precis-cli 0.1.3__py3-none-any.whl

app/shared/domain/transforms/strip.py

@@ -0,0 +1,73 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+@fileoverview Strip 转换运行器
+
+功能概述:
+- 去除字符串首尾空白字符
+- 支持自定义要去除的字符集合
+
+参数:
+    chars: 要去除的字符集合（默认 None，去除所有空白）
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+import pandas as pd
+
+from .base import TransformRunner, stringify_preserve_null
+
+
+class StripRunner(TransformRunner):
+    """@classdesc 字符串去除空白转换运行器"""
+
+    def execute(
+        self,
+        df: pd.DataFrame,
+        input_column: str,
+        params: dict[str, Any],
+        output_columns: list[str],
+    ) -> pd.DataFrame:
+        """
+        @methoddesc 执行 去除空白转换
+
+        业务用途:
+        - TransformRunner 协议的标准入口，由 transform 节点调用
+        - 读取 params 中的转换参数，对 input_column 应用转换，输出到 output_columns
+
+        参数:
+            df: 源 DataFrame
+            input_column: 输入列名
+            params: 转换参数字典
+            output_columns: 目标输出列名列表
+
+        返回:
+            转换后的 DataFrame
+        """
+        chars = params.get("chars", None)
+
+        if input_column not in df.columns:
+            raise ValueError(f"输入列不存在: {input_column}")
+
+        if not output_columns:
+            raise ValueError("Strip 需要至少一个 output_columns")
+
+        output_col = output_columns[0]
+        df[output_col] = stringify_preserve_null(df[input_column]).str.strip(chars)
+
+        return df

app/shared/domain/transforms/substring.py

@@ -0,0 +1,92 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+@fileoverview Substring 转换运行器
+
+功能概述:
+- 从字符串中提取子串
+- 支持 start（起始位置）、end（结束位置）、length（长度）三种参数组合
+
+参数:
+    start: 起始位置（从0开始，负数表示从末尾倒数）
+    end: 结束位置（可选，不包含）
+    length: 截取长度（可选，优先级高于 end）
+
+输入输出:
+    输入: "11010119900307889X"
+    参数: {"start": 0, "length": 17}
+    输出: "11010119900307889"
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+import pandas as pd
+
+from .base import TransformRunner, stringify_preserve_null
+
+
+class SubstringRunner(TransformRunner):
+    """@classdesc 子串提取转换运行器"""
+
+    def execute(
+        self,
+        df: pd.DataFrame,
+        input_column: str,
+        params: dict[str, Any],
+        output_columns: list[str],
+    ) -> pd.DataFrame:
+        """
+        @methoddesc 执行 子字符串转换
+
+        业务用途:
+        - TransformRunner 协议的标准入口，由 transform 节点调用
+        - 读取 params 中的转换参数，对 input_column 应用转换，输出到 output_columns
+
+        参数:
+            df: 源 DataFrame
+            input_column: 输入列名
+            params: 转换参数字典
+            output_columns: 目标输出列名列表
+
+        返回:
+            转换后的 DataFrame
+        """
+        start = params.get("start", 0)
+        end = params.get("end")
+        length = params.get("length")
+
+        if input_column not in df.columns:
+            raise ValueError(f"输入列不存在: {input_column}")
+
+        output_col = output_columns[0] if output_columns else "substring"
+
+        def _substring(value: str) -> str:
+            s = str(value)
+            _start = start if start is not None else 0
+
+            if length is not None:
+                return s[_start : _start + length]
+            elif end is not None:
+                return s[_start:end]
+            else:
+                return s[_start:]
+
+        # 空值透传为 None：None 不进入切片（原 astype(str) 会切出 "nan" 片段）
+        df[output_col] = stringify_preserve_null(df[input_column]).apply(lambda s: None if s is None else _substring(s))
+
+        return df

app/shared/domain/transforms/upper_case.py

@@ -0,0 +1,70 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+@fileoverview UpperCase 转换运行器
+
+功能概述:
+- 将字符串转换为大写
+
+参数:
+    无
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+import pandas as pd
+
+from .base import TransformRunner, stringify_preserve_null
+
+
+class UpperCaseRunner(TransformRunner):
+    """@classdesc 字符串转大写转换运行器"""
+
+    def execute(
+        self,
+        df: pd.DataFrame,
+        input_column: str,
+        params: dict[str, Any],
+        output_columns: list[str],
+    ) -> pd.DataFrame:
+        """
+        @methoddesc 执行 转大写转换
+
+        业务用途:
+        - TransformRunner 协议的标准入口，由 transform 节点调用
+        - 读取 params 中的转换参数，对 input_column 应用转换，输出到 output_columns
+
+        参数:
+            df: 源 DataFrame
+            input_column: 输入列名
+            params: 转换参数字典
+            output_columns: 目标输出列名列表
+
+        返回:
+            转换后的 DataFrame
+        """
+        if input_column not in df.columns:
+            raise ValueError(f"输入列不存在: {input_column}")
+
+        if not output_columns:
+            raise ValueError("UpperCase 需要至少一个 output_columns")
+
+        output_col = output_columns[0]
+        df[output_col] = stringify_preserve_null(df[input_column]).str.upper()
+
+        return df

app/shared/domain/transforms/weighted_sum.py

@@ -0,0 +1,104 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+@fileoverview WeightedSum 转换运行器
+
+功能概述:
+- 对输入值按权重数组求和
+- 支持两种输入格式：
+  1. 纯数字字符串（如 "110101"），逐字符转数字后加权
+  2. 逗号分隔数字（如 "1,1,0,1,0,1"），split 后加权
+
+参数:
+    weights: 权重数组（如 [7,9,10,5,8,4,2,1,6,3,7,9,10,5,8,4,2]）
+
+输入输出:
+    输入: "11010119900307889" 或 "1,1,0,1,0,1,1,9,9,0,0,3,0,7,8,8,9"
+    输出: 266（加权和）
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+import pandas as pd
+
+from .base import TransformRunner, stringify_preserve_null
+
+
+class WeightedSumRunner(TransformRunner):
+    """@classdesc 加权求和转换运行器"""
+
+    def execute(
+        self,
+        df: pd.DataFrame,
+        input_column: str,
+        params: dict[str, Any],
+        output_columns: list[str],
+    ) -> pd.DataFrame:
+        """
+        @methoddesc 执行 加权求和转换
+
+        业务用途:
+        - TransformRunner 协议的标准入口，由 transform 节点调用
+        - 读取 params 中的转换参数，对 input_column 应用转换，输出到 output_columns
+
+        参数:
+            df: 源 DataFrame
+            input_column: 输入列名
+            params: 转换参数字典
+            output_columns: 目标输出列名列表
+
+        返回:
+            转换后的 DataFrame
+        """
+        weights = params.get("weights", [])
+        if not weights:
+            raise ValueError("WeightedSum 需要 weights 参数")
+
+        if input_column not in df.columns:
+            raise ValueError(f"输入列不存在: {input_column}")
+
+        output_col = output_columns[0] if output_columns else "weighted_sum"
+
+        def _weighted_sum(value: str) -> int:
+            # 逗号分隔或纯字符串都支持
+            if "," in str(value):
+                digits = [d.strip() for d in str(value).split(",")]
+            else:
+                digits = list(str(value))
+
+            # §1.24: 位数不齐报错——原实现超位静默截断（身份证校验位算错）、
+            # 不足位无校验，校验和数字对不上但无提示。典型场景位数恒定，不齐即配置/数据错误。
+            if len(digits) > len(weights):
+                raise ValueError(f"输入位数({len(digits)})超过权重位数({len(weights)}): '{value}'")
+            if len(digits) < len(weights):
+                raise ValueError(f"输入位数({len(digits)})不足权重位数({len(weights)}): '{value}'")
+
+            total = 0
+            for i, d in enumerate(digits):
+                try:
+                    total += int(d) * weights[i]
+                except (ValueError, TypeError):
+                    continue
+            return total
+
+        # 空值透传为 None，不把 "nan" 求和成 0
+        df[output_col] = stringify_preserve_null(df[input_column]).apply(
+            lambda s: None if s is None else _weighted_sum(s)
+        )
+
+        return df

app/shared/domain/validation_constraints.py

@@ -0,0 +1,69 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+@fileoverview 数据约束兼容导出模块
+
+功能概述:
+- 作为约束系统的兼容层，保留同名符号导出
+- 避免因约束实现拆分带来的外部导入路径变更和破坏性影响
+- 集中导出所有约束类型和条件注册表
+
+架构设计:
+- 代理导出模式: 实际实现已迁移至 app.shared.domain.constraints 包
+- 本模块仅做 re-export，无独立业务逻辑
+
+输入示例:
+    from app.shared.domain.validation_constraints import NotNullConstraint, UniqueConstraint, CONDITION_REGISTRY
+
+输出示例:
+    # 可直接使用所有约束类和 register_condition 装饰器
+"""
+
+# 1. 项目内部导入
+# 从新的约束包中导入所有约束类，保持向后兼容
+# 外部代码仍可通过本模块导入，无需修改导入路径
+from app.shared.domain.constraints import (
+    AllowedValuesConstraint,
+    CharsetConstraint,
+    CompositeConstraint,
+    ConditionalConstraint,
+    Constraint,
+    DateLogicConstraint,
+    ForeignKeyConstraints,
+    NotNullConstraint,
+    RangeConstraint,
+    ScriptedConstraint,
+    UniqueConstraint,
+)
+from app.shared.domain.constraints.condition_registry import CONDITION_REGISTRY, register_condition
+
+# __all__ 显式声明本模块对外暴露的符号
+# 作用：1. 控制 from module import * 的行为；2. 明确模块公共 API 边界
+__all__ = [
+    "AllowedValuesConstraint",
+    "CharsetConstraint",
+    "CompositeConstraint",
+    "CONDITION_REGISTRY",
+    "ConditionalConstraint",
+    "Constraint",
+    "DateLogicConstraint",
+    "ForeignKeyConstraints",
+    "NotNullConstraint",
+    "RangeConstraint",
+    "ScriptedConstraint",
+    "UniqueConstraint",
+    "register_condition",
+]

app/shared/services/__init__.py

@@ -0,0 +1,62 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+@fileoverview 共享服务层包初始化模块
+
+功能概述:
+- 聚合并暴露校验服务相关的类和函数
+- 统一导出校验器、执行器、校验结果等核心组件
+"""
+
+from .validation import (
+    BaseValidator,
+    ChunkedDataLoader,
+    ChunkedValidationResult,
+    CompositeValidator,
+    ConstraintAdapter,
+    DateLogicValidator,
+    MemoryMonitor,
+    PreCheck,
+    UnifiedValidationService,
+    ValidationResult,
+    ValidationType,
+    load_file_data,
+)
+from .validation.engine import validate_full_dataset
+from .validation.executor import (
+    ValidationExecutor,
+    ValidationOptions,
+    create_executor,
+)
+
+__all__ = [
+    "ValidationType",
+    "ValidationResult",
+    "UnifiedValidationService",
+    "load_file_data",
+    "BaseValidator",
+    "ConstraintAdapter",
+    "PreCheck",
+    "CompositeValidator",
+    "DateLogicValidator",
+    "ValidationExecutor",
+    "ValidationOptions",
+    "create_executor",
+    "validate_full_dataset",
+    "MemoryMonitor",
+    "ChunkedDataLoader",
+    "ChunkedValidationResult",
+]

app/shared/services/ai/__init__.py

@@ -0,0 +1,48 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+@fileoverview AI 服务统一导出模块
+
+功能概述:
+- 向后兼容地暴露配置生成器核心接口
+- 将 llm.config_generator 的公共 API 聚合到 app.shared.services.ai 命名空间
+
+架构设计:
+- 作为兼容层存在，避免历史调用方修改导入路径
+- 通过 __all__ 明确控制对外暴露的符号
+
+输入示例:
+    from app.shared.services.ai import generate_full_config_v2, GenerateOptions
+
+输出示例:
+    可直接调用 generate_full_config_v2(options) 生成 V2 项目配置
+"""
+
+from app.shared.services.llm.config_generator import (
+    V2_MANIFEST_FILENAME,
+    GenerateOptions,
+    ProfilingOptions,
+    expand_data_input_paths,
+    generate_full_config_v2,
+)
+
+__all__ = [
+    "ProfilingOptions",
+    "GenerateOptions",
+    "generate_full_config_v2",
+    "expand_data_input_paths",
+    "V2_MANIFEST_FILENAME",
+]

app/shared/services/ai/agent/__init__.py

@@ -0,0 +1,40 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Agent 内核模块"""
+
+from app.shared.services.ai.agent.executor import AgentExecutor
+from app.shared.services.ai.agent.memory import AgentMemory
+from app.shared.services.ai.agent.tool_registry import ToolRegistry
+from app.shared.services.ai.agent.types import (
+    AgentMetrics,
+    AgentPlan,
+    AgentResult,
+    AgentTurn,
+    ToolCall,
+    ToolResult,
+)
+
+__all__ = [
+    "AgentExecutor",
+    "AgentMemory",
+    "AgentMetrics",
+    "AgentPlan",
+    "AgentResult",
+    "AgentTurn",
+    "ToolCall",
+    "ToolRegistry",
+    "ToolResult",
+]

app/shared/services/ai/agent/chat_tools/__init__.py

@@ -0,0 +1,47 @@
+# SPDX-License-Identifier: Apache-2.0
+#
+# Copyright 2026 Precis Team
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Chat mini-agent 专用工具模块
+
+6 个工具构成"查-改-验"核心闭环：
+- ReadProjectTool:    读取项目概览(查)
+- ReadTableTool:      读取表数据样本(查)
+- ApplyActionsTool:   执行配置修改(改)
+- ValidateTableTool:  执行数据校验(验)
+- ReadCanvasTool:     读取画布上实际显示的节点快照(查)
+    区别于 read_project 读项目配置文件——解决"配置里有但画布上没有"的同步鸿沟
+- AskUserTool:        向用户追加提问(交互)
+
+这些工具与 generation 路径的工具集完全独立：
+- 面向"增量改现有项目"场景，而非"从零生成完整配置"
+- 不依赖 ConfigGenerationService，仅依赖确认存在的公共后端入口
+- 通过 asyncio.to_thread 包装同步入口，避免阻塞事件循环
+"""
+
+from app.shared.services.ai.agent.chat_tools.apply_actions import ApplyActionsTool
+from app.shared.services.ai.agent.chat_tools.ask_user import AskUserTool
+from app.shared.services.ai.agent.chat_tools.read_canvas import ReadCanvasTool
+from app.shared.services.ai.agent.chat_tools.read_project import ReadProjectTool
+from app.shared.services.ai.agent.chat_tools.read_table import ReadTableTool
+from app.shared.services.ai.agent.chat_tools.validate_table import ValidateTableTool
+
+__all__ = [
+    "ApplyActionsTool",
+    "AskUserTool",
+    "ReadCanvasTool",
+    "ReadProjectTool",
+    "ReadTableTool",
+    "ValidateTableTool",
+]