Error Precompiling PyPlot in latest Julia 1.0.0-rc1.1

I just installed the latest Julia 1.0.0-rc1.1 and stumbled upon the following errors when installing PyPlot:

julia> using PyPlot
[ Info: Precompiling PyPlot [d330b81b-6aea-500a-939a-2ce795aea3ee]
WARNING: could not import Base.done into PyCall
WARNING: could not import Base.mimewritable into PyPlot
WARNING: could not import Base.linspace into Colors
ERROR: LoadError: UndefVarError: done not defined
Stacktrace:
 [1] getproperty(::Module, ::Symbol) at .\sysimg.jl:13
 [2] top-level scope at none:0
 [3] include at .\boot.jl:317 [inlined]
 [4] include_relative(::Module, ::String) at .\loading.jl:1038
 [5] include(::Module, ::String) at .\sysimg.jl:29
 [6] top-level scope at none:2
 [7] eval at .\boot.jl:319 [inlined]
 [8] eval(::Expr) at .\client.jl:389
 [9] top-level scope at .\none:3
in expression starting at C:\Users\Seif\.julia\packages\LaTeXStrings\qycA\src\LaTeXStrings.jl:64
ERROR: LoadError: Failed to precompile LaTeXStrings [b964fa9f-0449-5b57-a5c2-d3ea65f4040f] to C:\Users\Seif\.julia\compiled\v1.0\LaTeXStrings\H4HGh.ji.
Stacktrace:
 [1] error(::String) at .\error.jl:33
 [2] macro expansion at .\logging.jl:313 [inlined]
 [3] compilecache(::Base.PkgId, ::String) at .\loading.jl:1184
 [4] _require(::Base.PkgId) at .\logging.jl:311
 [5] require(::Base.PkgId) at .\loading.jl:852
 [6] macro expansion at .\logging.jl:311 [inlined]
 [7] require(::Module, ::Symbol) at .\loading.jl:834
 [8] include at .\boot.jl:317 [inlined]
 [9] include_relative(::Module, ::String) at .\loading.jl:1038
 [10] include(::Module, ::String) at .\sysimg.jl:29
 [11] top-level scope at none:2
 [12] eval at .\boot.jl:319 [inlined]
 [13] eval(::Expr) at .\client.jl:389
 [14] top-level scope at .\none:3
in expression starting at C:\Users\Seif\.julia\packages\PyPlot\jXCX\src\PyPlot.jl:295
ERROR: Failed to precompile PyPlot [d330b81b-6aea-500a-939a-2ce795aea3ee] to C:\Users\Seif\.julia\compiled\v1.0\PyPlot\oatAj.ji.
Stacktrace:
 [1] error(::String) at .\error.jl:33
 [2] macro expansion at .\logging.jl:313 [inlined]
 [3] compilecache(::Base.PkgId, ::String) at .\loading.jl:1184
 [4] _require(::Base.PkgId) at .\logging.jl:311
 [5] require(::Base.PkgId) at .\loading.jl:852
 [6] macro expansion at .\logging.jl:311 [inlined]
 [7] require(::Module, ::Symbol) at .\loading.jl:834

Does this mean that PyPlot is not ready for 1.0-rc1 yet and needs fixes or some other thing on my side?

The former, PyPlot is not yet ready for 1.0. We’ll publish and maintain a list of packages that are 1.0-ready once 1.0 is out.

Thank you Stefan, It is a nice idea to have a link listing 1.0 ready packages.

I removed the tag “First Steps” since using the v1.0 release client is not something a beginner user should be doing. The release client announcement noted that v1.0-RC is the wild west for package developers to find out what actually works: not a good place to be unless you’re planning to PR to fix things. v0.7 is much safer since instead of erroring you’ll get some depwarns.

3 Likes

Thanks, Chris, I hesitated a lot before choosing the tag and initially wrote Usage. Totally agree.

No worries, I just wanted to make sure it’s clear for anyone who stumbles on this.

So, why you released 1.0 ? I tought it would have come a bit (few months) after 0.7…
Sorry guys, but I personally think from a user point of view it has been a mistake.
I just installed it and come out with tons of small errors… Julia 1.0 is not ready for beginners, not of course for Julia itself, but for its pkg ecosystem…

My issue has been solved by just running ] up after only two days of my original post. Can you give me an example of a language with similar activity or speed of improvement? Now, it’s only 8 days since the release of Julia 1.0 and almost all major packages have been updated. If you want to live with a cutting-edge version, bear the small pain that comes with it only a month or so, or a previous version will be fine for you. I don’t think releasing 1.0 was a mistake, if they didn’t, we would’ve been waiting forever. It’s just a few months and every thing will be perfect. That was the right thing to do in the long run.

1 Like

yes, but users read the news on reddit or even less techy sites in the hours/days after the annunce, try it all excited (2 guys in my lab did it), can’t do basic things and abbandon… even if in a couple of weeks everything works, it is already too late, it’s too late for people non involved in the project…
In the eyes of many users 1.0 shoudn’t be a cutting edge version, but a reasonable “stable” one (and it is, the problem is mostly in releasing the updated packages, I agree that it is a matter of few days)

1 Like