refactor(impact): CodeParser 多语言就绪——语言适配器注册表（仅 Java，行为不变）
- BaseLanguageParser 抽象基类 + JavaLanguageParser（迁移全部 Java 正则，输出逐字节一致） - LANGUAGE_PARSERS 注册表 + register_language_parser 扩展点（新增语言=加适配器类并注册，下游零改动） - CodeParser 分发器：parse(root, language=None) 自动按扩展名探测 / 显式语言覆盖 / 多语言合并（language 逗号连接）/ 无源码或语言不支持抛 CodeParseError - source_aggregator 新增 existing_system_language 透传；config.py 探测扩展名补 .py/.ts/.go/.cs - 删除自动探测下不可达死分支；全量 362 passed / 99.28%（基线 351/99.27%） - 门禁复跑 PASS：summary 与 MVP 基线一致（16/5/8/3/50/0），概要设计书 13 章
This commit is contained in:
@@ -17,7 +17,8 @@ ENV_PREFIX = "GENESIS_"
|
||||
class ServerConfig(BaseModel):
|
||||
max_upload_mb: int = 100
|
||||
allowed_extensions: list[str] = Field(
|
||||
default_factory=lambda: [".xlsx", ".xls", ".docx", ".pptx", ".java", ".xml", ".yml"]
|
||||
default_factory=lambda: [".xlsx", ".xls", ".docx", ".pptx", ".java", ".xml", ".yml",
|
||||
".py", ".ts", ".go", ".cs"]
|
||||
)
|
||||
|
||||
|
||||
|
||||
@@ -1,7 +1,11 @@
|
||||
"""CodeParser：Java 项目源码解析 → CodeStructure（Impact Agent MVP）。
|
||||
"""CodeParser：多语言源码解析 → CodeStructure（Impact Agent）。
|
||||
|
||||
识别 @RestController/@Controller、@Service、@Entity/@Table 及方法级路由映射，
|
||||
输出供 ExistingSystemExplorer 组装 ExistingSystemInfo 的结构化清单。
|
||||
通过语言适配器注册表（LANGUAGE_PARSERS）分发：每种开发语言一个适配器
|
||||
（BaseLanguageParser 子类），定义文件扩展名、类名识别、角色（控制器/服务/实体）
|
||||
与端点提取。当前内置 Java 适配器（Spring Boot / JPA / MyBatis-Plus）。
|
||||
|
||||
新增语言 = 实现一个适配器类并调用 register_language_parser 注册，
|
||||
下游（ExistingSystemExplorer / ImpactAgent / Writer）零改动。
|
||||
"""
|
||||
from __future__ import annotations
|
||||
|
||||
@@ -18,27 +22,7 @@ from genesis.data_models import (
|
||||
|
||||
|
||||
class CodeParseError(Exception):
|
||||
"""既有系统解析失败（路径无效或非 Java 源码项目）。"""
|
||||
|
||||
|
||||
_JAVA_EXT = ".java"
|
||||
|
||||
_CLASS_RE = re.compile(
|
||||
r"(?:public\s+|abstract\s+|final\s+)?(?:class|interface|enum|record)\s+(\w+)"
|
||||
)
|
||||
_TABLE_RE = re.compile(r"@(?:Table|TableName)\s*\(\s*(?:name\s*=\s*)?[\"']([^\"']+)[\"']")
|
||||
_CLASS_MAPPING_RE = re.compile(r"@RequestMapping\s*\(\s*[\"']([^\"']+)[\"']")
|
||||
_METHOD_MAPPING_RE = re.compile(
|
||||
r"@(Get|Post|Put|Delete|Patch|Request)Mapping\s*(?:\(\s*[\"']([^\"']*)[\"'])?"
|
||||
)
|
||||
_METHOD_DECL_RE = re.compile(
|
||||
r"(?:public|private|protected|)\s+(?:static\s+|final\s+|synchronized\s+)*"
|
||||
r"[\w<>\[\],.?]+\s+(\w+)\s*\("
|
||||
)
|
||||
_FIELD_DECL_RE = re.compile(
|
||||
r"(?:private|public|protected)\s+[\w<>\[\],]+\s+(\w+)\s*;"
|
||||
)
|
||||
_IMPORT_RE = re.compile(r"^import\s+([\w.]+);", re.MULTILINE)
|
||||
"""既有系统解析失败（路径无效或无可识别源码）。"""
|
||||
|
||||
|
||||
def _relative(path: Path, root: Path) -> str:
|
||||
@@ -49,81 +33,100 @@ def _read(path: Path) -> str:
|
||||
return path.read_text(encoding="utf-8", errors="ignore")
|
||||
|
||||
|
||||
class CodeParser:
|
||||
"""解析 Java 项目目录，输出 CodeStructure（控制器/服务/实体/端点/模块）。"""
|
||||
class BaseLanguageParser:
|
||||
"""语言适配器基类：定义扩展名与单文件解析契约。
|
||||
|
||||
def parse(self, root_path: str | Path) -> CodeStructure:
|
||||
root = Path(root_path)
|
||||
if not root.is_dir():
|
||||
raise CodeParseError(f"既有系统路径无效或不存在: {root_path}")
|
||||
子类需设置 language / extensions，并实现 parse_file：
|
||||
返回 dict（keys: imports / classes / controllers / services / entities / endpoints）。
|
||||
"""
|
||||
|
||||
java_files = sorted(p for p in root.rglob(f"*{_JAVA_EXT}") if p.is_file())
|
||||
if not java_files:
|
||||
raise CodeParseError(f"未找到 Java 源码: {root_path}")
|
||||
language: str = ""
|
||||
extensions: tuple[str, ...] = ()
|
||||
|
||||
controllers: list[ControllerInfo] = []
|
||||
services: list[ServiceInfo] = []
|
||||
entities: list[EntityInfo] = []
|
||||
endpoints: list[EndpointInfo] = []
|
||||
classes: list[dict] = []
|
||||
raw_imports: list[dict] = []
|
||||
def source_files(self, root: Path) -> list[Path]:
|
||||
"""返回本语言适配器覆盖的源文件（排序 + 去重，保证确定性）。"""
|
||||
files: list[Path] = []
|
||||
for ext in self.extensions:
|
||||
files.extend(p for p in root.rglob(f"*{ext}") if p.is_file())
|
||||
return sorted(set(files))
|
||||
|
||||
for path in java_files:
|
||||
text = _read(path)
|
||||
rel = _relative(path, root)
|
||||
class_name = self._class_name(text)
|
||||
def parse_file(self, text: str, rel: str) -> dict:
|
||||
"""解析单个源文件，返回该文件对 CodeStructure 各分层的贡献。"""
|
||||
raise NotImplementedError
|
||||
|
||||
imports = _IMPORT_RE.findall(text)
|
||||
raw_imports.append({"path": rel, "imports": imports})
|
||||
|
||||
if not class_name:
|
||||
# package-info.java 等无类声明文件：仅登记 imports，不参与要素提取
|
||||
continue
|
||||
class JavaLanguageParser(BaseLanguageParser):
|
||||
"""Java（Spring Boot / JPA / MyBatis-Plus）适配器。"""
|
||||
|
||||
is_controller = "@RestController" in text or "@Controller" in text
|
||||
is_service = "@Service" in text
|
||||
# 既有系统实体可能用 JPA @Entity 或 MyBatis-Plus @TableName 标注
|
||||
is_entity = "@Entity" in text or "@TableName" in text
|
||||
language = "java"
|
||||
extensions = (".java",)
|
||||
|
||||
if is_controller:
|
||||
controllers.append(self._parse_controller(text, rel, class_name, endpoints))
|
||||
elif is_service:
|
||||
services.append(self._parse_service(text, rel, class_name))
|
||||
elif is_entity:
|
||||
entities.append(self._parse_entity(text, rel, class_name))
|
||||
_CLASS_RE = re.compile(
|
||||
r"(?:public\s+|abstract\s+|final\s+)?(?:class|interface|enum|record)\s+(\w+)"
|
||||
)
|
||||
_TABLE_RE = re.compile(r"@(?:Table|TableName)\s*\(\s*(?:name\s*=\s*)?[\"']([^\"']+)[\"']")
|
||||
_CLASS_MAPPING_RE = re.compile(r"@RequestMapping\s*\(\s*[\"']([^\"']+)[\"']")
|
||||
_METHOD_MAPPING_RE = re.compile(
|
||||
r"@(Get|Post|Put|Delete|Patch|Request)Mapping\s*(?:\(\s*[\"']([^\"']*)[\"'])?"
|
||||
)
|
||||
_METHOD_DECL_RE = re.compile(
|
||||
r"(?:public|private|protected|)\s+(?:static\s+|final\s+|synchronized\s+)*"
|
||||
r"[\w<>\[\],.?]+\s+(\w+)\s*\("
|
||||
)
|
||||
_FIELD_DECL_RE = re.compile(
|
||||
r"(?:private|public|protected)\s+[\w<>\[\],]+\s+(\w+)\s*;"
|
||||
)
|
||||
_IMPORT_RE = re.compile(r"^import\s+([\w.]+);", re.MULTILINE)
|
||||
|
||||
classes.append({"class_name": class_name, "path": rel})
|
||||
def parse_file(self, text: str, rel: str) -> dict:
|
||||
class_name = self._class_name(text)
|
||||
imports = self._IMPORT_RE.findall(text)
|
||||
out: dict = {
|
||||
"imports": imports,
|
||||
"classes": [],
|
||||
"controllers": [],
|
||||
"services": [],
|
||||
"entities": [],
|
||||
"endpoints": [],
|
||||
}
|
||||
if not class_name:
|
||||
# package-info.java 等无类声明文件：仅登记 imports，不参与要素提取
|
||||
return out
|
||||
|
||||
modules = self._modules(root, java_files)
|
||||
out["classes"].append({"class_name": class_name, "path": rel})
|
||||
|
||||
return CodeStructure(
|
||||
root_path=str(root),
|
||||
language="java",
|
||||
modules=modules,
|
||||
classes=classes,
|
||||
controllers=controllers,
|
||||
services=services,
|
||||
entities=entities,
|
||||
endpoints=endpoints,
|
||||
raw_imports=raw_imports,
|
||||
)
|
||||
is_controller = "@RestController" in text or "@Controller" in text
|
||||
is_service = "@Service" in text
|
||||
# 既有系统实体可能用 JPA @Entity 或 MyBatis-Plus @TableName 标注
|
||||
is_entity = "@Entity" in text or "@TableName" in text
|
||||
|
||||
if is_controller:
|
||||
ctrl, endpoints = self._parse_controller(text, rel, class_name)
|
||||
out["controllers"].append(ctrl)
|
||||
out["endpoints"] = endpoints
|
||||
elif is_service:
|
||||
out["services"].append(self._parse_service(text, rel, class_name))
|
||||
elif is_entity:
|
||||
out["entities"].append(self._parse_entity(text, rel, class_name))
|
||||
return out
|
||||
|
||||
@staticmethod
|
||||
def _class_name(text: str) -> str | None:
|
||||
m = _CLASS_RE.search(text)
|
||||
m = JavaLanguageParser._CLASS_RE.search(text)
|
||||
return m.group(1) if m else None
|
||||
|
||||
@staticmethod
|
||||
def _parse_controller(
|
||||
text: str, rel: str, class_name: str, endpoints: list[EndpointInfo]
|
||||
) -> ControllerInfo:
|
||||
text: str, rel: str, class_name: str
|
||||
) -> tuple[ControllerInfo, list[EndpointInfo]]:
|
||||
base_path = ""
|
||||
m = _CLASS_MAPPING_RE.search(text)
|
||||
m = JavaLanguageParser._CLASS_MAPPING_RE.search(text)
|
||||
if m:
|
||||
base_path = m.group(1)
|
||||
|
||||
ctrl_endpoints: list[str] = []
|
||||
for m in _METHOD_MAPPING_RE.finditer(text):
|
||||
endpoints: list[EndpointInfo] = []
|
||||
for m in JavaLanguageParser._METHOD_MAPPING_RE.finditer(text):
|
||||
verb, sub = m.group(1).upper(), m.group(2) or ""
|
||||
if verb == "REQUEST":
|
||||
verb = "ANY"
|
||||
@@ -139,18 +142,21 @@ class CodeParser:
|
||||
)
|
||||
)
|
||||
|
||||
return ControllerInfo(
|
||||
name=class_name,
|
||||
class_name=class_name,
|
||||
path=rel,
|
||||
base_path=base_path,
|
||||
endpoints=ctrl_endpoints,
|
||||
source_uri=rel,
|
||||
return (
|
||||
ControllerInfo(
|
||||
name=class_name,
|
||||
class_name=class_name,
|
||||
path=rel,
|
||||
base_path=base_path,
|
||||
endpoints=ctrl_endpoints,
|
||||
source_uri=rel,
|
||||
),
|
||||
endpoints,
|
||||
)
|
||||
|
||||
@staticmethod
|
||||
def _parse_service(text: str, rel: str, class_name: str) -> ServiceInfo:
|
||||
methods = list(dict.fromkeys(_METHOD_DECL_RE.findall(text)))
|
||||
methods = list(dict.fromkeys(JavaLanguageParser._METHOD_DECL_RE.findall(text)))
|
||||
return ServiceInfo(
|
||||
name=class_name,
|
||||
class_name=class_name,
|
||||
@@ -162,10 +168,10 @@ class CodeParser:
|
||||
@staticmethod
|
||||
def _parse_entity(text: str, rel: str, class_name: str) -> EntityInfo:
|
||||
table_name = None
|
||||
m = _TABLE_RE.search(text)
|
||||
m = JavaLanguageParser._TABLE_RE.search(text)
|
||||
if m:
|
||||
table_name = m.group(1)
|
||||
fields = list(dict.fromkeys(_FIELD_DECL_RE.findall(text)))
|
||||
fields = list(dict.fromkeys(JavaLanguageParser._FIELD_DECL_RE.findall(text)))
|
||||
return EntityInfo(
|
||||
name=class_name,
|
||||
class_name=class_name,
|
||||
@@ -175,12 +181,93 @@ class CodeParser:
|
||||
source_uri=rel,
|
||||
)
|
||||
|
||||
|
||||
# ---------- 语言适配器注册表 ----------
|
||||
|
||||
LANGUAGE_PARSERS: dict[str, type[BaseLanguageParser]] = {}
|
||||
|
||||
|
||||
def register_language_parser(name: str, parser_cls: type[BaseLanguageParser]) -> None:
|
||||
"""注册语言适配器（未来扩展点：新增语言只需实现并注册，下游零改动）。"""
|
||||
LANGUAGE_PARSERS[name] = parser_cls
|
||||
|
||||
|
||||
register_language_parser(JavaLanguageParser.language, JavaLanguageParser)
|
||||
|
||||
|
||||
class CodeParser:
|
||||
"""按语言分发解析既有系统目录，输出 CodeStructure。"""
|
||||
|
||||
def parse(self, root_path: str | Path, language: str | None = None) -> CodeStructure:
|
||||
root = Path(root_path)
|
||||
if not root.is_dir():
|
||||
raise CodeParseError(f"既有系统路径无效或不存在: {root_path}")
|
||||
|
||||
if language is not None:
|
||||
cls = LANGUAGE_PARSERS.get(language)
|
||||
if cls is None:
|
||||
raise CodeParseError(
|
||||
f"不支持的源码语言: {language}（支持: {', '.join(sorted(LANGUAGE_PARSERS))}）"
|
||||
)
|
||||
parser_cls = [cls]
|
||||
else:
|
||||
# 自动探测：仅保留在根目录下确有源文件的语言
|
||||
present = [
|
||||
name for name in sorted(LANGUAGE_PARSERS)
|
||||
if LANGUAGE_PARSERS[name]().source_files(root)
|
||||
]
|
||||
if not present:
|
||||
raise CodeParseError(
|
||||
f"未找到可识别的源码（支持: {', '.join(sorted(LANGUAGE_PARSERS))}）: {root_path}"
|
||||
)
|
||||
parser_cls = [LANGUAGE_PARSERS[name] for name in present]
|
||||
|
||||
controllers: list[ControllerInfo] = []
|
||||
services: list[ServiceInfo] = []
|
||||
entities: list[EntityInfo] = []
|
||||
endpoints: list[EndpointInfo] = []
|
||||
classes: list[dict] = []
|
||||
raw_imports: list[dict] = []
|
||||
all_files: list[Path] = []
|
||||
|
||||
for cls in parser_cls:
|
||||
parser = cls()
|
||||
files = parser.source_files(root)
|
||||
if not files:
|
||||
# 自动探测的 present 过滤已保证有文件；此处仅显式 language 时可达
|
||||
raise CodeParseError(f"未找到 {parser.language} 源码: {root_path}")
|
||||
all_files.extend(files)
|
||||
for path in files:
|
||||
text = _read(path)
|
||||
rel = _relative(path, root)
|
||||
out = parser.parse_file(text, rel)
|
||||
raw_imports.append({"path": rel, "imports": out["imports"]})
|
||||
controllers.extend(out["controllers"])
|
||||
services.extend(out["services"])
|
||||
entities.extend(out["entities"])
|
||||
endpoints.extend(out["endpoints"])
|
||||
classes.extend(out["classes"])
|
||||
|
||||
modules = self._modules(root, all_files)
|
||||
langs = [cls.language for cls in parser_cls]
|
||||
return CodeStructure(
|
||||
root_path=str(root),
|
||||
language=",".join(langs) if len(langs) > 1 else langs[0],
|
||||
modules=modules,
|
||||
classes=classes,
|
||||
controllers=controllers,
|
||||
services=services,
|
||||
entities=entities,
|
||||
endpoints=endpoints,
|
||||
raw_imports=raw_imports,
|
||||
)
|
||||
|
||||
@staticmethod
|
||||
def _modules(root: Path, java_files: list[Path]) -> list[str]:
|
||||
"""顶层目录中凡包含 Java 源码者视为一个模块（按名排序，保证确定性）。"""
|
||||
def _modules(root: Path, files: list[Path]) -> list[str]:
|
||||
"""顶层目录中凡包含源码者视为一个模块（按名排序，保证确定性）。"""
|
||||
mods = {
|
||||
p.relative_to(root).parts[0]
|
||||
for p in java_files
|
||||
for p in files
|
||||
if len(p.relative_to(root).parts) > 1
|
||||
}
|
||||
return sorted(mods)
|
||||
|
||||
@@ -45,11 +45,14 @@ class SourceParser:
|
||||
write_instruction_paths: list[str | Path] | None = None,
|
||||
rule_paths: list[str | Path] | None = None,
|
||||
existing_system_path: str | Path | None = None,
|
||||
existing_system_language: str | None = None,
|
||||
) -> StructuredSource:
|
||||
"""扩展名校验先于存在性校验（不存在的文件若扩展名未知将抛出 ValueError 而非 FileNotFoundError）。
|
||||
|
||||
existing_system_path：既有系统源码目录（追加/改修场景）。提供时解析为
|
||||
ExistingSystemInfo（门控通过 → 进入影响调查）；未提供/解析失败 → 保持 None。
|
||||
existing_system_language：既有系统源码开发语言（如 "java"）。默认 None 表示
|
||||
由 CodeParser 按扩展名自动探测；显式指定时按该语言解析（多语言支持扩展点）。
|
||||
"""
|
||||
|
||||
requirement_paths = requirement_paths or []
|
||||
@@ -77,7 +80,7 @@ class SourceParser:
|
||||
|
||||
existing_system = None
|
||||
if existing_system_path is not None:
|
||||
code = CodeParser().parse(existing_system_path)
|
||||
code = CodeParser().parse(existing_system_path, language=existing_system_language)
|
||||
existing_system = ExistingSystemExplorer().explore(code)
|
||||
|
||||
return StructuredSource(
|
||||
|
||||
Reference in New Issue
Block a user