`pip install squigglepy` fails in Google Colab notebook

[erwald]

How to reproduce:

1. Create a new Colab notebook
2. Run !pip install squigglepy in a cell.

I'd expect squigglepy to get installed (this used to work). Instead I get this:

Collecting squigglepy
  Downloading squigglepy-0.27-py3-none-any.whl (44 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 44.6/44.6 kB 1.4 MB/s eta 0:00:00
Collecting msgspec<0.16.0,>=0.15.1 (from squigglepy)
  Downloading msgspec-0.15.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (183 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 183.3/183.3 kB 5.6 MB/s eta 0:00:00
Collecting numpy<2.0.0,>=1.24.3 (from squigglepy)
  Downloading numpy-1.26.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (18.2 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 18.2/18.2 MB 74.5 MB/s eta 0:00:00
Collecting pathos<0.4.0,>=0.3.0 (from squigglepy)
  Downloading pathos-0.3.1-py3-none-any.whl (82 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 82.1/82.1 kB 7.8 MB/s eta 0:00:00
Requirement already satisfied: scipy<2.0.0,>=1.10.1 in /usr/local/lib/python3.10/dist-packages (from squigglepy) (1.11.3)
Requirement already satisfied: tqdm<5.0.0,>=4.65.0 in /usr/local/lib/python3.10/dist-packages (from squigglepy) (4.66.1)
Collecting ppft>=1.7.6.7 (from pathos<0.4.0,>=0.3.0->squigglepy)
  Downloading ppft-1.7.6.7-py3-none-any.whl (56 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 56.8/56.8 kB 3.9 MB/s eta 0:00:00
Collecting dill>=0.3.7 (from pathos<0.4.0,>=0.3.0->squigglepy)
  Downloading dill-0.3.7-py3-none-any.whl (115 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 115.3/115.3 kB 10.0 MB/s eta 0:00:00
Collecting pox>=0.3.3 (from pathos<0.4.0,>=0.3.0->squigglepy)
  Downloading pox-0.3.3-py3-none-any.whl (29 kB)
Collecting multiprocess>=0.70.15 (from pathos<0.4.0,>=0.3.0->squigglepy)
  Downloading multiprocess-0.70.15-py310-none-any.whl (134 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 134.8/134.8 kB 14.2 MB/s eta 0:00:00
Installing collected packages: ppft, pox, numpy, msgspec, dill, multiprocess, pathos, squigglepy
  Attempting uninstall: numpy
    Found existing installation: numpy 1.23.5
    Uninstalling numpy-1.23.5:
      Successfully uninstalled numpy-1.23.5
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
lida 0.0.10 requires fastapi, which is not installed.
lida 0.0.10 requires kaleido, which is not installed.
lida 0.0.10 requires python-multipart, which is not installed.
lida 0.0.10 requires uvicorn, which is not installed.
cupy-cuda11x 11.0.0 requires numpy<1.26,>=1.20, but you have numpy 1.26.1 which is incompatible.
numba 0.56.4 requires numpy<1.24,>=1.18, but you have numpy 1.26.1 which is incompatible.
Successfully installed dill-0.3.7 msgspec-0.15.1 multiprocess-0.70.15 numpy-1.26.1 pathos-0.3.1 pox-0.3.3 ppft-1.7.6.7 squigglepy-0.27

And if I then run import squigglepy in a cell, I get:

---------------------------------------------------------------------------

AttributeError                            Traceback (most recent call last)

[<ipython-input-2-e8bb0f348dea>](https://localhost:8080/#) in <cell line: 1>()
----> 1 import squigglepy

5 frames

[/usr/local/lib/python3.10/dist-packages/numpy/testing/_private/utils.py](https://localhost:8080/#) in <module>
     55 IS_PYSTON = hasattr(sys, "pyston_version_info")
     56 HAS_REFCOUNT = getattr(sys, 'getrefcount', None) is not None and not IS_PYSTON
---> 57 HAS_LAPACK64 = numpy.linalg._umath_linalg._ilp64
     58 
     59 _OLD_PROMOTION = lambda: np._get_promotion_state() == 'legacy'

AttributeError: module 'numpy.linalg._umath_linalg' has no attribute '_ilp64'

[peterhurford]

@erwald After you pip install numpy in the Google colab notebook, try restarting the runtime (Runtime -> restart session). This got it to work for me.

[erwald]

@peterhurford this is fixed now, right? I'll close.