Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-09-21 Thread 'bed...@yahoo.com' via Clojure
This is excellent news! Thank you very much! On Monday, September 21, 2020 at 7:39:25 PM UTC-7 Alex Miller wrote: > These changes are in the new 1.10.1.693 prerelease of clj if you want to > test it. Assuming things look good, I'll promote that to a stable release > soon. > > In this release,

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-09-21 Thread Alex Miller
These changes are in the new 1.10.1.693 prerelease of clj if you want to test it. Assuming things look good, I'll promote that to a stable release soon. In this release, things will be ordered :extra-paths (in order, if used), then :paths (in order, if used), then libs. That reverts the changes

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-09-14 Thread 'Alex Miller' via Clojure
Just FYI, we have a plan to address this and it should be in the next stable version. > On Sep 14, 2020, at 1:00 AM, 'bed...@yahoo.com' via Clojure > wrote: > >  > I couldn't agree more. > > It really boils down to the simple fact that class loading in the JVM - for > the standard classload

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-09-13 Thread 'bed...@yahoo.com' via Clojure
I couldn't agree more. It really boils down to the simple fact that class loading in the JVM - for the standard classloaders - is well-defined. If a build tool is not able to reflect this, it is an unreliable build tool. It doesn't matter if anyone thinks CLASSPATH order shouldn't matter from

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-09-13 Thread Matching Socks
Too fragile. This reminds me of the notion of "situated programming", featured in the talk by Rich Hickey: you and your programs operate in the middle of a bizarre and changing situation. For Clojure, the Java ecosystem is part of that situation. Even if some jars do not overlap today, you

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-12 Thread Sean Corfield
Alan, > 4. I could possibly try to replicate your proposed experiment explicitly, but I no longer have easy access to 1.10.1.645 since Homebrew has been fixed. I did find the `brew-install` repo on GH, but am not certain how to replicate the broken install of *.645. Per the clojure/homebrew-tool

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-12 Thread Alan Thompson
Hi Alex, 1. Great news that the Homebrew team has responded to your request to point only to stable versions. 2. The resources directory is the contains the path `resources/public/index.html`. The local one is definitely not the one being served in my original example (2nd half) re 1.10.1.645 wh

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-12 Thread Sean Corfield
I'd like to point out that tools.deps.alpha can build a differently-ordered classpath from the exact same input files when run on a different machine (i.e., the same source/deps, the same version of t.d.a./CLI). This has bitten me a couple of times with depstar which I use for building JARs/uberja

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-12 Thread Thomas Heller
> > None of these libraries are broken. They just include resources. Also, I > don't think it is realistic to tell library authors to please move certain > files out of the way because my build tool randomizes my classpath. That is > not going to happen. People will keep including things like

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'bed...@yahoo.com' via Clojure
Ok, let's reset as we start talking about different things now. None of these libraries are broken. They just include resources. Also, I don't think it is realistic to tell library authors to please move certain files out of the way because my build tool randomizes my classpath. That is not goi

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'Alex Miller' via Clojure
FYI, I've been working with the homebrew team today and homebrew-core now points to the latest stable clojure tools version and will track that automatically going forward. -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'Alex Miller' via Clojure
On Tue, Aug 11, 2020 at 6:41 PM 'bed...@yahoo.com' via Clojure < clojure@googlegroups.com> wrote: > Just 2 quick points before I go back to migrate to shadow-cljs & leiningen > ;) > > "just does not seem well defined " > This is not a line of argument you want to pursue when we are talking > about

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'Alex Miller' via Clojure
Can you change that resources file and see if what you're looking at changes to double check? I did actually check a bunch of jars from the prior message and did some searching for that message. Or you could even dump the classpath with -Spath, then move resources to the front, then use -Scp (which

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'bed...@yahoo.com' via Clojure
Just 2 quick points before I go back to migrate to shadow-cljs & leiningen ;) "just does not seem well defined " This is not a line of argument you want to pursue when we are talking about maven, who has had a stable order for what feels like decades now. If you need a current link: https://mav

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread Alan Thompson
Hi - I just tried your suggestion and no joy: ~/work/tmp810/xanadu > clj -e "((requiring-resolve ' clojure.java.io/resource) \"public/index.html\")" DEPRECATED: Libs must be qualified, change deps-ancient => deps-ancient/deps-ancient (deps.edn) DEPRECATED: Libs must be qualified, change reagent

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'Alex Miller' via Clojure
On Tue, Aug 11, 2020 at 3:01 PM 'bed...@yahoo.com' via Clojure < clojure@googlegroups.com> wrote: > Here's some maven-specific discussion: > https://stackoverflow.com/questions/793054/maven-classpath-order-issue. > They have a defined order since 2.0.9. and declaration order is considered > for tr

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'bed...@yahoo.com' via Clojure
Here's some maven-specific discussion: https://stackoverflow.com/questions/793054/maven-classpath-order-issue. They have a defined order since 2.0.9. and declaration order is considered for transitive dependencies conflict. Intellij's Dependencies tab in Module settings: You can re-order the de

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'Alex Miller' via Clojure
Just to beat the horse... On Tue, Aug 11, 2020 at 1:05 PM 'bed...@yahoo.com' via Clojure < clojure@googlegroups.com> wrote: > Changing the order of classpath entry and not having paths be at the > beginning of the CLASSPATH will break existing code that uses `(io/resource > ...)` to load resource

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread 'bed...@yahoo.com' via Clojure
Changing the order of classpath entry and not having paths be at the beginning of the CLASSPATH will break existing code that uses `(io/resource ...)` to load resources from the CLASSPATH. While in theory the order of the classpath should not matter, in practice it does, often involuntarily. Fig

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-11 Thread Thomas Heller
This might be a good incentive for people to keep their published .jar files clean. Unfortunately many published CLJS libs contain the rather common "public" folder with "public/index.html" and often compiled .js artifacts which aren't actually ever used. I do however think that it is useful to

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-10 Thread 'Alex Miller' via Clojure
Bunch of things here... Clojure maintains its own brew tap and a "stable" release that you can obtain with `brew install clojure/tools/clojure` (the brew conventions automatically find the prior repo based on that). That tap also includes prerelease unstable versions that can be obtained with "

Re: Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-10 Thread Alan Thompson
P.S. There seems to be no *`clojure --version`* flag. Should this be added to the command line tool? On Mon, Aug 10, 2020 at 4:58 PM Alan Thompson wrote: > Hi. Just helped a colleague debug a vexing problem on a CLJS project > using Figwheel.Main. > > If we do *`brew install clojure/tools/cl

Classpath bug re Clojure 1.10.1.645 when using Figwheel.Main

2020-08-10 Thread Alan Thompson
Hi. Just helped a colleague debug a vexing problem on a CLJS project using Figwheel.Main. If we do *`brew install clojure/tools/clojure`*, it works: ~/work/tmp810/xanadu > clj --help Version: *1.10.1.561* Usage: clojure [dep-opt*] [--] [init-opt*] [main-opt] [arg*] clj [dep-opt*] [--