segfault after importing CSV and Images, PyPlot from a separate module
- 主要语言
- Julia
- 星标
- 488
- 派生
- 90
- PR 合并指标
- 30 天内没有已合并 PR
描述
Hi,
I get the following segfault
```
signal (11): Segmentation fault: 11
in expression starting at /MyProjectDirectory/Test.jl:4
_platform_strcmp at /usr/lib/system/libsystem_platform.dylib (unknown line)
```
when I run `/MyProjectDirectory/Test.jl` which contains the code
```
push!(LOAD_PATH, "/MyProjectDirectory")
import CSV
using MyModule
```
that uses `/MyProjectDirectory/MyModule.jl` which contains the code
```
module MyModule
using Images
import PyPlot
end
```
I don't know if this is an issue to do with CSV, Images, PyPlot or Julia itself. This is a minimal working example. I'm using Julia 1.0.1 on MacOS Mojave. The segfault goes away if I don't load any one of these three packages. There is still a segault if I start Julia with `--check-bounds=yes`.
I hope this can be resolved soon! Segfaults are frightening...
These are the versions of the packages in question + important dependencies
```
Images v0.16.0
PyCall v1.18.5+ #Master (https://github.com/JuliaPy/PyCall.jl.git)
PyPlot v2.6.3
CSV v0.4.1
```
I have also posted this issue on the [`Images.jl` github](https://github.com/JuliaImages/Images.jl/issues/759) and the [`CSV.jl` github](https://github.com/JuliaData/CSV.jl/issues/328).
Best,
Max
贡献指南
这个仓库没有索引到贡献指南
调研方向
Reproduce the minimal example in /MyProjectDirectory/Test.jl with the module in /MyProjectDirectory/MyModule.jl on Julia 1.0.1 and macOS Mojave. Start by testing the CSV, Images, and PyPlot imports individually and in combination, then compare the package versions listed in the issue and review the linked Images.jl and CSV.jl reports. Done means identifying a reproducible package interaction and documenting which component owns the failure.
由索引模型根据 Issue 内容生成。
评估
- 技术栈
- julia, python
- 领域
- data-visualization
- Issue 类型
- 缺陷
- 难度
- 4/5
- 预计耗时
- 3-5 天
- 活跃度
- 停滞
- 描述清晰度
- 需要澄清
- 新手友好度
- 25/100