Fail to import speasy

[Beforerr]

I did not know what changed but suddenly I could not import speasy (in a new environment).

julia> Speasy.PythonCall.pyimport("speasy")
ERROR: Python: AttributeError: 'CacheItem' object has no attribute 'spz_name'
Python stacktrace:
 [1] load_inventory_file
   @ speasy.data_providers.generic_archive ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/data_providers/generic_archive/__init__.py:72
 [2] build_inventory
   @ speasy.data_providers.generic_archive ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/data_providers/generic_archive/__init__.py:89
 [3] _inventory
   @ speasy.core.dataprovider ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/core/dataprovider.py:55
 [4] wrapped
   @ speasy.core.proxy ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/core/proxy/__init__.py:156
 [5] update_inventory
   @ speasy.core.dataprovider ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/core/dataprovider.py:65
 [6] __init__
   @ speasy.core.dataprovider ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/core/dataprovider.py:50
 [7] __init__
   @ speasy.data_providers.generic_archive ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/data_providers/generic_archive/__init__.py:79
 [8] init_archive
   @ ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/core/requests_scheduling/request_dispatch.py:85
 [9] init_providers
   @ ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/core/requests_scheduling/request_dispatch.py:109
 [10] <module>
   @ ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/core/requests_scheduling/request_dispatch.py:114
 [11] <module>
   @ ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/core/requests_scheduling/__init__.py:2
 [12] <module>
   @ ~/projects/psp_conjunction/.CondaPkg/.pixi/envs/default/lib/python3.14/site-packages/speasy/__init__.py:21
Stacktrace:
 [1] pythrow()
   @ PythonCall.Core ~/.julia/packages/PythonCall/mkWc2/src/Core/err.jl:77
 [2] errcheck
   @ ~/.julia/packages/PythonCall/mkWc2/src/Core/err.jl:10 [inlined]
 [3] pyimport(m::String)
   @ PythonCall.Core ~/.julia/packages/PythonCall/mkWc2/src/Core/builtins.jl:1458
 [4] top-level scope
   @ REPL[5]:1

[Beforerr]

A simple check like below would fix the issue, but

                if dataset and hasattr(dataset, "spz_name"):
                    parent.__dict__[dataset.spz_name()] = dataset

It seems that we need to make make_dataset_index more robust (may be give a warning when it fails to get spz_name?)

[jeandet]

@Beforerr I think the issue is that you are using a released version of Speasy after trying the main branch where some non backward compatible changes have been made (the type of some cache entries has changed).
You can either clear your cache or use the main branch until we publish the next release (likely 2nd week of November).

[Beforerr]

Thank you @jeandet for clarifying the issue. I think I have mixed version in my computer so this unfortunately happens. Just wondering what is the recommended way to clear the cache, like drop_matching_entries(r'.*')?

[jeandet]

@Beforerr , that would work, if you want something more selective, you can filter only for the faulty function:

spz.core.cache.drop_matching_entries(r'__internal__/CacheCall/.*make_dataset_index.*')