Some devices don't ship with iproute2 compatible tools, but may ship
with net-tools compatible ones.
Unify code w/ SDL's similar codepaths for its NetworkMgr:isConnected
implementation
- New dogear icons in Mosaic cover view to indicate each
possible book status: 'Reading', 'On hold', 'Finished'
- Progress bar redesigned to be floating, taller, thicker
bordered in Mosaic cover view
- Don't show progress bar if book is finished
- FakeCover bottom text (filename) adjusted to not overlap
with progress bar and dogear if they exist
- Mosaic book shortcut letter moved from bottom left to
top left
Only available with EPUBs containing 2 or more fragments,
and a file size large enough to ensure a cache file is used.
The idea is simply, on any rendering setting change, to
skip the rerendering of the full book and to defer any
rerendering to the moment we draw a DocFragment, and
render only it.
So, on a setting change, only the fragment containing the
current page will be rerendered, and the new fragments we
may cross while turning pages.
When having done so, KOReader is in a degraded state (the
full page count is incorrect, the ToC is invalid...).
So, a full rerendering is needed, and one will happen
in the background, and when the user is idle, we reload
seamlessly and quickly from the cache file it has made.
ReaderFlipping will show some icons in the top left
corner to let it know at which steps in this procress
we are.
Includes:
- Cache: minor simplifications
- Cache: improve cache file reproducibility
- CHM: minor cleanup
- LVStream: minor cleanups
- LVStream: fix and improve `LVZipDecodeStream` implementation
- LVStream: disable CRC checks for ZIP streams
- CSS: optimize :nth-child/of-type() and :nth-last-child/of-type()
- Page splitting: fix oversight in flags handling
- Update _doc_rendering_hash also when loading from cache
- getDocumentRenderingHash(extended): add parameter
- Toggable support for partial rerenderings of individual DocFragments
Also:
- util.isSubProcessDone(): add 'wait' option
- Also add getppid and some contants to ffi/posix.
Also don't propagate tap/hold on disabled menu items.
(When top and bottom menu displayed at the same time,
tap on a disabled top menu item could active some
bottom toggle change.)
Re #10124
----
Also includes:
* Resize the Zoom-to-Page icon to match the size of its brethren
It was one pixel smaller on all sides, for... some reason?
(Possibly I borked the resize the first time around, don't remember if I
worked on that one).
The previous iteration, besides failing to handle leaf-only input,
was relying on splitFilePathName, which just doesn't do what's required
to incrementally build the directory tree the right way around ;).
This should more closely match mkdir -p, i.e., it will *fail* if any
part (or all of it) of the path exists but is *not* a directory.
Re #10074
- filemanagerutil.resetDocumentSettings()'s doc_settings:close() -> doc_settings:flush()
- Remove current_status from filemanagerutil.getStatusButtonsRow() args, get it inside from file
- Move genStatusButton() inside filemanagerutil.getStatusButtonsRow()
- Move "Reset settings" button generation to filemanagerutil
- Rename "Reset settings" button to "Reset" and update popup box text
- Disable "Reset settings" for file if it's currently open in Collections (same as History)
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
