defaults.persistent.lua
We were pcall'ing the parsing, but not the execution...
The funky Lua syntax quirks means that it is possible to pass the former
but not the latter ;).
Fix#9700, de-facto regression since #9546
Get rid of the doc & seqtext fields, as they are not actually used (nor
are they particularly useful, the event handler's name should be pretty
self-explanatory).
Also, tweak the key_events documentation to highlight the quirks of the
API, especially as far as array nesting is involved...
Random drive-by cleanup of the declarations of key_events & ges_events
to re-use the existing instance object (now that we know they're sane
;p) for tables with a single member (less GC pressure).
* Added a new plugin external-keyboard. It listens to USB events. When keyboard is plugged in or plugged out, it updates device and input configuration accordingly.
* Added new fake events UsbDevicePlugIn and UsbDevicePlugOut that are emitted when a device is connected to a book reader that plays the role of USB host. The usage of the existing events UsbPlugIn and UsbPlugOut has not changed - they are used when a reader is connected to a host. The koreader-base has a related PR for those events.
* Did a small refactoring of initialization for the modules FocusManager and InputText. They check device keyboard capabilities on their when the module is first loaded and store it. Some of the initialization code has been extracted into functions, so that we can re-initialize them when keyboard is (dis)connected.
* Initial implementation centered around text input, and tested with USB keyboards on devices with OTG support.
* Said OTG shenanigans are so far supported on devices with debugfs & the chipidea driver, or sunxi devices.
If you've ever enabled the main loop debugging, you'll know that
actually dumping the full window stack was *hilarious*.
Just print table counts, it's often good enough to debug what's
happening in the exceedingly rare cases you need this ;).
Also, it'll actually be readable, unlike the previous insanity ^^.
This PR makes zh_keyboard's number pad symbol mode instead of shift mode, so that numbers are shown when the keyboard is called expecting numeral input.
* Iterate over varargs directly via select if possible
* Use table.pack otherwise (https://github.com/koreader/koreader-base/pull/1535).
* This allows us to simplify a few Logger calls, as logger now handles nil values.
This PR adds support for a Thai keyboard. Layout mostly copied from my phone, with difference in diacritics placements: phone has only one key for all of them, while here we have a couple more. Swipe in different directions needed for inputing them.
The ultimate goal is for toast widgets (i.e., Notification when flagged as such) to:
* Not stop event propagation
* Close themselves when the event was emitted by user input.
Instead of doing event filtering in UIManager, we simply overload the onGesture & onKey* handlers in Notification to do just that, and just make sure UIManager will *send* those events to toasts, but without affecting the usual semantics of top widget selection and event propagation (which is as simple as just calling handleEvent on them unchecked ;p).
Thanks to @poire-z for the brainstorming in https://github.com/koreader/koreader/issues/9594 ;).
This also happens to fix a bug in which we might have looped on the top widget twice, because of an array vs. hash mishap ;).
specifically requested.
Using a custom message would effectively prevent the event message from
showing up, which is... bad.
(In fact, I'm half of the mind to get rid of the option to hide the
event messages, because it's a goddamned terrible idea).
Also, make sure references are actually dropped,
no matter how the widget is closed, by relying on onCloseWidget ;).
Enable the "long-press-on-close" trick on the actual Close button,
too, instead of only on the title bar's cross.
Fix: https://github.com/koreader/koreader/pull/9586#issuecomment-1272332275
* Persist: support serpent, and use by default over dump (as we assume consistency > readability in Persist).
* Logger/Dbg: Use serpent instead of dump to dump tables (it's slightly more compact, honors __tostring, and will tag tables with their ref, which can come in handy when debugging).
* Dbg: Don't duplicate Logger's log function, just use it directly.
* Fontlist/ConfigDialog: Use serpent for the debug dump.
* Call `os.setlocale(C, "numeric")` on startup instead of peppering it around dump calls. It's process-wide, so it didn't make much sense.
* Trapper: Use LuaJIT's serde facilities instead of dump. They're more reliable in the face of funky input, much faster, and in this case, the data never makes it to human eyes, so a human-readable format didn't gain us anything.
Basically:
* Use `extend` for class definitions
* Use `new` for object instantiations
That includes some minor code cleanups along the way:
* Updated `Widget`'s docs to make the semantics clearer.
* Removed `should_restrict_JIT` (it's been dead code since https://github.com/koreader/android-luajit-launcher/pull/283)
* Minor refactoring of LuaSettings/LuaData/LuaDefaults/DocSettings to behave (mostly, they are instantiated via `open` instead of `new`) like everything else and handle inheritance properly (i.e., DocSettings is now a proper LuaSettings subclass).
* Default to `WidgetContainer` instead of `InputContainer` for stuff that doesn't actually setup key/gesture events.
* Ditto for explicit `*Listener` only classes, make sure they're based on `EventListener` instead of something uselessly fancier.
* Unless absolutely necessary, do not store references in class objects, ever; only values. Instead, always store references in instances, to avoid both sneaky inheritance issues, and sneaky GC pinning of stale references.
* ReaderUI: Fix one such issue with its `active_widgets` array, with critical implications, as it essentially pinned *all* of ReaderUI's modules, including their reference to the `Document` instance (i.e., that was a big-ass leak).
* Terminal: Make sure the shell is killed on plugin teardown.
* InputText: Fix Home/End/Del physical keys to behave sensibly.
* InputContainer/WidgetContainer: If necessary, compute self.dimen at paintTo time (previously, only InputContainers did, which might have had something to do with random widgets unconcerned about input using it as a baseclass instead of WidgetContainer...).
* OverlapGroup: Compute self.dimen at *init* time, because for some reason it needs to do that, but do it directly in OverlapGroup instead of going through a weird WidgetContainer method that it was the sole user of.
* ReaderCropping: Under no circumstances should a Document instance member (here, self.bbox) risk being `nil`ed!
* Kobo: Minor code cleanups.
And do so only for the active layouts.
This prevents loading a potentially large amount of data without even having navigated to said menu (or having the layout enabled).