3.10.0: Add QZSS time scale -- possibly a breaking change

[ChristopherRabotin]

What's next?

This is probably the last release of version 3.x. There's a fair amount of small bugs here and there in 3.x, and these aren't trivial to fix. The architecture changes introduced by @gwbres and I for the development of 4.x should fix all of those in one go.

Also, version 4.0 will likely switch to the Mozilla Public License from Apache 2: if you have concerns about this, please head over to #274 to voice them. From the user point of view, there should be absolutely no change, even if you're selling code that uses an unmodified version of Hifitime.

What's Changed

• support qzss time scale by @gwbres in support qzss time scale #228
• Update 4.0.0-dev to latest master by @ChristopherRabotin in Update 4.0.0-dev to latest master #243
• Integrate 3.8.6 into 4.0.0-dev by @ChristopherRabotin in Integrate 3.8.6 into 4.0.0-dev #264
• BREAKING CHANGE: 4.0.0-dev branch become 3.10.0 by @ChristopherRabotin in BREAKING CHANGE: 4.0.0-dev branch become 3.10.0 #277
• Draft: replace std::time:SystemTime with web_time::SystemTime by @karpfediem in Draft: replace std::time:SystemTime with web_time::SystemTime #279
• Update reqwest requirement from 0.11 to 0.12 by @dependabot in Update reqwest requirement from 0.11 to 0.12 #286
• Fix panic caused by arithmetic overflow by @cardigan1008 in Fix panic caused by arithmetic overflow #283
• Fix panic caused by array out of bounds by @cardigan1008 in Fix panic caused by array out of bounds #284
• Fix panic caused by unwrap by @cardigan1008 in Fix panic caused by unwrap #285

New Contributors

• @karpfediem made their first contribution in Draft: replace std::time:SystemTime with web_time::SystemTime #279
• @cardigan1008 made their first contribution in Fix panic caused by arithmetic overflow #283

Full Changelog: 3.9.0...3.10.0

---

This discussion was created from the release 3.10.0: Add QZSS time scale -- possibly a breaking change.

[gwbres]

Hello Chris, this looks very promising 🚀

Right now I'm working on PPP performances enhancement, I want to conclude this topic prior taking the leap.
I'm also unsure what the best strategy would be to take the leap, considering hifitime is core library to us, and impacts at every level

[gwbres]

Hello @ChristopherRabotin,

I cannot put the new Hifitime to the test using my RTK solver (the ultimate usecase) until nyx gets updated.
Would it be possible to propose a new nyx-dev branch ?
All I can say is I have updated all my crates manually, bypassed the RTK stuff, ran the RINEX CI and it did pass 🚀 🚀 🚀
Considering we have about 230 tests at the moment, that is quite something to say

[ChristopherRabotin]

Ask and you shall receive: nyx-space/nyx#294 !

[gwbres]

Will hifitime V4 come with more friendly Serdes Input/output ?

[ChristopherRabotin]

Indeed! In fact, that change is already merged to master as of yesterday.

…

On Sat, Apr 27, 2024, 13:02 gwbres ***@***.***> wrote: Will hifitime V4 come with more friendly Serdes Input/output ? — Reply to this email directly, view it on GitHub <#293 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABEZV2ARH5IYY2F37WIPKI3Y7PY6HAVCNFSM6AAAAABF53V3SWVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TENBXGM2DI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[gwbres]

Ok cool that is a good thing.
For example I use JSON configuration files in my application and the previous behavior kind of forbid using Durations in those, due to the impossibility to understand the value