I've run into a behaviour change that was actually already present in alpha20 – with the CLJ-99 patch in place, {min,max}-key now return the first argument with the minimum/maximum key, whereas previously they returned the last such argument.
The new behaviour seems like the more natural one, but this is a breaking change, so I filed https://dev.clojure.org/jira/browse/CLJ-2247 to track this (with a patch that takes the "default" approach of restoring established behaviour). Cheers, Michał On 3 October 2017 at 21:11, Beau Fabry <imf...@gmail.com> wrote: > We've been using 1.9 in a small app for a while with no issues. After > upgrading schema to the latest version (with the PR above) I've also > successfully run our larger codebase with 1.9. > > On Tuesday, October 3, 2017 at 4:41:14 AM UTC-7, stuart....@gmail.com > wrote: >> >> Hi Mark, >> >> I think this approach totally makes sense, and the alpha naming exists to >> inform this kind of decision-making. >> >> For libraries where the use of spec does not have to be user-facing, I am >> putting specs in separate (Clojure) namespaces, and loading them in such a >> way that they can coexist with non (or maybe different) spec environments. >> But that is extra work for sure. >> >> Stu >> >> On Mon, Oct 2, 2017 at 3:35 PM, Mark Engelberg <mark.en...@gmail.com> >> wrote: >> >>> On Mon, Oct 2, 2017 at 7:55 AM, Stuart Halloway <stuart....@gmail.com> >>> wrote: >>> >>>> Hi David, >>>> >>>> Spec will be in alpha for a while. That is part of the point of it >>>> being a separate library. Can you say more about what problems this is >>>> causing? >>>> >>>> Stu >>>> >>>> >>> As a library maintainer, I am forced to upgrade and release my library >>> any time something I depend upon makes a breaking change. I don't get paid >>> for maintaining open source libraries, it's something I do in my spare >>> time, so I prefer to do it on my own schedule. When an underlying library >>> makes a breaking change, I get dozens of urgent requests from people who >>> need me to cut a new release ASAP, and by Murphy's Law, that often happens >>> when I have very little time to do it. It's a nuisance. >>> >>> Clojure is pretty good about not making breaking changes, but it happens >>> from time to time. Clojurescript is less good about not making breaking >>> changes, and therefore, maintaining Clojurescript libraries is more of a >>> headache. On the plus side, Clojurescript users seem to care very little >>> about backwards compatibility (most keep up with the latest version), so >>> sometimes it is easier to make a change to keep up with a change in >>> Clojurescript than one in Clojure, where I am expected to not only support >>> the latest breaking change, but also the last several releases. >>> >>> Anything that is labeled as "alpha" is waving a big red flag that there >>> could be breaking changes at any time with little warning. For my >>> libraries which depend on spec, there's no way I'm going to bring them out >>> of alpha status until spec comes out of alpha status. If I make an >>> official release of something that depends on spec, then I'm going to be on >>> the hook to rapidly cut a new release every time spec changes, which could >>> be at any time. I don't want that hassle. I don't want to make a promise >>> to the community to maintain a stable product if the thing I depend upon >>> has not made a similar promise. When spec reaches a point where the API >>> will not be changing, or rather, when we know that new changes will only be >>> additive, I can begin to trust that it won't be a huge maintenance headache >>> to release something based on spec. >>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "Clojure" group. >>> To post to this group, send email to clo...@googlegroups.com >>> Note that posts from new members are moderated - please be patient with >>> your first post. >>> To unsubscribe from this group, send email to >>> clojure+u...@googlegroups.com >>> For more options, visit this group at >>> http://groups.google.com/group/clojure?hl=en >>> --- >>> You received this message because you are subscribed to the Google >>> Groups "Clojure" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to clojure+u...@googlegroups.com. >>> For more options, visit https://groups.google.com/d/optout. >>> >> >> -- > You received this message because you are subscribed to the Google > Groups "Clojure" group. > To post to this group, send email to clojure@googlegroups.com > Note that posts from new members are moderated - please be patient with > your first post. > To unsubscribe from this group, send email to > clojure+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/clojure?hl=en > --- > You received this message because you are subscribed to the Google Groups > "Clojure" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to clojure+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.