Includes:
- Text: fix possible overflow with "white-space: pre"
- Fonts: getFontFileNameAndFaceIndex(): return family type
- Fonts: account for _bias in Font hash
- epub.css, fb2.css: remove hardcoded monospace font names
- Fonts: allow scaling monospace fonts
- Fonts: allow adjusting fallback font sizes to x-height
- HTML parser: trust xml encoding before html charset
In our list of fonts, show a "M" indicator alongside
monospace fonts, and allow long-press to select the one
to be use for "font-family: monospace".
Add 2 options to Font settings, to scale monospace fonts
by a percentage, and to adjust fallback fonts to their
x-height.
In some versions of the /ebrmain/config/extensions.cfg file there is a
comment #ebrcfg. This will not match the regex and crashes koreader as
the table is empty.
Add a check to ensure the table is not empty while reading extensions
from the default file.
This is useful for the sysstat plugin not to have 00:00 for suspend time :)
Something similar should probably be done for the other canSuspend platforms.
This is mostly needed because opening the menu with swipe
is actually done by handling a pan south event, and so the
final swipe event is another event, that is then handled
by TouchMenu, which would straight away close itself.
Disable redundant or incompatible tweaks when enabling some.
This makes the menu more dynamic to show what's impacted,
save some re-renderings, and may avoid the user trying
combinations that would have no effect.
Includes:
- CSS: add "-cr-hint: late" to increase selectors specificity
- CSS: add a few non-static "-cr-only-if:" values
- getStyledImageSize(): fix missing images when gRenderDPI=0
- Text: fix possible crash with floats and <br/>
- LVImg: fix crash on GIF images without color table
Also update thirdparty/fribidi to 1.0.12.
Related to https://github.com/koreader/koreader/issues/8482
Changes:
- New light driver for Onyx devices that should work on modern ones. Thanks to @bb010g!
- Onyx Nova 2 migrated to the new driver.
Added devices:
- iReader r6800
- Onyx Magicbook
- Onyx Note Air 2
- Onyx Nova Air C
- Sony DPT-CP1
* AutoSuspend: Use the canSuspend devcap check instead of reinventing the wheel.
* Device & UIManager: Cleanup canSuspend devcap check related stuff to avoid boilerplate code.
(It also now defaults to no, and is explicitly set by device implementations where supported).
* AutoSuspend: Re-engage suspend/shutdown timers when fully charged.
This restores the existing behavior pre #9036
(c.f., https://github.com/koreader/koreader/pull/9258#issuecomment-1167672356)
* SDL: Unbreak the fake suspend behavior so that it actually works.
Tweak the default screensaver message to remind users that Power is bound to F2.
(Fix#9262)
* AutoSuspend: Re-engage suspend/shutdown timers on unplug.
This matters on Kobo, because the unexpected wakeup guard might have stopped the suspend timer.
Schedule wifi keep-alive on wifi start but do not reschedule when
connection is lost. This will make sure this code is not run when
wifi is disabled again saving some battery and cpu cycles.
Commonize fullscreen code across different devices.
Add fullscreen toggle on SDL desktop targets.
Expose the toggle to the dispatcher and map it to F11.
Hide touchmenu when resizing.
This is a maintenance PR.
With scheduling the next autodim_check on every tap, we can keep the device longer in standby mode. So we have more power savings.
The long story: The current version works like that:
Let's say I set the idle-time to 3 min and I need 2.5 min to read a page.
Start at 0 min (next scheduled check is at/in 3 min)
Read and tap to turn page after 2.5 min (next scheduled check is at 3 min - in 0.5 min)
At 3 min there is a scheduled_check (wakeup) -> reschedule check in 3 min.
Read and tap to turn page after 2.5 min (no wakeup)
Then after 1 min there is a wakeup -> reschedule.
---> So I get an additional wakeup on every second page.
This PR solves the problem: During normal reading I don't get any wakeup during reading. (The only drawback is, that one function is unscheduled/scheduled on every tap, but that is no big deal).
* Get title, author and number of pages from document's metadata
* Add exportable_title parsed from document's title, used in export file generation.
* JSON: export number of pages
