aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2021-12-29Cursor: fix down mode motion events for subsurfacesIsaac Freund1-62/+77
There are a couple of TODOs here that are not worth resolving until after moving to the scene graph api.
2021-12-24docs: update zig version required in readmeHugo Machet1-1/+1
2021-12-24build: update to zig version 0.9.0Isaac Freund62-379/+345
2021-12-22docs: soften the README disclaimer a bitIsaac Freund1-3/+4
2021-12-22river: remove wlroots 0.14 workaroundsIsaac Freund6-34/+5
Since updating to wlroots 0.15 we can remove several workarounds we had in place due to wlroots improvements.
2021-12-22Subsurface: add commit listener on create if mappedIsaac Freund1-0/+2
Subsurfaces may already be mapped when create is called, in which case we must add the commit listener.
2021-12-22Cursor: properly handle clients setting the cursor imageIsaac Freund1-2/+6
The new code to dedup XcursorManager.setCursorImage() calls for efficiency currently doesn't handle clients setting the cursor properly. This commit corrects this oversight.
2021-12-22ci: update for wlroots 0.15.0Isaac Freund2-25/+36
2021-12-21layer-shell: fix regression from wlroots 0.15 updateIsaac Freund2-13/+13
The way wlroots handles the layer surface configure flow has changed a bit and river's logic needs updating.
2021-12-21Cursor: dedup XcursorManager.setCursorImage() callsIsaac Freund1-6/+25
wlroots doesn't avoid re-setting the same cursor image so this is relatively expensive to call repeatedly if nothing has changed.
2021-12-21river: update to wlroots 0.15.0Isaac Freund14-85/+79
2021-12-15common/flags: make argFlag() return a sliceIsaac Freund3-9/+11
We always pass the result of this to mem.span() currently, no need for the code duplication.
2021-12-10Cursor: improve output focus handling on button pressLeon Henrik Plickat1-1/+10
This patch allows to focus outputs by clicking on the empty background and by clicking on layer surfaces without keyboard interactivity. This makes it possible to use the cursor to focus outputs with no visible views. This also fixes problems with pointer interactive layer surfaces (for example launchers and docks) on non-focused outputs.
2021-11-29layer-shell: tighten up handling of large marginsIsaac Freund1-16/+6
2021-11-23Cursor: add workaround in pointer drag terminationIsaac Freund2-1/+22
Currently wlroots sends use the drag destroy event before sending the wl_data_device.leave event to the client, which makes things a bit awkward. My patch fixing this has been merged to wlroots master so we can remove this when upgrading to wlroots 0.15, but until then this workaround will fix the issue.
2021-11-19Cursor: fix pointer drags with focus-follows-cursorIsaac Freund2-23/+50
2021-11-11docs: remove reference to /etc/river/initIsaac Freund1-1/+3
We used to look in /etc/river/init if no init at ~/.config/river/init or $XDG_CONFIG_HOME/river/init was found but this feature was removed. It seems that we forgot to remove this mention of the old behavior however.
2021-11-08docs: fix wlroots link in readmeWannes De Meyer1-1/+1
2021-11-03docs: clarify description of dependenciesIsaac Freund1-3/+4
2021-11-03build: bump version to 0.2.0-devIsaac Freund1-1/+1
2021-11-03build: bump version to 0.1.0Isaac Freund1-1/+1
2021-11-02build: fix trailing newline in version stringIsaac Freund4-4/+7
2021-11-02docs: remove AUTHORS fileIsaac Freund1-6/+0
We haven't managed to keep this at all up to date, which defeats the purpose. Please refer to the git history for a complete list of contributors.
2021-11-01completions/fish: simplifyFollieHiyuki1-95/+70
2021-11-01docs: update URLs for migration to riverwm github orgIsaac Freund6-10/+10
2021-11-01common: use -h instead of -helpIsaac Freund6-12/+12
This doesn't really matter that much as unrecognized options will still trigger a help message to be printed, but -h is much more standard so lets make the predictable choice here while sticking to only single '-' flags.
2021-11-01docs: assorted fixes/cleanupsIsaac Freund3-8/+8
2021-10-31docs: standardize on "layout generator"Isaac Freund6-8/+8
This is likely more clear than "layout client" to most users.
2021-10-30Cursor: fix move/resize with high poll rate/low dpi miceIsaac Freund1-9/+32
2021-10-30View: fix checks to respect client side move/resizeIsaac Freund3-5/+5
2021-10-30view: respect client size resize of floating viewsIsaac Freund1-3/+8
mpv for example has key bindings to set the window size to a multiple of the video resolution. This is a valid use case for client-size resizing of the view and river should respect this if the view is floating.
2021-10-20Xwayland: always send configure if requestedIsaac Freund1-4/+9
This seems to fix an issue with mouse input for steam if steam is not started on an output at 0,0. X11 is pretty spooky.
2021-10-20View: send enter/leave events to full surface treeIsaac Freund1-5/+22
2021-10-20Cursor: only trigger focus-follow-cursor on motionIsaac Freund1-17/+18
This greatly improves the UX of this feature, as views moving under a stationary cursor (as happens during the zoom command for example) will no longer trigger focus change.
2021-10-12Xwayland: honor fullscreen requestsIsaac Freund2-5/+11
2021-10-11pointer-constraints: fix coordinatesIsaac Freund2-12/+29
Currently the implementation treats the x/y coordinates of View.State.box as layout coordinates instead of output-relative. This causes issues when using an output not at 0,0.
2021-10-06contrib: recommend using a venv for layout.pyIsaac Freund1-8/+11
This avoids needing to run pwayland's scanner with root privileges and install the river-layout bindings to system directories.
2021-10-06contrib: make layout.py more pythonicMichal Siedlaczek1-22/+19
2021-10-06contrib: add python layoutLeon Henrik Plickat1-0/+142
2021-10-04docs: change Enter to ReturnEuro201791-2/+2
xkbcommon uses `Return` to indicate this key, `Enter` isn't a valid keysym name. Co-authored-by: Euro <no>
2021-10-04Decrease layout demand timeout to 100msLeon Henrik Plickat1-1/+1
Layout generators are generally pretty fast. The timeout is only reached when the generator is faulty / stuck. In that case, freezing for 1 second is simply bad UX.
2021-10-02Server: fix use after free in deinit()Isaac Freund1-1/+1
The input manager and seats must be destroyed before the display is destroyed as they need to destroy their timer event sources used for key repeat.
2021-10-02View: fix unmap/destroy control flowIsaac Freund12-68/+64
Currently the view destruction sequence is started as soon as a view is unmapped. However, this is incorrect as a client may map the view again instead of destroying it. Instead, only start the view destruction sequence when the underlying xdg toplevel or xwayland surface is destroyed.
2021-10-02doc: update my (Isaac's) email addressIsaac Freund3-3/+3
2021-10-02View: add fullscreen check missing in previous commitIsaac Freund1-7/+9
2021-09-30View: resize if moved between outputs while fullscreenIsaac Freund1-0/+7
2021-09-30Subsurface: remove commit listener in destroy()Isaac Freund1-0/+2
Currently if destroy() is called while a subsurface is mapped a dangling commit listener is left behind. This is obivously a problem, so check if the subsurface is mapped in destroy() and remove the listener if needed.
2021-09-27ci: fix xbps invocationIsaac Freund1-6/+6
This avoids errors like: util-linux-common-2.37.2_1 in transaction breaks installed pkg `libfdisk-2.37.1_1' util-linux-common-2.37.2_1 in transaction breaks installed pkg `libsmartcols-2.37.1_1' libuuid-2.37.2_1 in transaction breaks installed pkg `libfdisk-2.37.1_1' libuuid-2.37.2_1 in transaction breaks installed pkg `util-linux-2.37.1_1' libblkid-2.37.2_1 in transaction breaks installed pkg `libfdisk-2.37.1_1' libblkid-2.37.2_1 in transaction breaks installed pkg `util-linux-2.37.1_1' libmount-2.37.2_1 in transaction breaks installed pkg `util-linux-2.37.1_1' Transaction aborted due to unresolved dependencies.
2021-09-27Cursor: implement surfaceAt() for XwaylandUnmanagedIsaac Freund1-6/+19
2021-09-18xdg-toplevel: fix configure serial checkingIsaac Freund2-7/+26
Currently if another configure is in flight after the one we are tracking the serial of and the client acks the second configure as well (or only the second configure) before committing, we will never realize the configure we are tracking has been acked. Instead, listen for the ack_configure signal and set a bool that we can check on surface commit. This probably isn't an issue that would actually be hit by well behaved clients as river doesn't send redundant configure events. However, having correct code is always better even if it's slightly more complex.
You can read more about the author.