot is specified as the focus, the focus window is dynamically taken to be the root window of whatever screen the pointer is on at each keyboard event. In this case, the revert-to argument is ignored.

This request generates FocusIn and FocusOut events.

The specified focus window must be viewable at the time of the request (or a Match error results). If the focus window later becomes not viewable, the new focus window depends on the revert-to argument. If revert-to is Parent, the focus reverts to the parent (or the closest viewable ancestor) and the new revert-to value is taken to be None. If revert-to is PointerRoot or None, the focus reverts to that value. When the focus reverts, FocusIn and FocusOut events are generated, but the last-focus-change time is not affected.

This request returns the current focus state.

This request returns a bit vector for the logical state of the keyboard. Each bit set to 1 indicates that the corresponding key is currently pressed. The vector is represented as 32 bytes. Byte N (from 0) contains the bits for keys 8N to 8N + 7 with the least significant bit in the byte representing key 8N. Note that the logical state of a device (as seen by means of the protocol) may lag the physical state if device event processing is frozen.

This request loads the specified font, if necessary, and associates identifier fid with it. The font name should use the ISO Latin-1 encoding, and uppercase and lowercase do not matter. When the characters “?” and “*” are used in a font name, a pattern match is performed and any matching font is used. In the pattern, the “?” character (octal value 77) will match any single character, and the “*” character (octal value 52) will match any number of characters. A structured format for font names is specified in the X.Org standard X Logical Font Description Conventions.

Fonts are not associated with a particular screen and can be stored as a component of any graphics context.

This request deletes the association between the resource ID and the font. The font itself will be freed when no other resource references it.

This request returns logical information about a font. If a gcontext is given for font, the currently contained font is used.

The draw-direction is just a hint and indicates whether most char-infos have a positive, LeftToRight, or a negative, RightToLeft, character-width metric. The core protocol defines no support for vertical text.

If min-byte1 and max-byte1 are both zero, then min-char-or-byte2 specifies the linear character index corresponding to the first element of char-infos, and max-char-or-byte2 specifies the linear character index of the last element. If either min-byte1 or max-byte1 are nonzero, then both min-char-or-byte2 and max-char-or-byte2 will be less than 256, and the 2-byte character index values corresponding to char-infos element N (counting from 0) are:

	byte1 = N/D + min-byte1
	byte2 = N\\D + min-char-or-byte2

where:

	D = max-char-or-byte2 - min-char-or-byte2 + 1
	/ = integer division
	\\ = integer modulus

If char-infos has length zero, then min-bounds and max-bounds will be identical, and the effective char-infos is one filled with this char-info, of length:

	L = D * (max-byte1 - min-byte1 + 1)

That is, all glyphs in the specified linear or matrix range have the same information, as given by min-bounds (and max-bounds). If all-chars-exist is True, then all characters in char-infos have nonzero bounding boxes.

The default-char specifies the character that will be used when an undefined or nonexistent character is used. Note that default-char is a CARD16, not CHAR2B. For a font using 2-byte matrix format, the default-char has byte1 in the most significant byte and byte2 in the least significant byte. If the default-char itself specifies an undefined or nonexistent character, then no printing is performed for an undefined or nonexistent character.

The min-bounds and max-bounds contain the minimum and maximum values of each individual CHARINFO component over all char-infos (ignoring nonexistent characters). The bounding box of the font (that is, the smallest rectangle enclosing the shape obtained by superimposing all characters at the same origin [x,y]) has its upper-left coordinate at:

	[x + min-bounds.left-side-bearing, y - max-bounds.ascent]

with a width of:

	max-bounds.right-side-bearing - min-bounds.left-side-bearing

and a height of:

	max-bounds.ascent + max-bounds.descent

The font-ascent is the logical extent of the font above the baseline and is used for determining line spacing. Specific characters may extend beyond this. The font-descent is the logical extent of the font at or below the baseline and is used for determining line spacing. Specific characters may extend beyond this. If the baseline is at Y-coordinate y, then the logical extent of the font is inclusive between the Y-coordinate values (y - font-ascent) and (y + font-descent - 1).

A font is not guaranteed to have any properties. The interpretation of the property value (for example, INT32, CARD32) must be derived from a priori knowledge of the property. A basic set of font properties is specified in the X.Org standard X Logical Font Description Conventions.

For a character origin at [x,y], the bounding box of a character (that is, the smallest rectangle enclosing the character's shape), described in terms of CHARINFO components, is a rectangle with its upper-left corner at:

	[x + left-side-bearing, y - ascent]

with a width of:

	right-side-bearing - left-side-bearing

and a height of:

	ascent + descent

and the origin for the next character is defined to be:

	[x + character-width, y]

Note that the baseline is logically viewed as being just below nondescending characters (when descent is zero, only pixels with Y-coordinates less than y are drawn) and that the origin is logically viewed as being coincident with the left edge of a nonkerned character (when left-side-bearing is zero, no pixels with X-coordinate less than x are drawn).

Note that CHARINFO metric values can be negative.

A nonexistent character is represented with all CHARINFO components zero.

The interpretation of the per-character attributes field is server-dependent.

This request returns the logical extents of the specified string of characters in the specified font. If a gcontext is given for font, the currently contained font is used. The draw-direction, font-ascent, and font-descent are the same as described in QueryFont. The overall-ascent is the maximum of the ascent metrics of all characters in the string, and the overall-descent is the maximum of the descent metrics. The overall-width is the sum of the character-width metrics of all characters in the string. For each character in the string, let W be the sum of the character-width metrics of all characters preceding it in the string, let L be the left-side-bearing metric of the character plus W, and let R be the right-side-bearing metric of the character plus W. The overall-left is the minimum L of all characters in the string, and the overall-right is the maximum R.

For fonts defined with linear indexing rather than 2-byte matrix indexing, the server will interpret each CHAR2B as a 16-bit number that has been transmitted most significant byte first (that is, byte1 of the CHAR2B is taken as the most significant byte).

Characters with all zero metrics are ignored. If the font has no defined default-char, then undefined characters in the string are also ignored.

This request returns a list of available font names (as controlled by the font search path; see SetFontPath request) that match the pattern. At most, max-names names will be returned. The pattern should use the ISO Latin-1 encoding, and uppercase and lowercase do not matter. In the pattern, the “?” character (octal value 77) will match any single character, and the “*” character (octal value 52) will match any number of characters. The returned names are in lowercase.

This request is similar to ListFonts, but it also returns information about each font. The information returned for each font is identical to what QueryFont would return except that the per-character metrics are not returned. Note that this request can generate multiple replies. With each reply, replies-hint may provide an indication of how many more fonts will be returned. This number is a hint only and may be larger or smaller than the number of fonts actually returned. A zero value does not guarantee that no more fonts will be returned. After the font replies, a reply with a zero-length name is sent to indicate the end of the reply sequence.

This request defines the search path for font lookup. There is only one search path per server, not one per client. The interpretation of the strings is operating-system-dependent, but the strings are intended to specify directories to be searched in the order listed.

Setting the path to the empty list restores the default path defined for the server.

As a side effect of executing this request, the server is guaranteed to flush all cached information about fonts for which there currently are no explicit resource IDs allocated.

The meaning of an error from this request is system specific.

This request returns the current search path for fonts.

This request creates a pixmap and assigns the identifier pid to it. The width and height must be nonzero (or a Value error results). The depth must be one of the depths supported by the root of the specified drawable (or a Value error results). The initial contents of the pixmap are undefined.

It is legal to pass an InputOnly window as a drawable to this request.

This request deletes the association between the resource ID and the pixmap. The pixmap storage will be freed when no other resource references it.

This request creates a graphics context and assigns the identifier cid to it. The gcontext can be used with any destination drawable having the same root and depth as the specified drawable; use with other drawables results in a Match error.

The value-mask and value-list specify which components are to be explicitly initialized. The context components are:

In graphics operations, given a source and destination pixel, the result is computed bitwise on corresponding bits of the pixels; that is, a Boolean operation is performed in each bit plane. The plane-mask restricts the operation to a subset of planes, so the result is:

	((src FUNC dst) AND plane-mask) OR (dst AND (NOT plane-mask))

Range checking is not performed on the values for foreground, background, or plane-mask. They are simply truncated to the appropriate number of bits.

The meanings of the functions are:

The line-width is measured in pixels and can be greater than or equal to one, a wide line, or the special value zero, a thin line.

Wide lines are drawn centered on the path described by the graphics request. Unless otherwise specified by the join or cap style, the bounding box of a wide line with endpoints [x1, y1], [x2, y2] and width w is a rectangle with vertices at the following real coordinates:

	[x1-(w*sn/2), y1+(w*cs/2)], [x1+(w*sn/2), y1-(w*cs/2)],
	[x2-(w*sn/2), y2+(w*cs/2)], [x2+(w*sn/2), y2-(w*cs/2)]

The sn is the sine of the angle of the line and cs is the cosine of the angle of the line. A pixel is part of the line (and hence drawn) if the center of the pixel is fully inside the bounding box, which is viewed as having infinitely thin edges. If the center of the pixel is exactly on the bounding box, it is part of the line if and only if the interior is immediately to its right (x increasing direction). Pixels with centers on a horizontal edge are a special case and are part of the line if and only if the interior or the boundary is immediately below (y increasing direction) and if the interior or the boundary is immediately to the right (x increasing direction). Note that this description is a mathematical model describing the pixels that are drawn for a wide line and does not imply that trigonometry is required to implement such a model. Real or fixed point arithmetic is recommended for computing the corners of the line endpoints for lines greater than one pixel in width.

Thin lines (zero line-width) are nominally one pixel wide lines drawn using an unspecified, device-dependent algorithm. There are only two constraints on this algorithm. First, if a line is drawn unclipped from [x1,y1] to [x2,y2] and another line is drawn unclipped from [x1+dx,y1+dy] to [x2+dx,y2+dy], then a point [x,y] is touched by drawing the first line if and only if the point [x+dx,y+dy] is touched by drawing the second line. Second, the effective set of points comprising a line cannot be affected by clipping. Thus, a point is touched in a clipped line if and only if the point lies inside the clipping region and the point would be touched by the line when drawn unclipped.

Note that a wide line drawn from [x1,y1] to [x2,y2] always draws the same pixels as a wide line drawn from [x2,y2] to [x1,y1], not counting cap-style and join-style. Implementors are encouraged to make this property true for thin lines, but it is not required. A line-width of zero may differ from a line-width of one in which pixels are drawn. In general, drawing a thin line will be faster than drawing a wide line of width one, but thin lines may not mix well aesthetically with wide lines because of the different drawing algorithms. If it is desirable to obtain precise and uniform results across all displays, a client should always use a line-width of one, rather than a line-width of zero.

The line-style defines which sections of a line are drawn:

The cap-style defines how the endpoints of a path are drawn:

The join-style defines how corners are drawn for wide lines:

For a line with coincident endpoints (x1=x2, y1=y2), when the cap-style is applied to both endpoints, the semantics depends on the line-width and the cap-style:

For a line with coincident endpoints (x1=x2, y1=y2), when the join-style is applied at one or both endpoints, the effect is as if the line was removed from the overall path. However, if the total path consists of (or is reduced to) a single point joined with itself, the effect is the same as when the cap-style is applied at both endpoints.

The tile/stipple represents an infinite two-dimensional plane with the tile/stipple replicated in all dimensions. When that plane is superimposed on the drawable for use in a graphics operation, the upper-left corner of some instance of the tile/stipple is at the coordinates within the drawable specified by the tile/stipple origin. The tile/stipple and clip origins are interpreted relative to the origin of whatever destination drawable is specified in a graphics request.

The tile pixmap must have the same root and depth as the gcontext (or a Match error results). The stipple pixmap must have depth one and must have the same root as the gcontext (or a Match error results). For fill-style Stippled (but not fill-style OpaqueStippled), the stipple pattern is tiled in a single plane and acts as an additional clip mask to be ANDed with the clip-mask. Any size pixmap can be used for tiling or stippling, although some sizes may be faster to use than others.

The fill-style defines the contents of the source for line, text, and fill requests. For all text and fill requests (for example, PolyText8, PolyText16, PolyFillRectangle, FillPoly, and PolyFillArc) as well as for line requests with line-style Solid, (for example, PolyLine, PolySegment, PolyRectangle, PolyArc ) and for the even dashes for line requests with line-style OnOffDash or DoubleDash:

For the odd dashes for line requests with line-style DoubleDash:

The dashes value allowed here is actually a simplified form of the more general patterns that can be set with SetDashes. Specifying a value of N here is equivalent to specifying the two element list [N, N] in SetDashes. The value must be nonzero (or a Value error results). The meaning of dash-offset and dashes are explained in the SetDashes request.

The clip-mask restricts writes to the destination drawable. Only pixels where the clip-mask has bits set to 1 are drawn. Pixels are not drawn outside the area covered by the clip-mask or where the clip-mask has bits set to 0. The clip-mask affects all graphics requests, but it does not clip sources. The clip-mask origin is interpreted relative to the origin of whatever destination drawable is specified in a graphics request. If a pixmap is specified as the clip-mask, it must have depth 1 and have the same root as the gcontext (or a Match error results). If clip-mask is None, then pixels are always drawn, regardless of the clip origin. The clip-mask can also be set with the SetClipRectangles request.

For ClipByChildren, both source and destination windows are additionally clipped by all viewable InputOutput children. For IncludeInferiors, neither source nor destination window is clipped by inferiors. This will result in including subwindow contents in the source and drawing through subwindow boundaries of the destination. The use of IncludeInferiors with a source or destination window of one depth with mapped inferiors of differing depth is not illegal, but the semantics is undefined by the core protocol.

The fill-rule defines what pixels are inside (that is, are drawn) for paths given in FillPoly requests. EvenOdd means a point is inside if an infinite ray with the point as origin crosses the path an odd number of times. For Winding, a point is inside if an infinite ray with the point as origin crosses an unequal number of clockwise and counterclockwise directed path segments. A clockwise directed path segment is one that crosses the ray from left to right as observed from the point. A counter-clockwise segment is one that crosses the ray from right to left as observed from the point. The case where a directed line segment is coincident with the ray is uninteresting because one can simply choose a different ray that is not coincident with a segment.

For both fill rules, a point is infinitely small and the path is an infinitely thin line. A pixel is inside if the center point of the pixel is inside and the center point is not on the boundary. If the center point is on the boundary, the pixel is inside if and only if the polygon interior is immediately to its right (x increasing direction). Pixels with centers along a horizontal edge are a special case and are inside if and only if the polygon interior is immediately below (y increasing direction).

The arc-mode controls filling in the PolyFillArc request.

The graphics-exposures flag controls GraphicsExposure event generation for CopyArea and CopyPlane requests (and any similar requests defined by extensions).

The default component values are:

Storing a pixmap in a gcontext might or might not result in a copy being made. If the pixmap is later used as the destination for a graphics request, the change might or might not be reflected in the gcontext. If the pixmap is used simultaneously in a graphics request as both a destination and as a tile or stipple, the results are not defined.

It is quite likely that some amount of gcontext information will be cached in display hardware and that such hardware can only cache a small number of gcontexts. Given the number and complexity of components, clients should view switching between gcontexts with nearly identical state as significantly more expensive than making minor changes to a single gcontext.

This request changes components in gc. The value-mask and value-list specify which components are to be changed. The values and restrictions are the same as for CreateGC.

Changing the clip-mask also overrides any previous SetClipRectangles request on the context. Changing dash-offset or dashes overrides any previous SetDashes request on the context.

The order in which components are verified and altered is server-dependent. If an error is generated, a subset of the components may have been altered.

This request copies components from src-gc to dst-gc. The value-mask specifies which components to copy, as for CreateGC. The two gcontexts must have the same root and the same depth (or a Match error results).

This request sets dash-offset and dashes in gc for dashed line styles. Dashes cannot be empty (or a Value error results). Specifying an odd-length list is equivalent to specifying the same list concatenated with itself to produce an even-length list. The initial and alternating elements of dashes are the even dashes; the others are the odd dashes. Each element specifies a dash length in pixels. All of the elements must be nonzero (or a Value error results). The dash-offset defines the phase of the pattern, specifying how many pixels into dashes the pattern should actually begin in any single graphics request. Dashing is continuous through path elements combined with a join-style but is reset to the dash-offset between each sequence of joined lines.

The unit of measure for dashes is the same as in the ordinary coordinate system. Ideally, a dash length is measured along the slope of the line, but implementations are only required to match this ideal for horizontal and vertical lines. Failing the ideal semantics, it is suggested that the length be measured along the major axis of the line. The major axis is defined as the x axis for lines drawn at an angle of between -45 and +45 degrees or between 135 and 225 degrees from the x axis. For all other lines, the major axis is the y axis.

For any graphics primitive, the computation of the endpoint of an individual dash only depends on the geometry of the primitive, the start position of the dash, the direction of the dash, and the dash length.

For any graphics primitive, the total set of pixels used to render the primitive (both even and odd numbered dash elements) with DoubleDash line-style is the same as the set of pixels used to render the primitive with Solid line-style.

For any graphics primitive, if the primitive is drawn with OnOffDash or DoubleDash line-style unclipped at position [x,y] and again at position [x+dx,y+dy], then a point [x1,y1] is included in a dash in the first instance if and only if the point [x1+dx,y1+dy] is included in the dash in the second instance. In addition, the effective set of points comprising a dash cannot be affected by clipping. A point is included in a clipped dash if and only if the point lies inside the clipping region and the point would be included in the dash when drawn unclipped.

This request changes clip-mask in gc to the specified list of rectangles and sets the clip origin. Output will be clipped to remain contained within the rectangles. The clip origin is interpreted relative to the origin of whatever destination drawable is specified in a graphics request. The rectangle coordinates are interpreted relative to the clip origin. The rectangles should be nonintersecting, or graphics results will be undefined. Note that the list of rectangles can be empty, which effectively disables output. This is the opposite of passing None as the clip-mask in CreateGC and ChangeGC.

If known by the client, ordering relations on the rectangles can be specified with the ordering argument. This may provide faster operation by the server. If an incorrect ordering is specified, the server may generate a Match error, but it is not required to do so. If no error is generated, the graphics results are undefined. UnSorted means that the rectangles are in arbitrary order. YSorted means that the rectangles are nondecreasing in their Y origin. YXSorted additionally constrains YSorted order in that all rectangles with an equal Y origin are nondecreasing in their X origin. YXBanded additionally constrains YXSorted by requiring that, for every possible Y scanline, all rectangles that include that scanline have identical Y origins and Y extents.

This request deletes the association between the resource ID and the gcontext and destroys the gcontext.

The x and y coordinates are relative to the window's origin and specify the upper-left corner of the rectangle. If width is zero, it is replaced with the current width of the window minus x. If height is zero, it is replaced with the current height of the window minus y. If the window has a defined background tile, the rectangle is tiled with a plane-mask of all ones and function of Copy and a subwindow-mode of ClipByChildren. If the window has background None, the contents of the window are not changed. In either case, if exposures is True, then one or more exposure events are generated for regions of the rectangle that are either visible or are being retained in a backing store.

It is a Match error to use an InputOnly window in this request.

This request combines the specified rectangle of src-drawable with the specified rectangle of dst-drawable. The src-x and src-y coordinates are relative to src-drawable's origin. The dst-x and dst-y are relative to dst-drawable's origin, each pair specifying the upper-left corner of the rectangle. The src-drawable must have the same root and the same depth as dst-drawable (or a Match error results).

If regions of the source rectangle are obscured and have not been retained in backing store or if regions outside the boundaries of the source drawable are specified, then those regions are not copied, but the following occurs on all corresponding destination regions that are either visible or are retained in backing-store. If the dst-drawable is a window with a background other than None, these corresponding destination regions are tiled (with plane-mask of all ones and function Copy) with that background. Regardless of tiling and whether the destination is a window or a pixmap, if graphics-exposures in gc is True, then GraphicsExposure events for all corresponding destination regions are generated.

If graphics-exposures is True but no GraphicsExposure events are generated, then a NoExposure event is generated.

GC components: function, plane-mask, subwindow-mode, graphics-exposures, clip-x-origin, clip-y-origin, clip-mask

The src-drawable must have the same root as dst-drawable (or a Match error results), but it need not have the same depth. The bit-plane must have exactly one bit set to 1 and the value of bit-plane must be less than %2 sup n% where n is the depth of src-drawable (or a Value error results). Effectively, a pixmap of the same depth as dst-drawable and with size specified by the source region is formed using the foreground/background pixels in gc (foreground everywhere the bit-plane in src-drawable contains a bit set to 1, background everywhere the bit-plane contains a bit set to 0), and the equivalent of a CopyArea is performed, with all the same exposure semantics. This can also be thought of as using the specified region of the source bit-plane as a stipple with a fill-style of OpaqueStippled for filling a rectangular area of the destination.

GC components: function, plane-mask, foreground, background, subwindow-mode, graphics-exposures, clip-x-origin, clip-y-origin, clip-mask

This request combines the foreground pixel in gc with the pixel at each point in the drawable. The points are drawn in the order listed.

The first point is always relative to the drawable's origin. The rest are relative either to that origin or the previous point, depending on the coordinate-mode.

GC components: function, plane-mask, foreground, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

This request draws lines between each pair of points (point[i], point[i+1]). The lines are drawn in the order listed. The lines join correctly at all intermediate points, and if the first and last points coincide, the first and last lines also join correctly.

For any given line, no pixel is drawn more than once. If thin (zero line-width) lines intersect, the intersecting pixels are drawn multiple times. If wide lines intersect, the intersecting pixels are drawn only once, as though the entire PolyLine were a single filled shape.

The first point is always relative to the drawable's origin. The rest are relative either to that origin or the previous point, depending on the coordinate-mode.

When either of the two lines involved in a Bevel join is neither vertical nor horizontal, then the slope and position of the line segment defining the bevel join edge is implementation dependent. However, the computation of the slope and distance (relative to the join point) only depends on the line width and the slopes of the two lines.

GC components: function, plane-mask, line-width, line-style, cap-style, join-style, fill-style, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background, tile, stipple, tile-stipple-x-origin, tile-stipple-y-origin, dash-offset, dashes

For each segment, this request draws a line between [x1, y1] and [x2, y2]. The lines are drawn in the order listed. No joining is performed at coincident endpoints. For any given line, no pixel is drawn more than once. If lines intersect, the intersecting pixels are drawn multiple times.

GC components: function, plane-mask, line-width, line-style, cap-style, fill-style, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background, tile, stipple, tile-stipple-x-origin, tile-stipple-y-origin, dash-offset, dashes

This request draws the outlines of the specified rectangles, as if a five-point PolyLine were specified for each rectangle:

	[x,y] [x+width,y] [x+width,y+height] [x,y+height] [x,y]

The x and y coordinates of each rectangle are relative to the drawable's origin and define the upper-left corner of the rectangle.

The rectangles are drawn in the order listed. For any given rectangle, no pixel is drawn more than once. If rectangles intersect, the intersecting pixels are drawn multiple times.

GC components: function, plane-mask, line-width, line-style, cap-style, join-style, fill-style, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background, tile, stipple, tile-stipple-x-origin, tile-stipple-y-origin, dash-offset, dashes

This request draws circular or elliptical arcs. Each arc is specified by a rectangle and two angles. The angles are signed integers in degrees scaled by 64, with positive indicating counterclockwise motion and negative indicating clockwise motion. The start of the arc is specified by angle1 relative to the three-o'clock position from the center of the rectangle, and the path and extent of the arc is specified by angle2 relative to the start of the arc. If the magnitude of angle2 is greater than 360 degrees, it is truncated to 360 degrees. The x and y coordinates of the rectangle are relative to the origin of the drawable. For an arc specified as [x,y,w,h,a1,a2], the origin of the major and minor axes is at [x+(w/2),y+(h/2)], and the infinitely thin path describing the entire circle/ellipse intersects the horizontal axis at [x,y+(h/2)] and [x+w,y+(h/2)] and intersects the vertical axis at [x+(w/2),y] and [x+(w/2),y+h]. These coordinates are not necessarily integral; that is, they are not truncated to discrete coordinates.

For a wide line with line-width lw, the ideal bounding outlines for filling are given by the two infinitely thin paths consisting of all points whose perpendicular distance from a tangent to the path of the circle/ellipse is equal to lw/2 (which may be a fractional value). When the width and height of the arc are not equal and both are nonzero, then the actual bounding outlines are implementation dependent. However, the computation of the shape and position of the bounding outlines (relative to the center of the arc) only depends on the width and height of the arc and the line-width.

The cap-style is applied the same as for a line corresponding to the tangent of the circle/ellipse at the endpoint. When the angle of an arc face is not an integral multiple of 90 degrees, and the width and height of the arc are both are nonzero, then the shape and position of the cap at that face is implementation dependent. However, for a Butt cap, the face is defined by a straight line, and the computation of the position (relative to the center of the arc) and the slope of the line only depends on the width and height of the arc and the angle of the arc face. For other cap styles, the computation of the position (relative to the center of the arc) and the shape of the cap only depends on the width and height of the arc, the line-width, the angle of the arc face, and the direction (clockwise or counter clockwise) of the arc from the endpoint.

The join-style is applied the same as for two lines corresponding to the tangents of the circles/ellipses at the join point. When the width and height of both arcs are nonzero, and the angle of either arc face is not an integral multiple of 90 degrees, then the shape of the join is implementation dependent. However, the computation of the shape only depends on the width and height of each arc, the line-width, the angles of the two arc faces, the direction (clockwise or counter clockwise) of the arcs from the join point, and the relative orientation of the two arc center points.

For an arc specified as [x,y,w,h,a1,a2], the angles must be specified in the effectively skewed coordinate system of the ellipse (for a circle, the angles and coordinate systems are identical). The relationship between these angles and angles expressed in the normal coordinate system of the screen (as measured with a protractor) is as follows:

	skewed-angle = atan(tan(normal-angle) * w/h) + adjust

The skewed-angle and normal-angle are expressed in radians (rather than in degrees scaled by 64) in the range [0,2*PI). The atan returns a value in the range [-PI/2,PI/2]. The adjust is:

The arcs are drawn in the order listed. If the last point in one arc coincides with the first point in the following arc, the two arcs will join correctly. If the first point in the first arc coincides with the last point in the last arc, the two arcs will join correctly. For any given arc, no pixel is drawn more than once. If two arcs join correctly and the line-width is greater than zero and the arcs intersect, no pixel is drawn more than once. Otherwise, the intersecting pixels of intersecting arcs are drawn multiple times. Specifying an arc with one endpoint and a clockwise extent draws the same pixels as specifying the other endpoint and an equivalent counterclockwise extent, except as it affects joins.

By specifying one axis to be zero, a horizontal or vertical line can be drawn.

Angles are computed based solely on the coordinate system, ignoring the aspect ratio.

GC components: function, plane-mask, line-width, line-style, cap-style, join-style, fill-style, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background, tile, stipple, tile-stipple-x-origin, tile-stipple-y-origin, dash-offset, dashes

This request fills the region closed by the specified path. The path is closed automatically if the last point in the list does not coincide with the first point. No pixel of the region is drawn more than once.

The first point is always relative to the drawable's origin. The rest are relative either to that origin or the previous point, depending on the coordinate-mode.

The shape parameter may be used by the server to improve performance. Complex means the path may self-intersect. Contiguous coincident points in the path are not treated as self-intersection.

Nonconvex means the path does not self-intersect, but the shape is not wholly convex. If known by the client, specifying Nonconvex over Complex may improve performance. If Nonconvex is specified for a self-intersecting path, the graphics results are undefined.

Convex means that for every pair of points inside the polygon, the line segment connecting them does not intersect the path. If known by the client, specifying Convex can improve performance. If Convex is specified for a path that is not convex, the graphics results are undefined.

GC components: function, plane-mask, fill-style, fill-rule, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background, tile, stipple, tile-stipple-x-origin, tile-stipple-y-origin

This request fills the specified rectangles, as if a four-point FillPoly were specified for each rectangle:

	[x,y] [x+width,y] [x+width,y+height] [x,y+height]

The x and y coordinates of each rectangle are relative to the drawable's origin and define the upper-left corner of the rectangle.

The rectangles are drawn in the order listed. For any given rectangle, no pixel is drawn more than once. If rectangles intersect, the intersecting pixels are drawn multiple times.

GC components: function, plane-mask, fill-style, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background, tile, stipple, tile-stipple-x-origin, tile-stipple-y-origin

For each arc, this request fills the region closed by the infinitely thin path described by the specified arc and one or two line segments, depending on the arc-mode. For Chord, the single line segment joining the endpoints of the arc is used. For PieSlice, the two line segments joining the endpoints of the arc with the center point are used.

For an arc specified as [x,y,w,h,a1,a2], the origin of the major and minor axes is at [x+(w/2),y+(h/2)], and the infinitely thin path describing the entire circle/ellipse intersects the horizontal axis at [x,y+(h/2)] and [x+w,y+(h/2)] and intersects the vertical axis at [x+(w/2),y] and [x+(w/2),y+h]. These coordinates are not necessarily integral; that is, they are not truncated to discrete coordinates.

The arc angles are interpreted as specified in the PolyArc request. When the angle of an arc face is not an integral multiple of 90 degrees, then the precise endpoint on the arc is implementation dependent. However, for Chord arc-mode, the computation of the pair of endpoints (relative to the center of the arc) only depends on the width and height of the arc and the angles of the two arc faces. For PieSlice arc-mode, the computation of an endpoint only depends on the angle of the arc face for that endpoint and the ratio of the arc width to arc height.

The arcs are filled in the order listed. For any given arc, no pixel is drawn more than once. If regions intersect, the intersecting pixels are drawn multiple times.

GC components: function, plane-mask, fill-style, arc-mode, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background, tile, stipple, tile-stipple-x-origin, tile-stipple-y-origin

This request combines an image with a rectangle of the drawable. The dst-x and dst-y coordinates are relative to the drawable's origin.

If Bitmap format is used, then depth must be one (or a Match error results), and the image must be in XY format. The foreground pixel in gc defines the source for bits set to 1 in the image, and the background pixel defines the source for the bits set to 0.

For XYPixmap and ZPixmap, the depth must match the depth of the drawable (or a Match error results). For XYPixmap, the image must be sent in XY format. For ZPixmap, the image must be sent in the Z format defined for the given depth.

The left-pad must be zero for ZPixmap format (or a Match error results). For Bitmap and XYPixmap format, left-pad must be less than bitmap-scanline-pad as given in the server connection setup information (or a Match error results). The first left-pad bits in every scanline are to be ignored by the server. The actual image begins that many bits into the data. The width argument defines the width of the actual image and does not include left-pad.

GC components: function, plane-mask, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background

This request returns the contents of the given rectangle of the drawable in the given format. The x and y coordinates are relative to the drawable's origin and define the upper-left corner of the rectangle. If XYPixmap is specified, only the bit planes specified in plane-mask are transmitted, with the planes appearing from most significant to least significant in bit order. If ZPixmap is specified, then bits in all planes not specified in plane-mask are transmitted as zero. Range checking is not performed on plane-mask; extraneous bits are simply ignored. The returned depth is as specified when the drawable was created and is the same as a depth component in a FORMAT structure (in the connection setup), not a bits-per-pixel component. If the drawable is a window, its visual type is returned. If the drawable is a pixmap, the visual is None.

If the drawable is a pixmap, then the given rectangle must be wholly contained within the pixmap (or a Match error results). If the drawable is a window, the window must be viewable, and it must be the case that, if there were no inferiors or overlapping windows, the specified rectangle of the window would be fully visible on the screen and wholly contained within the outside edges of the window (or a Match error results). Note that the borders of the window can be included and read with this request. If the window has a backing store, then the backing-store contents are returned for regions of the window that are obscured by noninferior windows; otherwise, the returned contents of such obscured regions are undefined. Also undefined are the returned contents of visible regions of inferiors of different depth than the specified window. The pointer cursor image is not included in the contents returned.

This request is not general-purpose in the same sense as other graphics-related requests. It is intended specifically for rudimentary hardcopy support.

The x and y coordinates are relative to the drawable's origin and specify the baseline starting position (the initial character origin). Each text item is processed in turn. A font item causes the font to be stored in gc and to be used for subsequent text. Switching among fonts does not affect the next character origin. A text element delta specifies an additional change in the position along the x axis before the string is drawn; the delta is always added to the character origin. Each character image, as defined by the font in gc, is treated as an additional mask for a fill operation on the drawable.

All contained FONTs are always transmitted most significant byte first.

If a Font error is generated for an item, the previous items may have been drawn.

For fonts defined with 2-byte matrix indexing, each STRING8 byte is interpreted as a byte2 value of a CHAR2B with a byte1 value of zero.

GC components: function, plane-mask, fill-style, font, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

GC mode-dependent components: foreground, background, tile, stipple, tile-stipple-x-origin, tile-stipple-y-origin

This request is similar to PolyText8, except 2-byte (or 16-bit) characters are used. For fonts defined with linear indexing rather than 2-byte matrix indexing, the server will interpret each CHAR2B as a 16-bit number that has been transmitted most significant byte first (that is, byte1 of the CHAR2B is taken as the most significant byte).

The x and y coordinates are relative to the drawable's origin and specify the baseline starting position (the initial character origin). The effect is first to fill a destination rectangle with the background pixel defined in gc and then to paint the text with the foreground pixel. The upper-left corner of the filled rectangle is at:

	[x, y - font-ascent]

the width is:

	overall-width

and the height is:

	font-ascent + font-descent

The overall-width, font-ascent, and font-descent are as they would be returned by a QueryTextExtents call using gc and string.

The function and fill-style defined in gc are ignored for this request. The effective function is Copy, and the effective fill-style Solid.

For fonts defined with 2-byte matrix indexing, each STRING8 byte is interpreted as a byte2 value of a CHAR2B with a byte1 value of zero.

GC components: plane-mask, foreground, background, font, subwindow-mode, clip-x-origin, clip-y-origin, clip-mask

This request is similar to ImageText8, except 2-byte (or 16-bit) characters are used. For fonts defined with linear indexing rather than 2-byte matrix indexing, the server will interpret each CHAR2B as a 16-bit number that has been transmitted most significant byte first (that is, byte1 of the CHAR2B is taken as the most significant byte).

This request creates a colormap of the specified visual type for the screen on which the window resides and associates the identifier mid with it. The visual type must be one supported by the screen (or a Match error results). The initial values of the colormap entries are undefined for classes GrayScale, PseudoColor, and DirectColor. For StaticGray, StaticColor, and TrueColor, the entries will have defined values, but those values are specific to the visual and are not defined by the core protocol. For StaticGray, StaticColor, and TrueColor, alloc must be specified as None (or a Match error results). For the other classes, if alloc is None, the colormap initially has no allocated entries, and clients can allocate entries.

If alloc is All, then the entire colormap is allocated writable. The initial values of all allocated entries are undefined. For GrayScale and PseudoColor, the effect is as if an AllocColorCells request returned all pixel values from zero to N - 1, where N is the colormap-entries value in the specified visual. For DirectColor, the effect is as if an AllocColorPlanes request returned a pixel value of zero and red-mask, green-mask, and blue-mask values containing the same bits as the corresponding masks in the specified visual. However, in all cases, none of these entries can be freed with FreeColors.

This request deletes the association between the resource ID and the colormap and frees the colormap storage. If the colormap is an installed map for a screen, it is uninstalled (see UninstallColormap request). If the colormap is defined as the colormap for a window (by means of CreateWindow or ChangeWindowAttributes), the colormap for the window is changed to None, and a ColormapNotify event is generated. The protocol does not define the colors displayed for a window with a colormap of None.

This request has no effect on a default colormap for a screen.

This request creates a colormap of the same visual type and for the same screen as src-cmap, and it associates identifier mid with it. It also moves all of the client's existing allocations from src-cmap to the new colormap with their color values intact and their read-only or writable characteristics intact, and it frees those entries in src-cmap. Color values in other entries in the new colormap are undefined. If src-cmap was created by the client with alloc All (see CreateColormap request), then the new colormap is also created with alloc All, all color values for all entries are copied from src-cmap, and then all entries in src-cmap are freed. If src-cmap was not created by the client with alloc All, then the allocations to be moved are all those pixels and planes that have been allocated by the client using either AllocColor, AllocNamedColor, AllocColorCells, or AllocColorPlanes and that have not been freed since they were allocated.

This request makes this colormap an installed map for its screen. All windows associated with this colormap immediately display with true colors. As a side effect, additional colormaps might be implicitly installed or uninstalled by the server. Which other colormaps get installed or uninstalled is server-dependent except that the required list must remain installed.

If cmap is not already an installed map, a ColormapNotify event is generated on every window having cmap as an attribute. In addition, for every other colormap that is installed or uninstalled as a result of the request, a ColormapNotify event is generated on every window having that colormap as an attribute.

At any time, there is a subset of the installed maps that are viewed as an ordered list and are called the required list. The length of the required list is at most M, where M is the min-installed-maps specified for the screen in the connection setup. The required list is maintained as follows. When a colormap is an explicit argument to InstallColormap, it is added to the head of the list; the list is truncated at the tail, if necessary, to keep the length of the list to at most M. When a colormap is an explicit argument to UninstallColormap and it is in the required list, it is removed from the list. A colormap is not added to the required list when it is installed implicitly by the server, and the server cannot implicitly uninstall a colormap that is in the required list.

Initially the default colormap for a screen is installed (but is not in the required list).

If cmap is on the required list for its screen (see InstallColormap request), it is removed from the list. As a side effect, cmap might be uninstalled, and additional colormaps might be implicitly installed or uninstalled. Which colormaps get installed or uninstalled is server-dependent except that the required list must remain installed.

If cmap becomes uninstalled, a ColormapNotify event is generated on every window having cmap as an attribute. In addition, for every other colormap that is installed or uninstalled as a result of the request, a ColormapNotify event is generated on every window having that colormap as an attribute.

This request returns a list of the currently installed colormaps for the screen of the specified window. The order of colormaps is not significant, and there is no explicit indication of the required list (see InstallColormap request).

This request allocates a read-only colormap entry corresponding to the closest RGB values provided by the hardware. It also returns the pixel and the RGB values actually used. Multiple clients requesting the same effective RGB values can be assigned the same read-only entry, allowing entries to be shared.

This request looks up the named color with respect to the screen associated with the colormap. Then, it does an AllocColor on cmap. The name should use the ISO Latin-1 encoding, and uppercase and lowercase do not matter. The exact RGB values specify the true values for the color, and the visual values specify the values actually used in the colormap.

The number of colors must be positive, and the number of planes must be nonnegative (or a Value error results). If C colors and P planes are requested, then C pixels and P masks are returned. No mask will have any bits in common with any other mask or with any of the pixels. By ORing together masks and pixels, C*%2 sup P% distinct pixels can be produced; all of these are allocated writable by the request. For GrayScale or PseudoColor, each mask will have exactly one bit set to 1; for DirectColor, each will have exactly three bits set to 1. If contiguous is True and if all masks are ORed together, a single contiguous set of bits will be formed for GrayScale or PseudoColor, and three contiguous sets of bits (one within each pixel subfield) for DirectColor. The RGB values of the allocated entries are undefined.

The number of colors must be positive, and the reds, greens, and blues must be nonnegative (or a Value error results). If C colors, R reds, G greens, and B blues are requested, then C pixels are returned, and the masks have R, G, and B bits set, respectively. If contiguous is True, then each mask will have a contiguous set of bits. No mask will have any bits in common with any other mask or with any of the pixels. For DirectColor, each mask will lie within the corresponding pixel subfield. By ORing together subsets of masks with pixels, C*%2 sup R+G+B% distinct pixels can be produced; all of these are allocated writable by the request. The initial RGB values of the allocated entries are undefined. In the colormap, there are only C*%2 sup R% independent red entries, C*%2 sup G% independent green entries, and C*%2 sup B% independent blue entries. This is true even for PseudoColor. When the colormap entry for a pixel value is changed using StoreColors or StoreNamedColor, the pixel is decomposed according to the masks and the corresponding independent entries are updated.

The plane-mask should not have any bits in common with any of the pixels. The set of all pixels is produced by ORing together subsets of plane-mask with the pixels. The request frees all of these pixels that were allocated by the client (using AllocColor, AllocNamedColor, AllocColorCells, and AllocColorPlanes). Note that freeing an individual pixel obtained from AllocColorPlanes may not actually allow it to be reused until all of its related pixels are also freed. Similarly, a read-only entry is not actually freed until it has been freed by all clients, and if a client allocates the same read-only entry multiple times, it must free the entry that many times before the entry is actually freed.

All specified pixels that are allocated by the client in cmap are freed, even if one or more pixels produce an error. A Value error is generated if a specified pixel is not a valid index into cmap. An Access error is generated if a specified pixel is not allocated by the client (that is, is unallocated or is only allocated by another client) or if the colormap was created with all entries writable (using an alloc value of All in CreateColormap). If more than one pixel is in error, it is arbitrary as to which pixel is reported.