* don't delete general refs from jobs that we cancel
though we still disconnect the refs if we can
* get job remove working in all cases
we apparently need to manually handle DestroyBuilding jobs
everything else we should let cancel_job handle
* update changelog
* ensure job items are disassociated from the job
when the job is removed. the new df-provided ``cancel_job()`` doesn't do
this for us whereas the old custom implementation did.
ref: #2028
* remove trailing whitespace
* Clean up general refs before removing job
Because the game method doesn't do it itself
* Fix typo in var name
* clean up code
* update changelog
This contains a fix on Windows that checks for the executable bit according to
Git, instead of according to Windows, which reduces false-positive
"check-executables-have-shebangs" failures.
* use new focus subsystem in widgets.EditField
* always eat the enter key if we have an on_submit
* add modal attribute
* give EditFields a default height of 1
so they can be autoarranged
* implement keyboard focus subsystem
* Fix error in focus group combining
* documentation for the inputToSubviews decision
* modify unit tests to catch that last bug
* fix wrong `Label.frame_body.x2` value
`update_scroll_inset` might change `frame_inset`, i.e. we need to `computeFrame` with the new values.
* add tests for Label
* add missing `local`, remove code in comments
* move `TestFramedScreen` outside test functions
* test jsoncpp upgrade
* use new json library target name
* don't remap the output dirs
* undo warnings at the source
* set new defaults for jsoncpp
* fix typo in new options
* fix signed comparison mismatch warning
* address random(?) compile failures
saying our std::atomic is not initialized in Debug.cpp
* update example init files
replace onLoad.init-example with documentation on how to create scripts that run on world/map load
it was confusing to show it being loaded with sc-script since it gets autorun anyway if it is just named properly
* update changelog
* add quickfort keybinding
* move standard tweaks from dreamfort init to main
* Added (chain) for [CHAIN_METAL_TEXT] armours in gui/materials.lua used by gui/create-item-- again (oops)
* Added customRawData utility
* Oops, whitespace
* Revised rawStringsFieldNames
* Dialed down on lua trickery and fixed wrongly formatted changelog entry
* Fixed changelog in wrong place and made customRawData a proper module
* Fixed not caching not-present tags, revised examples and fixed error
* Fixed whitespace. Changing settings in editor!
* customRawData docs
* Added getCreatureTag for respecting caste tags, "fixed" bizarre caching error (quotes because I don't even know what was causing it) and updated docs
* Added line limiting for docs, I guess
* Added missing string convert argument
* docs indent fix, code block fix, and revision
* Major revision
* gdi, docs error
* Another? But... huh.
* ...
* Made requested changes
* Whoops
* Rearrange docs lines
* Followed example, should fix linter issues
* fix typo. linted offline this time......
* Make it so that last instance of tag is what is read from
* Added requested change
* eventful key change
* i to lenArgs
* change eventful key
* add test for broken caste selection
* Major redesign
* tags --> tokens
* Added plant growth behaviour and did some requested changes
* More error handling
* fix docs
* Added basic error suppression
* Docs clarification.
* Docs registering example and fix error
* Strip errors on frame after onWorldLoad, not on map load
* Revert "Strip errors on frame after onWorldLoad, not on map load"
This reverts commit e20a0ef8d3743f79d961077f46910b77b16f36b9.
* Revert "Docs registering example and fix error"
This reverts commit 9c848c54c3f84e0ecc1dc421137c8a8b4a52280d.
* Revert "Docs clarification."
This reverts commit 6b4b6a1aa40c50398504f37ecf1ff0f93d6459b1.
* Revert "Added basic error suppression"
This reverts commit d11cb1438cf1e56ff700469e944f0b9af64651d7.
* Use more eventful key more consistent with other files
* use onStateChange instead of eventful and remove redundant utils require
* Code review stuff
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update docs/Lua API.rst
committing a suggestion
Co-authored-by: Alan <lethosor@users.noreply.github.com>
* Prepend examples with DFHACK_
* Remove unused parameters
* Use new ensure_key global
* Named a couple of unnamed arguments (untested)
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Alan <lethosor@users.noreply.github.com>
* WIP: add scroll icons to Label widget
It's an opt-out. The icons are rendered in the right-most column of the 1st and last row. They are only rendered when text can actually be scrolled in the corresponding direction.
WIP: Currently, the icons might overlay text characters, there is no mechanism preventing it
* gui.lua: expose the `parse_inset()` function
* refactor Label's scroll icon code
* since `render_scroll_icons` only works with a label, it's now a class function
* `update_scroll_inset` ensures `frame_inset.r` or `.l` is at least 1, according to `show_scroll_icons`
* `show_scroll_icons` has 4 possible values: `false` for no icons, `left` for icons on the first column on the left (also ensuring `frame_inset.l >= 1`), `right` - last column on the right, `DEFAULT_NIL` - same as `right` if text height greater than `frame_body.height`, else same as `false`.
* make `render_scroll_icons` always draw icons
The check now happens in `onRenderFrame`
* draw frame's background
calling `Label.super.onRenderFrame(self, dc, rect)` makes frame's background invisible for some reason
* remove trailing spaces
* fix scroll icons placed far above/below text
With `Label.frame_inset = 1` the text could be vertically centered with plenty of space below and above,
but not all rendered. Before this change, the scroll icons would be at the very top and bottom of the frame
instead of near the first and last rendered text line.
* always `update_scroll_inset` to react to resized window
* draw scroll icons next to text
* update `Lua API.rst` with new `Label` parameters
* move comment separator up
This way every scroll related parameter is in one group
* list default values for new parameters in docs
* add missing description of `Label.scroll_keys`
In cases where the cached `GIT_EXECUTABLE` does not exist, `git-describe.cmake`
would fail silently, resulting in a `git-describe.h` with missing definitions
that caused `DFHackVersion.cpp` to fail to compile.
This change produces a more descriptive error earlier in the build process.
* route 'Loading script...' messages through the debug logging framework
* they are now controllable via debugfilter. to turn off, add this to your dfhack.init file: debugfilter set Warning core script
* clarify debug logging docs
* prevent DEBUG messages from being compiled out of the binary
* recolor INFO messages so they blend into existing console output
* add configuration interface for log message header elements so they are individually configurable. all default to off
* Add functions reverse-engineered from ambushing unit code
* Fix whitespace
* Fix debug_showambush check
* Remove getOuterContainerRef from Lua API
Don't think this works properly without allocating a new specific_ref. More trouble that it's worth.
* Fixed tile visibility check
* I don't think gamemode or gametype are ever NULL
* Minor tweaks to documentation
* Reimplement getOuterContainerRef for Lua; fix some comments
* Update Units.cpp and changelog
* Update Units.cpp
* Update changelog.txt
My understanding is that iterators (.begin()/end()) are slightly faster with
pre-increment instead of post. Something about avoiding a copy if I recall correctly.
I replaced all `([:alpha:]+)\+\+` with `++$1` so non-iterators were switched as well - luckily there is no impact to pre/post aside from with iterators.
- handler freq now works on an individual handler basis (set to 0 to fire as frequently as possible)
- tick events no longer need to be re-registered after every eventful tick
settled on `find()` since the string method of the same name has the
same semantics. other options: `str_find()` since it fails on non-string
or `match()` (but that might imply that the pattern needs to match the
entire string).
this fixes the issue where the alt modstate was getting stuck on systems
that don't send standard keyup events after alt-tab. for example, in KDE
Plasma (on Gentoo, at least), the keyup event when alt is released after
alt-tab contains an incorrect keycode (NUMLOCK instead of L_ALT), which
was preventing us from correctly clearing the alt modstate.
`dismiss_on_submit = false` makes it easier to implement toggleable options, where we don't want the list to disappear on selection. Like work order conditions' trait selection.
lethosor
DFHack-Urist via GitHub Actions
Quietust
Timur Kelman
Myk
quarque2
Josh Cooper
Ryan Williams
Tim Siegel
Tachytaenius
Josh Cooper
Alan
Kelly Kinkade