FIX-#7246: Pin pyarrow>=10.0.1 as pandas 2.2.* does (#7247)

Signed-off-by: Anatoly Myachev <[email protected]>

Author: anmyachev

environment-dev.yml

@@ -14,7 +14,7 @@ dependencies:
   # optional dependencies
   # ray==2.5.0 broken: https://github.com/conda-forge/ray-packages-feedstock/issues/100
   - ray-core>=2.1.0,!=2.5.0
-  - pyarrow>=7.0.0
+  - pyarrow>=10.0.1
   # workaround for https://github.com/conda/conda/issues/11744
   - grpcio!=1.45.*
   - grpcio!=1.46.*

modin/core/io/column_stores/parquet_dispatcher.py

@@ -210,11 +210,7 @@ def row_groups_per_file(self):
 
     @functools.cached_property
     def files(self):
-        try:
-            files = self.dataset.files
-        except AttributeError:
-            # compatibility at least with 3.0.0 <= pyarrow < 8.0.0
-            files = self.dataset._dataset.files
+        files = self.dataset.files
         return self._get_files(files)
 
     def to_pandas_dataframe(

requirements-dev.txt

@@ -8,7 +8,7 @@ psutil>=5.8.0
 ## optional dependencies
 # ray==2.5.0 broken: https://github.com/conda-forge/ray-packages-feedstock/issues/100
 ray>=2.1.0,!=2.5.0
-pyarrow>=7.0.0
+pyarrow>=10.0.1
 dask[complete]>=2.22.0
 distributed>=2.22.0
 xarray>=2022.12.0

requirements/env_unidist_linux.yml

@@ -14,7 +14,7 @@ dependencies:
   - psutil>=5.8.0
 
   # optional dependencies
-  - pyarrow>=7.0.0
+  - pyarrow>=10.0.1
   - xarray>=2022.12.0
   - jinja2>=3.1.2
   - scipy>=1.10.0

requirements/env_unidist_win.yml

@@ -14,7 +14,7 @@ dependencies:
   - psutil>=5.8.0
 
   # optional dependencies
-  - pyarrow>=7.0.0
+  - pyarrow>=10.0.1
   - xarray>=2022.12.0
   - jinja2>=3.1.2
   - scipy>=1.10.0

requirements/requirements-no-engine.yml

@@ -11,7 +11,7 @@ dependencies:
   - psutil>=5.8.0
 
   # optional dependencies
-  - pyarrow>=7.0.0
+  - pyarrow>=10.0.1
   - xarray>=2022.12.0
   - jinja2>=3.1.2
   - scipy>=1.10.0

setup.py

@@ -7,7 +7,7 @@
 
 dask_deps = ["dask>=2.22.0", "distributed>=2.22.0"]
 # ray==2.5.0 broken: https://github.com/conda-forge/ray-packages-feedstock/issues/100
-ray_deps = ["ray>=2.1.0,!=2.5.0", "pyarrow>=7.0.0"]
+ray_deps = ["ray>=2.1.0,!=2.5.0", "pyarrow>=10.0.1"]
 mpi_deps = ["unidist[mpi]>=0.2.1"]
 consortium_standard_deps = ["dataframe-api-compat>=0.2.7"]
 spreadsheet_deps = ["modin-spreadsheet>=0.1.0"]