From 58cdab6adc72c377e2460f90b05e90b9140ddf29 Mon Sep 17 00:00:00 2001
From: Thomas Grainger <tagrain@gmail.com>
Date: Sat, 7 Oct 2023 12:47:37 -0700
Subject: [PATCH] fixes for numpy2 (#886)

* np.lib.pad was removed in numpy2

* skip rapidjson when json dumping np.float64 on numpy 2

* pin cython<3 on pandas build
---
 .github/workflows/main.yaml   |  2 +-
 fastparquet/test/test_json.py | 22 +++++++++++++---------
 fastparquet/writer.py         |  4 ++--
 3 files changed, 16 insertions(+), 12 deletions(-)

diff --git a/.github/workflows/main.yaml b/.github/workflows/main.yaml
index 6836e358..b8321e35 100644
--- a/.github/workflows/main.yaml
+++ b/.github/workflows/main.yaml
@@ -120,7 +120,7 @@ jobs:
       - name: pip-install
         shell: bash -l {0}
         run: |
-          pip install Cython
+          pip install 'Cython<3'
           pip install hypothesis
           pip install pytest-localserver pytest-xdist pytest-asyncio
           pip install -e . --no-deps # Install fastparquet
diff --git a/fastparquet/test/test_json.py b/fastparquet/test/test_json.py
index e40c4e68..e52c8a5b 100644
--- a/fastparquet/test/test_json.py
+++ b/fastparquet/test/test_json.py
@@ -4,6 +4,7 @@
 
 import numpy as np
 import pytest
+from packaging.version import Version
 
 from fastparquet.json import (
     JsonCodecError,
@@ -25,15 +26,15 @@ def _clear_cache():
 
 
 @pytest.mark.parametrize(
-    "data",
+    "data,has_float64",
     [
-        None,
-        [1, 1, 2, 3, 5],
-        [1.23, -3.45],
-        [np.float64(0.12), np.float64(4.56)],
-        [[1, 2, 4], ["x", "y", "z"]],
-        {"k1": "value", "k2": "à/è", "k3": 3},
-        {"k1": [1, 2, 3], "k2": [4.1, 5.2, 6.3]},
+        (None, False),
+        ([1, 1, 2, 3, 5], False),
+        ([1.23, -3.45], False),
+        ([np.float64(0.12), np.float64(4.56)], True),
+        ([[1, 2, 4], ["x", "y", "z"]], False),
+        ({"k1": "value", "k2": "à/è", "k3": 3}, False),
+        ({"k1": [1, 2, 3], "k2": [4.1, 5.2, 6.3]}, False),
     ],
 )
 @pytest.mark.parametrize(
@@ -44,7 +45,10 @@ def _clear_cache():
     "decoder_module, decoder_class",
     list(_codec_classes.items()),
 )
-def test_engine(encoder_module, encoder_class, decoder_module, decoder_class, data):
+def test_engine(encoder_module, encoder_class, decoder_module, decoder_class, data, has_float64):
+    if encoder_module == "rapidjson" and has_float64 and Version(np.__version__).major >= 2:
+        pytest.skip(reason="rapidjson cannot json dump np.float64 on numpy 2")
+
     pytest.importorskip(encoder_module)
     pytest.importorskip(decoder_module)
 
diff --git a/fastparquet/writer.py b/fastparquet/writer.py
index 8c8e1364..d439ad5a 100644
--- a/fastparquet/writer.py
+++ b/fastparquet/writer.py
@@ -243,7 +243,7 @@ def convert(data, se):
         elif type == parquet_thrift.Type.BOOLEAN:
             # TODO: with our own bitpack writer, no need to copy for
             #  the padding
-            padded = np.lib.pad(data.values, (0, 8 - (len(data) % 8)),
+            padded = np.pad(data.values, (0, 8 - (len(data) % 8)),
                                 'constant', constant_values=(0, 0))
             out = np.packbits(padded.reshape(-1, 8)[:, ::-1].ravel())
         elif dtype.name in typemap:
@@ -265,7 +265,7 @@ def convert(data, se):
                 elif type == parquet_thrift.Type.BOOLEAN:
                     # TODO: with our own bitpack writer, no need to copy for
                     #  the padding
-                    padded = np.lib.pad(data.values, (0, 8 - (len(data) % 8)),
+                    padded = np.pad(data.values, (0, 8 - (len(data) % 8)),
                                         'constant', constant_values=(0, 0))
                     out = np.packbits(padded.reshape(-1, 8)[:, ::-1].ravel())
                 else: