jarvis-ai-assistant 0.3.30__py3-none-any.whl → 0.7.6__py3-none-any.whl

jarvis/jarvis_c2rust/loaders.py

@@ -0,0 +1,207 @@
+# -*- coding: utf-8 -*-
+"""
+C2Rust 转译器数据加载器
+"""
+from __future__ import annotations
+
+import json
+import time
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Tuple
+
+from jarvis.jarvis_c2rust.constants import C2RUST_DIRNAME, SYMBOLS_JSONL
+from jarvis.jarvis_c2rust.models import FnRecord
+
+
+class _DbLoader:
+    """读取 symbols.jsonl 并提供按 id/name 查询与源码片段读取"""
+
+    def __init__(self, project_root: Path) -> None:
+        self.project_root = project_root.resolve()
+        self.data_dir = self.project_root / C2RUST_DIRNAME
+
+        self.symbols_path = self.data_dir / SYMBOLS_JSONL
+        # 统一流程：仅使用 symbols.jsonl，不再兼容 functions.jsonl
+        if not self.symbols_path.exists():
+            raise FileNotFoundError(
+                f"在目录下未找到 symbols.jsonl: {self.data_dir}"
+            )
+
+        self.fn_by_id: Dict[int, FnRecord] = {}
+        self.name_to_id: Dict[str, int] = {}
+        self._load()
+
+    def _load(self) -> None:
+        """
+        读取统一的 symbols.jsonl。
+        不区分函数与类型定义，均加载为通用记录（位置与引用信息）。
+        """
+        def _iter_records_from_file(path: Path):
+            try:
+                with path.open("r", encoding="utf-8") as f:
+                    idx = 0
+                    for line in f:
+                        line = line.strip()
+                        if not line:
+                            continue
+                        try:
+                            obj = json.loads(line)
+                        except Exception:
+                            continue
+                        idx += 1
+                        yield idx, obj
+            except FileNotFoundError:
+                return
+
+        # 加载所有符号记录（函数、类型等）
+        for idx, obj in _iter_records_from_file(self.symbols_path):
+            fid = int(obj.get("id") or idx)
+            nm = obj.get("name") or ""
+            qn = obj.get("qualified_name") or ""
+            fp = obj.get("file") or ""
+            refs = obj.get("ref")
+            # 统一使用列表类型的引用字段
+            if not isinstance(refs, list):
+                refs = []
+            refs = [c for c in refs if isinstance(c, str) and c]
+            sr = int(obj.get("start_line") or 0)
+            sc = int(obj.get("start_col") or 0)
+            er = int(obj.get("end_line") or 0)
+            ec = int(obj.get("end_col") or 0)
+            lr = obj.get("lib_replacement") if isinstance(obj.get("lib_replacement"), dict) else None
+            rec = FnRecord(
+                id=fid,
+                name=nm,
+                qname=qn,
+                file=fp,
+                start_line=sr,
+                start_col=sc,
+                end_line=er,
+                end_col=ec,
+                refs=refs,
+                lib_replacement=lr,
+            )
+            self.fn_by_id[fid] = rec
+            if nm:
+                self.name_to_id.setdefault(nm, fid)
+            if qn:
+                self.name_to_id.setdefault(qn, fid)
+
+    def get(self, fid: int) -> Optional[FnRecord]:
+        return self.fn_by_id.get(fid)
+
+    def get_id_by_name(self, name_or_qname: str) -> Optional[int]:
+        return self.name_to_id.get(name_or_qname)
+
+    def read_source_span(self, rec: FnRecord) -> str:
+        """按起止行读取源码片段（忽略列边界，尽量完整）"""
+        try:
+            p = Path(rec.file)
+            # 若记录为相对路径，基于 project_root 解析
+            if not p.is_absolute():
+                p = (self.project_root / p).resolve()
+            if not p.exists():
+                return ""
+            lines = p.read_text(encoding="utf-8", errors="replace").splitlines()
+            s = max(1, rec.start_line)
+            e = min(len(lines), max(rec.end_line, s))
+            # Python 索引从0开始，包含终止行
+            chunk = "\n".join(lines[s - 1 : e])
+            return chunk
+        except Exception:
+            return ""
+
+
+class _SymbolMapJsonl:
+    """
+    JSONL 形式的符号映射管理：
+    - 每行一条记录，支持同名函数的多条映射（用于处理重载/同名符号）
+    - 记录字段：
+      {
+        "c_name": "<简单名>",
+        "c_qname": "<限定名，可为空字符串>",
+        "c_file": "<源文件路径>",
+        "start_line": <int>,
+        "end_line": <int>,
+        "module": "src/xxx.rs 或 src/xxx/mod.rs",
+        "rust_symbol": "<Rust函数名>",
+        "updated_at": "YYYY-MM-DDTHH:MM:SS"
+      }
+    - 提供按名称（c_name/c_qname）查询、按源位置判断是否已记录等能力
+    """
+
+    def __init__(self, jsonl_path: Path) -> None:
+        self.jsonl_path = jsonl_path
+        self.records: List[Dict[str, Any]] = []
+        # 索引：名称 -> 记录列表索引
+        self.by_key: Dict[str, List[int]] = {}
+        # 唯一定位（避免同名冲突）：(c_file, start_line, end_line, c_qname or c_name) -> 记录索引列表
+        self.by_pos: Dict[Tuple[str, int, int, str], List[int]] = {}
+        self._load()
+
+    def _load(self) -> None:
+        self.records = []
+        self.by_key = {}
+        self.by_pos = {}
+        # 读取 JSONL
+        if self.jsonl_path.exists():
+            try:
+                with self.jsonl_path.open("r", encoding="utf-8") as f:
+                    for line in f:
+                        line = line.strip()
+                        if not line:
+                            continue
+                        try:
+                            obj = json.loads(line)
+                        except Exception:
+                            continue
+                        self._add_record_in_memory(obj)
+            except Exception:
+                pass
+
+    def _add_record_in_memory(self, rec: Dict[str, Any]) -> None:
+        idx = len(self.records)
+        self.records.append(rec)
+        for key in [rec.get("c_name") or "", rec.get("c_qname") or ""]:
+            k = str(key or "").strip()
+            if not k:
+                continue
+            self.by_key.setdefault(k, []).append(idx)
+        pos_key = (str(rec.get("c_file") or ""), int(rec.get("start_line") or 0), int(rec.get("end_line") or 0), str(rec.get("c_qname") or rec.get("c_name") or ""))
+        self.by_pos.setdefault(pos_key, []).append(idx)
+
+    def has_symbol(self, sym: str) -> bool:
+        return bool(self.by_key.get(sym))
+
+    def get(self, sym: str) -> List[Dict[str, Any]]:
+        idxs = self.by_key.get(sym) or []
+        return [self.records[i] for i in idxs]
+
+    def get_any(self, sym: str) -> Optional[Dict[str, Any]]:
+        recs = self.get(sym)
+        return recs[-1] if recs else None
+
+    def has_rec(self, rec: FnRecord) -> bool:
+        key = (str(rec.file or ""), int(rec.start_line or 0), int(rec.end_line or 0), str(rec.qname or rec.name or ""))
+        return bool(self.by_pos.get(key))
+
+    def add(self, rec: FnRecord, module: str, rust_symbol: str) -> None:
+        obj = {
+            "c_name": rec.name or "",
+            "c_qname": rec.qname or "",
+            "c_file": rec.file or "",
+            "start_line": int(rec.start_line or 0),
+            "end_line": int(rec.end_line or 0),
+            "module": str(module or ""),
+            "rust_symbol": str(rust_symbol or (rec.name or f"fn_{rec.id}")),
+            "updated_at": time.strftime("%Y-%m-%dT%H:%M:%S", time.localtime()),
+        }
+        # 先写盘，再更新内存索引
+        try:
+            self.jsonl_path.parent.mkdir(parents=True, exist_ok=True)
+            with self.jsonl_path.open("a", encoding="utf-8") as f:
+                f.write(json.dumps(obj, ensure_ascii=False) + "\n")
+        except Exception:
+            pass
+        self._add_record_in_memory(obj)
+

jarvis/jarvis_c2rust/models.py

@@ -0,0 +1,28 @@
+# -*- coding: utf-8 -*-
+"""
+C2Rust 转译器数据模型
+"""
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any, Dict, List, Optional
+
+
+@dataclass
+class FnRecord:
+    id: int
+    name: str
+    qname: str
+    file: str
+    start_line: int
+    start_col: int
+    end_line: int
+    end_col: int
+    refs: List[str]
+    # 额外元信息（来自 symbols/items）：函数签名、返回类型与参数（可选）
+    signature: str = ""
+    return_type: str = ""
+    params: Optional[List[Dict[str, str]]] = None
+    # 来自库替代阶段的上下文元数据（若存在）
+    lib_replacement: Optional[Dict[str, Any]] = None
+