andlabs-ui/todo.md

90 lines
6.4 KiB
Markdown
Raw Normal View History

2014-02-12 09:58:05 -06:00
so I don't forget:
2014-03-06 18:47:25 -06:00
- Window.SizeToFit() or WIndow.OptimalSize() (use: `Window.SetOptimalSize())`) for sizing a window to the control's interest
2014-02-12 09:58:05 -06:00
- Control.Show()/Control.Hide()
2014-02-13 16:04:57 -06:00
- Groupbox
2014-03-06 19:12:05 -06:00
- see if we really need to track errors on a lot of places that report errors
- Window.Show()/Window.Hide() report errors due to UpdateWindow(), which can fail, but that is only called when the window is first opened: split that functionality out
2014-03-07 19:47:07 -06:00
- it appears GTK+ and Cocoa both either don't provide a convenient way to grab errors or you're not supposed to; I assume you're supposed to just assume everything works... but on Windows we check errors for functions that return errors, and there's no guarantee that only certian errors will be returned...
- character-limited entry fields, numeric entry fields, multiline entry fields
2014-02-14 14:00:59 -06:00
- possible rename of LineEdit?
- especially for password fields - NewPasswordEntry()?
2014-03-03 14:47:12 -06:00
- [Windows, Mac OS X] should ListBox have a border style?
- [Windows] a different border on LineEdits?
2014-03-06 19:12:05 -06:00
- padding and spacing in Stack
- change Listbox constructor so that there's a separate constructor for each variant, rather than passing in parameters
2014-02-14 19:42:26 -06:00
- allow Combobox to have initial settings
2014-02-15 13:26:37 -06:00
- Combobox and Listbox insertions and deletions should allow bulk (...string)
- Combobox/Listbox.DeleteAll
- Combobox/Listbox.Select (with Listbox.Select allowing bulk)
- Listbox.SelectAll
- have Listbox.Delete() panic on invalid index; it does not yet due to the Mac OS X signaling issue mentioned under "super important"
2014-03-06 18:47:25 -06:00
- same for other methods that take indices, like the Stack and Grid stretchy methods
2014-02-17 01:55:59 -06:00
- make the Windows implementation of message boxes run on uitask
- ensure MsgBoxError can run if initialization failed if things change ever
- should Labels be selectable?
- should message box text be selectable on all platforms or only on those that make it the default?
- Listbox/Combobox.Index(n)
2014-03-07 14:08:01 -06:00
- Index(n) is the name used by reflect.Value; use a different one?
2014-02-12 09:54:58 -06:00
important things:
2014-03-01 14:23:49 -06:00
- because the main event loop is not called if initialization fails, it is presently impossible for MsgBoxError() to work if UI initialization fails; this basically means we cannot allow initializiation to fail on Mac OS X if we want to be able to report UI init failures to the user with one
- figure out where to auto-place windows in Cocoa (also window coordinates are still not flipped properly so (0,0) on screen is the bottom-left)
2014-03-06 18:47:25 -06:00
- also provide a method to center windows; Cocoa provides one for us but
2014-03-01 21:31:28 -06:00
- I think Cocoa NSButton text is not vertically aligned properly...?
- and listbox item text is too low?
2014-03-06 18:47:25 -06:00
- NSPopUpButton does allow no initial selection ([b setSelectedIndex:-1]); use it
2014-03-07 17:33:51 -06:00
- need to use it /after/ adding initial items, otherwise it won't work
2014-03-06 18:47:25 -06:00
- find out if I can do the same with the ListBoxes
- NSComboBox scans the entered text to see if it matches one of the items and returns the index of that item if it does; find out how to suppress this so that it returns -1 unless the item was chosen from the list (like the other platforms)
- some Cocoa controls don't seem to resize correctly: Buttons have space around the edges and don't satisfy stretchiness
2014-03-06 18:47:25 -06:00
- there's no GTK+ or Cocoa error handling whatsoever; we need to figure out how it works
- I know how Cocoa error handling works: it uses Cocoa exceptions; need to figure out how to catch and handle them somehow
- make sure GTK+ documentation version point differences (x in 4.3.x) don't matter
2014-02-24 13:56:11 -06:00
- button sizes and LineEdit sizes on Windows seem too big; Comboboxes have margins
- Cocoa has similar margining issues (like on Comboboxes)
2014-02-24 13:56:11 -06:00
- sometimes the size of the drop-down part of a Combobox becomes 0 or 1 or some other impossibly small value on Windows
2014-03-01 03:46:42 -06:00
- make gcc (Unix)/clang (Mac OS X) pedantic about warnings/errors; also -Werror
- make sure scrollbars in Listbox work identically on all platforms (specifically the existence and autohiding of both horizontal and vertical scrollbars)
- pin down this behavior; also note non-editability
2014-03-06 21:01:44 -06:00
- GTK+ windows cannot be resized smaller than their controls's current sizes in their current positions; find out how to overrule that so they can be freely resized
super ultra important things:
- the windows build appears to be unstable:
2014-03-08 16:30:34 -06:00
- 64-bit crashes in malloc in wine with heap corruption warnings aplenty during DLL loading; in windows 7 it works fine
- 32-bit: it works, but if I save the class name converted to UTF-16 beforehand, wine indicates that the class name is replaced with the window title, so something there is wrong...
2014-02-15 13:03:46 -06:00
- handle in-library panics (internal errors) by reporting them to the user
- david wendt is telling me he's getting frequent crashes on his end with the GTK+ amd64 build...
TODO re-evaluate; I think I fixed them all ages ago now
- occasionally I get
panic: error sending message to message loop to call function: Invalid thread ID.
when starting up the windows/386 build; race in ui()/msgloop()?
2014-03-06 18:47:25 -06:00
- happens the first time I run a new build in wine; also if my computer is running too slowly when running in wine
2014-03-07 14:08:01 -06:00
- GTK+: stderr is flooded with
```
(test:17575): Gdk-CRITICAL **: gdk_device_ungrab: assertion 'GDK_IS_DEVICE (device)' failed
(test:17575): Gtk-CRITICAL **: gtk_device_grab_remove: assertion 'GDK_IS_DEVICE (device)' failed
```
figure out why
2014-03-08 16:30:34 -06:00
- Mac OS X 32-bit builds don't call runtime.sighandler() on a Cocoa exception like the 64-bit builds do
2014-03-08 20:41:30 -06:00
- does array index out of bounds on 32-bit Cocoa not generate a SIGTRAP like invalid selector does? need to experiment... or is something clobbering SIGTRAP? I didn't catch it in my tests here...
2014-02-12 10:34:10 -06:00
important things:
2014-02-15 12:09:51 -06:00
- make specific wording in documentation consistent (make/create, etc.)
- document minor details like wha thappens on specific events so that they are guaranteed to work the same on all platforms (are there any left?)
2014-03-07 17:33:51 -06:00
- make passing of parameters and type conversions of parameters to uitask on Windows consistent: explicit _WPARAM(xxx)/_LPARAM(xxx)/uintptr(xxx), for example
- do this for type signatures in exported functions: (err error) or just error?
2014-03-08 16:30:34 -06:00
- do this for the names of GTK+ helper functions (gtkXXX or gXXX)
2014-02-12 10:34:10 -06:00
2014-02-12 09:54:58 -06:00
far off:
- localization
2014-02-12 10:34:36 -06:00
- strip unused constants from the Windows files
- combine more Windows files; rename some?
2014-03-03 14:37:02 -06:00
- tab stops
2014-02-13 16:04:57 -06:00
maybe:
- rename Stack to Box?
2014-02-15 13:26:37 -06:00
- make Combobox and Listbox satisfy sort.Interface?
2014-02-24 23:34:21 -06:00
- indeterminate progress bars (not supported on Windows 2000)
- should a noneditable Combobox be allowed to return to unselected mode by the user?