fix: DrHider — CSV in ZIP, double-encode fix, multipart parser

deploy/convert_server.py

@@ -217,28 +217,64 @@ class Handler(BaseHTTPRequestHandler):
     def _handle_drhider(self):
         """Обфускация: multipart files → LLM-NER → ZIP."""
         from services.drhider import obfuscate_files
-        from services.llm_client import HttpxLLMClient
-        from services.upload import parse_multipart
 
         length = int(self.headers.get("Content-Length", 0))
         raw = self.rfile.read(length)
         ctype = self.headers.get("Content-Type", "")
 
-        files = parse_multipart(raw, ctype)
+        # Парсим multipart вручную
+        import cgi, re
+        _, params = cgi.parse_header(ctype)
+        boundary = params.get("boundary", "").encode()
+        if not boundary:
+            self._json({"ok": False, "error": "No multipart boundary"}, 400)
+            return
+
+        files = []
+        # Разбиваем по boundary
+        parts = raw.split(b"--" + boundary)
+        for part in parts:
+            if b"Content-Disposition" not in part:
+                continue
+            try:
+                header_end = part.index(b"\r\n\r\n")
+            except ValueError:
+                continue
+            headers_raw = part[:header_end].decode("latin-1", errors="replace")
+            content = part[header_end + 4:]
+            if content.endswith(b"\r\n"):
+                content = content[:-2]
+            # Извлечь filename
+            m = re.search(r'filename="([^"]*)"', headers_raw)
+            if not m:
+                continue
+            fname = m.group(1)
+            if not fname:
+                continue
+            files.append((fname, content, ""))
+
         if not files:
             self._json({"ok": False, "error": "Нет файлов"}, 400)
             return
 
-        file_list = [(f.filename, f.content, f.content_type) for f in files]
+        # LLM-клиент (совместимый с drhider.LLMClient протоколом: complete(prompt)->str)
+        class _VMLLM:
+            def complete(self, prompt):
+                import httpx
+                r = httpx.post(
+                    os.environ.get("LLM_URL", "https://api.aillm.ru/v1/chat/completions"),
+                    json={"model": os.environ.get("LLM_MODEL", "gpt-oss-120b"),
+                          "messages": [{"role": "user", "content": prompt}],
+                          "max_tokens": 8000, "temperature": 0.1},
+                    headers={"Authorization": f"Bearer {os.environ.get('LLM_KEY', '')}",
+                             "Content-Type": "application/json"},
+                    timeout=120
+                )
+                r.raise_for_status()
+                return r.json()["choices"][0]["message"]["content"]
 
         try:
-            llm = HttpxLLMClient(
+            zip_data, _csv = obfuscate_files(files, llm_client=_VMLLM())
-                os.environ.get("LLM_API_URL", "https://api.aillm.ru/v1/chat/completions"),
-                os.environ.get("LLM_API_KEY", ""),
-                os.environ.get("LLM_MODEL", "gpt-oss-120b")
-            )
-            zip_data, _csv = obfuscate_files(file_list, llm_client=llm)
-
             self.send_response(200)
             self.send_header("Content-Type", "application/zip")
             self.send_header("Content-Disposition", 'attachment; filename="drhider_output.zip"')

deploy/services/drhider.py

@@ -218,12 +218,13 @@ class TwoPassObfuscator:
                 else:
                     # .doc или другой — замена по тексту
                     txt = all_texts.get(fname, '')
-                    obf_content = self._replace_in_text(txt, fname).encode('utf-8')
+                    obf_content = self._replace_in_text(txt, fname)
 
                 results.append((fname, obf_content))
 
             # Собираем ZIP + CSV
-            return self._build_zip(results), self._build_mapping_csv()
+            csv_str = self._build_mapping_csv()
+            return self._build_zip(results, csv_str), csv_str
 
         finally:
             shutil.rmtree(self._workdir, ignore_errors=True)
@@ -362,12 +363,14 @@ class TwoPassObfuscator:
 
     # --- Сборка выдачи ---
 
-    def _build_zip(self, files: List[Tuple[str, bytes]]) -> bytes:
+    def _build_zip(self, files: List[Tuple[str, bytes]], mapping_csv: str = "") -> bytes:
-        """Собрать ZIP с обфусцированными файлами."""
+        """Собрать ZIP с обфусцированными файлами + mapping.csv."""
         buf = io.BytesIO()
         with zipfile.ZipFile(buf, 'w', zipfile.ZIP_DEFLATED) as zf:
             for fname, content in files:
                 zf.writestr(fname, content)
+            if mapping_csv:
+                zf.writestr("mapping.csv", mapping_csv.encode('utf-8'))
         return buf.getvalue()
 
     def _build_mapping_csv(self) -> str:

site/services/drhider.py

@@ -218,12 +218,13 @@ class TwoPassObfuscator:
                 else:
                     # .doc или другой — замена по тексту
                     txt = all_texts.get(fname, '')
-                    obf_content = self._replace_in_text(txt, fname).encode('utf-8')
+                    obf_content = self._replace_in_text(txt, fname)
 
                 results.append((fname, obf_content))
 
             # Собираем ZIP + CSV
-            return self._build_zip(results), self._build_mapping_csv()
+            csv_str = self._build_mapping_csv()
+            return self._build_zip(results, csv_str), csv_str
 
         finally:
             shutil.rmtree(self._workdir, ignore_errors=True)
@@ -362,12 +363,14 @@ class TwoPassObfuscator:
 
     # --- Сборка выдачи ---
 
-    def _build_zip(self, files: List[Tuple[str, bytes]]) -> bytes:
+    def _build_zip(self, files: List[Tuple[str, bytes]], mapping_csv: str = "") -> bytes:
-        """Собрать ZIP с обфусцированными файлами."""
+        """Собрать ZIP с обфусцированными файлами + mapping.csv."""
         buf = io.BytesIO()
         with zipfile.ZipFile(buf, 'w', zipfile.ZIP_DEFLATED) as zf:
             for fname, content in files:
                 zf.writestr(fname, content)
+            if mapping_csv:
+                zf.writestr("mapping.csv", mapping_csv.encode('utf-8'))
         return buf.getvalue()
 
     def _build_mapping_csv(self) -> str: