在GCP JupyterLab导入BigQuery库时遇AttributeError问题求助

问题重现 #

执行以下导入代码时触发错误：

from google.cloud import storage
from google.cloud import bigquery

import pandas as pd
from sklearn import datasets

错误信息 #

AttributeError                            Traceback (most recent call last)
/tmp/ipykernel_5450/3407211047.py in 
1 from google.cloud import storage
2 from google.cloud import bigquery
3
4 import pandas as pd
5 from sklearn import datasets

/opt/conda/lib/python3.7/site-packages/google/cloud/bigquery/__init__.py in 
33 __version__ = bigquery_version.__version__
34
35 from google.cloud.bigquery.client import Client
36 from google.cloud.bigquery.dataset import AccessEntry
37 from google.cloud.bigquery.dataset import Dataset

/opt/conda/lib/python3.7/site-packages/google/cloud/bigquery/client.py in 
74
75
76 from google.cloud.bigquery import _job_helpers
77 from google.cloud.bigquery._job_helpers import make_job_id as _make_job_id
78 from google.cloud.bigquery._helpers import _get_sub_prop

/opt/conda/lib/python3.7/site-packages/google/cloud/bigquery/_job_helpers.py in 
22 from google.api_core import retry as retries
23
24 from google.cloud.bigquery import job
25
26 # Avoid circular imports

/opt/conda/lib/python3.7/site-packages/google/cloud/bigquery/job/__init__.py in 
25 from google.cloud.bigquery.job.base import TransactionInfo
26 from google.cloud.bigquery.job.base import UnknownJob
27 from google.cloud.bigquery.job.copy_ import CopyJob
28 from google.cloud.bigquery.job.copy_ import CopyJobConfig
29 from google.cloud.bigquery.job.copy_ import OperationType

/opt/conda/lib/python3.7/site-packages/google/cloud/bigquery/job/copy_.py in 
19 from google.cloud.bigquery.encryption_configuration import EncryptionConfiguration
20 from google.cloud.bigquery import _helpers
21 from google.cloud.bigquery.table import TableReference
22
23 from google.cloud.bigquery.job.base import _AsyncJob

/opt/conda/lib/python3.7/site-packages/google/cloud/bigquery/table.py in 
55 import google.cloud._helpers  # type: ignore
56 from google.cloud.bigquery import _helpers
57 from google.cloud.bigquery import _pandas_helpers
58 from google.cloud.bigquery.exceptions import LegacyBigQueryStorageError
59 from google.cloud.bigquery.schema import _build_schema_resource

/opt/conda/lib/python3.7/site-packages/google/cloud/bigquery/_pandas_helpers.py in 
49     date_dtype_name = time_dtype_name = ""  # Use '' rather than None because pytype
50
51 pyarrow = _helpers.PYARROW_VERSIONS.try_import()
52
53 try:

AttributeError: module 'google.cloud.bigquery._helpers' has no attribute 'PYARROW_VERSIONS'

已尝试的无效操作 #

• 安装pyarrow：

  conda install -c conda-forge pyarrow
  

• 升级相关库：

  pip install --upgrade --force-reinstall google-cloud-bigquery-storage
  pip install --upgrade google-cloud-bigquery
  pip install --upgrade google-cloud-storage
  

解决方案 #

这个错误源于google-cloud-bigquery与依赖库（如pyarrow、google-api-core）版本不兼容，或conda与pip混用导致的依赖冲突。可按以下步骤解决：

方法1：统一用pip管理依赖，重装指定版本组合 #

1. 卸载冲突包：

   pip uninstall -y google-cloud-bigquery google-cloud-bigquery-storage pyarrow google-api-core
   

2. 安装适配Python3.7的兼容版本：

   pip install google-cloud-bigquery==3.10.0 google-cloud-bigquery-storage==2.20.0 pyarrow==12.0.1 google-api-core==2.11.1
   

方法2：清理conda与pip的冲突环境 #

若使用conda环境，避免混用两种包管理器：

1. 移除pip安装的相关包：

   pip uninstall -y google-cloud-bigquery google-cloud-bigquery-storage
   

2. 用conda安装兼容版本：

   conda install -c conda-forge google-cloud-bigquery=3.10.0 google-cloud-bigquery-storage=2.20.0 pyarrow=12.0.1
   

方法3：创建全新conda环境（推荐） #

1. 创建Python3.7专属环境：

   conda create -n gcp-bq-env python=3.7 -y
   

2. 激活环境：

   conda activate gcp-bq-env
   

3. 一键安装所有依赖：

   conda install -c conda-forge google-cloud-bigquery google-cloud-storage pandas scikit-learn pyarrow -y
   

完成操作后，重启JupyterLab内核，重新执行导入代码即可解决问题。

内容的提问来源于stack exchange，提问作者lisha