@fabrao fix for the error: https://github.com/mauricioszabo/repl-tooling/pull/67. Will publish a new version soon-ish 🙂
so, what is the origin of problem?
To be honest, .getPath
of Java returns a /
on the beginning of the file.
Have no idea why other people didn't have this problem. I assume most people are working in MacOS or Linux 😄
It can be !!!
Also, on load-file
the "origin" of the file gets corrected: is only when Java needs to resolve the filename (which is necessary when we use "refresh")
So, I´m now Windows Chlorine bug finder
Thanks
@fabrao Just published a new version. Thanks for helping me debug this issue, just tested on windows and seems that its working now 🙂
When do you think you´ll start with debug?
What do you want in a debugger?
(I think step debuggers are a terrible idea)
https://clojurians.slack.com/archives/CF02KPG4U/p1586403154136900
I tested the fixes from new version and it´s ok now
I used sometimes cider-debug-defun-at-point
in Emacs
it´s like gdb
Probably not very soon. There are a lot of things I want to do before: get rid of UNREPL, better support for parsing big EDNs, avoid locking Atom when too much text is injected, and so on....
I think debugger is a high level of complexity, correct?
Yeah, debuggers are really hard to get right -- although any "Clojure" debugger is really just building on top of the existing JVM debugging system.
But I really think they're a wrong-headed approach to solving problems. I see a lot of people, in various languages, reach for a step debugger/breakpoints without actually thinking scientifically about what the problem might be and how to diagnose it. Step debuggers encourage random tinkering 😞
i appreciate having the option though to examine the actual values. we don't always have the luxury of interfacing with code we have written ourselves.
I only use it if I´m tired and having to solve some problem
REPL all the things 🙂