Reading statistics #226

I added the remaining time in chapter in #298 based on word count per page left in chapter and a manually inputted WPM reading speed.
Also added reading time per book in #309 .
These could possibly be combined to get a better estimate of the WPM but for now people can get their average wpm from something like koreader or a test on a website and that should be more than good enough.

Also see #1600 in the Issues section.

Are any of the current maintainers interested / willing to shepherd a consolidated reading stats PR/feature through to merge into the main code base?

This would be phenomenal.
Some basic stats such as time left in book or time spent reading book would be amazing.

One consideration for a consolidated reading-stats design: it may be worth aligning the data model and granularity with what KOReader already persists (e.g. per-book aggregates plus per-page or per-session records with start time, duration, page index, total pages — the kind of data that tools like KoInsight or KoShelf ingest from KOReader’s statistics.sqlite for dashboards and cross-device views).

CrossPoint already lives in an ecosystem with KOReader sync; if stats are stored in a compatible or easy-to-map shape, a future export/sync path could reuse the same pipelines (or share a server) instead of maintaining a one-off JSON schema that only fits CrossPoint. That does not mean copying every KOReader feature, but schema compatibility (or a thin mapping layer) could pay off for users who read on multiple devices.

The crossink fork has reading statistic in it, I think perhaps taking a look at that work and porting it here could be work worth doing. What do you guys think?

It counts milliseconds to do the tracking. Might have some battery life impact but I don't think its significant, but I haven't done an extensive research on this fork.

@jerojero That'd be great. CrossPet has a great layout for Reading statistics.

For anyone interested in some minimal time tracking: #1815

While I like the idea of a more comprehensive reading tracking feature, I wanted to start tracking time now. This PR could either be a starting point for more in-depth tracking or just a way to track time for now which could later be manually transferred to a new system.

Would folk be willing to take the Crossink implementation of reading stats as a starting point?

The dev has a solid feature set and has kept in-sync with Crosspoint for the most part and I think it'd be a good starting point. It won't cover all the requests/etc we've had but we would at least be able to port a stable form of stats that cover the general case which we can later iterate based on community feedback.

As soon as we're able to release v1.3.0, I'll pin this thread again

I also think that lightweight reading stats would be a great addition to crosspoint! As mentioned, it could be behind a toggle so that people who don't want it, don't have to see it.