aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeLines
* meson/PKGBUILD: bump version to 1.1.01.1.0releases/1.1Daniel Eklöf2020-01-15-2/+2
|
* ci: checkout/clone (doh!)Daniel Eklöf2020-01-14-2/+2
|
* ci: no 'git' available by default, need to install itDaniel Eklöf2020-01-14-0/+1
|
* ci; no pushd/popd availableDaniel Eklöf2020-01-14-2/+2
|
* input: reset XKB compose state on keyboard leaveDaniel Eklöf2020-01-14-0/+1
|
* ci: checkout tllist/fcft subprojectsDaniel Eklöf2020-01-14-0/+4
|
* Revert "input: generate key down events on keyboard-enter"Daniel Eklöf2020-01-14-9/+1
| | | | This reverts commit 84739d78c626ad12543417dc1e3038abd6a4e2d5.
* input: generate key down events on keyboard-enterDaniel Eklöf2020-01-13-1/+9
|
* input: reset modifier state on keyboard-leaveDaniel Eklöf2020-01-13-0/+5
|
* render: resize: TIOCSWINSZ: don't include padding in ws_{x,y}pixelDaniel Eklöf2020-01-12-2/+2
|
* term: shorten application_synchronized_updates -> app_sync_updatesDaniel Eklöf2020-01-12-26/+26
|
* term: enable_application_synchronized_updates: not a noop if already enabledDaniel Eklöf2020-01-12-3/+0
| | | | | If synchronized updates are already enabled when we receive a request to enable them, extend the timeout.
* dcs: bsu/esu: add link to specDaniel Eklöf2020-01-12-0/+4
|
* dcs: abort() in bsu/esuDaniel Eklöf2020-01-12-2/+4
| | | | | Since I haven't seen this in the wild, let's abort() to make sure we catch when an application does this, so we can verify it really works.
* Merge branch 'synchronized-updates'Daniel Eklöf2020-01-12-29/+124
|\
| * term: app sync updates: disable delayed rendering timers once onlyDaniel Eklöf2020-01-12-8/+9
| | | | | | | | | | Disable the delayed rendering timers when enabling application synchronized updates, not every time we've received ptmx data.
| * term/render: move {enable,disable}_application_synchronized_updates()Daniel Eklöf2020-01-12-38/+37
| | | | | | | | From render -> terminal
| * render: add a timeout for application synchronized updatesDaniel Eklöf2020-01-12-8/+60
| | | | | | | | | | This ensures we can recover from a crashing (or bad behaving) application that sends a BSU but then never sends an ESU.
| * render: rename render_{enable,disable}_refresh()Daniel Eklöf2020-01-12-7/+9
| |
| * term: rename refresh_prohibited -> application_synchronized_updatesDaniel Eklöf2020-01-12-7/+7
| |
| * term: ptmx: cancel, and don't re-arm, delay timers when refresh has been disabledDaniel Eklöf2020-01-12-22/+34
| |
| * dcs: bsu/esu: call render_{disable,enable}_refresh()Daniel Eklöf2020-01-12-6/+8
| |
| * render: add render_{enable,disable}_refresh()Daniel Eklöf2020-01-12-0/+27
|/ | | | | | | Calling render_disable_refresh() causes update requests to that terminal to be ignored. Calling render_enable_refresh() re-enables updates.
* dcs: abort() in bsu/esuDaniel Eklöf2020-01-12-0/+2
| | | | I **really** want to know if someone is using these...
* dcs: put: don't append data if buffer resize failedDaniel Eklöf2020-01-12-1/+2
|
* dcs: use default param value of 0 in log messageDaniel Eklöf2020-01-12-1/+1
|
* Merge branch 'dcs'Daniel Eklöf2020-01-12-0/+106
|\
| * dcs: initial handling of DCS in generalDaniel Eklöf2020-01-12-0/+106
|/ | | | | | | | | | | | | | | | | | | | | | Add data structure to term->vt. This structure tracks the free-form data that is passed-through, and the handler to call at the end. Intermediates and parameters are collected by the normal VT parser. Then, when we enter the passthrough state, we call dcs_hook(). This function checks the intermediate(s) and parameters, and selects the appropriate unhook handler (and optionally does some execution already). In passthrough mode, we simply append strings to an internal buffer. This might have to be changed in the future, if we need to support a DCS that needs to execute as we go. In unhook (i.e. when the DCS is terminated), we execute the unhook handler. As a proof-of-concept, handlers for BSU/ESU (Begin/End Synchronized Update) has been added (but are left unimplemented).
* render: don't bother trying to lookup glyph for empty cellsDaniel Eklöf2020-01-12-1/+2
| | | | | Depending on the font, it may not have an entry for code point '0', which will cause us to try to load fallback fonts.
* Merge branch 'switch-based-keymap-lookup'Daniel Eklöf2020-01-11-156/+176
|\
| * input: commentsDaniel Eklöf2020-01-11-45/+57
| |
| * input: keymap_lookup: only do lookup, don't sent to clientDaniel Eklöf2020-01-11-11/+12
| |
| * keymap: remove keymap tableDaniel Eklöf2020-01-11-89/+0
| | | | | | | | Lookup is now done in code, using a switch-statement
| * input: add a switch-based function that translates XKB keysym to escape dataDaniel Eklöf2020-01-11-19/+105
| |
| * input: break out keymap lookup to a separate functionDaniel Eklöf2020-01-11-25/+35
|/
* README: installation: s/debug/release/Daniel Eklöf2020-01-11-1/+1
|
* README: install::other: do a release build, and mention profile guidedDaniel Eklöf2020-01-11-2/+6
|
* fcft/tllist: remove git submodulesDaniel Eklöf2020-01-11-18/+13
|
* keymap: add KP enterDaniel Eklöf2020-01-11-0/+2
| | | | | This makes the keypad enter key emit "\eOM" in keypad application mode.
* term/wayland: use MONOTONIC timersDaniel Eklöf2020-01-10-6/+6
|
* README: some terminal emulators call "server mode" "daemon mode"Daniel Eklöf2020-01-10-2/+2
|
* keymap: add escapes for keypad 'begin'Daniel Eklöf2020-01-10-0/+7
| | | | This is usually the '5' key, in non-num lock mode.
* Revert "terminal: use edge-triggered FDM handler for the PTY"Daniel Eklöf2020-01-10-16/+16
| | | | This reverts commit f5e27b49eb53ec64a592236680deee0c68c32f0f.
* Revert "selection: use edge-triggered FDM handlers"Daniel Eklöf2020-01-10-3/+3
| | | | This reverts commit 5bac3cfa2cb6a1e7b6eb0012c134bd2e844fcf61.
* Revert "server: use edge-triggered FDM handler for the client connections"Daniel Eklöf2020-01-10-13/+13
| | | | This reverts commit 9141b8a6573f973f97aa1615a8cc0c488cfd77e7.
* server: use edge-triggered FDM handler for the client connectionsDaniel Eklöf2020-01-10-13/+13
|
* server: be more resilient against bad behaving clientsDaniel Eklöf2020-01-10-3/+16
| | | | | | | | | | | | | | | Make both the server listening socket and the connecting client sockets non-blocking. Then, when reading the initial length of the setup packet, handle read errors from partial reads separately. Assume the client writes all four bytes of the 'length' field at once, and bail out if we are unable to read those 4 bytes. Limit the maximum setup packet size to 128K. This is to prevent clients from pretending the setup packet is insanely large, causing us to fail to malloc.
* terminal: make the event FD used to delay shutdown non-blockingDaniel Eklöf2020-01-10-1/+1
|
* terminal: use edge-triggered FDM handler for the PTYDaniel Eklöf2020-01-10-16/+16
| | | | | | | | This means that the read handler now has to keep reading until we get EAGAIN. But it also means we don't have to flip the EPOLLOUT flag in the FDM handler back and forth when writing to the PTY.
* selection: use edge-triggered FDM handlersDaniel Eklöf2020-01-10-3/+3
|
* wayland; fixupDaniel Eklöf2020-01-10-0/+1
|
* wayland: just to be sure, check return value of wl_display_read_events()Daniel Eklöf2020-01-10-1/+4
|
* wayland: make wayland socket non-blockingDaniel Eklöf2020-01-10-0/+6
|
* fdm: fix log messageDaniel Eklöf2020-01-10-1/+1
|
* fdm: verify FDs are non-blocking (debug builds only)Daniel Eklöf2020-01-10-0/+10
|
* fcft: bumpDaniel Eklöf2020-01-09-0/+0
|
* completions: zsh: add empty completions for options with required argumentDaniel Eklöf2020-01-09-1/+1
|
* selection: finalize: ignore whether selections are 'enabled' or notDaniel Eklöf2020-01-06-3/+0
| | | | | | This fixes an issue where a "forced" selection (shift being pressed while slave is tracking mouse events) would not finalize if the user released shift *before* releasing the mouse button.
* selection: update: don't dirty cells that don't change stateDaniel Eklöf2020-01-06-21/+50
| | | | | | | | | | | | | | | | Previously when updating a selection, we would unmark *all* cells in the old selection, and then mark all cells in the new selection. This caused *all* cells to be dirtied and thus re-rendered. Avoid this, by adding a temporary state to the cells' selected state. Before unmarking the old selection, pre-mark the new selection using a temporary state. When unmarking the old selection, ignore cells in this temporary state. When marking the new selection, ignore cells in this temporary state (except clearing the temporary state).
* selection: foreach_selection_*: must wrap row numberDaniel Eklöf2020-01-05-2/+6
| | | | | Fixes occasional crashes when we're at the end of the grid and selection wraps around.
* search: rename render.search_offset -> render.search_glyph_offsetDaniel Eklöf2020-01-05-8/+8
|
* Merge branch 'search-render-improvements'Daniel Eklöf2020-01-05-6/+29
|\
| * render: ensure cursor is always visible in the search boxDaniel Eklöf2020-01-05-6/+29
|/ | | | | | | | | | | | | | | | | | Maintain a view 'offset' (which glyph from the search string to start rendering at). This defines the start of the viewable area. The end is the offset + the search box size (which is limited to the window size). Adjust this offset whenever the cursor moves outside the viewable area. For now, this is always done in the same way: set the offset to the cursor position. This means that when we're entering text at the end of the search criteria (i.e. the normal case; we're simply typing), and the search box reaches the window size, the cursor will jump to the start of the search box, which will be empty. This could be confusing, but let's go with for now.
* README: add 'git submodule update --init' step to how-to-installDaniel Eklöf2020-01-05-1/+8
|
* render: xcursor: remove render_xcursor_refresh()Daniel Eklöf2020-01-04-18/+17
|
* doc: slight change of wording for mouse-drag-to-selectDaniel Eklöf2020-01-04-5/+5
|
* README: abuse of definition list to get multiple paragraphsDaniel Eklöf2020-01-04-5/+5
|
* README: let's see if codeberg understands these paragraphs belong togetherDaniel Eklöf2020-01-04-1/+1
|
* renderer: destroy: actually free the renderer instance (doh!)Daniel Eklöf2020-01-04-0/+2
|
* wayland: wayl_flush: re-initialize fd array before each pollDaniel Eklöf2020-01-04-5/+13
|
* wayland: add commentDaniel Eklöf2020-01-04-0/+4
|
* wayland: only call wl_display_prepare_read() if we read any eventsDaniel Eklöf2020-01-04-3/+4
| | | | | | That is, only call it if we have POLLIN. If not, then we never read any events and we still hold the read lock and thus we shouldn't try to acquire it again.
* wayland: use a low priority FDM hook to flush the wayland socketDaniel Eklöf2020-01-04-3/+14
| | | | | | | | This way, we don't have to manually insert flushes in code paths that may execute outside of a wl_display_dispatch_pending(). (Those that execute inside a wl_display_dispatch_pending() are subject to the flush performed at the end of the normal wayland FDM handler).
* fdm: add hook prioritiesDaniel Eklöf2020-01-04-17/+69
| | | | | | | | | | There are now three hook priorities: low, normal, high. High priority hooks are executed *first*. Normal next, and last the low priority hooks. The renderer's terminal refresh hook now runs as a normal priority hook.
* render: throttle xcursor updatesDaniel Eklöf2020-01-04-76/+136
| | | | | | | | | | | | | | | | | | | | | | | With a bad behaving client (e.g. 'less' with mouse support enabled), we can end up with a *lot* of xcursor updates (so much, that we flooded the wayland socket before we implemented a blocking wayl_flush()). Since there's little point in updating the cursor more than once per frame interval, use frame callbacks to throttle the updates. This works more or lesslike normal terminal refreshes: render_xcursor_set() stores the last terminal (window) that had (and updated) the cursor. The renderer's FDM hook checks if we have such a pending terminal set, and if so, tries to refresh the cursor. This is done by first checking if we're already waiting for a callback from a previous cursor update, and if so we do nothing; the callback will update the cursor for the next frame. If we're *not* already waiting for a callback, we update the cursor immediately.
* input: scroll: call term_mouse_up() once, after all term_mouse_down()Daniel Eklöf2020-01-04-0/+1
| | | | | Note that scroll button release events are not actually sent to the client, but this makes the code more correct, logically.
* wayland: wl_display_flush() never blocksDaniel Eklöf2020-01-04-4/+46
| | | | | | | | | | | | | | | | | | | Since it doesn't block, we need to detect EAGAIN failures and ensure we actually flush everything. If we don't, we sooner or later end up in a wayland client library call that aborts due to the socket buffer being full. Ideally, we'd simply enable POLLOUT in the FDM. However, we cannot write *anything* to the wayland socket until we've actually managed to send everything. This means enabling POLLOUT in the FDM wont work since we may (*will*) end up trying to write more data to it before we've flushed it. So, add a wrapper function, wayl_flush(), that acts as a blocking variant of wl_display_flush(), by detecting EAGAIN failiures and calling poll() itself, on the wayland socket only, until all data has been sent.
* render: remove debug loggingDaniel Eklöf2020-01-04-5/+2
|
* Merge branch 'render-refresh-optimize'Daniel Eklöf2020-01-04-11/+125
|\
| * term: more comments in the 'render' sub-structDaniel Eklöf2020-01-04-3/+4
| |
| * render: attrs_to_font() is not used outside of render.cDaniel Eklöf2020-01-04-1/+1
| |
| * render: trigger terminal refreshes in an FDM hookDaniel Eklöf2020-01-04-8/+68
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In some cases, we end up calling render_refresh() multiple times in the same FDM iteration. This means will render the first update immediately, and then set the 'pending' flag, causing the updated content to be rendered in the next frame. This can cause flicker, or flashes, since we're presenting one or more intermediate frames until the final content is shown. Not to mention that it is inefficient to render multiple frames like this. Fix by: * render_refresh() only sets a flag in the terminal * install an FDM hook; this hook loops all terminals and executes what render_refresh() _used_ to do (that is, render immediately if we're not waiting for a frame callback, otherwise set 'pending' flag). for all terminals that have the 'refresh_needed' flag set.
| * fdm: add support for hooksDaniel Eklöf2020-01-04-0/+53
|/ | | | | Hooks are functions executed just before going into a poll(). Or just after executing all FD handlers, if you like.
* doc: document how to do block selectionsDaniel Eklöf2020-01-04-0/+2
|
* doc: add descriptions for ctrl+w and ctrl+shift+w in scrollback searchDaniel Eklöf2020-01-04-0/+9
|
* doc: add headers to keyboard shortcuts for normal+scrollback searchDaniel Eklöf2020-01-04-2/+3
|
* README: mention that ctrl+mouse creates a block selectionDaniel Eklöf2020-01-04-0/+2
|
* PKGBUILD: remove old *.gcda filesDaniel Eklöf2020-01-04-1/+2
|
* Merge branch 'block-selection'Daniel Eklöf2020-01-04-127/+267
|\
| * selection: min_bufsize_for_extraction: add one extra column for *all* rowsDaniel Eklöf2020-01-04-7/+4
| |
| * selection: rename selection_cell_count -> min_bufsize_for_extractionDaniel Eklöf2020-01-04-2/+2
| |
| * selection: cell count now adds one extra column per row, for \nDaniel Eklöf2020-01-04-3/+11
| | | | | | | | | | | | | | | | When we extract text, we may insert '\n' at the end of each line (or last column of selection, for block selections). These newlines doesn't occupy any physical cells, and thus we must **make** room for them in the extraction buffer.
| * selection: fix bug where first column on all rows but the first was lostDaniel Eklöf2020-01-04-10/+12
| | | | | | | | | | | | | | | | | | | | When extracting text from the selection, we lost the first column on all rows but the first. This is because the algorithm changed slightly when we moved to foreach_selection(); the end-of-line detection is now done on the first column of the new line, instead of the last column on the previous line.
| * selection: simplify extraction by converting to UTF-8 at the endDaniel Eklöf2020-01-04-21/+25
| | | | | | | | | | Instead of converting a cell at a time to UTF-8, copy the wide characters as is, and then convert everything at once at the end.
| * selection: extract_selection now uses foreach_selected()Daniel Eklöf2020-01-04-57/+105
| | | | | | | | | | This "automagically" adds support for block selections, since foreach_selected already handles block selections.
| * term: mouse_grabbed: ctrl may be pressed too (for block selection)Daniel Eklöf2020-01-04-1/+1
| |
| * selection: selection_cancel() now sets 'kind' to SELECTION_NONEDaniel Eklöf2020-01-04-1/+6
| |
| * selection: store cell 'selected' state in the cells' attributesDaniel Eklöf2020-01-04-78/+129
| | | | | | | | | | | | | | | | | | Instead of having the renderer calculate, for each cell, whether that cell is currently selected or not, make selection_update() mark/unmark the selected cells. The renderer now only has to look at the cells' 'selected' attribute. This makes the renderer both smaller and faster.
| * render: coord_is_selected: handle block selectionsDaniel Eklöf2020-01-03-8/+22
| |
| * selection: track selection type; normal or block selectionDaniel Eklöf2020-01-03-7/+18
|/