summaryrefslogtreecommitdiff
path: root/XIproto.h
AgeCommit message (Collapse)AuthorFilesLines
2014-03-10inputproto: Allow library users to avoid having the 'Pointer' typedef declaredKeith Packard1-2/+4
'Pointer' collides with too many other application names, so stop using it locally and allow applications to avoid having it in the API. Signed-off-by: Keith Packard <keithp@keithp.com> Reviewed-by: Eric Anholt <eric@anholt.net>
2009-04-22Define the Cursor datasize correctlyBenjamin Close1-0/+2
On 64 bit machines, without Cursor defined Xlib would allocate 64 bits rather than 32 to any structs using Cursor. This led to data not correctly being available on the wire hence the Xserver would do strange things. We hence define Cursor to what it should be and make sure we undefine it after we've finished to users of XIproto.h aren't affected Fix-by: Peter Hutterer <peter.hutterer@who-t.net> Signed-off-by: Benjamin Close <Benjamin.Close@clearchain.com> Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2009-04-19Add GrabDevice and UngrabDevice XI2 requests.Peter Hutterer1-1/+1
2009-04-10Revert "Add major/minor version as supported by client to ↵Peter Hutterer1-6/+1
GetExtensionVersionReq." This reverts commit f6e41306f76de966884d4b72c5fb5e5d6d534ce4. Sending the supported version hidden in another request is potentially dangerous, so let's not do it. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2009-03-20Remove XI2 requests from XIproto.hPeter Hutterer1-336/+2
All requests been moved to XI2proto.h. Only ExtendedGrabDevice is gone for good.
2009-02-26Whitespace cleanups.Peter Hutterer1-533/+533
Yep. Slow day today.
2008-10-31PropertyNotify, move deviceid back to last byte.Peter Hutterer1-6/+6
This way, it can be type-cast to deviceKeyButtonPointer to extract the deviceid, which is (aside from time) the only thing it has in common with those anyway.
2008-10-31Remove window access protocol requests.Peter Hutterer1-56/+4
This is a bad idea. It didn't provide security and you can get the same functionality as you did with normal event registration.
2008-10-15Undef Atom after we're done so we don't pollute users of XIproto.hJulien Cristau1-0/+1
2008-10-15Make sure Atoms are defined as CARD32.Peter Hutterer1-0/+1
2008-09-19Remove Configure/QueryDeviceProperty.Peter Hutterer1-67/+15
2008-08-15Protect against C++ includes.Peter Hutterer1-0/+4
2008-08-15Remove RCS tags, typo fix.Peter Hutterer1-4/+1
2008-08-15Back out Device Properties from XI 2, push into XI 1.5.Peter Hutterer1-157/+200
2008-07-29Add DeviceControlChanged define.Peter Hutterer1-1/+1
This value is used for the devchange field in the DevicePresenceNotify event when a device's control has been modified.
2008-07-09Set IEVENTS back to 18, got set to 8 inadvertantly.Peter Hutterer1-1/+1
2008-07-09Add XI device property requests and replies.Peter Hutterer1-4/+147
New requests: ListDeviceProperties ... list all props of a device QueryDeviceProperty ... query meta-information about a property ChangeDeviceProperty ... change the content of a property DeleteDeviceProperty ... delete a property GetDeviceProperty ... retrieve a property New event: DevicePropertyChangedNotify ... the given property on the device has changed
2008-04-26Add major/minor version as supported by client to GetExtensionVersionReq.Peter Hutterer1-1/+6
This sort-of breaks old clients. Behaviour to be assumed is that if nbytes is 0, major/minorVersion is set and specifies the version as supported by the client. If nbytes is non-zero, the request is trailed by the extension name (INAME) and major/minorVersion is undefined. This is the behaviour of pre-MPX clients. And then there may be clients who found that no other extension uses this request and supplying a name wasn't actually necessary since it was XI anyway. These clients will break. Tough luck. Read the man pages next time.
2008-04-25Add #define IREQUESTS 45. Specifies the number of requests in XI.Peter Hutterer1-0/+1
2008-03-04Add deviceid to QueryDevicePointer reply.Peter Hutterer1-1/+1
Doesn't hurt, we have padding left over anyway.
2008-02-25Squash opcode range for MPX XI requests.Peter Hutterer1-6/+6
This removes the opcode holes that were left by the excessive request removal of the last weeks.
2008-02-25Remove GetPairedPointer, paired device can be found through ListInputDevices.Peter Hutterer1-31/+0
2008-02-25Remove "ungrab" from ExtendedGrabDevice request, remove XUngrabExtDevice().Peter Hutterer1-2/+2
That's what UngrabDevice is for, it does the same anyway.
2008-02-14Remove GrabAccessControl and FakeDeviceData.Peter Hutterer1-54/+0
Both aren't thought out enough to justify their inclusion in the first version of MPX.
2008-02-12Remove RawDeviceEvent - for now anyway.Peter Hutterer1-30/+0
Wasn't quite as thought-out as it should be. Throwing it out for now, to get the rest of MPX more stable.
2008-02-12Remove pairingChangedNotify event.Peter Hutterer1-23/+0
I swear I already removed that before... Anyway, we don't need it anymore, since pairings can't be changed anyway. Hooray for the device hierarchy.
2008-02-12Remove "shared" field from QueryDevicePointer.Peter Hutterer1-1/+1
If it's a slave device, it's shared, if it's a master device it has its own cursor. No need for this field.
2008-01-25Add a device id to XiSelectEvent.Peter Hutterer1-0/+3
2007-11-13Remove RegisterPairingClient, deprecated with the device hierarchy now.Peter Hutterer1-30/+0
2007-11-13Add DeviceClassesChangedEvent.Peter Hutterer1-0/+29
2007-11-08Add DeviceHierarchyChangedEvent.Peter Hutterer1-25/+24
Uses same event type as the now removed PointerKeyboardPairingChangedNotify. (removing the RandomStringEvent too, should have been gone a while ago)
2007-10-19Add XChangeDeviceHierarchy and its components.Peter Hutterer1-7/+47
2007-10-17xDeviceInfo: add "attached" field (replace previous padding).Peter Hutterer1-2/+2
If use is set to IsXExtensionPointer/Keyboard/Devices, attached indicates the device ID of the master device it is attached to. If the device is floating, attached is set to IsFloating.
2007-05-17Move deviceid around in deviceEnterNotify, make room for detail field.Peter Hutterer1-3/+4
2007-05-17Add event_type to RawDeviceEvent to store matching core event type.Peter Hutterer1-1/+3
2007-05-14Add ExtendedGrabRequest and the matching reply.Peter Hutterer1-0/+41
2007-05-02Add xFakeDeviceDataReqPeter Hutterer1-0/+24
2007-05-02Providing a device id for a RawDeviceEvent may not be a bad idea.Peter Hutterer1-1/+1
2007-05-01Add XGE support and event types for RawDeviceEvent and PairingChanged event.Peter Hutterer1-7/+81
2007-04-27Merge branch 'master' into mpxPeter Hutterer1-1/+1
2007-04-24Add flags to be used for DevicePrensence's devchange field.Peter Hutterer1-1/+1
2007-03-22add GetClientPointer request and reply.Peter Hutterer1-0/+59
add GetPairedPointer request and reply. move declaration of _XiGetDevicePresenceNotifyEvent out of the macro and wrap it between extern "C". Otherwise C++ code won't be able to find it.
2007-03-09add SetClientPointer request.Peter Hutterer1-3/+23
fix typos and wrong names for access function declarations.
2007-02-22add access control requests.Peter Hutterer1-3/+85
fix wrong field lengths for RegisterPairing request and reply.
2007-02-21add xRegisterPairingClient request and replyPeter Hutterer1-0/+31
2007-02-08add flags field to deviceEnterNotify structPeter Hutterer1-3/+2
add same_screen, focus to XDeviceCrossingEvent struct
2007-01-29add ChangePointerKeyboardPairing requestPeter Hutterer1-1/+40
add pairingChangedNotify event
2007-01-10add deviceEnterNotify event, DeviceEnterNotify, DeviceLeaveNotify supportPeter Hutterer1-1/+31
add MPX Major/Minor version numbers
2007-01-08 add ChangeDeviceCursor requestPeter Hutterer1-0/+18
2006-12-20 add QueryDevicePointer request + replyPeter Hutterer1-0/+62
add WarpDevicePointer request