Andrew Walbran

To confuse matters, looking at another board, the chips are different: on this one U2 is "BOP BS." and U5 is "TP09. 1822 H1".

Answering my own question, with some help from the Telegram group, it looks like it's a [BMI120](https://datasheet.lcsc.com/szlcsc/1912111437_Bosch-Sensortec-BMI120_C437657.pdf), or more likely a [clone of it](https://www.aliexpress.com/item/1005001856758988.html). It has I2C slave address 0x68,...

> Would this PR fix #2071 ? I think so, can you test it?

I've updated this PR to include the code to try at runtime to find the new `localtime_rz` and friends and use them if they are available. If not, it will...

> > I disagree that "the current solution in chrono is fine for most users". Use of the system property as an API is not great, but depending on the...

I've added a feature, PTAL. Unfortunately it doesn't seem to be possible with cargo to completely avoid the `iana-time-zone` dependency without the feature, because it's still used unconditionally on other...

> I'm a maintainer of `iana-time-zone`. What I would like to know is: Why aren't you reaching out to [`android_system_properties`](https://github.com/nical/android_system_properties/)? The library is used in [44.5k](https://github.com/nical/android_system_properties/network/dependents) projects, so if you...

I added a feature flag in 3923d5c to revert to the old behaviour, this is awaiting review.

I don't see why not, but it might be difficult to avoid depending on a particular async executor, which is a rather application-specific choice.

I would also like to be able to have tests decide at runtime whether to be ignored. In the environment I'm working in, tests are built (along with the rest...