feat: зеркало кода ВМ — deploy/ из contractor

2026-06-27 19:35:53 +04:00
parent de074c8f99
commit 623cd9140f
30 changed files with 4805 additions and 0 deletions
@@ -0,0 +1,118 @@
+"""Upload service — accept file, store to DB, parse."""
+import uuid, base64, json, io, cgi, os
+from db import documents, contracts, supplements
+from .parse import parse_file
+
+MAX_FILE_SIZE = 100 * 1024 * 1024  # 100 MB
+
+
+def handle_upload(rfile, content_type, content_length):
+    """Parse multipart upload, store in DB, return result dict."""
+    # Validate content_length
+    try:
+        cl = int(content_length)
+    except (TypeError, ValueError):
+        return {"ok": False, "error": "invalid content-length"}
+    if cl <= 0:
+        return {"ok": False, "error": "empty request"}
+    if cl > MAX_FILE_SIZE:
+        return {"ok": False, "error": f"file too large (max {MAX_FILE_SIZE // 1024 // 1024}MB)"}
+
+    body = rfile.read(cl)
+
+    environ = {
+        "REQUEST_METHOD": "POST",
+        "CONTENT_TYPE": content_type,
+        "CONTENT_LENGTH": str(content_length),
+    }
+    fs = cgi.FieldStorage(fp=io.BytesIO(body), environ=environ, keep_blank_values=True)
+
+    filename = None
+    file_data = None
+    contract_id = ""
+
+    if "files" in fs:
+        item = fs["files"]
+        if isinstance(item, list):
+            item = item[0]
+        filename = os.path.basename(item.filename) if item.filename else None
+        if filename and (".." in filename or "/" in filename or "\\" in filename):
+            return {"ok": False, "error": "invalid filename"}
+        file_data = item.file.read() if hasattr(item, "file") else item.value
+        if isinstance(file_data, str):
+            file_data = file_data.encode("utf-8")
+
+    if "contract_id" in fs:
+        contract_id = fs.getfirst("contract_id", "")
+        # Validate UUID
+        if contract_id:
+            try:
+                uuid.UUID(contract_id)
+            except (ValueError, AttributeError):
+                contract_id = ""
+
+    batch_id = fs.getfirst("batch_id", None) if "batch_id" in fs else None
+    # Validate UUID
+    if batch_id:
+        try:
+            uuid.UUID(batch_id)
+        except (ValueError, AttributeError):
+            batch_id = None
+
+    # zip_source — имя родительского ZIP-архива (для визуальной группировки)
+    zip_source = None
+    if "zip_source" in fs:
+        raw_zip = fs.getfirst("zip_source", "")
+        if raw_zip:
+            zip_source = os.path.basename(raw_zip)[:255]  # защита от path traversal + лимит
+
+    if not filename or not file_data:
+        return {"ok": False, "error": "no file in request"}
+
+    mime = _mime_for(filename)
+    b64 = base64.b64encode(file_data).decode()
+
+    if contract_id:
+        try:
+            supplements.delete_by_document(contract_id, filename)
+        except Exception:
+            pass  # old record may not exist or FK issue — proceed with insert
+
+    doc = documents.insert(filename, mime, b64, batch_id=batch_id, zip_source=zip_source)
+
+    if not contract_id:
+        from datetime import datetime
+        now = datetime.now()
+        c = contracts.insert(f"б/н {now.strftime('%Y%m%d')}-{now.strftime('%H%M')}")
+        contract_id = c["id"]
+        supp_type = "initial"
+    else:
+        supp_type = "additional"
+
+    supplements.insert(contract_id, doc["id"], supp_type)
+
+    parsed = parse_file(filename, file_data)
+    if parsed and parsed.get("status") == "parsed":
+        documents.set_parsed(doc["id"], parsed["elements"])
+    elif parsed and parsed.get("status") == "error":
+        documents.set_error(doc["id"], parsed.get("error", "parse failed"))
+
+    return {
+        "ok": True,
+        "doc_id": doc["id"],
+        "contract_id": contract_id,
+        "filename": filename,
+        "size": len(file_data),
+        "parsed": parsed,
+    }
+
+
+def _mime_for(filename):
+    ext = filename.rsplit(".", 1)[-1].lower() if "." in filename else ""
+    mime_map = {
+        "pdf": "application/pdf",
+        "docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+        "doc": "application/msword",
+        "zip": "application/zip",
+    }
+    return mime_map.get(ext, "application/octet-stream")