summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2014-04-23More future plans.Pietro Gagliardi
2014-04-22More future plans.Pietro Gagliardi
2014-04-21More future plans.Pietro Gagliardi
2014-04-20More TODOs (because I can't think right now).Pietro Gagliardi
2014-04-20More future plans and TODOs.Pietro Gagliardi
2014-04-19More future plans.Pietro Gagliardi
2014-04-18More future plans.Pietro Gagliardi
2014-04-18More future plans. This is really turning into a The Old New Thing articles ↵Pietro Gagliardi
list...
2014-04-17More future plans.Pietro Gagliardi
2014-04-16More future plans (and some link gathering, too).Pietro Gagliardi
2014-04-16More future plans.Pietro Gagliardi
2014-04-15Settled window sizing for GTK+; moved the respective TODO to future plans.Pietro Gagliardi
2014-04-14Added click-to-focus to GTK+ Areas and documented that click-to-focus ↵Pietro Gagliardi
happens; also TODOs related to that.
2014-04-14Removed the Windows Area resizing glitches TODOs... as they seemed to have ↵Pietro Gagliardi
gone away on their own (GDI+ issue that no longer applies since I dropped GDI+?).
2014-04-14More TODOs.Pietro Gagliardi
2014-04-14More TODOs.Pietro Gagliardi
2014-04-14Clarified the previous commit.Pietro Gagliardi
2014-04-14Ensured that pressing Modifiers+Key/ExtKey does not produce an event with ↵Pietro Gagliardi
just the Key/ExtKey.
2014-04-13Wrote non-working code to get Windows to have Window sizes not count window ↵Pietro Gagliardi
decoration. Technically it does work, but it doesn't seem to fix Area scrollbars... which reminds me, the default configuration for the keyboard test exhibits the "last pixel" but mentioned in one of the TODOs in area_windows.go on both scrollbars, so let's fix that first anyway. I'll move the TODO for window decoration to the future plans after I get a definite answer about GTK+'s behavior.
2014-04-13Fixed Mac OS X so that window sizes do not include the titlebar and frame.Pietro Gagliardi
2014-04-13Fixed Area test time label weirdness by making Labels truncate their text ↵Pietro Gagliardi
instead of word-wrapping on all platforms. This doesn't explain GTK+/Wayland, alas.
2014-04-13Fixed weird Area time label truncation on Windows; more TODOs and future ↵Pietro Gagliardi
plans related to that.
2014-04-13Removed winbuttonsize/; we're not going to use that for a while (the sizing ↵Pietro Gagliardi
we have now is fine; not optimal, but it works, for the most part). (We can always pull it back later.)
2014-04-13Renamed unmigrated/ to experiments/ as I have now migrated everything and ↵Pietro Gagliardi
that folder now contains experimental code.
2014-04-13Migrated cursors_windows.go and icons_windows.go.Pietro Gagliardi
2014-04-13Removed unmigrated/main_windows.go; that test program has long since been ↵Pietro Gagliardi
unneeded.
2014-04-13Rolled back the Windows horizontally scrolling Listbox changes because I ↵Pietro Gagliardi
can't seem to get it to work on Mac OS X (and keeping that code there is inefficient); removed the guarantee of horizontal scrollbars in Listbox and made that a future plan. Both the Windows and the non-working Mac OS X code is in unmigrated/hscrolllistbox.go (and the unmigrated/ folder will be renamed shortly).
2014-04-13Consolidated NSScrollView code on Mac OS X to be all in one place.Pietro Gagliardi
2014-04-13More shuffling of TODOs and future plans.Pietro Gagliardi
2014-04-13Added horizontal scrollbars to Listboxes on Windows. Also fixed typos and ↵Pietro Gagliardi
bugs in Windows sysData.selectedTexts().
2014-04-13Pinned down horizontal scrollbar behavior on Listboxes; also reworded the ↵Pietro Gagliardi
Area scrollbar behavior sentence a bit to match. Now to actually implement the changes...
2014-04-13More TODO/future plan shuffling.Pietro Gagliardi
2014-04-13More TODOs and future plans. Also finally removed the TODO about incorrect ↵Pietro Gagliardi
Area redraw on Windows.
2014-04-12More TODOs and future plans.Pietro Gagliardi
2014-04-12Fixed Listboxes on Mac OS X having an initial selection.Pietro Gagliardi
2014-04-12Fixed Comboboxes on Mac OS X having an initial selection. This also lays the ↵Pietro Gagliardi
groundwork for adding Combobox/Listbox.Select() as a public function...
2014-04-12Verified that the Listbox item text is vertically positioned properly on Mac ↵Pietro Gagliardi
OS X.
2014-04-12Removed the GDI+ dependency on Windows since we no longer use it.Pietro Gagliardi
2014-04-12Updated README to reflect the new Area image.RGBA changes.Pietro Gagliardi
2014-04-12Made Areas on Windows flicker-free.Pietro Gagliardi
2014-04-12Fixed ARGB rendering on Windows. Now to make it flicker-free...Pietro Gagliardi
2014-04-12Documented the thread safety of AreaHandler and Area.SetSize().Pietro Gagliardi
2014-04-12Switched to ARGB drawing in Areas on GTK+.Pietro Gagliardi
2014-04-11Rewrote Area code on Windows to use alpha-premultiplied RGB and only use GDI ↵Pietro Gagliardi
functions to do it... and it doesn't quite work right yet.
2014-04-11Added a general-purpose function to produce a native-endian ARGB ↵Pietro Gagliardi
representation of the image. I thought cairo image surfaces would work, but they're not designed for reading in images, the cairo functions for reading image data were introduced in the next version (1.12), and GDK-Pixbuf only does non-alpha-premultiplied... This can be used on Windows though.
2014-04-11Switched to alpha-premultiplied image data for Areas on Mac OS X. This is ↵Pietro Gagliardi
the easiest change; it involves removing one bit flag from the NSBitmapImageRep constructor (the other changes are indentation changes).
2014-04-11Changed from using image.NRGBA to image.RGBA (premultiplied alpha) in ↵Pietro Gagliardi
Area.Paint() since it'll be easier to work with on Windows after applying what Treeki/Ninjifox suggested I do instead of what I am doing now.
2014-04-11Added CS_HREDRAW and CS_VREDRAW to our Area on Windows, just to be safe.Pietro Gagliardi
2014-04-11Attempted to fix the Windows transparency issues; we're not quite there... ↵Pietro Gagliardi
flicker...
2014-04-10Removed TODO about Windows -1 conversions, as the Go spec specifically says ↵Pietro Gagliardi
that signed values are sign-extended to the full size of the type to convert to (and even shows an example of this), so the code we have was correct, not just working out of luck.