I maintained chrono for about half a decade, and the specific reason I took on maintenance was because the API sucked and I hoped I could make it better.
Turns out, the reason people wanted to use it was because the API was stable and every slight step towards improvement was met with... resistance. I stopped maintaining it because the burnout from seeing what it could be and not seeing any path to get there was too much for me.
I'm extremely grateful to burntsushi for writing the datetime library rust deserves. Chrono is stable and is maintained (though semi-deprecated) by an extremely reliable rustacean -- you don't need to leave, but you'll probably be happier if you do.
lifthrasiir 23 hours ago [-]
As the original author of Chrono, thank you so much for your service. It was always a learning experience for me to design and maintain Chrono (that's why the API is not very good, despite my best efforts) and I too had been burnt out from doing so, so it was fortunate that the right people stepped up and kept the torch burning for so long time.
quodlibetor 19 hours ago [-]
I shouldn't have used the word sucked -- personal jargon that I use all the time to describe my own code and try not to apply to other people's -- it was created pre-1.0 and we learned a lot as a community in the 10 years you spent building it.
Thank you so much for building something that was obviously good enough to sustain the community for 12 years <3
hermitdev 19 hours ago [-]
As someone who has worked with implementing date/time functionality before, I'm empathize. Getting APIs "right" is hard, and getting it wrong is easy because there's so many ways to get it wrong. Especially when you start getting into something like handling something like daylight saving time, which is inherently political and changes arbitrarily over time. I've spent most of my career in finance. I distinctly remember near shouting matches over what it meant to add "1 day" to a datetime. Does it mean adding 24 hours? Or, does it mean give me the same time of day tomorrow? What happens if tomorrow is the other side of a DST change? These seems like trivial questions, but matter if you're talking about, say physical delivery of oil or gas at a set rate per hour.
lifthrasiir 18 hours ago [-]
I'm fine, actually. There is a joke that every successful programming language creator would spend next decades fixing faults in one's language. I feel that applies to every successful software as well...
democracy 18 hours ago [-]
I learnt not to say "shit code" in public since quite often the history would point at my changes after investigation lol
masklinn 9 hours ago [-]
Honestly it’s the best case of having called something shit code, it means you’ve improved enough to recognise its faults, and at most you’re hurting your own feelings and nobody else’s.
the__alchemist 21 hours ago [-]
Thanks to you both. I am giving benefit of doubt to those saying Jiff is even better, and I am trying Jiff today, but make no mistake: Chrono is by far, the best datetime library I've used. If Jiff ends up being people's #1, I hope they realize that Chrono is probably the #2 of any datetime library in existance. Thank you so much. It is dramatically better than Python's and JS's (official and unofficial) ones. I can rely on it for std and embedded tasks, and trust it to act as I expect.
BrandoElFollito 20 hours ago [-]
Why is it so much better than, say, Arrow?
the__alchemist 19 hours ago [-]
Arrow has no date type. No time type either! Using their recommended approach of "use datetimes for everything" is a poor representation of data in many scenarios and leads to surprise behavior. Chrono's sins, as enumerated in the Jiff documents, aren't on the same tier.
Unless you're talking about Arrow, the Apache Parquet interface library in either Python or Rust - that's a different matter; great software!
BrandoElFollito 9 hours ago [-]
I am trying to find a case where a time without a date would be useful. The only one I can think about are analog monotonic clocks.
In any other case I can think of, you need to attach the time to a date and a timezone. You can display it bare in the context of the viewer (who has an implicit or explicit timezone and time reference) but any manipulation without these will not end well.
Would you have some examples for the poor representation of data and surprise behaviour?
the__alchemist 7 hours ago [-]
TLDR: Think about how to represent various things in a struct field[s]. Sometimes the time or date information isn't part of state, and adding it is an overconstraint.
What datetime is your annual birthday celebration? When's Christmas this year? What time is your Monday alarm clock set for? Wen's the daily standup?
There are so many, and sometimes Arrow/Moment's abstraction is OK. Other times it's awkward as it overconstrains. Sometimes it has surprising behavior.
Or, ask yourself why Jiff and Chrono, and python datetime etc have separate date, time, and datetime types.
Datetimes are useful, and I use them more than individual dates and times. The latter 2 are still useful, and datetime is a sloppy or buggy fit.
My best guess of your (and the arrow and moment creator's mentality here): They have been hit by bugs of the kind you're saying, involving timezones, or things that should be datetimes + timezone, but were using native datetimes, or dates etc. The response was to use datetimes for everything, ignoring other cases. Does that sound accurate?
steveklabnik 23 hours ago [-]
Thanks for helping maintain Chrono. Burnout is the worst, I hope you're feeling better these days.
Arch485 22 hours ago [-]
I'm really curious - what makes you say chrono's API sucks?
I've used it in several projects, and to me it seems just fine. Granted, I don't do a lot of cross-time zone work; I exclusively use UTC whenever possible.
lifthrasiir 20 hours ago [-]
A commonly cited reason is that it mixes Option<T> and Result<T, E> for error handling. It was designed well before Rust's API convention is settled, and I also wrongly assumed that some features like procedural macros would be delivered soon (which was why `ymd` and such panic on error, error handling would happen in the macros then).
burntsushi 24 hours ago [-]
<3
thatsabadlook 16 hours ago [-]
Hope you are doing well sushi
ninkendo 3 hours ago [-]
> API was stable and every slight step towards improvement was met with... resistance
(Putting my flame suit on here, but...) I think that's a good thing, and I wish this were the norm with more libraries in the open source world. Once you hit a stable version, breaking changes should happen approximately never. Improvements that need breaking changes should happen under a new name, with a new project, so that the old project can still exist and receive updates.
I still say SemVer has done more harm than good for the industry... it normalized the idea that "it's ok to break compatibility, just bump the major version" and then you have libraries on version 26 that break everything every time the author decides to feng shuei the code some more. Every major version bump should be thought of as a sort of "grave marker" indicating a place in history where the author messed up and had to admit the old design was literally broken, and every downstream dependent had to go through some amount of pain as a result. It shouldn't be normalized.
You can say "nobody's forcing you to upgrade" but, with the amount of vulnerabilities out there, yes, you often are forced to upgrade certain dependencies.
You can say "just maintain the older versions too", and that would be absolutely great, if anyone really did this. In practice anything but the most recent major is neglected.
You can say "I agree and that's why I never release 1.0", and I'd say that's a reasonable approach, but that just changes the problem to "too many packages are pre-1.0" and we'd have the same issue.
You can say "that degree of compatibility is too difficult to expect out of open source maintainers", and my response is that there ought to be fewer open source maintainers. (For real. I'd rather have 1/100th the number of crates in Cargo, if the ones that were there were all maintained by people who were serious about backwards compatibility.)
You say "the API sucked and I hoped I could make it better", but the API is the library to anyone using it, and changing it means you're literally changing what the library is, which means all existing code using the library has to change to meet your new opinion on what the better API is (an opinion many wouldn't share, including and especially the original authors.) Why not just start a new project instead?
ddahlen 21 hours ago [-]
Thanks for maintaining such a core tool, it has been really helpful for me!
joel2oliver 22 hours ago [-]
Thanks. I shipped a couple of things on chrono and never had to think about it. That was you.
thatsabadlook 16 hours ago [-]
I am extremely grateful that you maintained chrono. Used it at several companies. Thank you.
Dagonfly 1 days ago [-]
> When should I use Jiff? Jiff isn’t always the right thing to use.
I really appreciate when a project calmly acknowledges that it's not the right tool for every use case, and is not trying to be. That's professional engineering.
Especially in the AI-age, the crucial skill will be what NOT to implement.
BrandoElFollito 20 hours ago [-]
I always use a time library (in Python and TS). If you work on time intervals in seconds, fine, you can use what is built-in. But I favor consistency and everything time is a time library object for me.
Outside of this not using a timer library is going for a world of pain
Sebb767 12 hours ago [-]
From [0]:
> Unlike “Comparison with other Rust datetime crates,” this document is opinionated.
> [...]
> Jiff is pronounced like “gif” with a soft “g,” as in “gem.”
Burntsushi really did not hold back with this document being opinionated :-) (Though they are right, of course).
It’s so good like all the other burntsushi libs. Really hope it will be adopted more and more by the wider ecosystem over chrono and co.
dicke 1 days ago [-]
He is incredibly kind and gracious too. I've made some contributions to jiff, and Andrew is so easy to work with.
AeroNotix 23 hours ago [-]
Andrew had a pretty decent impact on the course of my career and I doubt he even remembers it or even thought of it as an act of good he did.
In the very early Go days (<0.9) he responded to a comment of mine going into why Go was worth looking into, quite passionately and enthusiastically which set me off on learning it and where it could be applied and that changed the kinds of roles I went after.
Ever since it's a bright little nugget whenever I see his name pop up and have tried to follow his work ever since.
burntsushi 23 hours ago [-]
Wow, I don't remember that! I'm happy to have had that impact. :-)
1 days ago [-]
ddahlen 1 days ago [-]
This looks interesting!
Time is such a pain, I write code for simulating orbits of asteroids/stuff in the solar system, and time and time conversions play a major roll in it. I often need accuracy at the mega-year to giga-year scales, but also sometimes micro-second relative.
I saw on this page that you at least mention TAI time, how are you handling time scales in general? Like would it be hard to add some other timescales like UT1?
Right now I have to do all of this manually and I have been using chrono as just a convenience wrapper so that my users can convert ISO time strings into my internal representation.
Edit: Oh no! I just found out chrono is being semi-deprecated, looks like I'm going to be switching...
burntsushi 1 days ago [-]
Jiff only supports the Unix time scale. It sounds like you probably want `hifitime` instead! I believe it was designed with your use case in mind. See: https://docs.rs/hifitime/
ddahlen 1 days ago [-]
Thanks for making useful tools!
Ah yeah hifitime, I tried it out but its internal representation doesn't work well for my long term simulations, the integer representation ends up increasing the numerical noise in my simulations when I do really sensitive long term ones.
pbrowne011 1 days ago [-]
Nice! I appreciated the documentation for tzdb on avoiding embedding and the thought process behind using "rearguard" data [1]. Some related GitHub issues are also worth reading [2] [3].
I do wish they could use the system database on windows. Windows does maintain this data, but it's not in the same format.
The C++ date-time library goes and loads the system copy of ICU (which is updated via windows update) and pulls the tz data from ICU's calendar api.
the__alchemist 1 days ago [-]
I will give this a shot, however... Rust is the only language I've used where I want to use the established one, chrono. I will roll my own in JS, for example, or begrudgingly use or wrap the python.datetime one (And not touch the third party python ones which do awful things inspired by Moment.js like conflating dates , times, and datetimes, under the guise dates and times don't have use on their own), but Chrono actually is comparatively nice (S-tier relative to others), both std and no_std.
masklinn 1 days ago [-]
> I will roll my own in JS
Hopefully Safari will soon have temporal implemented and we can leave that behind (it's in 27.3b so not out of the question).
And as randomblock linked, chrono is deprecating itself.
the__alchemist 1 days ago [-]
Ty for the details! That is about as clear as anyone could hope!
This is still very surprising given the context; if the top datetime library of any language is getting soft-replaced with something nicer, woe be to those in other languages!
I love Arrow in Python - I need consistent time handling because dates are weird and conflating everything as a date is awesome.
poulpy123 1 days ago [-]
I would have expected a modern language to have an extensive datetime library in its standard library
lambda 1 days ago [-]
The problem with this model is that then you try to build all of the libraries that you anticipate users are going to need in one big go before you release version 1 of you language. Because you're trying to do a lot at once, they don't ask get the proper attention they deserve. Then, because of stability guarantees, you're stuck only making backwards compatible changes in the future (or you risk either constant breaking changes, or big changes where a lot changes at once forking the ecosystem, like Python 2 to 3). So if you made any mistakes in the library, you're stuck with them, only being about to deprecate but never really remove them.
The alternative is to keep a relatively small standard library, and a good package manager, to make it easier for people to iterate on API design in third party packages. Major version changes in a single library can have much less impact accross the ecosystem than standard library packages. Eventually you may get a de facto standard, and sometimes they are then later incorporated into the standard library.
This model isn't without its downsides; it can sometimes mean fragmentation when standardization would help, sometimes these de facto standard libraries can become important enough that they run into the same ecosystem fragmentation issues on major releases. But it allows for more flexibility.
The latter model is what the Rust project decided on for everything outside of the most fundamental time types. Dates and times and timezones are notoriously hard to model in code, and easy to create API designs that are full of footguns, so this is a place where that extra freedom to iterate was more important than standardizing on an API early on.
That iteration has led to a number of different libraries, including time, chrono, and now jiff, allowing for exploring the space of API design to really get it right rather than being stuck with what was available early on.
Sharlin 21 hours ago [-]
Jiff demonstrates the exact reason Rust doesn't ship a datetime library - at least not yet. It is extremely well understood that datetimes are one of the hardest abstractions to get right, and it took many years and several third-party attempts to get to this point. It would've been an extremely severe mistake to include datetime in std and then be forced to maintain a likely flawed API for all eternity. Luckily Rust devs knew and know better.
dminik 1 days ago [-]
Any modern language recognizes that some things are inherently complex and will always have tradeoffs that certain users won't be satisfied with and that will be locked in forever. Such modern languages then let users pick whatever approach they actually need by implementing sensible package management.
darksaints 1 days ago [-]
A default standard encompassing the most common usecases is a completely reasonable approach. Having a date/time library in the standard library does not preclude other libraries, it just makes it the default and keeps ecosystem integration sane. Not to mention the massive potential reduction in supply chain attacks.
kbolino 24 hours ago [-]
In this particular matter, I can't help but be biased against your perspective due to Go. In that language, the desire to meet the "most common usescases" has led to an IMO pathologically bad attitude towards doing this right. In Rust, only the bare primitives are exposed -- Instant, SystemTime, Duration -- but these at least allow for others to build better abstractions. Whereas, in Go, all of the primitives are kept hidden behind a "good enough" (it's not) abstract API, and you have to break out O/S-specific and/or unstable code if you want to do things better, especially where monotonic time is concerned.
darksaints 23 hours ago [-]
Java had the same problem, but then they then fixed with `java.time`, which I believe should be the benchmark for off-the-shelf date/time functionality. Doing it right absolutely is a management and leadership problem, and management/leadership can fail, but when it is done right the result is so much better than the bazaar of 20 half-baked mutually incompatible solutions.
kbolino 23 hours ago [-]
I don't agree that Java had the same problem. Java has had System.currentTimeMillis() since the beginning, and System.nanoTime() since 2005. Meanwhile, in Go, time.Now().Unix() performs a pointless timezone lookup and time.Since(foo) only uses the monotonic clock if foo has monotonic time, which is a hidden property of time.Time values that you can't interrogate.
Of course, Go can always backtrack and expose the primitives, and Rust can always add a higher-level API, but as user of the languages, I'd rather have the latter problem than the former. It doesn't help that Go's zoned-time API isn't very good either.
stouset 22 hours ago [-]
The gold standard these days is JavaScript’s Temporal, which Jiff and java.time.temporal are descendants of.
nick__m 20 hours ago [-]
you are wrong on ancestry, java.time is descendant of joda-time and temporal the descendant of a js port of joda-time.
masklinn 24 hours ago [-]
Having a date/time library in the standard library actually makes it very difficult for other libraries to take off, and when it's deeply deficient (which is generally the case, turns out time is quite hard) it's actively detrimental to users of the ecosystem, both because it's hell to use and because it teaches them to do things incorrectly. Which then makes an innovative library have a harder time take off, because they need to fix most of their possible users.
sunshowers 1 days ago [-]
A poorly designed one, maybe. My understanding is that the Temporal model is a very recent invention.
masklinn 1 days ago [-]
In fairness, Temporal largely follows the model originally defined by Joda, and refined in java.time. So I'd say that Java is probably one of the few languages with a well designed datetime library in the standard library.
Temporal does improve some things (it has more strict zoning / less defaults, and improves serialization through RFC 9557), but some of the divergences are not necessarily improvements and rather divergences of opinion (e.g. java.time wilfully made non-ISO calendars second-class, Temporal makes them first class more like Joda, and it's one of the things Colebourne — the lead author of Joda and java.time — strongly disagrees with[1], but the Temporal authors have pretty compelling reasons for it[2]).
And Java needed two very flawed attempts before finally getting it more-or-less right (and as you said, it was a third-party library in the end that was adopted rather than one designed by the core developers.)
vips7L 1 days ago [-]
I really enjoy working with java.time. It's one of the best additions to the standard library.
procaryote 22 hours ago [-]
Java has a pretty good approach of promoting the more popular library for a thing to standard library status, with relatively minor changes. It's an easy way to get a really good standard library over time, as it's already proven in practice
imtringued 4 hours ago [-]
If the Rust team followed that advice it would have two datetime libraries by now.
tredre3 1 days ago [-]
Rust doesn't even have an async runtime in its standard library, despite async being part of the language. They believe increased reliance on third party dependencies is better for the ecosystem. Things like supply chain risks or disparate documentation or conflicting dependencies or enormous .cargo are seen as less of a problem than having a built-in API that could go stale one day.
JoshTriplett 23 hours ago [-]
> They believe increased reliance on third party dependencies is better for the ecosystem.
Not at all. We just want to avoid having the standard library be "where code goes to die", as Python folks have put it.
We're working on adding more popular APIs from the ecosystem to the standard library. See https://github.com/rust-lang/std-replacement-data/ for a data set tracking what crates have been obsoleted by additions to the standard library; we hope to increase that over time.
hmry 23 hours ago [-]
I don't know about some of those points (maybe you could expand on them)
> disparate documentation
I've never experienced this while using a Rust library. They all use rustdoc hosted on docs.rs (or locally with $ cargo doc). They all look identical and there's a single global search bar that searches all your dependencies.
> conflicting dependencies
Since you can install two different versions of the same library, this is only an issue if two of your dependencies both expose a transitive dependency with different versions in their external APIs. (E.g. libA returns a Foo 1.0 and libB expects an argument of Foo 2.0) Only library I've ever run into this with is raw-window-handle.
> supply chain risks
> enormous .cargo
These I agree with. Although honestly the supply chain risk is not as big with these foundational libraries as it would first appear, since they are almost all maintained by the same 5 rust compiler developers. Really the only difference is that they're not in the rust-lang github organization and not shipped precompiled with the standard library
afdbcreid 1 days ago [-]
As I've said in other places, supply chain risks have nothing to do with things not being in std, only with them not being official - and things can be (and are) official without being included in std.
Documentation certainly has nothing to do with it.
Other factors you mentioned are more convincing.
solidsnack9000 1 days ago [-]
I'm not sure this makes sense for a systems programming language -- but I can see it for an application programming language.
For a systems programming language, Rust is relatively easy to write applications in -- but I suspect standard library choices (and other choices) will be driven by the systems side for the foreseeable future.
kjuulh 1 days ago [-]
Jiff is very nice to work with. It is always a pleasure working with something crafted by burntsushi
brianwitte 1 days ago [-]
the jiff pronunciation meta got my brain working hard for second. love it
the__alchemist 1 days ago [-]
With how my brain pronounces short names and acronyms etc vs the commonly-accepted ones, this one has to be something like like "Geth".
> Jiff is pronounced like “gif”
This would have been a hilarious line if he left off the second part!
Sytten 14 hours ago [-]
I did the chrono 0.1 -> time 0.3 -> chrono 0.4 dance, hopefully jiff is the last. I need to add support for it in diesel if not already. Pretty sure we have corners of llrt where we still use chrono.
masklinn 9 hours ago [-]
There’s a jiff-diesel, but in burntsushi’s own words
> Yeah I had a really hard time making the jiff-diesel and jiff-sqlx crates. Part of the problem is the very poor datetime support in the underlying databases. The other part of the problem is that I'm not a user of either diesel or sqlx, so there are some unknown unknowns for me.
IME time is one of those things that a lot of developers try to wish away the complexity. The hard part is defining what you even mean by time, but if nobody asks the question then you don't have to provide an answer.
Strings are also similarly gnarly topic that I find a lot of devs try to ignore.
Levitating 16 hours ago [-]
Looks like great software.
BetterThanSober 14 hours ago [-]
Time is so intuitive to us humans (mostly) yet feels so excruciatingly complicated to manipulate in computer
I really, really, hate Monday
Sparkle-san 1 days ago [-]
> Jiff is pronounced like “gif” with a soft “g,” as in “gem.”
Woah now, let's not get too controversial.
MadameMinty 1 days ago [-]
Out of pure spite, I will pronounce Jiff like "gif", with a hard "g", as in "good".
dmit 1 days ago [-]
The man's name is "burnt sushi". I suspect he's fine with controversy. :)
flufluflufluffy 1 days ago [-]
if I ever use this lib i'm gonna be scowling with this in the back of my mind the whole time xD
Graphics Interchange Format. Guh-raphics. GIF
jitl 20 hours ago [-]
By this logic, "JPEG" should have the P pronouced like it is in "Photographic", like "jay-feg"
notstayinglong 19 hours ago [-]
To playfully go full circle, the original Greek pronunciation of ph was as an aspirated p, which is how English pronounces all p's (except when preceded by an s), so we're back to "jay-peg".
Fun linguistics experience for native English speakers: hold an open palm in front of your mouth, fairly close, and say 'spit', then say 'pit'. Much more air hits your hand when saying 'pit' than when saying 'spit'. That's the aspiration.
19 hours ago [-]
berkes 21 hours ago [-]
Graphics. Like that long-necked animal?
arjie 21 hours ago [-]
My favourite part about BurntSushi software was that the structure of the program was enough that even a novice at the language could change things. Very well laid out. I had a fork of many of his applications for personal use and it was always trivial to jump in and edit. Excellent stuff.
> Jiff is pronounced like “gif” with a soft “g,” as in “gem.”
This is classic anti-trust monopoly leveraging. Lucky I'm on this side of the debate :)
dizzant 19 hours ago [-]
> Jiff is pronounced like “gif” with a soft “g,” as in “gem.”
A masterclass in viral outrage bait. Not a Rustacean, I clicked the link exclusively for this nugget and will probably tell my Rust-loving colleagues about it tomorrow.
For the record, I like the converse better:
> Jiff is pronounced with a hard “g”, as in “gif”.
kodebach 18 hours ago [-]
Always fascinating how some languages still struggle with datetime stuff, when Java (or the JVM in general) basically solved it over a decade ago. And I say "solved" because everything seems to eventually settle on the same API as java.time.
gosen 1 days ago [-]
I thought it's the right moment to finally do something with Rust, but I work on my "J" pronunciation first
NetOpWibby 1 days ago [-]
Love seeing the correct pronunciation of gif
echelon 12 hours ago [-]
Gif as in giraffe, not graphic.
Giraffical Interchange Format.
24 hours ago [-]
PowerElectronix 1 days ago [-]
Weird, you can't ever pronounce Jiff as gif.
fstephany 1 days ago [-]
Jiff will probably be the default datetime library for a lot of Rustaceans (and LLMs) in the years to come.
Thank you for your work on this burntsushi!
kccqzy 19 hours ago [-]
The one thing I don’t like in most datetime libraries is that for timezone names they all just defer to IANA time zone database. This database is a wonderful compendium of all time zones historically in use, but it is absolutely programmer oriented and not user oriented. There are two main problems: (1) it’s based on large cities; (2) it captures history faithfully.
The first problem is the bigger problem. If you are writing an app that asks the user for a zoned time, the user will want something like Nov 16 7pm eastern time. But EST is of course highly ambiguous and so is EDT. (The database actually gives EST to Panama Time and leaves EDT undefined.) The correct solution is either EST5EDT or America/New_York. Neither is user friendly. The first string EST5EDT looks like gibberish to the average user. The second raises concerns in all other cities sharing the same timezone as New York. Why not have an entry for Philadelphia or Washington D.C.? Some even say it is inherently classist to pick the city with the biggest population, not the most important by some other metric. Is New York more deserving than our nation’s political capital? Why is Los Angeles more deserving than San Francisco, other than the fact that the maintainer of the database is physically in Los Angeles? This is an actual criticism in iOS calendar app on the now defunct Apple forums. If you want the user to pick a timezone you better make sure Eastern Time is an allowed option, and maybe also include all the cities and municipalities in the zone; that’s what the newest iOS does. But in any case almost every developer now hard codes a mapping from Eastern Time to America/New_York if their audience is U.S. This feels like duplicative work better done in a library. The fact that a new library reaches 1.0 and doesn’t plan to add this might mean a lack of ambition to tackle this problem.
The second problem is the fact that the timezone definitions change according to the actual rules in effect in history. This is sometimes not what you want. Indeed I notice that most if not all date libraries use the proleptic Gregorian calendar (our current calendar system extended backwards); no library that I know of automatically switches to Julian calendars before the cutoff. Why this inconsistency? I’d argue that when most humans express a zoned time they expect the current rules to be extended backwards. People forget the daylight saving rules change in the 2000s when talking about times before that change and expect the current rules.
Not to be a detractor here for this library given the hard work of burntsushi, but we should aim higher.
masklinn 9 hours ago [-]
> This feels like duplicative work better done in a library.
While not necessarily wrong, it should probably be a separate library entirely: there are projects maintaining GIS boundaries for the IANA timezones.
kccqzy 5 minutes ago [-]
After some more thought, I agree this should be a separate library. I did some search, and I found that ICU library seems to contain all the timezones in the world and their display names. https://cldr.unicode.org/translation/time-zones-and-city-nam...
I found the icu::TimeZoneNames class. The docstring even describes my exact problem:
> For example, people in the United States refer the zone used by the east part of North America as "Eastern Time". The tz database contains multiple time zones "America/New_York", "America/Detroit", "America/Montreal" and some others that belong to "Eastern Time". However, assigning different display names to these time zones does not make much sense for most of people.
IANA time zone identifiers are baked into RFC 9557, which extends RFC 3339. For sure, IANA identifiers are generally for programmers.
> The second problem is the fact that the timezone definitions change according to the actual rules in effect in history. This is sometimes not what you want. Indeed I notice that sometimes all date libraries use the proleptic Gregorian calendar meaning our current calendar system extended backwards; no library that I know of automatically switches to Julian calendars before the cutoff. Why this inconsistency? I’d argue that when most humans express a zoned time they expect the current rules to be extended backwards. People forget the daylight saving rules change in the 2000s when talking about times before that change and expect the current rules.
Good question. One part of this answer is that the IANA tzdb does not only contain daylight saving time transitions. It also includes transitions from one offset to another, perhaps because daylight saving time was stopped or started. The other part of this answer is, I would guess, more pragmatic. Namely, we have structured data delineating time zone transitions back at least a century. If you go far enough back, then indeed, time zone transitions stop happening. We, however, do not, as far as I'm aware, have shared structured data delineating changes in calendar. This would also need to be organized by both location and time. For example, the Julian calendar was not dropped in favor of the Gregorian instantaneously everywhere. Some countries, like Russia, continued using the Julian calendar into the 20th century.
masklinn 2 hours ago [-]
> For example, the Julian calendar was not dropped in favor of the Gregorian instantaneously everywhere. Some countries, like Russia, continued using the Julian calendar into the 20th century.
An other fun one is that some of the transition schemes were... imaginative, let's say. Famously Sweden didn't want a hard break so in 1699 they decided that they would just skip leap days until they were on Gregorian. So they skipped the one in 1700, then promptly forgot the 1704 and 1708 ones because they were busy with the Great Northern War, as a result the scheme was cancelled by reverting to the Julian calendar, which they did by adding a second leap day in 1712 (which thus had a February 30th, and 367 days). Sweden ended up doing the switch at once in 1753.
That's calendaring rather than timezoning tho. Timezones really only became a factor once timekeeping was precise enough and times had to be communicated and synchronised across vast distances, which really came about with the railways (first formal "time zone" was the Railway Time of England's Great Western Railway). Before that, cities and villages oft had their individual time based on apparent solar time at the location (aka a sundial). For instance Paris Mean Time was 9 minutes and 21 seconds ahead of London.
3ph3m3re 10 hours ago [-]
As a rust hobbyist coming from Java, I feel a bit sad that there is no standard library for date handling like java.time API.
Do you seasoned rustaceans may explain why we don’t have such a standard library?
It’s a genuine question from an ignorant :)
imtringued 3 hours ago [-]
Mistakes in the stdlib are eternal.
capitol_ 10 hours ago [-]
It's very very easy to miss something or design it wrong when building a complex API.
Just look at the original java time stuff Vs those that came in java 8.
And once something is in the standard library it can never change.
echelon 17 hours ago [-]
> Jiff is pronounced like “gif” with a soft “g,” as in “gem.”
What a cardinal sin! Instantly distrust this project.
(I'm kidding, Andrew. But this is clearly the wrong pronunciation! :P)
Taiizor 9 hours ago [-]
[flagged]
jheriko 1 days ago [-]
[dead]
1 days ago [-]
Rendered at 18:57:03 GMT+0000 (Coordinated Universal Time) with Vercel.
Turns out, the reason people wanted to use it was because the API was stable and every slight step towards improvement was met with... resistance. I stopped maintaining it because the burnout from seeing what it could be and not seeing any path to get there was too much for me.
I'm extremely grateful to burntsushi for writing the datetime library rust deserves. Chrono is stable and is maintained (though semi-deprecated) by an extremely reliable rustacean -- you don't need to leave, but you'll probably be happier if you do.
Thank you so much for building something that was obviously good enough to sustain the community for 12 years <3
Unless you're talking about Arrow, the Apache Parquet interface library in either Python or Rust - that's a different matter; great software!
In any other case I can think of, you need to attach the time to a date and a timezone. You can display it bare in the context of the viewer (who has an implicit or explicit timezone and time reference) but any manipulation without these will not end well.
Would you have some examples for the poor representation of data and surprise behaviour?
What datetime is your annual birthday celebration? When's Christmas this year? What time is your Monday alarm clock set for? Wen's the daily standup?
There are so many, and sometimes Arrow/Moment's abstraction is OK. Other times it's awkward as it overconstrains. Sometimes it has surprising behavior.
Or, ask yourself why Jiff and Chrono, and python datetime etc have separate date, time, and datetime types.
Datetimes are useful, and I use them more than individual dates and times. The latter 2 are still useful, and datetime is a sloppy or buggy fit.
My best guess of your (and the arrow and moment creator's mentality here): They have been hit by bugs of the kind you're saying, involving timezones, or things that should be datetimes + timezone, but were using native datetimes, or dates etc. The response was to use datetimes for everything, ignoring other cases. Does that sound accurate?
I've used it in several projects, and to me it seems just fine. Granted, I don't do a lot of cross-time zone work; I exclusively use UTC whenever possible.
(Putting my flame suit on here, but...) I think that's a good thing, and I wish this were the norm with more libraries in the open source world. Once you hit a stable version, breaking changes should happen approximately never. Improvements that need breaking changes should happen under a new name, with a new project, so that the old project can still exist and receive updates.
I still say SemVer has done more harm than good for the industry... it normalized the idea that "it's ok to break compatibility, just bump the major version" and then you have libraries on version 26 that break everything every time the author decides to feng shuei the code some more. Every major version bump should be thought of as a sort of "grave marker" indicating a place in history where the author messed up and had to admit the old design was literally broken, and every downstream dependent had to go through some amount of pain as a result. It shouldn't be normalized.
You can say "nobody's forcing you to upgrade" but, with the amount of vulnerabilities out there, yes, you often are forced to upgrade certain dependencies.
You can say "just maintain the older versions too", and that would be absolutely great, if anyone really did this. In practice anything but the most recent major is neglected.
You can say "I agree and that's why I never release 1.0", and I'd say that's a reasonable approach, but that just changes the problem to "too many packages are pre-1.0" and we'd have the same issue.
You can say "that degree of compatibility is too difficult to expect out of open source maintainers", and my response is that there ought to be fewer open source maintainers. (For real. I'd rather have 1/100th the number of crates in Cargo, if the ones that were there were all maintained by people who were serious about backwards compatibility.)
You say "the API sucked and I hoped I could make it better", but the API is the library to anyone using it, and changing it means you're literally changing what the library is, which means all existing code using the library has to change to meet your new opinion on what the better API is (an opinion many wouldn't share, including and especially the original authors.) Why not just start a new project instead?
I really appreciate when a project calmly acknowledges that it's not the right tool for every use case, and is not trying to be. That's professional engineering.
Especially in the AI-age, the crucial skill will be what NOT to implement.
Outside of this not using a timer library is going for a world of pain
> Unlike “Comparison with other Rust datetime crates,” this document is opinionated. > [...] > Jiff is pronounced like “gif” with a soft “g,” as in “gem.”
Burntsushi really did not hold back with this document being opinionated :-) (Though they are right, of course).
[0] https://docs.rs/jiff/latest/jiff/_documentation/design/index...
In the very early Go days (<0.9) he responded to a comment of mine going into why Go was worth looking into, quite passionately and enthusiastically which set me off on learning it and where it could be applied and that changed the kinds of roles I went after.
Ever since it's a bright little nugget whenever I see his name pop up and have tried to follow his work ever since.
Time is such a pain, I write code for simulating orbits of asteroids/stuff in the solar system, and time and time conversions play a major roll in it. I often need accuracy at the mega-year to giga-year scales, but also sometimes micro-second relative.
I saw on this page that you at least mention TAI time, how are you handling time scales in general? Like would it be hard to add some other timescales like UT1?
Right now I have to do all of this manually and I have been using chrono as just a convenience wrapper so that my users can convert ISO time strings into my internal representation.
Edit: Oh no! I just found out chrono is being semi-deprecated, looks like I'm going to be switching...
Ah yeah hifitime, I tried it out but its internal representation doesn't work well for my long term simulations, the integer representation ends up increasing the numerical noise in my simulations when I do really sensitive long term ones.
[1] https://docs.rs/jiff-tzdb/latest/jiff_tzdb/#data-generation
[2] https://github.com/BurntSushi/jiff/issues/278
[3] https://github.com/BurntSushi/jiff/issues/279
The C++ date-time library goes and loads the system copy of ICU (which is updated via windows update) and pulls the tz data from ICU's calendar api.
Hopefully Safari will soon have temporal implemented and we can leave that behind (it's in 27.3b so not out of the question).
> Chrono actually is comparatively nice
burntsushi has written pretty extensively about:
1. comparisons to existing datetime libraries (https://docs.rs/jiff/latest/jiff/_documentation/comparison/i...), chrono's original author also commented (https://github.com/BurntSushi/jiff/issues/63)
2. why they felt it was a good idea to write a new library (https://docs.rs/jiff/latest/jiff/_documentation/design/index...)
3. why they didn't wish to contribute to either chrono or time (https://docs.rs/jiff/latest/jiff/_documentation/design/index...)
And as randomblock linked, chrono is deprecating itself.
This is still very surprising given the context; if the top datetime library of any language is getting soft-replaced with something nicer, woe be to those in other languages!
I found switching to Jiff from Chrono quite easy.
The alternative is to keep a relatively small standard library, and a good package manager, to make it easier for people to iterate on API design in third party packages. Major version changes in a single library can have much less impact accross the ecosystem than standard library packages. Eventually you may get a de facto standard, and sometimes they are then later incorporated into the standard library.
This model isn't without its downsides; it can sometimes mean fragmentation when standardization would help, sometimes these de facto standard libraries can become important enough that they run into the same ecosystem fragmentation issues on major releases. But it allows for more flexibility.
The latter model is what the Rust project decided on for everything outside of the most fundamental time types. Dates and times and timezones are notoriously hard to model in code, and easy to create API designs that are full of footguns, so this is a place where that extra freedom to iterate was more important than standardizing on an API early on.
That iteration has led to a number of different libraries, including time, chrono, and now jiff, allowing for exploring the space of API design to really get it right rather than being stuck with what was available early on.
Of course, Go can always backtrack and expose the primitives, and Rust can always add a higher-level API, but as user of the languages, I'd rather have the latter problem than the former. It doesn't help that Go's zoned-time API isn't very good either.
Temporal does improve some things (it has more strict zoning / less defaults, and improves serialization through RFC 9557), but some of the divergences are not necessarily improvements and rather divergences of opinion (e.g. java.time wilfully made non-ISO calendars second-class, Temporal makes them first class more like Joda, and it's one of the things Colebourne — the lead author of Joda and java.time — strongly disagrees with[1], but the Temporal authors have pretty compelling reasons for it[2]).
[1] https://blog.joda.org/2009/11/why-jsr-310-isn-joda-time_4941... https://github.com/tc39/proposal-temporal/issues/5 https://github.com/tc39/proposal-temporal/issues/1454
[2] https://tc39.es/proposal-temporal/docs/calendar-review#why-a...
Not at all. We just want to avoid having the standard library be "where code goes to die", as Python folks have put it.
We're working on adding more popular APIs from the ecosystem to the standard library. See https://github.com/rust-lang/std-replacement-data/ for a data set tracking what crates have been obsoleted by additions to the standard library; we hope to increase that over time.
> disparate documentation
I've never experienced this while using a Rust library. They all use rustdoc hosted on docs.rs (or locally with $ cargo doc). They all look identical and there's a single global search bar that searches all your dependencies.
> conflicting dependencies
Since you can install two different versions of the same library, this is only an issue if two of your dependencies both expose a transitive dependency with different versions in their external APIs. (E.g. libA returns a Foo 1.0 and libB expects an argument of Foo 2.0) Only library I've ever run into this with is raw-window-handle.
> supply chain risks
> enormous .cargo
These I agree with. Although honestly the supply chain risk is not as big with these foundational libraries as it would first appear, since they are almost all maintained by the same 5 rust compiler developers. Really the only difference is that they're not in the rust-lang github organization and not shipped precompiled with the standard library
Documentation certainly has nothing to do with it.
Other factors you mentioned are more convincing.
For a systems programming language, Rust is relatively easy to write applications in -- but I suspect standard library choices (and other choices) will be driven by the systems side for the foreseeable future.
> Jiff is pronounced like “gif”
This would have been a hilarious line if he left off the second part!
> Yeah I had a really hard time making the jiff-diesel and jiff-sqlx crates. Part of the problem is the very poor datetime support in the underlying databases. The other part of the problem is that I'm not a user of either diesel or sqlx, so there are some unknown unknowns for me.
https://www.reddit.com/r/rust/comments/1xx7v4v/comment/19sp2...
So I’m sure they’d be glad for assistance.
Strings are also similarly gnarly topic that I find a lot of devs try to ignore.
I really, really, hate Monday
Woah now, let's not get too controversial.
Graphics Interchange Format. Guh-raphics. GIF
Fun linguistics experience for native English speakers: hold an open palm in front of your mouth, fairly close, and say 'spit', then say 'pit'. Much more air hits your hand when saying 'pit' than when saying 'spit'. That's the aspiration.
> Jiff is pronounced like “gif” with a soft “g,” as in “gem.”
This is classic anti-trust monopoly leveraging. Lucky I'm on this side of the debate :)
A masterclass in viral outrage bait. Not a Rustacean, I clicked the link exclusively for this nugget and will probably tell my Rust-loving colleagues about it tomorrow.
For the record, I like the converse better:
> Jiff is pronounced with a hard “g”, as in “gif”.
Giraffical Interchange Format.
Thank you for your work on this burntsushi!
The first problem is the bigger problem. If you are writing an app that asks the user for a zoned time, the user will want something like Nov 16 7pm eastern time. But EST is of course highly ambiguous and so is EDT. (The database actually gives EST to Panama Time and leaves EDT undefined.) The correct solution is either EST5EDT or America/New_York. Neither is user friendly. The first string EST5EDT looks like gibberish to the average user. The second raises concerns in all other cities sharing the same timezone as New York. Why not have an entry for Philadelphia or Washington D.C.? Some even say it is inherently classist to pick the city with the biggest population, not the most important by some other metric. Is New York more deserving than our nation’s political capital? Why is Los Angeles more deserving than San Francisco, other than the fact that the maintainer of the database is physically in Los Angeles? This is an actual criticism in iOS calendar app on the now defunct Apple forums. If you want the user to pick a timezone you better make sure Eastern Time is an allowed option, and maybe also include all the cities and municipalities in the zone; that’s what the newest iOS does. But in any case almost every developer now hard codes a mapping from Eastern Time to America/New_York if their audience is U.S. This feels like duplicative work better done in a library. The fact that a new library reaches 1.0 and doesn’t plan to add this might mean a lack of ambition to tackle this problem.
The second problem is the fact that the timezone definitions change according to the actual rules in effect in history. This is sometimes not what you want. Indeed I notice that most if not all date libraries use the proleptic Gregorian calendar (our current calendar system extended backwards); no library that I know of automatically switches to Julian calendars before the cutoff. Why this inconsistency? I’d argue that when most humans express a zoned time they expect the current rules to be extended backwards. People forget the daylight saving rules change in the 2000s when talking about times before that change and expect the current rules.
Not to be a detractor here for this library given the hard work of burntsushi, but we should aim higher.
While not necessarily wrong, it should probably be a separate library entirely: there are projects maintaining GIS boundaries for the IANA timezones.
I found the icu::TimeZoneNames class. The docstring even describes my exact problem:
> For example, people in the United States refer the zone used by the east part of North America as "Eastern Time". The tz database contains multiple time zones "America/New_York", "America/Detroit", "America/Montreal" and some others that belong to "Eastern Time". However, assigning different display names to these time zones does not make much sense for most of people.
I then looked at the icu::TimeZoneNames::find member function and it seemed to do exactly what I proposed. https://unicode-org.github.io/icu-docs/apidoc/dev/icu4c/clas... Except it’s marked as internal use only. I don’t know what’s going on.
> The second problem is the fact that the timezone definitions change according to the actual rules in effect in history. This is sometimes not what you want. Indeed I notice that sometimes all date libraries use the proleptic Gregorian calendar meaning our current calendar system extended backwards; no library that I know of automatically switches to Julian calendars before the cutoff. Why this inconsistency? I’d argue that when most humans express a zoned time they expect the current rules to be extended backwards. People forget the daylight saving rules change in the 2000s when talking about times before that change and expect the current rules.
Good question. One part of this answer is that the IANA tzdb does not only contain daylight saving time transitions. It also includes transitions from one offset to another, perhaps because daylight saving time was stopped or started. The other part of this answer is, I would guess, more pragmatic. Namely, we have structured data delineating time zone transitions back at least a century. If you go far enough back, then indeed, time zone transitions stop happening. We, however, do not, as far as I'm aware, have shared structured data delineating changes in calendar. This would also need to be organized by both location and time. For example, the Julian calendar was not dropped in favor of the Gregorian instantaneously everywhere. Some countries, like Russia, continued using the Julian calendar into the 20th century.
An other fun one is that some of the transition schemes were... imaginative, let's say. Famously Sweden didn't want a hard break so in 1699 they decided that they would just skip leap days until they were on Gregorian. So they skipped the one in 1700, then promptly forgot the 1704 and 1708 ones because they were busy with the Great Northern War, as a result the scheme was cancelled by reverting to the Julian calendar, which they did by adding a second leap day in 1712 (which thus had a February 30th, and 367 days). Sweden ended up doing the switch at once in 1753.
That's calendaring rather than timezoning tho. Timezones really only became a factor once timekeeping was precise enough and times had to be communicated and synchronised across vast distances, which really came about with the railways (first formal "time zone" was the Railway Time of England's Great Western Railway). Before that, cities and villages oft had their individual time based on apparent solar time at the location (aka a sundial). For instance Paris Mean Time was 9 minutes and 21 seconds ahead of London.
Do you seasoned rustaceans may explain why we don’t have such a standard library?
It’s a genuine question from an ignorant :)
Just look at the original java time stuff Vs those that came in java 8.
And once something is in the standard library it can never change.
What a cardinal sin! Instantly distrust this project.
(I'm kidding, Andrew. But this is clearly the wrong pronunciation! :P)