On Monza, the lift frame can report these for *multiple contacts* without an id or a slot, which is super-duper invalid -_-".
(And we have ABS_MT_POSITION_X & ABS_MT_POSITION_Y for the "saner" frames, plus these devices generally don't support pens, so we don't make use of pressure).
(Also, switched to a dedicated handler like for the phoenix quirks, to avoid bogging down the standard handler with hacks for broken drivers).
Fix#11910
I don't even remember how badly things broke (at least on old devices) without it, despite it making absolutely no sense at all (state-extended just flips a global that dictates whether some things get flagged as wakeup sources or not).
So, don't rock the boat too much: we don't remove it, but instead of using a sleep, we use a task deadline instead, which ensures we'll keep processing input events in the right order in the meantime. We'll already have neutered input by this point, so we'll only process power events anyway.
That means that the only iffy things are potentially *when* and *where* we have to potentially cancel that task. Resume makes sense, of course, and we log an info message to make the log flow clear; but we also do so in suspend... just in case. With a warning log because that probably indicates something fishy went on.
Also cleanup the comments while I'm there, and actually rewrite the wakeup_count stuff properly so it could actually theoretically be used if ntx kernels were actually reliable. Spoiler alert: they're not, this is still horribly broken on at least Mk < 7. Works just fine on a Forma, though, so, yay.
Fix#12325
* UIManager: Let the fb backend deal with Kaleido wfm promotion. This fixes a number of quirks that poisoned the refresh queue with spurious full-screen refreshes. See https://github.com/koreader/koreader-base/pull/1865 for more details.
* This also means we now disable Kaleido waveform modes when color rendering is disabled (remember to trash your thumbnail cache if you don't want to mix color w/ grayscale thumbnails, though).
* UIManager: Merge refreshes that share an edge, because that was driving me nuts (and would have most likely been merged by the kernel anyway). A perfect test-case is the FM, which trips two separate refreshes because of its title bar.
* ReaderFlipping: Use sensible dimensions, so that we only refresh the icon's region.
* ReaderBookmark: Only refresh the dogear instead of the whole page when toggling bookmarks.
* NetworkSetting: Make it a real boy, so it consistently refreshes properly on dismiss instead of relying on UIManager saving the day.
* Kobo: Aggressively prevent *both* suspend & standby while MTK devices are plugged-in, as both will horribly implode the kernel (we previously only prevent standby while charging).
* Kobo: Switch to 8bpp on B&W MTK devices (or when color rendering is disabled on Kaleido panels).
Devices with a single target might want to specify it in `Device.ota_model`
Devices with multiple targets want to override the function or to specify `ota_model` variants for each target.
Switch to a new `input.fdopen` API & wrapper so we can keep the fds opened by `fbink_input_scan` instead of closing them to re-open them right after that...
This should hopefully help on racy zForce devices that attempt to handle power management when opening/closing the device. We know this sometimes horribly fail to re-activate the IR grid (c.f., our manual activation on resume), but this apparently could also happen here (re: #11844) because of the quick succession of open->close->open.
* Kobo: Drop a bunch of if ladder crap and switch to auto-detection of input devices via fbink_input
* Kindle: Drop an even larger bundle of crap to do the same ;p. (re: #11392)
* ExternalKeyboard: Switch to fbink_input to whitelist keyboards instead of the manual parsing of caps via its FindKeyboard class
* Input: Extended open/close wrappers to handle logging & tracking of dupe open/close calls.
* Support the Clara B&W, Clara Colour & Libra Colour
* Enable HW dithering on *all* the Kobo MTK devices
* Enforce 32bpp instead of 8bpp for Kobo devices with a color panel (the driver doesn't actually support 8bpp anyway)
* Enable standby support on MTK (whenever possible, i.e., not when plugged in, as that is horribly, horribly broken).
* Enforce the dedicated "color" waveform mode for image content in ScreenSaver, ImageViewer & Reader.
* Fix charging LED support on MTK
* Tweak the frontlight ramp on MTK + LM3630 so that it actually ramps smoothly
@hius07 mentioned something to that effect a while back, makes sense.
Unlike the set of checkmarks in the dev settings, this one flips both
debug + verbose at once, *and* asks for a restart for framebuffer's
sake.
Also update the "Report a bug" spiel to request verbose debug logs.
This reverts commit e7e0d2edb6.
Whatever's actually going on with nightmode, this isn't it.
This was independently confirmed to behave as expected,
and the issue that prompted this was *not* fixed by this commit.
* Enforce a minimal standby timer for the first standby after a resume.
* On Kobo, sleep a bit before standby.
This aims to alleviate race conditions causing visible refresh glitches on sunxi, especially when using an extremely low standby timer (i.e., below the defaults).
As early as turnOnWifi.
Implement it on hasWifiManager platforms, preventing useless
connectivity checks to run when they're obviously never going to work
because you're out of range of your AP.
Also implemented a flag to notify the backend if the connection attempt
was interactive or not.
Right now, interactive is extremely restricted, it basically means the
menu checkmark, or a gesture.
The intent being that for stuff like the beforeWifiAction framework, we
don't want the backend to spawn extra UI.
Specifically, for hasWifiManager platforms, we no longer spawn the AP
scan list on failure unless the caller was interactive.
TL;DR: beforeWifiAction is now much less obnoxious when you're obviously
not able to connect.
This is semantically more correct, and should prevent platform-specific
weirdness if an ordering concern ever comes up (e.g., like it did for
the viewport stuff).
Re #10743
Note that this only makes faulty switches slightly less annoying: for a
stuck switch, instead of a string of page turns, you'll get a single
missed page turn on the tap that actually releases the stuck contact...
* Notification: Drop the fencing from #10083; it never actually helped, and had subtle side-effects we could do without.
* VirtualKeyBoard: Flash on close, otherwise, some of the fast refresh glitches may be burned into the working buffer until a flash. Making sure we flash ourselves prevent it from sticking around on the page ;).
* util: Move `writeToSysfs` to base (i.e., `ffi/util`), as we need it there (and it actually makes more sense there anyway ;p).
* Bump base for https://github.com/koreader/koreader-base/pull/1645, which is where the actual workaround (hopefully) lives.
Re #8414, #9806, #10558
* Enable before_wifi_action & after_wifi_action on hasWifiToggle platforms (which is basically all of 'em except naked SDL).
* Decouple restoreWifiAsync from hasWifiManger, because we can do that on other platforms (namely, Kindle. Probably PB, too, but WiFi is already a mess there, and I can't test it).
* Implement restoreWifiAsync on Kindle.
* Properly flag rM as hasWifiManager & hasFastWifiStatusQuery, because it is actually both of those (it uses our wpa_supplicant backend).
* Update the KOSync checks to take these changes into account, to properly disable auto_sync if necessary.
* Really made the Network* event signaling consistent. For realz this time.
* In an effort to make the whole beforeWifiAction framework somewhat usable there, we now assume connectivity is always available on !hasWifiToggle platforms...
Make sure we only send Suspend/Resume events when we *actually* suspend/resume. This is done via the Device `_beforeSuspend`/`_afterResume` methods, and those were called by the *input handlers*, not the PM logic; which means they would fire, while the PM logic could actually take a smarter decision and *not* do what the event just sent implied ;).
(i.e., sleep with a cover -> suspend + actual suspend, OK; but if you then resume with a button -> input assumes resume, but PM will actually suspend again!).
Existing design issue made more apparent by #9448 ;).
Also fixes/generalizes a few corner-cases related to screen_saver_lock handling (e.g., don't allow USBMS during a lock).
And deal with the fallout of the main change to the Kobo frontlight ramp behavior ;).
Allow disabling the hall efect sensor via the sysfs knob, so the kindle system wont sleep & wake the device
for those of use that stay in koreader, are caseless and have get spurious wakeups
* Rewrite the loop mechanism to use scheduled tasks instead of a single blocking-ish subprocess.
* Change the actual logic to be more pleasing to the eye, especially on newer devices, as those *may* natively ramp on set; and fix a bad interaction with that behavior that could lead to no ramp at all on ramp down.
* Simplify Generic's Suspend handling to deal with the refresh ordering in a saner manner. The screensaver might be visible a tad longer than before this change before the frontlight actually ramps off.
Much easier to deal with thanks to the cleanup work done in #10062 ;).
* `carrier` is set to 1 as soon as the device is *administratively* up (in practice, as soon as we run `ifconfig up`). This is perfectly fine for `isWifiOn`, but absolutely not for `isConnected`, because we are not, actually, connected to *anything*, no attempt at associating has even been made at that point. Besides being semantically wrong, in practice, this will horribly break the connectivity check, because it expects that `isConnected` means we can talk to at least the LAN.
* Delving into the Linux docs reveals that `operstate` looks like a better candidate, as it reflects *operational status*; for Wi-Fi, that means associated and successfully authenticated. That's... closer, but still not it, because we still don't have an IP, so we technically can't talk to anything other than the AP.
* So, I've brought out the big guns (`getifaddrs`), and replicated a bit of code that I already use in the USBNetwork hack on Kindle, to detect whether we actually have an IP assigned. (Other approaches, like `/proc/net/route`, may not be entirely fool-proof, and/or get complicated when IPv6 enters the fray (which it does, on Kobo, Mk. 8+ devices are IPv6-enabled)).
TL;DR: Bunch of C via ffi, and `isConnected` now returns true only when the device is operationally up *and* we have an IP assigned.
Pulls in https://github.com/koreader/koreader-base/pull/1579 & https://github.com/koreader/lj-wpaclient/pull/10
cervantes kindle kobo remarkable: use sysfs carrier file to determine connection state
cleanup hasWifiManager checks
gateway check: use ip if available
Fixes: #10087Closes: #10092
Fix#10020, Regression since #10008
(hasMultiTouch is set by `Generic.init`, but we use it before that;
so just set it properly ourselves in our superclass, because we don't
actually need to rely on Generic's autodetection).
Generic is responsible for setting up the translation via input hooks;
since our own input hooks involve swapping or mirroring X/Y, we need to
run our own hooks *first*, so that the viewport translation actually
does the right thing...
Probably broken for a good long while, I'd just assumed the inaccuracies
on the H2O were due to the IR grid... :/.
Reported @ https://www.mobileread.com/forums/showthread.php?t=351340
Regression since #9902, because I'd forgotten that the default Kobo block *does* set those...
Reported by @ptrm on Gitter, thanks!
Take two, essentially revert 8a16518918
We have defaults for most of those in the Kobo superclass, so these can never really be `nil`.
* Get rid of the `canToggleGSensor` Device cap, it's now mandatory for `hasGSensor` devices. (This means Kindles can now toggle the gyro, fix#9136).
* This also means that `Device:toggleGSensor` is now implemented by `Generic`.
* Update the Screen & Gyro rotation constants to be clearer (c.f., https://github.com/koreader/koreader-base/pull/1568) (/!\ This might conceivably break some `rotation_map` user-patches).
* Input: Move the platform-specific gyro handling to Device implementations, and let Input only handle a single, custom protocol (`EV_MSC:MSC_GYRO`).
* Input: Refine the `rotation_map` disable method implemented in 43b021d37c. Instead of directly poking at the internal field, use a new method, `disableRotationMap` (/!\ Again, this might break some `rotation_map` user-patches).
* Input: Minor tweaks to event adjust hooks to make them more modular, allowing the Kobo implementation to build and use a single composite hook. API compatibility maintained with wrappers.
This should be fairly unintrusive, but still noticeable enough without having to watch the battery gauge like a hawk, or resort to actual monitoring via kernel logs and/or usb meter and/or multimeter...
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.
* UIManager: Support more specialized update modes for corner-cases:
* A2, which we'll use for the VirtualKeyboards keys (they'd... inadvertently switched to UI with the highlight refactor).
* NO_MERGE variants of ui & partial (for sunxi). Use `[ui]` in ReaderHighlight's popup, because of a Sage kernel bug that could otherwise make it translucent, sometimes completely so (*sigh*).
* UIManager: Assorted code cleanups & simplifications.
* Logger & dbg: Unify logging style, and code cleanups.
* SDL: Unbreak suspend/resume outside of the emulator (fix#9567).
* NetworkMgr: Cache the network status, and allow it to be queried. (Used by AutoSuspend to avoid repeatedly poking the system when computing the standby schedule delay).
* OneTimeMigration: Don't forget about `NETWORK_PROXY` & `STARDICT_DATA_DIR` when migrating `defaults.persistent.lua` (fix#9573)
* WakeupMgr: Workaround an apparent limitation of the RTC found on i.MX5 Kobo devices, where setting a wakealarm further than UINT16_MAX seconds in the future would apparently overflow and wraparound... (fix#8039, many thanks to @yfede for the extensive deep-dive and for actually accurately pinpointing the issue!).
* Kobo: Handle standby transitions at full CPU clock speeds, in order to limit the latency hit.
* UIManager: Properly quit on reboot & exit. This ensures our exit code is preserved, as we exit on our own terms (instead of being killed by the init system). This is important on platforms where exit codes are semantically meaningful (e.g., Kobo).
* UIManager: Speaking of reboot & exit, make sure the Screensaver shows in all circumstances (e.g., autoshutdown, re: #9542)), and that there aren't any extraneous refreshes triggered. (Additionally, fix a minor regression since #9448 about tracking this very transient state on Kobo & Cervantes).
* Kindle: ID the upcoming Scribe.
* Bump base (https://github.com/koreader/koreader-base/pull/1524)
* Handle the power button input device
* Handle the charging LED properly
* Handle the new Wi-Fi chip properly
* Handle frontlight warmth adjustments
* Handle the battery gauge properly
* Bump base (a lot of third-party updates) [https://github.com/koreader/koreader-base/pull/1523]
* Bump android-luajit-launcher to match the LuaJIT update [https://github.com/koreader/android-luajit-launcher/pull/386]
Fix#9552 (many thanks to @Bartvelp for bearing with my stupid tests ;)).
* Android: Make sure sdcv can find the STL
* DocCache: Be less greedy when serializing to disk, and only do that for the *current* document ;).
* CanvasContext: Explicitly document API quirks.
* Fontlist: Switch the on-disk Persist format to zstd (it's ever so slightly faster).
* Bump base for https://github.com/koreader/koreader-base/pull/1515 (fix#9506)
