Releases v0.12.2 (#256)

bin/copyright.py

@@ -6,6 +6,7 @@
 
 import argparse
 import datetime
+import difflib
 import os.path
 import re
 import subprocess  # nosec
@@ -97,7 +98,8 @@ def main() -> None:
                     else:
                         if not no_git_log:
                             print(
-                                f"No log found with git on '{file_name}' (the next messages will be hidden)."
+                                f"No log found with git on '{file_name}' "
+                                "(the next messages will be hidden)."
                             )
                             no_git_log = True
                     used_year = CURRENT_YEAR
@@ -118,7 +120,7 @@ def main() -> None:
 
         with open(file_name, "r", encoding="utf-8") as file_obj:
             content = file_obj.read()
-            file_success, content = update_file(
+            file_success, new_content = update_file(
                 content,
                 used_year,
                 one_date_re,
@@ -131,9 +133,13 @@ def main() -> None:
             )
         if not file_success:
             success = False
+            file_diff_lines = difflib.unified_diff(
+                content.splitlines(keepends=True), new_content.splitlines(keepends=True)
+            )
+            file_diff = "".join(file_diff_lines)
             with open(file_name, "w", encoding="utf-8") as file_obj:
-                file_obj.write(content)
-            print(f"Fixing copyright in '{file_name}'")
+                file_obj.write(new_content)
+            print(f"Copyright updated in '{file_name}'.\n{file_diff}")
 
     if not success:
         sys.exit(1)

docs/source/installation-ext.rst

@@ -8,29 +8,37 @@
 `地址 <https://pip.pypa.io/en/stable/installing/>`_ 安装，推荐使用 `阿里云镜像 <http://mirrors.aliyun.com/help/pypi>`_
 加快下载速度。
 
-接着确保 setuptools 的版本：
+接着升级 pip 和 setuptools 的版本：
 
 .. code-block:: sh
 
-    pip install setuptools>=3.0
+    pip install -U pip setuptools
 
-接着就可以安装 PyODPS：
+此后可以安装 PyODPS：
 
 .. code-block:: sh
 
     pip install pyodps
 
+如果安装时出现 `urllib3 v2.0 only supports OpenSSL 1.1.1+` 的报错，需要先安装一个兼容旧版
+OpenSSL 的 urllib3 版本
+
+.. code-block:: sh
+
+    pip install urllib3\<2.0
+
+此后再安装 PyODPS。
+
 检查安装完成：
 
 .. code-block:: sh
 
     python -c "from odps import ODPS"
 
-
 如果使用的python不是系统默认的python版本，安装完pip则可以：
 
 .. code-block:: sh
 
-    /home/tops/bin/python2.7 -m pip install setuptools>=3.0
+    /home/tops/bin/python2.7 -m pip install -U pip setuptools
 
 其余步骤类似。

docs/source/locale/en/LC_MESSAGES/installation-ext.po

@@ -8,25 +8,24 @@ msgid ""
 msgstr ""
 "Project-Id-Version: PyODPS 0.7.16\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2024-03-15 15:54+0800\n"
+"POT-Creation-Date: 2024-12-17 11:11+0800\n"
 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
 "Language-Team: LANGUAGE <[email protected]>\n"
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=utf-8\n"
 "Content-Transfer-Encoding: 8bit\n"
-"Generated-By: Babel 2.12.1\n"
+"Generated-By: Babel 2.16.0\n"
 
 #: ../../source/installation-ext.rst:5
 msgid "安装指南"
 msgstr "Installation instructions"
 
 #: ../../source/installation-ext.rst:7
 msgid ""
-"如果能访问外网，推荐使用 pip 安装。较新版本的 Python 通常自带 pip。如果你"
-"的 Python 不包含 pip，可以参考 `地址 <https://pip.pypa.io/en/stable/"
-"installing/>`_ 安装，推荐使用 `阿里云镜像 <http://mirrors.aliyun.com/help"
-"/pypi>`_ 加快下载速度。"
+"如果能访问外网，推荐使用 pip 安装。较新版本的 Python 通常自带 pip。如果你的 Python 不包含 pip，可以参考 `地址 "
+"<https://pip.pypa.io/en/stable/installing/>`_ 安装，推荐使用 `阿里云镜像 "
+"<http://mirrors.aliyun.com/help/pypi>`_ 加快下载速度。"
 msgstr ""
 "We recommend that you use pip to install Python on MaxCompute (PyODPS) "
 "when you have access to the Internet. pip is often installed with latest "
@@ -36,42 +35,59 @@ msgstr ""
 "<https://pip.pypa.io/en/stable/installing/>`_ ."
 
 #: ../../source/installation-ext.rst:11
-msgid "接着确保 setuptools 的版本："
+msgid "接着升级 pip 和 setuptools 的版本："
 msgstr ""
-"Make sure that you install the correct versions of setuptools by the "
-"following command."
+"Make sure that you upgrade your pip and setuptools to the latest version "
+"by the following command."
 
 #: ../../source/installation-ext.rst:13
-msgid "pip install setuptools>=3.0"
+msgid "pip install -U pip setuptools"
 msgstr ""
 
 #: ../../source/installation-ext.rst:17
-msgid "接着就可以安装 PyODPS："
+msgid "此后可以安装 PyODPS："
 msgstr "Install PyODPS by using the following code:"
 
 #: ../../source/installation-ext.rst:19
 msgid "pip install pyodps"
 msgstr ""
 
 #: ../../source/installation-ext.rst:23
+msgid ""
+"如果安装时出现 `urllib3 v2.0 only supports OpenSSL 1.1.1+` 的报错，需要先安装一个兼容旧版 "
+"OpenSSL 的 urllib3 版本"
+msgstr ""
+"If you meet the error `urllib3 v2.0 only supports OpenSSL 1.1.1+` during"
+" installation, you need to install an older urllib3 version which is "
+"compatible with older OpenSSL versions."
+
+#: ../../source/installation-ext.rst:26
+msgid "pip install urllib3\\<2.0"
+msgstr ""
+
+#: ../../source/installation-ext.rst:30
+msgid "此后再安装 PyODPS。"
+msgstr "Then PyODPS can be installed as expected."
+
+#: ../../source/installation-ext.rst:32
 msgid "检查安装完成："
 msgstr "Check whether the installation has been completed:"
 
-#: ../../source/installation-ext.rst:25
+#: ../../source/installation-ext.rst:34
 msgid "python -c \"from odps import ODPS\""
 msgstr ""
 
-#: ../../source/installation-ext.rst:30
+#: ../../source/installation-ext.rst:38
 msgid "如果使用的python不是系统默认的python版本，安装完pip则可以："
 msgstr ""
 "If your version of Python is not the default version, run the following "
 "code after installing pip:"
 
-#: ../../source/installation-ext.rst:32
-msgid "/home/tops/bin/python2.7 -m pip install setuptools>=3.0"
+#: ../../source/installation-ext.rst:40
+msgid "/home/tops/bin/python2.7 -m pip install -U pip setuptools"
 msgstr ""
 
-#: ../../source/installation-ext.rst:36
+#: ../../source/installation-ext.rst:44
 msgid "其余步骤类似。"
 msgstr "Repeat these steps to install your version of Python."
 

docs/source/locale/en/LC_MESSAGES/options.po

@@ -8,23 +8,21 @@ msgid ""
 msgstr ""
 "Project-Id-Version: PyODPS 0.8.0\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2024-07-27 19:27+0800\n"
+"POT-Creation-Date: 2024-11-30 15:32+0800\n"
 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
 "Language-Team: LANGUAGE <[email protected]>\n"
 "MIME-Version: 1.0\n"
 "Content-Type: text/plain; charset=utf-8\n"
 "Content-Transfer-Encoding: 8bit\n"
-"Generated-By: Babel 2.12.1\n"
+"Generated-By: Babel 2.16.0\n"
 
 #: ../../source/options.rst:5
 msgid "配置选项"
 msgstr "Configuration"
 
 #: ../../source/options.rst:8
-msgid ""
-"PyODPS 提供了一系列的配置选项，可通过 ``odps.options`` 获得，如下面的例子"
-"："
+msgid "PyODPS 提供了一系列的配置选项，可通过 ``odps.options`` 获得，如下面的例子："
 msgstr ""
 "PyODPS provides a series of configuration options, which can be obtained "
 "through ``odps.options``. Here is a simple example:"
@@ -34,11 +32,10 @@ msgid ""
 "from odps import options\n"
 "# 设置所有输出表的生命周期（lifecycle 选项）\n"
 "options.lifecycle = 30\n"
-"# 使用 Tunnel 下载 string 类型时使用 bytes（tunnel.string_as_binary 选项"
-"）\n"
+"# 使用 Tunnel 下载 string 类型时使用 bytes（tunnel.string_as_binary 选项）\n"
 "options.tunnel.string_as_binary = True\n"
-"# PyODPS DataFrame 用 ODPS 执行时，参照下面 dataframe 相关配置，sort 时"
-"设置 limit 到一个比较大的值\n"
+"# PyODPS DataFrame 用 ODPS 执行时，参照下面 dataframe 相关配置，sort 时设置 limit 到一个比较大的值"
+"\n"
 "options.df.odps.sort.limit = 100000000"
 msgstr ""
 "from odps import options\n"
@@ -110,6 +107,16 @@ msgstr "LogView holding time (hours)"
 msgid "24"
 msgstr ""
 
+#: ../../source/options.rst:1
+msgid "use_legacy_logview"
+msgstr ""
+
+#: ../../source/options.rst:1
+msgid "使用旧版 LogView 地址，None 表示自动根据 Endpoint 处理"
+msgstr ""
+"Use legacy LogView address. If None is specified, LogView will be "
+"selected given MaxCompute endpoint."
+
 #: ../../source/options.rst:1
 msgid "quota_name"
 msgstr ""
@@ -123,9 +130,7 @@ msgid "local_timezone"
 msgstr ""
 
 #: ../../source/options.rst:1
-msgid ""
-"使用的时区，None 表示不处理，True 表示本地时区，False 表示 UTC，也可用 "
-"pytz 的时区"
+msgid "使用的时区，None 表示不处理，True 表示本地时区，False 表示 UTC，也可用 pytz 的时区"
 msgstr ""
 "Used time zone. None indicates that PyODPS takes no actions, True "
 "indicates local time, and False indicates UTC. The time zone of pytz "
@@ -387,7 +392,7 @@ msgstr ""
 msgid "配置使用 Tunnel 所需的标签"
 msgstr "Tags when calling tunnel service"
 
-#: ../../source/options.rst:73
+#: ../../source/options.rst:74
 msgid "DataFrame 配置"
 msgstr "DataFrame configurations"
 

docs/source/options.rst

@@ -30,6 +30,7 @@ PyODPS 提供了一系列的配置选项，可通过 ``odps.options`` 获得，
    "default_project", "默认 Project", "None"
    "logview_host", "LogView 主机名", "None"
    "logview_hours", "LogView 保持时间（小时）", "24"
+   "use_legacy_logview", "使用旧版 LogView 地址，None 表示自动根据 Endpoint 处理", "None"
    "quota_name", "提交任务时使用的计算 Quota 名称", "None"
    "local_timezone", "使用的时区，None 表示不处理，True 表示本地时区，False 表示 UTC，也可用 pytz 的时区", "None"
    "lifecycle", "所有表生命周期", "None"

odps/_version.py

@@ -1,4 +1,4 @@
-# Copyright 1999-2024 Alibaba Group Holding Ltd.
+# Copyright 1999-2025 Alibaba Group Holding Ltd.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-version_info = (0, 12, 1, 1)
+version_info = (0, 12, 2)
 _num_index = max(idx if isinstance(v, int) else 0 for idx, v in enumerate(version_info))
 __version__ = ".".join(map(str, version_info[: _num_index + 1])) + "".join(
     version_info[_num_index + 1 :]

odps/accounts.py

@@ -1,4 +1,4 @@
-# Copyright 1999-2024 Alibaba Group Holding Ltd.
+# Copyright 1999-2025 Alibaba Group Holding Ltd.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -535,6 +535,8 @@ def sign_request(self, req, endpoint, region_name=None):
         url = req.url[len(endpoint) :]
         url_components = urlparse(unquote(url), allow_fragments=False)
         self._build_canonical_str(url_components, req)
+        if self.token is None:
+            raise TypeError("Cannot sign request with None bearer token")
         req.headers["x-odps-bearer-token"] = self.token
         if self._last_refresh_time:
             req.headers["x-pyodps-token-timestamp"] = str(self._last_refresh_time)
@@ -546,7 +548,7 @@ def __init__(self, credential_provider):
         self.provider = credential_provider
         super(CredentialProviderAccount, self).__init__(None, None, None)
 
-    def sign_request(self, req, endpoint, region_name=None):
+    def _refresh_credential(self):
         try:
             credential = self.provider.get_credential()
         except:
@@ -555,6 +557,9 @@ def sign_request(self, req, endpoint, region_name=None):
         self.access_id = credential.get_access_key_id()
         self.secret_access_key = credential.get_access_key_secret()
         self.sts_token = credential.get_security_token()
+
+    def sign_request(self, req, endpoint, region_name=None):
+        utils.call_with_retry(self._refresh_credential)
         return super(CredentialProviderAccount, self).sign_request(
             req, endpoint, region_name=region_name
         )

odps/apis/storage_api/storage_api.py

@@ -1,4 +1,4 @@
-# Copyright 1999-2024 Alibaba Group Holding Ltd.
+# Copyright 1999-2025 Alibaba Group Holding Ltd.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -20,6 +20,7 @@
 from enum import Enum
 from hashlib import md5
 from io import BytesIO, IOBase
+from typing import List, Union
 
 try:
     import pyarrow as pa
@@ -621,16 +622,21 @@ def __init__(
         table: Table,
         rest_endpoint: str = None,
         quota_name: str = None,
+        tags: Union[None, str, List[str]] = None,
     ):
-        if isinstance(odps, ODPS) and isinstance(table, Table):
-            self._odps = odps
-            self._table = table
-            self._quota_name = quota_name
-            self._rest_endpoint = rest_endpoint
-            self._tunnel_rest = None
-        else:
+        if not isinstance(odps, ODPS) or not isinstance(table, Table):
             raise ValueError("Please input odps configuration")
 
+        self._odps = odps
+        self._table = table
+        self._quota_name = quota_name
+        self._rest_endpoint = rest_endpoint
+        self._tunnel_rest = None
+
+        self._tags = tags or options.tunnel.tags
+        if isinstance(self._tags, str):
+            self._tags = self._tags.split(",")
+
     @property
     def table(self):
         return self._table
@@ -653,14 +659,10 @@ def _get_resource(self, *args) -> str:
         url = self._table.table_resource(endpoint=endpoint, force_schema=True)
         return "/".join([url] + list(args))
 
-    @staticmethod
-    def _fill_common_headers(raw_headers=None, tags=None):
+    def _fill_common_headers(self, raw_headers=None):
         headers = raw_headers or {}
-        tags = tags or options.tunnel.tags
-        if tags:
-            if isinstance(tags, str):
-                tags = tags.split(",")
-            headers["odps-tunnel-tags"] = ",".join(tags)
+        if self._tags:
+            headers["odps-tunnel-tags"] = ",".join(self._tags)
         return headers
 
     def create_read_session(