aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeLines
* meson: bump version to 1.10.01.10.0Daniel Eklöf2021-11-14-1/+1
|
* changelog: prepare for 1.10.0Daniel Eklöf2021-11-14-4/+11
|
* Merge branch 'user-notification-cleanup'Daniel Eklöf2021-11-14-44/+25
|\
| * user-notification: config: various small cleanupsCraig Barnes2021-11-14-44/+25
|/ | | | | | | * Rename user_notification_add() to user_notification_add_fmt() * Add new user_notification_add() helper function * Use xvasprintf() to replace user_notification_add_va() * Make better use of helper functions in config.c
* config: remove options that were deprecated in 1.8.xDaniel Eklöf2021-11-14-82/+3
| | | | | | | * [main].bell * [main].jump-label-letters * [main].url-launch * [main].osc8-underline
* changelog: fix minor grammatical mistakeCraig Barnes2021-11-14-1/+1
|
* themes: solarized-dark: add commented out selection fg/bg colorsDaniel Eklöf2021-11-13-18/+23
|
* themes: add solarized-dark-normal-brightsDaniel Eklöf2021-11-13-0/+31
| | | | | | | | | | | This is a custom version of solarized-dark, where the bright colors have been generated by increasing the regular colors’ luminance by 20%. In other words, this version does *not* have the solarized colors that normally occupy the bright colors. Thus, if you want the full solarized experience in e.g. editors, make sure to use editor-specific 24-bit themes.
* Merge branch 'configurable-dim-colors'Daniel Eklöf2021-11-13-9/+91
|\ | | | | | | Closes #776
| * config: add [colors].dim0-7Daniel Eklöf2021-11-13-9/+91
|/ | | | | | | | | | | | | | | | | | | | | | | | | This allows you to configure custom colors to be used when colors are being dimmed (`\E[2m`). It is implemented by color matching (just like bold-text-in-bright=palette-based); the color-to-be-dimmed is matched against the current color palette. If it matches one of the regular colors (colors 0-7), the corresponding “dim” color will be used. If it matches one of the bright colors (colors 8-15), the corresponding “regular” color will be used (but *only* if the “dim” color has been set). Otherwise, the color is dimmed by reducing its luminance. The default behavior, i.e. when dim0-7 hasn’t been configured, is to dim by reducing luminance for *all* colors. I.e. we don’t do any color matching at all. In particular, this means that dimming a bright color will *not* result in the corresponding “regular” color. Closes #776
* doc: foot.ini: try to clear up the meaning of (AA)RRGGBBDaniel Eklöf2021-11-13-30/+39
|
* config: scrollback.indicator.text: use value_to_wchars()Daniel Eklöf2021-11-13-13/+2
|
* Merge branch 'modify-other-keys-mode-2'Daniel Eklöf2021-11-13-32/+90
|\
| * input: add support for modifyOtherKeys=2Daniel Eklöf2021-11-13-32/+90
|/ | | | | | | | | | | | | | | | Similar to modifyOtherKeys=1 (foot’s default, and only, mode), except that: * All modifiers (and not just Ctrl) generate \E[27;m;n~ escapes * Regular keys (with modifiers) also generate \E[27;m;n~ escapes (for example, C-h no longer generates ^H, but \E[27;5;104~) For our keymap based lookups, this is handled by adding MOD_MODIFY_OTHER_KEYS_STATE<N> variants. For “generic” keys, we simply adjust the conditions for when to emit a \E[27;m;n~ escape - the only requirement is that at least one modifier is active.
* Merge branch 'no-scrollback-multiplier-on-alt-screen'Daniel Eklöf2021-11-13-4/+13
|\ | | | | | | Closes #787
| * input: don’t apply [scrollback].multiplier when alt screen is in useDaniel Eklöf2021-11-13-4/+13
|/ | | | Closes #787
* Merge branch 'config-refactor'Daniel Eklöf2021-11-13-803/+708
|\
| * changelog: boolean optionsDaniel Eklöf2021-11-13-0/+4
| |
| * config: value_to_wchars(): free the target variableDaniel Eklöf2021-11-13-34/+11
| |
| * config: appease compilerDaniel Eklöf2021-11-13-1/+1
| | | | | | | | | | | | | | | | ../../foot/config.c:154:50: error: format string is not a string literal (potentially insecure) [-Werror,-Wformat-security] log_msg(log_class, LOG_MODULE, file, lineno, formatted_msg); ^~~~~~~~~~~~~ ../../foot/config.c:154:50: note: treat the string as an argument to avoid this log_msg(log_class, LOG_MODULE, file, lineno, formatted_msg)
| * config: drop all warning logs from the tweak sectionDaniel Eklöf2021-11-13-53/+19
| |
| * config: value_to_enum(), return conversion success/failure as a boolDaniel Eklöf2021-11-13-79/+91
| | | | | | | | | | | | | | Result is set through a pointer passed as an argument. Note that we assume enums are 32-bit, and explicitly cast them to int*. Static asserts have been added to ensure this invariant holds.
| * config: parse_section_main(): explicit return from each branchDaniel Eklöf2021-11-13-25/+27
| | | | | | | | Add UNREACHABLE() at the bottom
| * config: add value_to_dimensions()Daniel Eklöf2021-11-13-14/+15
| |
| * config: add {str,value}_to_uint{16,32}()Daniel Eklöf2021-11-13-106/+92
| | | | | | | | | | This allows us to pass the final pointer directly to the conversion functions, as well as allowing us to print outside-range errors.
| * config: add value_to_str()Daniel Eklöf2021-11-13-18/+17
| | | | | | | | Takes a pointer to a char*, free:s it, and strdups the current value.
| * config: value_to_bool: return success/failDaniel Eklöf2021-11-13-26/+49
| | | | | | | | | | Write boolean value to a pointer passed as argument, return conversion success/failure
| * config: add value_to_enum()Daniel Eklöf2021-11-13-103/+107
| |
| * config: let value_to_double() log errorsDaniel Eklöf2021-11-13-13/+14
| |
| * config: let value_to_ulong() log errorsDaniel Eklöf2021-11-13-29/+16
| |
| * config: add LOG_CONTEXTUAL_{ERR,WARN,ERRNO}Daniel Eklöf2021-11-13-288/+163
| | | | | | | | | | | | These work like LOG_AND_NOTIFY_* but add contextual information: path of the conf file, line number, section name, key/option name and the value.
| * config: first argument to log_and_notify() is now a ‘struct context’Daniel Eklöf2021-11-13-139/+119
| |
| * config: rename str_to_spawn_template() -> value_to_spawn_template()Daniel Eklöf2021-11-13-20/+11
| |
| * config: rename str_to_fonts() -> value_to_fonts()Daniel Eklöf2021-11-13-10/+8
| |
| * config: rename str_to_wchars() -> value_to_wchars()Daniel Eklöf2021-11-13-19/+13
| |
| * config: rename str_to_double() -> value_to_double()Daniel Eklöf2021-11-13-28/+22
| | | | | | | | | | * str_to_double() -> value_to_double() * str_to_pt_or_px() -> value_to_pt_or_px()
| * config: rename str_to_ulong() -> value_to_ulong()Daniel Eklöf2021-11-13-47/+73
| | | | | | | | | | | | * str_to_ulong() -> value_to_ulong() * str_to_color() -> value_to_color() * str_to_two_colors() -> value_to_two_colors()
| * config: value_to_bool(): fixupDaniel Eklöf2021-11-13-1/+1
| |
| * config: rename str_to_bool() -> value_to_bool()Daniel Eklöf2021-11-13-19/+19
| | | | | | | | Change argument to a ‘struct context’ pointer
| * config: pass aggregated struct to top-level section handlersDaniel Eklöf2021-11-13-53/+138
|/
* main: improve handling of setlocale(3) return valueCraig Barnes2021-11-10-2/+6
| | | | | | | | | There's no need to call setlocale() twice, since the first call returns the same value as the second. We also handle a NULL return value explicitly and show a distinct error message for that case, since it typically indicates a misconfigured locale and not just a non-UTF8 locale.
* wayland: guard XDG startup notification with HAVE_XDG_ACTIVATIONDaniel Eklöf2021-11-09-0/+2
| | | | Fixes compilation on systems with “too old” wayland-protocols
* PKGBUILD: dropDaniel Eklöf2021-11-07-83/+11
| | | | | The bundled PKGBUILD has been pretty much identical to the one in the AUR package “foot-git” for a long time now.
* doc: foot.ini: mention “xkbcli interactive-wayland”Daniel Eklöf2021-11-07-0/+2
|
* Merge branch 'error-out-on-too-old-wl-seat-interface'Daniel Eklöf2021-11-07-0/+7
|\ | | | | | | Closes #779
| * wayland: error out if there aren’t any seats availableDaniel Eklöf2021-11-07-0/+7
|/ | | | | | | | | | | | If there aren’t any seats available, input is not possible (neither keyboard nor mouse input). Note that this is not the same thing as e.g. no keyboard being plugged in. In this case, there will be a seat, but without the keybaord capability. Input may not be possible _right now_, but may be later, if the user plugs in a keyboard. Closes #779
* Merge branch 'xdga-launch'Daniel Eklöf2021-11-03-11/+51
|\
| * foot/client: implement xdga client activationRonan Pigott2021-11-01-11/+51
|/ | | | | | | | | | | This is an application of the xdg activation protocol that will allow compositors to associate new foot toplevels with the command that launched them. footclient receives an activation token from the launcher which the compositor can use to track application startup. It passes the token to the foot server, which then activates the new window with the token to complete the startup sequence.
* config: format all (well, most) error messages in the same wayDaniel Eklöf2021-10-31-127/+120
| | | | path:lineno: [section].key: value: message desribing what’s wrong with value
* config: use "[section].key" syntax when logging errorsDaniel Eklöf2021-10-31-55/+55
|
* config: improve error messages for invalid section keysDaniel Eklöf2021-10-31-9/+18
|
* install: remove text describing obsolete differences between PKGBUILDsDaniel Eklöf2021-10-31-9/+3
|
* readme: add GitHub Sponsors linkDaniel Eklöf2021-10-31-0/+6
|
* Merge branch 'client-sendall'Daniel Eklöf2021-10-31-7/+28
|\
| * client: use MSG_NOSIGNAL in send(3)Daniel Eklöf2021-10-30-1/+1
| | | | | | | | | | | | | | MSG_NOSIGNAL Requests not to send the SIGPIPE signal if an attempt to send is made on a stream-oriented socket that is no longer connected. The [EPIPE] error shall still be returned.
| * client: handle partial writes in send(3)Daniel Eklöf2021-10-29-7/+28
|/
* Merge branch 'csd-border'Daniel Eklöf2021-10-29-8/+97
|\
| * render: csd_part: we don’t use the ‘src’ pixman imageDaniel Eklöf2021-10-29-3/+0
| |
| * foot.ini: csd.border-color: <color> -> <csd.color>Daniel Eklöf2021-10-29-1/+1
| |
| * csd: add support for a visible borderDaniel Eklöf2021-10-27-7/+99
|/ | | | | | | | | | | | | | | | | | | | | | | | When we’re using CSDs, we’ve up until now rendered a 5px invisible border. This border handles interactive resizing. I.e. hovering it changes the mouse cursor, and mouse button events are used to start an interactive resize. This patch makes it possible to color part of (or the entire) border, with a configurable color. To facilitate this, two new options have been added: * csd.border-width * csd.border-color border-width defaults to 0, resulting in the look we’re used to. border-color defaults to the title bar color. If the title bar color hasn’t been set, it defaults to the default foreground color (just like the title bar color does). This means that, setting border-width but not border-color, results in a border that blends with the title bar.
* meson: install themes to $datadir/foot/themesDaniel Eklöf2021-10-27-0/+2
|
* Merge branch 'scrolling-fixes'Daniel Eklöf2021-10-27-14/+21
|\
| * input: clean up mouse scroll handlingDaniel Eklöf2021-10-27-14/+21
|/ | | | | | | | | | * Allow scrolling on the normal (non-alt) screen, when application is grabbing the mouse (when user presses Shift). * Use term_mouse_grabbed() instead of explicitly checking for MOUSE_NONE tracking. * Remove mouse tracking check from cmd_scrollback_{up,down}. Caller is expected to have done the check. * Don’t scroll down on mouse wheel tilt events.
* Merge branch 'noop-key-binding'Daniel Eklöf2021-10-27-5/+18
|\ | | | | | | Closes #765
| * config: add [key-bindings].noop actionDaniel Eklöf2021-10-27-5/+18
|/ | | | | | | Key combinations assigned to this action will not be sent to the application. Closes #765
* doc: fix a typoDaniel Martí2021-10-23-1/+1
| | | | | I presume "can be used here as well" was meant. Spotted it while reading the docs.
* ci: also build release branchesDaniel Eklöf2021-10-23-4/+16
|
* foot/client: version info: add +/- assertionsDaniel Eklöf2021-10-23-4/+15
|
* term: bell: simplify if-statement - we don’t need two nested levelsDaniel Eklöf2021-10-22-11/+9
|
* doc: foot.5: we support “real” urgency through XDG activationDaniel Eklöf2021-10-22-9/+6
| | | | | | Foot has been signaling window urgency through the (new) “XDG activation” protocol for some time now. Update the documentation to reflect this.
* changeloge: style fixesDaniel Eklöf2021-10-22-5/+6
|
* term: ignore window title updates if title is un-changedDaniel Eklöf2021-10-22-0/+5
|
* Merge branch 'update-input-serial-on-each-keypress'Daniel Eklöf2021-10-21-0/+4
|\ | | | | | | Closes #753
| * input: update serial on each key press/release and mouse buttonDaniel Eklöf2021-10-20-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The serial is used when copying/pasting data from the clipboard. Up until now, we’ve used the serial from the keyboard/mouse enter events. This works in most cases, but breaks in the following example: $ wl-copy WLCOPY /* Ctrl+Shift+v works fine (pastes "WLCOPY") */ $ printf "\033]52;c;eHl6\a" /* Ctrl+Shift+v pastes "WLCOPY" instead of "xyz" */ Shifting focus away and then back to the foot window, and re-executing the printf works, suggesting the “enter” serial is no longer valid after another process(?) has copied something to the clipboard. Updating the serial on key press/release (and the corresponding mouse serial on mouse button events) seems to fix this. I’ve also tested that “normal” copy/paste operations, within the same foot instance, and between foot and other applications, are still working. In at least river (wlroots based), and GNOME/mutter. Closes #753
* | config: error out on invalid grapheme-width-method valuesDaniel Eklöf2021-10-21-0/+7
| |
* | render: tab -> spacesDaniel Eklöf2021-10-20-6/+6
|/
* Merge branch 'parse-rgb-strlen'Daniel Eklöf2021-10-20-4/+5
|\
| * osc: use STRLEN() macro to make parse_rgb() more self-documentingCraig Barnes2021-10-20-4/+5
| | | | | | | | | | The empty literals in the macro are to ensure the argument itself is a string literal, so it can't be used on anything else.
* | Merge branch 'warn-not-to-use-single-letter-key-bindings-in-url-mode'Daniel Eklöf2021-10-20-1/+10
|\ \
| * | doc: foot.5: warn user to be careful with single letter key bindings in url modeDaniel Eklöf2021-10-20-1/+10
| | |
* | | Merge branch 'osc-bel'Daniel Eklöf2021-10-20-6/+19
|\ \ \ | |/ / |/| |
| * | osc: use BEL terminator in OSC replies to BEL-terminated OSC queriesCraig Barnes2021-10-20-6/+19
| |/ | | | | | | | | | | | | | | | | This matches the documented (and observed) behavior in xterm: > XTerm accepts either BEL or ST for terminating OSC sequences, and > when returning information, uses the same terminator used in a query -- https://invisible-island.net/xterm/ctlseqs/ctlseqs.html#h3-Operating-System-Commands
* | Merge branch 'clipboard-convert-carriage-ret-to-new-line'Daniel Eklöf2021-10-20-0/+7
|\ \ | |/ |/| | | Closes #752
| * selection: convert \r -> \n when reading clipboard dataDaniel Eklöf2021-10-19-0/+7
|/ | | | | | | | This fixes an issue where pasting (using e.g. OSC-52) in client applications that doesn’t do this conversion themselves, like tmux, doesn’t work. Closes #752
* Merge branch 'nordiq'Daniel Eklöf2021-10-18-0/+25
|\
| * add Nordiq themearmin2021-10-18-0/+25
|/
* pgo: full-inner: add --override tweak.grapheme-shaping=noDaniel Eklöf2021-10-16-1/+8
| | | | | | | | | Grapheme shaping is now enabled by default in foot. However, when generating the profiling data in PGO builds, this results in skewed optimizations. The end result is worse benchmark results regardless of whether grapheme-shaping is enabled or not (when running the benchmarks).
* doc: foot.1: reporting bugs: move “search existing issues”Daniel Eklöf2021-10-16-2/+4
|
* readme: reporting bugs: move “search existing issues”Daniel Eklöf2021-10-16-2/+4
|
* doc: foot.1: reporting bugs: re-order and tweakDaniel Eklöf2021-10-16-5/+8
| | | | | * Tell people to search existing bugs first (open + closed) * Re-order instructions: most important ones first
* readme: reporting bugs: re-order and tweakDaniel Eklöf2021-10-16-6/+8
| | | | | * Tell people to search existing bugs first (open + closed) * Re-order instructions: most important ones first
* Merge branch 'enable-grapheme-shaping-by-default'Daniel Eklöf2021-10-16-16/+17
|\
| * config: enable grapheme shaping by defaultDaniel Eklöf2021-10-13-16/+17
|/ | | | | | | | This changes the default value of tweak.grapheme-shaping to “yes”, thus enabling grapheme shaping by default. It also changes the default value of tweak.grapheme-width-method to “wcswidth”, for maximum compatibility with terminal applications.
* Merge branch 'generate-version-handle-repo-not-having-any-tags'Daniel Eklöf2021-10-11-6/+18
|\
| * ci: run “foot --version” and “footclient --version” after each buildDaniel Eklöf2021-10-11-0/+10
| |
| * generate-version: handle git repo not having any tagsDaniel Eklöf2021-10-11-6/+8
| |
* | doc: ctlseq: improve text formatting in CSI and SGR sectionsCraig Barnes2021-10-11-10/+10
| | | | | | | | | | | | | | | | Using underlining for parameters allows the angle brackets to be removed while still keeping a visual separation between literals and parameters. The removes any uncertainty about whether the angle brackets are part of the sequence or not. It also mirrors the formatting used further down in the document.
* | doc: ctlseq: fix SGR table headingCraig Barnes2021-10-11-2/+2
|/ | | | | The other tables that contain only params also use the heading "Parameter" rather than "Sequence".
* ci: limit builds to the master branch (and pulls targeting it)Daniel Eklöf2021-10-10-0/+4
|
* ci: codeberg CIDaniel Eklöf2021-10-10-0/+88
|