Releases v0.12.1 (#254)

Author: wjsi

.github/workflows/pypi-cd.yml

@@ -77,10 +77,3 @@ jobs:
           echo "repository=$PYPI_REPO"  >> ~/.pypirc
           python -m pip install twine
           python -m twine upload -r pypi --skip-existing wheelhouse/*
-
-      - name: Upload artifacts to github
-        uses: actions/upload-artifact@v4
-        if: ${{ always() }}
-        with:
-          name: wheels
-          path: ./wheelhouse

.pre-commit-config.yaml

@@ -0,0 +1,110 @@
+minimum_pre_commit_version: 2.15.0
+# we use *| at the beginning of lines to make sure patterns can be written
+#  in multiple lines with added spaces ignored
+exclude: "^benchmarks/.*
+*|^bin/.*
+*|^cupid/.*
+*|^docs/.*
+*|^misc/.*
+*|^odps/mars_extension/.*
+*|^odps/df/.*
+*|^odps/internal/ml/.*
+*|^odps/internal/static/.*
+*|^odps/internal/xlib/.*
+*|^odps/lib/.*
+*|^odps/ml/.*
+*|^odps/static/.*"
+# reserve "manual" for relatively slow hooks which we still want to run in CI
+default_stages: [
+  commit,
+  merge-commit,
+  push,
+  prepare-commit-msg,
+  commit-msg,
+  post-checkout,
+  post-commit,
+  post-merge,
+  post-rewrite
+]
+ci:
+    autofix_prs: false
+repos:
+- repo: local
+  hooks:
+  - id: check-yaml
+    name: check-yaml
+    entry: check-yaml
+    language: python
+    exclude: ^operators/
+    types_or: [yaml]
+    additional_dependencies: [pre-commit-hooks==4.4.0]
+  - id: end-of-file-fixer
+    name: end-of-file-fixer
+    entry: end-of-file-fixer
+    language: python
+    additional_dependencies: [pre-commit-hooks==4.4.0]
+  - id: trailing-whitespace
+    name: trailing-whitespace
+    entry: trailing-whitespace-fixer
+    language: python
+    additional_dependencies: [pre-commit-hooks==4.4.0]
+  - id: check-illegal-windows-names
+    name: check illegal windows names
+    entry: Illegal windows filenames detected
+    language: fail
+    files: '(?i)((^|/)(\\|CON|PRN|AUX|NUL|COM[1-9]|LPT[1-9])(\.|/|$)|:)'
+- repo: local
+  hooks:
+  # Check that the copyright is configured
+  - id: copyright-check
+    name: copyright-check
+    entry: python bin/copyright.py
+    language: python
+    files: 'odps/.*'
+    exclude: '^odps/lib/.*|^odps/tunnel/pb/.*\.py$'
+    types_or: [python, pyi, cython]
+    additional_dependencies: [pyyaml]
+- repo: local
+  hooks:
+  # Replace quote replacer for black (which only supports Python >= 3.4)
+  - id: double-quote-string-fixer
+    name: double-quote-string-fixer
+    entry: python bin/string_fixer.py
+    language: python
+    files: 'odps/.*'
+    exclude: '^odps/lib/.*'
+    types_or: [python, pyi, cython]
+    args: [--replace-single]
+- repo: local
+  hooks:
+  - id: flake8
+    name: flake8
+    description: "Flake8"
+    entry: flake8
+    language: python
+    require_serial: true
+    types_or: [python, pyi]
+    additional_dependencies: [flake8==5.0.4]
+- repo: local
+  hooks:
+  # NOTE: we make `black` a local hook because if it's installed from
+  # PyPI (rather than from source) then it'll run twice as fast thanks to mypyc
+  - id: black
+    name: black
+    description: "Black"
+    entry: black
+    language: python
+    require_serial: true
+    types_or: [python, pyi]
+    additional_dependencies: [black==23.3.0]
+- repo: local
+  hooks:
+    - id: isort
+      name: isort
+      description: "isort"
+      entry: isort
+      language: python
+      require_serial: true
+      types_or: [python, pyi]
+      args: [--profile, black]
+      additional_dependencies: [isort==5.11.0]

benchmarks/perf_types.py

@@ -35,8 +35,28 @@
 def schema():
     pr = cProfile.Profile()
     pr.enable()
-    fields = ["bigint", "double", "datetime", "boolean", "string", "decimal"]
-    types = ["bigint", "double", "datetime", "boolean", "string", "decimal"]
+    fields = [
+        "bigint",
+        "double",
+        "datetime",
+        "boolean",
+        "string",
+        "decimal",
+        "array",
+        "map",
+        "struct",
+    ]
+    types = [
+        "bigint",
+        "double",
+        "datetime",
+        "boolean",
+        "string",
+        "decimal",
+        "array<bigint>",
+        "map<string, bigint>",
+        "struct<key:string, value:bigint>",
+    ]
     try:
         schema = TableSchema.from_lists(fields, types)
         schema.build_snapshot()
@@ -50,35 +70,53 @@ def schema():
 
 def test_set_record_field_bigint(schema):
     r = Record(schema=schema)
-    for i in range(10**6):
+    for _ in range(10**6):
         r["bigint"] = 2**63 - 1
 
 
 def test_set_record_field_double(schema):
     r = Record(schema=schema)
-    for i in range(10**6):
+    for _ in range(10**6):
         r["double"] = 0.0001
 
 
 def test_set_record_field_boolean(schema):
     r = Record(schema=schema)
-    for i in range(10**6):
+    for _ in range(10**6):
         r["boolean"] = False
 
 
 def test_set_record_field_string(schema):
     r = Record(schema=schema)
-    for i in range(10**6):
+    for _ in range(10**6):
         r["string"] = STRING_LITERAL
 
 
-def test_write_set_record_field_datetime(schema):
+def test_set_record_field_datetime(schema):
     r = Record(schema=schema)
-    for i in range(10**6):
+    for _ in range(10**6):
         r["datetime"] = datetime(2016, 1, 1)
 
 
 def test_set_record_field_decimal(schema):
     r = Record(schema=schema)
-    for i in range(10**6):
+    for _ in range(10**6):
         r["decimal"] = Decimal("1.111111")
+
+
+def test_set_record_field_array(schema):
+    r = Record(schema=schema)
+    for _ in range(10**6):
+        r["array"] = [2**63 - 1]
+
+
+def test_set_record_field_map(schema):
+    r = Record(schema=schema)
+    for _ in range(10**6):
+        r["map"] = {"data_key": 2**63 - 1}
+
+
+def test_set_record_field_struct(schema):
+    r = Record(schema=schema)
+    for _ in range(10**6):
+        r["struct"] = ("data_key", 2**63 - 1)

docs/source/base-sql.rst

@@ -174,11 +174,12 @@ PyODPS 默认不限制能够从 Instance 读取的数据规模，但 Project Own
 Instance 结果的读取，此时只能使用受限读取模式读取数据，在此模式下可读取的行数受到 Project 配置限制，通常为 10000 行。如果
 PyODPS 检测到读取 Instance 数据被限制，且 ``options.tunnel.limit_instance_tunnel`` 未设置，会自动启用受限读取模式。
 如果你的 Project 被保护，想要手动启用受限读取模式，可以为 ``open_reader`` 方法增加 ``limit=True`` 选项，或者设置
-``options.tunnel.limit_instance_tunnel = True`` 。
+``options.tunnel.limit_instance_tunnel = True``\ 。
 
 在部分环境中，例如 DataWorks，``options.tunnel.limit_instance_tunnel`` 可能默认被置为 True。此时，如果需要读取\
 所有数据，需要为 ``open_reader`` 增加参数 `tunnel=True, limit=False` 。需要注意的是，如果 Project 本身被保护，\
-这两个参数\ **不能**\ 解除保护，此时应联系 Project Owner 开放相应的读权限。
+这两个参数\ **不能**\ 解除保护，MaxCompute 也\ **不提供**\ 绕开该权限限制读取更多数据的方法。此时应联系 Project Owner
+开放相应的读权限。
 
 如果你所使用的 MaxCompute 只能支持旧 Result 接口，同时你需要读取所有数据，可将 SQL 结果写入另一张表后用读表接口读取
 （可能受到 Project 安全设置的限制）。